From 8feb8e896d77439146d2e2ab3d0ab55bb5baf5fc Mon Sep 17 00:00:00 2001 From: Yong Zhang Date: Tue, 29 May 2012 15:16:05 +0800 Subject: smp: Remove ipi_call_lock[_irq]()/ipi_call_unlock[_irq]() There is no user of those APIs anymore, just remove it. Signed-off-by: Yong Zhang Cc: ralf@linux-mips.org Cc: sshtylyov@mvista.com Cc: david.daney@cavium.com Cc: nikunj@linux.vnet.ibm.com Cc: paulmck@linux.vnet.ibm.com Cc: axboe@kernel.dk Cc: Andrew Morton Link: http://lkml.kernel.org/r/1338275765-3217-11-git-send-email-yong.zhang0@gmail.com Acked-by: Srivatsa S. Bhat Acked-by: Peter Zijlstra Signed-off-by: Thomas Gleixner --- kernel/smp.c | 20 -------------------- 1 file changed, 20 deletions(-) (limited to 'kernel') diff --git a/kernel/smp.c b/kernel/smp.c index d0ae5b24875e..29dd40a9f2f4 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -581,26 +581,6 @@ int smp_call_function(smp_call_func_t func, void *info, int wait) return 0; } EXPORT_SYMBOL(smp_call_function); - -void ipi_call_lock(void) -{ - raw_spin_lock(&call_function.lock); -} - -void ipi_call_unlock(void) -{ - raw_spin_unlock(&call_function.lock); -} - -void ipi_call_lock_irq(void) -{ - raw_spin_lock_irq(&call_function.lock); -} - -void ipi_call_unlock_irq(void) -{ - raw_spin_unlock_irq(&call_function.lock); -} #endif /* USE_GENERIC_SMP_HELPERS */ /* Setup configured maximum number of CPUs to activate */ -- cgit v1.2.3 From c00b275043adc14d668f36266b890f0c53d46640 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 29 May 2012 21:27:44 +0200 Subject: uprobes: Optimize is_swbp_at_addr() for current->mm Change is_swbp_at_addr() to try to avoid the costly read_opcode() if mm == current->mm, __copy_from_user_inatomic() should succeed in the likely case. Currently this optimization is not important, but we are going to add more is_swbp_at_addr(current->mm) callers. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120529192744.GA8057@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 985be4d80fe8..d0f5ec0dcdea 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -333,10 +333,20 @@ static int is_swbp_at_addr(struct mm_struct *mm, unsigned long vaddr) uprobe_opcode_t opcode; int result; + if (current->mm == mm) { + pagefault_disable(); + result = __copy_from_user_inatomic(&opcode, (void __user*)vaddr, + sizeof(opcode)); + pagefault_enable(); + + if (likely(result == 0)) + goto out; + } + result = read_opcode(mm, vaddr, &opcode); if (result) return result; - +out: if (is_swbp_insn(&opcode)) return 1; -- cgit v1.2.3 From a3d7bb47937b3a40b9f0c75655e97b3bb6407cbe Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 29 May 2012 21:27:59 +0200 Subject: uprobes: Change read_opcode() to use FOLL_FORCE set_orig_insn()->read_opcode() should not fail if the probed task did mprotect() after uprobe_register(), change it to use FOLL_FORCE. Without FOLL_WRITE this doesn't have any "side" effect but allows to read the !VM_READ memory. There is another reason for this change, we are going to use is_swbp_at_addr() from handle_swbp() which can race with another thread doing mprotect(). Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120529192759.GB8057@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index d0f5ec0dcdea..a0dbc87a2ec6 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -312,7 +312,7 @@ static int read_opcode(struct mm_struct *mm, unsigned long vaddr, uprobe_opcode_ void *vaddr_new; int ret; - ret = get_user_pages(NULL, mm, vaddr, 1, 0, 0, &page, NULL); + ret = get_user_pages(NULL, mm, vaddr, 1, 0, 1, &page, NULL); if (ret <= 0) return ret; -- cgit v1.2.3 From 3a9ea0520f38def4a3915b91f82455b749f07d88 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 29 May 2012 21:28:57 +0200 Subject: uprobes: Introduce find_active_uprobe() helper No functional changes. Move the "find uprobe" code from handle_swbp() to the new helper, find_active_uprobe(). Note: with or without this change, the find-active-uprobe logic is not exactly right. We can race with another thread which unmaps the memory with the valid uprobe before we take mm->mmap_sem. We can't find this uprobe simply because find_vma() fails. In this case we wrongly assume that this trap was not caused by uprobe and send the erroneous SIGTRAP. See the next changes. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120529192857.GC8057@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 47 ++++++++++++++++++++++++++++------------------- 1 file changed, 28 insertions(+), 19 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index a0dbc87a2ec6..eaf4d55fd424 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -1489,38 +1489,47 @@ static bool can_skip_sstep(struct uprobe *uprobe, struct pt_regs *regs) return false; } -/* - * Run handler and ask thread to singlestep. - * Ensure all non-fatal signals cannot interrupt thread while it singlesteps. - */ -static void handle_swbp(struct pt_regs *regs) +static struct uprobe *find_active_uprobe(unsigned long bp_vaddr) { + struct mm_struct *mm = current->mm; + struct uprobe *uprobe = NULL; struct vm_area_struct *vma; - struct uprobe_task *utask; - struct uprobe *uprobe; - struct mm_struct *mm; - unsigned long bp_vaddr; - uprobe = NULL; - bp_vaddr = uprobe_get_swbp_addr(regs); - mm = current->mm; down_read(&mm->mmap_sem); vma = find_vma(mm, bp_vaddr); - if (vma && vma->vm_start <= bp_vaddr && valid_vma(vma, false)) { - struct inode *inode; - loff_t offset; + if (vma && vma->vm_start <= bp_vaddr) { + if (valid_vma(vma, false)) { + struct inode *inode; + loff_t offset; - inode = vma->vm_file->f_mapping->host; - offset = bp_vaddr - vma->vm_start; - offset += (vma->vm_pgoff << PAGE_SHIFT); - uprobe = find_uprobe(inode, offset); + inode = vma->vm_file->f_mapping->host; + offset = bp_vaddr - vma->vm_start; + offset += (vma->vm_pgoff << PAGE_SHIFT); + uprobe = find_uprobe(inode, offset); + } } srcu_read_unlock_raw(&uprobes_srcu, current->uprobe_srcu_id); current->uprobe_srcu_id = -1; up_read(&mm->mmap_sem); + return uprobe; +} + +/* + * Run handler and ask thread to singlestep. + * Ensure all non-fatal signals cannot interrupt thread while it singlesteps. + */ +static void handle_swbp(struct pt_regs *regs) +{ + struct uprobe_task *utask; + struct uprobe *uprobe; + unsigned long bp_vaddr; + + bp_vaddr = uprobe_get_swbp_addr(regs); + uprobe = find_active_uprobe(bp_vaddr); + if (!uprobe) { /* No matching uprobe; signal SIGTRAP. */ send_sig(SIGTRAP, current, 0); -- cgit v1.2.3 From d790d34653ab20c74034902f5f0889bba807949a Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 29 May 2012 21:29:14 +0200 Subject: uprobes: Teach find_active_uprobe() to provide the "is_swbp" info A separate patch to simplify the review, and for the documentation. The patch adds another "int *is_swbp" argument to find_active_uprobe(), so far its only caller doesn't use this info. With this patch find_active_uprobe() additionally does: - if find_vma() + ->vm_start check fails, *is_swbp = -EFAULT - otherwise, if valid_vma() + find_uprobe() fails, it holds the result of is_swbp_at_addr(), can be negative too. The latter is only possible if we raced with another thread which did munmap/etc after we hit this bp. IOW. If find_active_uprobe(&is_swbp) returns NULL, the caller can look at is_swbp to figure out whether the current insn is bp or not, or detect the race with another thread if it is negative. Note: I think that performance-wise this change is fine. This adds is_swbp_at_addr(), but only if we raced with uprobe_unregister() or if we hit the "normal" int3 but this mm has uprobes as well. And even in this case the slow read_opcode() path is very unlikely, this insn recently triggered do_int3(), __copy_from_user_inatomic() shouldn't fail in the likely case. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120529192914.GD8057@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index eaf4d55fd424..ee3df704e78a 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -1489,7 +1489,7 @@ static bool can_skip_sstep(struct uprobe *uprobe, struct pt_regs *regs) return false; } -static struct uprobe *find_active_uprobe(unsigned long bp_vaddr) +static struct uprobe *find_active_uprobe(unsigned long bp_vaddr, int *is_swbp) { struct mm_struct *mm = current->mm; struct uprobe *uprobe = NULL; @@ -1497,7 +1497,6 @@ static struct uprobe *find_active_uprobe(unsigned long bp_vaddr) down_read(&mm->mmap_sem); vma = find_vma(mm, bp_vaddr); - if (vma && vma->vm_start <= bp_vaddr) { if (valid_vma(vma, false)) { struct inode *inode; @@ -1508,6 +1507,11 @@ static struct uprobe *find_active_uprobe(unsigned long bp_vaddr) offset += (vma->vm_pgoff << PAGE_SHIFT); uprobe = find_uprobe(inode, offset); } + + if (!uprobe) + *is_swbp = is_swbp_at_addr(mm, bp_vaddr); + } else { + *is_swbp = -EFAULT; } srcu_read_unlock_raw(&uprobes_srcu, current->uprobe_srcu_id); @@ -1526,9 +1530,10 @@ static void handle_swbp(struct pt_regs *regs) struct uprobe_task *utask; struct uprobe *uprobe; unsigned long bp_vaddr; + int is_swbp; bp_vaddr = uprobe_get_swbp_addr(regs); - uprobe = find_active_uprobe(bp_vaddr); + uprobe = find_active_uprobe(bp_vaddr, &is_swbp); if (!uprobe) { /* No matching uprobe; signal SIGTRAP. */ -- cgit v1.2.3 From 77fc4af1b59d12ab3b1467adf0a5204806853123 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 29 May 2012 21:29:28 +0200 Subject: uprobes: Change register_for_each_vma() to take mm->mmap_sem for writing Change register_for_each_vma() to take mm->mmap_sem for writing. This is a bit unfortunate but hopefully not too bad, this is the slow path anyway. This is needed to ensure that find_active_uprobe() can not race with uprobe_register() which adds the new bp at the same bp_vaddr, after find_uprobe() fails and before is_swbp_at_addr_fast() checks the memory. IOW, this is needed to ensure that if find_active_uprobe() returns NULL but is_swbp == true, we can safely assume that it was the "normal" int3 and we should send SIGTRAP. There is another reason for this change. We are going to replace uprobes_state->count with MMF_ flags set by register/unregister and cleared by find_active_uprobe(), and set/clear shouldn't race with each other. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120529192928.GE8057@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index ee3df704e78a..a2ed82b4808c 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -853,12 +853,12 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register) } mm = vi->mm; - down_read(&mm->mmap_sem); + down_write(&mm->mmap_sem); vma = find_vma(mm, (unsigned long)vi->vaddr); if (!vma || !valid_vma(vma, is_register)) { list_del(&vi->probe_list); kfree(vi); - up_read(&mm->mmap_sem); + up_write(&mm->mmap_sem); mmput(mm); continue; } @@ -867,7 +867,7 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register) vaddr != vi->vaddr) { list_del(&vi->probe_list); kfree(vi); - up_read(&mm->mmap_sem); + up_write(&mm->mmap_sem); mmput(mm); continue; } @@ -877,7 +877,7 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register) else remove_breakpoint(uprobe, mm, vi->vaddr); - up_read(&mm->mmap_sem); + up_write(&mm->mmap_sem); mmput(mm); if (is_register) { if (ret && ret == -EEXIST) -- cgit v1.2.3 From 56bb4cf6475d702d2fb00fc641aa6441097c0330 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 29 May 2012 21:29:47 +0200 Subject: uprobes: Teach handle_swbp() to rely on "is_swbp" rather than uprobes_srcu Currently handle_swbp() assumes that it can't race with unregister, so it roughly does: if (find_uprobe(vaddr)) process_uprobe(); else send_sig(SIGTRAP); This relies on the not-really-working uprobes_srcu code we are going to remove, see the next patch. With this patch we rely on the result of is_swbp_at_addr(bp_vaddr) if find_uprobe() fails. If is_swbp == 1, then we hit the normal int3, we should send SIGTRAP. If is_swbp == 0, we raced with uprobe_unregister(), we simply restart this insn again. The "difficult" case is is_swbp == -EFAULT, when we can't read this memory. In this case I think we should restart too, and this is more correct compared to the current code which sends SIGTRAP. Ignoring ENOMEM/etc from get_user_pages(), this can only happen if another thread unmaps this memory before find_active_uprobe() takes mmap_sem. It would be better to pretend it was unmapped before this insn was executed, restart, and get SIGSEGV. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120529192947.GF8057@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index a2ed82b4808c..1f02e3bbfc1d 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -1530,14 +1530,26 @@ static void handle_swbp(struct pt_regs *regs) struct uprobe_task *utask; struct uprobe *uprobe; unsigned long bp_vaddr; - int is_swbp; + int uninitialized_var(is_swbp); bp_vaddr = uprobe_get_swbp_addr(regs); uprobe = find_active_uprobe(bp_vaddr, &is_swbp); if (!uprobe) { - /* No matching uprobe; signal SIGTRAP. */ - send_sig(SIGTRAP, current, 0); + if (is_swbp > 0) { + /* No matching uprobe; signal SIGTRAP. */ + send_sig(SIGTRAP, current, 0); + } else { + /* + * Either we raced with uprobe_unregister() or we can't + * access this memory. The latter is only possible if + * another thread plays with our ->mm. In both cases + * we can simply restart. If this vma was unmapped we + * can pretend this insn was not executed yet and get + * the (correct) SIGSEGV after restart. + */ + instruction_pointer_set(regs, bp_vaddr); + } return; } -- cgit v1.2.3 From 778b032d96909690c19d84f8d17c13be65ed6f8e Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 29 May 2012 21:30:08 +0200 Subject: uprobes: Kill uprobes_srcu/uprobe_srcu_id Kill the no longer needed uprobes_srcu/uprobe_srcu_id code. It doesn't really work anyway. synchronize_srcu() can only synchronize with the code "inside" the srcu_read_lock/srcu_read_unlock section, while uprobe_pre_sstep_notifier() does srcu_read_lock() _after_ we already hit the breakpoint. I guess this probably works "in practice". synchronize_srcu() is slow and it implies synchronize_sched(), and the probed task enters the non- preemptible section at the start of exception handler. Still this is not right at least in theory, and task->uprobe_srcu_id blows task_struct. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Linus Torvalds Cc: Masami Hiramatsu Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120529193008.GG8057@redhat.com Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 - kernel/events/uprobes.c | 22 +++------------------- 2 files changed, 3 insertions(+), 20 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 6029d8c54476..6bd19655c1a7 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1569,7 +1569,6 @@ struct task_struct { #endif #ifdef CONFIG_UPROBES struct uprobe_task *utask; - int uprobe_srcu_id; #endif }; diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 1f02e3bbfc1d..8c5e043cd309 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -38,7 +38,6 @@ #define UINSNS_PER_PAGE (PAGE_SIZE/UPROBE_XOL_SLOT_BYTES) #define MAX_UPROBE_XOL_SLOTS UINSNS_PER_PAGE -static struct srcu_struct uprobes_srcu; static struct rb_root uprobes_tree = RB_ROOT; static DEFINE_SPINLOCK(uprobes_treelock); /* serialize rbtree access */ @@ -738,20 +737,14 @@ remove_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, loff_t vaddr) } /* - * There could be threads that have hit the breakpoint and are entering the - * notifier code and trying to acquire the uprobes_treelock. The thread - * calling delete_uprobe() that is removing the uprobe from the rb_tree can - * race with these threads and might acquire the uprobes_treelock compared - * to some of the breakpoint hit threads. In such a case, the breakpoint - * hit threads will not find the uprobe. The current unregistering thread - * waits till all other threads have hit a breakpoint, to acquire the - * uprobes_treelock before the uprobe is removed from the rbtree. + * There could be threads that have already hit the breakpoint. They + * will recheck the current insn and restart if find_uprobe() fails. + * See find_active_uprobe(). */ static void delete_uprobe(struct uprobe *uprobe) { unsigned long flags; - synchronize_srcu(&uprobes_srcu); spin_lock_irqsave(&uprobes_treelock, flags); rb_erase(&uprobe->rb_node, &uprobes_tree); spin_unlock_irqrestore(&uprobes_treelock, flags); @@ -1388,9 +1381,6 @@ void uprobe_free_utask(struct task_struct *t) { struct uprobe_task *utask = t->utask; - if (t->uprobe_srcu_id != -1) - srcu_read_unlock_raw(&uprobes_srcu, t->uprobe_srcu_id); - if (!utask) return; @@ -1408,7 +1398,6 @@ void uprobe_free_utask(struct task_struct *t) void uprobe_copy_process(struct task_struct *t) { t->utask = NULL; - t->uprobe_srcu_id = -1; } /* @@ -1513,9 +1502,6 @@ static struct uprobe *find_active_uprobe(unsigned long bp_vaddr, int *is_swbp) } else { *is_swbp = -EFAULT; } - - srcu_read_unlock_raw(&uprobes_srcu, current->uprobe_srcu_id); - current->uprobe_srcu_id = -1; up_read(&mm->mmap_sem); return uprobe; @@ -1656,7 +1642,6 @@ int uprobe_pre_sstep_notifier(struct pt_regs *regs) utask->state = UTASK_BP_HIT; set_thread_flag(TIF_UPROBE); - current->uprobe_srcu_id = srcu_read_lock_raw(&uprobes_srcu); return 1; } @@ -1691,7 +1676,6 @@ static int __init init_uprobes(void) mutex_init(&uprobes_mutex[i]); mutex_init(&uprobes_mmap_mutex[i]); } - init_srcu_struct(&uprobes_srcu); return register_die_notifier(&uprobe_exception_nb); } -- cgit v1.2.3 From 6be96a5c905178637ec06a5d456a76b2b304fca3 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 6 Jun 2012 19:12:30 -0700 Subject: cgroup: remove hierarchy_mutex It was introduced for memcg to iterate cgroup hierarchy without holding cgroup_mutex, but soon after that it was replaced with a lockless way in memcg. No one used hierarchy_mutex since that, so remove it. Signed-off-by: Li Zefan Signed-off-by: Tejun Heo --- Documentation/cgroups/cgroups.txt | 2 +- include/linux/cgroup.h | 17 ++------------- kernel/cgroup.c | 45 --------------------------------------- 3 files changed, 3 insertions(+), 61 deletions(-) (limited to 'kernel') diff --git a/Documentation/cgroups/cgroups.txt b/Documentation/cgroups/cgroups.txt index 8e74980ab385..e86faaea7d66 100644 --- a/Documentation/cgroups/cgroups.txt +++ b/Documentation/cgroups/cgroups.txt @@ -656,7 +656,7 @@ example in cpusets, no task may attach before 'cpus' and 'mems' are set up. void bind(struct cgroup *root) -(cgroup_mutex and ss->hierarchy_mutex held by caller) +(cgroup_mutex held by caller) Called when a cgroup subsystem is rebound to a different hierarchy and root cgroup. Currently this will only involve movement between diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index d3f5fba2c159..c90eaa803440 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -499,22 +499,9 @@ struct cgroup_subsys { #define MAX_CGROUP_TYPE_NAMELEN 32 const char *name; - /* - * Protects sibling/children links of cgroups in this - * hierarchy, plus protects which hierarchy (or none) the - * subsystem is a part of (i.e. root/sibling). To avoid - * potential deadlocks, the following operations should not be - * undertaken while holding any hierarchy_mutex: - * - * - allocating memory - * - initiating hotplug events - */ - struct mutex hierarchy_mutex; - struct lock_class_key subsys_key; - /* * Link to parent, and list entry in parent's children. - * Protected by this->hierarchy_mutex and cgroup_lock() + * Protected by cgroup_lock() */ struct cgroupfs_root *root; struct list_head sibling; @@ -602,7 +589,7 @@ int cgroup_attach_task_all(struct task_struct *from, struct task_struct *); * the lifetime of cgroup_subsys_state is subsys's matter. * * Looking up and scanning function should be called under rcu_read_lock(). - * Taking cgroup_mutex()/hierarchy_mutex() is not necessary for following calls. + * Taking cgroup_mutex is not necessary for following calls. * But the css returned by this routine can be "not populated yet" or "being * destroyed". The caller should check css and cgroup's status. */ diff --git a/kernel/cgroup.c b/kernel/cgroup.c index ceeafe874b3f..dec62f5936ef 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -1073,28 +1073,24 @@ static int rebind_subsystems(struct cgroupfs_root *root, BUG_ON(cgrp->subsys[i]); BUG_ON(!dummytop->subsys[i]); BUG_ON(dummytop->subsys[i]->cgroup != dummytop); - mutex_lock(&ss->hierarchy_mutex); cgrp->subsys[i] = dummytop->subsys[i]; cgrp->subsys[i]->cgroup = cgrp; list_move(&ss->sibling, &root->subsys_list); ss->root = root; if (ss->bind) ss->bind(cgrp); - mutex_unlock(&ss->hierarchy_mutex); /* refcount was already taken, and we're keeping it */ } else if (bit & removed_bits) { /* We're removing this subsystem */ BUG_ON(ss == NULL); BUG_ON(cgrp->subsys[i] != dummytop->subsys[i]); BUG_ON(cgrp->subsys[i]->cgroup != cgrp); - mutex_lock(&ss->hierarchy_mutex); if (ss->bind) ss->bind(dummytop); dummytop->subsys[i]->cgroup = dummytop; cgrp->subsys[i] = NULL; subsys[i]->root = &rootnode; list_move(&ss->sibling, &rootnode.subsys_list); - mutex_unlock(&ss->hierarchy_mutex); /* subsystem is now free - drop reference on module */ module_put(ss->module); } else if (bit & final_bits) { @@ -3917,37 +3913,6 @@ static void init_cgroup_css(struct cgroup_subsys_state *css, set_bit(CSS_CLEAR_CSS_REFS, &css->flags); } -static void cgroup_lock_hierarchy(struct cgroupfs_root *root) -{ - /* We need to take each hierarchy_mutex in a consistent order */ - int i; - - /* - * No worry about a race with rebind_subsystems that might mess up the - * locking order, since both parties are under cgroup_mutex. - */ - for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { - struct cgroup_subsys *ss = subsys[i]; - if (ss == NULL) - continue; - if (ss->root == root) - mutex_lock(&ss->hierarchy_mutex); - } -} - -static void cgroup_unlock_hierarchy(struct cgroupfs_root *root) -{ - int i; - - for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { - struct cgroup_subsys *ss = subsys[i]; - if (ss == NULL) - continue; - if (ss->root == root) - mutex_unlock(&ss->hierarchy_mutex); - } -} - /* * cgroup_create - create a cgroup * @parent: cgroup that will be parent of the new cgroup @@ -4008,9 +3973,7 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, ss->post_clone(cgrp); } - cgroup_lock_hierarchy(root); list_add(&cgrp->sibling, &cgrp->parent->children); - cgroup_unlock_hierarchy(root); root->number_of_cgroups++; err = cgroup_create_dir(cgrp, dentry, mode); @@ -4037,9 +4000,7 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, err_remove: - cgroup_lock_hierarchy(root); list_del(&cgrp->sibling); - cgroup_unlock_hierarchy(root); root->number_of_cgroups--; err_destroy: @@ -4247,10 +4208,8 @@ again: list_del_init(&cgrp->release_list); raw_spin_unlock(&release_list_lock); - cgroup_lock_hierarchy(cgrp->root); /* delete this cgroup from parent->children */ list_del_init(&cgrp->sibling); - cgroup_unlock_hierarchy(cgrp->root); list_del_init(&cgrp->allcg_node); @@ -4324,8 +4283,6 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss) * need to invoke fork callbacks here. */ BUG_ON(!list_empty(&init_task.tasks)); - mutex_init(&ss->hierarchy_mutex); - lockdep_set_class(&ss->hierarchy_mutex, &ss->subsys_key); ss->active = 1; /* this function shouldn't be used with modular subsystems, since they @@ -4452,8 +4409,6 @@ int __init_or_module cgroup_load_subsys(struct cgroup_subsys *ss) } write_unlock(&css_set_lock); - mutex_init(&ss->hierarchy_mutex); - lockdep_set_class(&ss->hierarchy_mutex, &ss->subsys_key); ss->active = 1; /* success! */ -- cgit v1.2.3 From 7eb9ba5ed312ec6ed9d22259c5da1acb7cf4bd29 Mon Sep 17 00:00:00 2001 From: Ananth N Mavinakayanahalli Date: Fri, 8 Jun 2012 15:02:57 +0530 Subject: uprobes: Pass probed vaddr to arch_uprobe_analyze_insn() On RISC architectures like powerpc, instructions are fixed size. Instruction analysis on such platforms is just a matter of (insn % 4). Pass the vaddr at which the uprobe is to be inserted so that arch_uprobe_analyze_insn() can flag misaligned registration requests. Signed-off-by: Ananth N Mavinakaynahalli Cc: michael@ellerman.id.au Cc: antonb@thinktux.localdomain Cc: Paul Mackerras Cc: benh@kernel.crashing.org Cc: peterz@infradead.org Cc: Srikar Dronamraju Cc: Jim Keniston Cc: oleg@redhat.com Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/20120608093257.GG13409@in.ibm.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/uprobes.h | 2 +- arch/x86/kernel/uprobes.c | 3 ++- kernel/events/uprobes.c | 2 +- 3 files changed, 4 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/arch/x86/include/asm/uprobes.h b/arch/x86/include/asm/uprobes.h index 1e9bed14f7ae..f3971bbcd1de 100644 --- a/arch/x86/include/asm/uprobes.h +++ b/arch/x86/include/asm/uprobes.h @@ -48,7 +48,7 @@ struct arch_uprobe_task { #endif }; -extern int arch_uprobe_analyze_insn(struct arch_uprobe *aup, struct mm_struct *mm); +extern int arch_uprobe_analyze_insn(struct arch_uprobe *aup, struct mm_struct *mm, unsigned long addr); extern int arch_uprobe_pre_xol(struct arch_uprobe *aup, struct pt_regs *regs); extern int arch_uprobe_post_xol(struct arch_uprobe *aup, struct pt_regs *regs); extern bool arch_uprobe_xol_was_trapped(struct task_struct *tsk); diff --git a/arch/x86/kernel/uprobes.c b/arch/x86/kernel/uprobes.c index dc4e910a7d96..36fd42091fa7 100644 --- a/arch/x86/kernel/uprobes.c +++ b/arch/x86/kernel/uprobes.c @@ -409,9 +409,10 @@ static int validate_insn_bits(struct arch_uprobe *auprobe, struct mm_struct *mm, * arch_uprobe_analyze_insn - instruction analysis including validity and fixups. * @mm: the probed address space. * @arch_uprobe: the probepoint information. + * @addr: virtual address at which to install the probepoint * Return 0 on success or a -ve number on error. */ -int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm) +int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm, unsigned long addr) { int ret; struct insn insn; diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 8c5e043cd309..b52376d02332 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -706,7 +706,7 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, if (is_swbp_insn((uprobe_opcode_t *)uprobe->arch.insn)) return -EEXIST; - ret = arch_uprobe_analyze_insn(&uprobe->arch, mm); + ret = arch_uprobe_analyze_insn(&uprobe->arch, mm, addr); if (ret) return ret; -- cgit v1.2.3 From b871a42b6091b720e82ddff237659534c525c25b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 11 Jun 2012 15:07:08 +0200 Subject: smpboot: Remove leftover declaration Signed-off-by: Thomas Gleixner --- kernel/smpboot.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel') diff --git a/kernel/smpboot.h b/kernel/smpboot.h index 80c0acfb8472..6ef9433e1c70 100644 --- a/kernel/smpboot.h +++ b/kernel/smpboot.h @@ -3,8 +3,6 @@ struct task_struct; -int smpboot_prepare(unsigned int cpu); - #ifdef CONFIG_GENERIC_SMP_IDLE_THREAD struct task_struct *idle_thread_get(unsigned int cpu); void idle_thread_set_boot_cpu(void); -- cgit v1.2.3 From 82ec90eac304e81b1389175b4dded7abecc678ef Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 17 May 2012 18:51:11 -0700 Subject: resources: allow adjust_resource() for resources with no parent If a resource has no parent, allow its start/end to be set arbitrarily as long as any children are still contained within the new range. [bhelgaas: changelog] Signed-off-by: Yinghai Lu Signed-off-by: Bjorn Helgaas --- kernel/resource.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/resource.c b/kernel/resource.c index e1d2b8ee76d5..dc8b47764443 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -722,14 +722,12 @@ int adjust_resource(struct resource *res, resource_size_t start, resource_size_t write_lock(&resource_lock); + if (!parent) + goto skip; + if ((start < parent->start) || (end > parent->end)) goto out; - for (tmp = res->child; tmp; tmp = tmp->sibling) { - if ((tmp->start < start) || (tmp->end > end)) - goto out; - } - if (res->sibling && (res->sibling->start <= end)) goto out; @@ -741,6 +739,11 @@ int adjust_resource(struct resource *res, resource_size_t start, resource_size_t goto out; } +skip: + for (tmp = res->child; tmp; tmp = tmp->sibling) + if ((tmp->start < start) || (tmp->end > end)) + goto out; + res->start = start; res->end = end; result = 0; -- cgit v1.2.3 From 8d240dd88cca33b704adf3fe281aa64b5aac2dd8 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Thu, 29 Mar 2012 19:11:40 +0200 Subject: ftrace: Remove a superfluous check register_ftrace_function() checks ftrace_disabled and calls __register_ftrace_function which does it again. Drop the first check and add the unlikely hint to the second one. Also, drop the label as John correctly notices. No functional change. Link: http://lkml.kernel.org/r/20120329171140.GE6409@aftab Cc: Borislav Petkov Cc: John Kacur Signed-off-by: Borislav Petkov Signed-off-by: Steven Rostedt --- kernel/trace/ftrace.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index a008663d86c8..b4f20fba09fc 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -312,7 +312,7 @@ static int remove_ftrace_list_ops(struct ftrace_ops **list, static int __register_ftrace_function(struct ftrace_ops *ops) { - if (ftrace_disabled) + if (unlikely(ftrace_disabled)) return -ENODEV; if (FTRACE_WARN_ON(ops == &global_ops)) @@ -4299,16 +4299,12 @@ int register_ftrace_function(struct ftrace_ops *ops) mutex_lock(&ftrace_lock); - if (unlikely(ftrace_disabled)) - goto out_unlock; - ret = __register_ftrace_function(ops); if (!ret) ret = ftrace_startup(ops, 0); - - out_unlock: mutex_unlock(&ftrace_lock); + return ret; } EXPORT_SYMBOL_GPL(register_ftrace_function); -- cgit v1.2.3 From 7374e82771c6d5a9af2080be46f64a5826c7efb1 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 31 May 2012 21:40:05 -0400 Subject: tracing: Register the ftrace internal events during early boot All trace events including ftrace internel events (like trace_printk and function tracing), register functions that describe how to print their output. The events may be recorded as soon as the ring buffer is allocated, but they are just raw binary in the buffer. The mapping of event ids to how to print them are held within a structure that is registered on system boot. If a crash happens in boot up before these functions are registered then their output (via ftrace_dump_on_oops) will be useless: Dumping ftrace buffer: --------------------------------- <...>-1 0.... 319705us : Unknown type 6 --------------------------------- This can be quite frustrating for a kernel developer trying to see what is going wrong. There's no reason to register them so late in the boot up process. They can be registered by early_initcall(). Reported-by: Peter Zijlstra Signed-off-by: Steven Rostedt --- kernel/trace/trace_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index df611a0e76c5..123b189c732c 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -1325,4 +1325,4 @@ __init static int init_events(void) return 0; } -device_initcall(init_events); +early_initcall(init_events); -- cgit v1.2.3 From efd68e7254503f3207805f674a1ea1d743f5dfe2 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Sun, 3 Jun 2012 22:04:33 -0700 Subject: devicetree: add helper inline for retrieving a node's full name The pattern (np ? np->full_name : "") is rather common in the kernel, but can also make for quite long lines. This patch adds a new inline function, of_node_full_name() so that the test for a valid node pointer doesn't need to be open coded at all call sites. Signed-off-by: Grant Likely Cc: Paul Mundt Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner --- arch/microblaze/pci/pci-common.c | 6 ++---- arch/powerpc/kernel/pci-common.c | 6 ++---- arch/powerpc/kernel/vio.c | 5 ++--- arch/powerpc/platforms/cell/iommu.c | 3 +-- arch/powerpc/platforms/pseries/iommu.c | 2 +- arch/sparc/kernel/of_device_64.c | 2 +- drivers/of/base.c | 2 +- drivers/of/irq.c | 2 +- include/linux/of.h | 10 ++++++++++ kernel/irq/irqdomain.c | 8 ++++---- 10 files changed, 25 insertions(+), 21 deletions(-) (limited to 'kernel') diff --git a/arch/microblaze/pci/pci-common.c b/arch/microblaze/pci/pci-common.c index ed22bfc5db14..ca8f6e769960 100644 --- a/arch/microblaze/pci/pci-common.c +++ b/arch/microblaze/pci/pci-common.c @@ -249,8 +249,7 @@ int pci_read_irq_line(struct pci_dev *pci_dev) } else { pr_debug(" Got one, spec %d cells (0x%08x 0x%08x...) on %s\n", oirq.size, oirq.specifier[0], oirq.specifier[1], - oirq.controller ? oirq.controller->full_name : - ""); + of_node_full_name(oirq.controller)); virq = irq_create_of_mapping(oirq.controller, oirq.specifier, oirq.size); @@ -1493,8 +1492,7 @@ static void __devinit pcibios_scan_phb(struct pci_controller *hose) struct pci_bus *bus; struct device_node *node = hose->dn; - pr_debug("PCI: Scanning PHB %s\n", - node ? node->full_name : ""); + pr_debug("PCI: Scanning PHB %s\n", of_node_full_name(node)); pcibios_setup_phb_resources(hose, &resources); diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c index 8e78e93c8185..886c254fd565 100644 --- a/arch/powerpc/kernel/pci-common.c +++ b/arch/powerpc/kernel/pci-common.c @@ -248,8 +248,7 @@ static int pci_read_irq_line(struct pci_dev *pci_dev) } else { pr_debug(" Got one, spec %d cells (0x%08x 0x%08x...) on %s\n", oirq.size, oirq.specifier[0], oirq.specifier[1], - oirq.controller ? oirq.controller->full_name : - ""); + of_node_full_name(oirq.controller)); virq = irq_create_of_mapping(oirq.controller, oirq.specifier, oirq.size); @@ -1628,8 +1627,7 @@ void __devinit pcibios_scan_phb(struct pci_controller *hose) struct device_node *node = hose->dn; int mode; - pr_debug("PCI: Scanning PHB %s\n", - node ? node->full_name : ""); + pr_debug("PCI: Scanning PHB %s\n", of_node_full_name(node)); /* Get some IO space for the new PHB */ pcibios_setup_phb_io_space(hose); diff --git a/arch/powerpc/kernel/vio.c b/arch/powerpc/kernel/vio.c index cb87301ccd55..63f72ede4341 100644 --- a/arch/powerpc/kernel/vio.c +++ b/arch/powerpc/kernel/vio.c @@ -1296,8 +1296,7 @@ static void __devinit vio_dev_release(struct device *dev) struct iommu_table *tbl = get_iommu_table_base(dev); if (tbl) - iommu_free_table(tbl, dev->of_node ? - dev->of_node->full_name : dev_name(dev)); + iommu_free_table(tbl, of_node_full_name(dev->of_node)); of_node_put(dev->of_node); kfree(to_vio_dev(dev)); } @@ -1509,7 +1508,7 @@ static ssize_t devspec_show(struct device *dev, { struct device_node *of_node = dev->of_node; - return sprintf(buf, "%s\n", of_node ? of_node->full_name : "none"); + return sprintf(buf, "%s\n", of_node_full_name(of_node)); } static ssize_t modalias_show(struct device *dev, struct device_attribute *attr, diff --git a/arch/powerpc/platforms/cell/iommu.c b/arch/powerpc/platforms/cell/iommu.c index b9f509a34c01..b6732004c882 100644 --- a/arch/powerpc/platforms/cell/iommu.c +++ b/arch/powerpc/platforms/cell/iommu.c @@ -552,8 +552,7 @@ static struct iommu_table *cell_get_iommu_table(struct device *dev) iommu = cell_iommu_for_node(dev_to_node(dev)); if (iommu == NULL || list_empty(&iommu->windows)) { printk(KERN_ERR "iommu: missing iommu for %s (node %d)\n", - dev->of_node ? dev->of_node->full_name : "?", - dev_to_node(dev)); + of_node_full_name(dev->of_node), dev_to_node(dev)); return NULL; } window = list_entry(iommu->windows.next, struct iommu_window, list); diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index 0915b1ad66ce..aab5fbc924e6 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -1051,7 +1051,7 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev) if (!pdn || !PCI_DN(pdn)) { printk(KERN_WARNING "pci_dma_dev_setup_pSeriesLP: " "no DMA window found for pci dev=%s dn=%s\n", - pci_name(dev), dn? dn->full_name : ""); + pci_name(dev), of_node_full_name(dn)); return; } pr_debug(" parent is %s\n", pdn->full_name); diff --git a/arch/sparc/kernel/of_device_64.c b/arch/sparc/kernel/of_device_64.c index 7a3be6f6737a..7bbdc26d9512 100644 --- a/arch/sparc/kernel/of_device_64.c +++ b/arch/sparc/kernel/of_device_64.c @@ -580,7 +580,7 @@ static unsigned int __init build_one_device_irq(struct platform_device *op, printk("%s: Apply [%s:%x] imap --> [%s:%x]\n", op->dev.of_node->full_name, pp->full_name, this_orig_irq, - (iret ? iret->full_name : "NULL"), irq); + of_node_full_name(iret), irq); if (!iret) break; diff --git a/drivers/of/base.c b/drivers/of/base.c index d9bfd49b1935..9282d4cc1091 100644 --- a/drivers/of/base.c +++ b/drivers/of/base.c @@ -1173,7 +1173,7 @@ static void of_alias_add(struct alias_prop *ap, struct device_node *np, ap->stem[stem_len] = 0; list_add_tail(&ap->link, &aliases_lookup); pr_debug("adding DT alias:%s: stem=%s id=%i node=%s\n", - ap->alias, ap->stem, ap->id, np ? np->full_name : NULL); + ap->alias, ap->stem, ap->id, of_node_full_name(np)); } /** diff --git a/drivers/of/irq.c b/drivers/of/irq.c index 9cf00602f566..ff8ab7b27373 100644 --- a/drivers/of/irq.c +++ b/drivers/of/irq.c @@ -255,7 +255,7 @@ int of_irq_map_raw(struct device_node *parent, const __be32 *intspec, skiplevel: /* Iterate again with new parent */ - pr_debug(" -> new parent: %s\n", newpar ? newpar->full_name : "<>"); + pr_debug(" -> new parent: %s\n", of_node_full_name(newpar)); of_node_put(ipar); ipar = newpar; newpar = NULL; diff --git a/include/linux/of.h b/include/linux/of.h index 2ec1083af7ff..1012377cae92 100644 --- a/include/linux/of.h +++ b/include/linux/of.h @@ -163,6 +163,11 @@ static inline int of_node_to_nid(struct device_node *np) { return -1; } #define of_node_to_nid of_node_to_nid #endif +static inline const char* of_node_full_name(struct device_node *np) +{ + return np ? np->full_name : ""; +} + extern struct device_node *of_find_node_by_name(struct device_node *from, const char *name); #define for_each_node_by_name(dn, name) \ @@ -303,6 +308,11 @@ const char *of_prop_next_string(struct property *prop, const char *cur); #else /* CONFIG_OF */ +static inline const char* of_node_full_name(struct device_node *np) +{ + return ""; +} + static inline bool of_have_populated_dt(void) { return false; diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 41c1564103f1..38c5eb839c92 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -448,7 +448,7 @@ unsigned int irq_create_mapping(struct irq_domain *domain, } pr_debug("irq %lu on domain %s mapped to virtual irq %u\n", - hwirq, domain->of_node ? domain->of_node->full_name : "null", virq); + hwirq, of_node_full_name(domain->of_node), virq); return virq; } @@ -477,7 +477,7 @@ unsigned int irq_create_of_mapping(struct device_node *controller, return intspec[0]; #endif pr_warning("no irq domain found for %s !\n", - controller->full_name); + of_node_full_name(controller)); return 0; } @@ -725,8 +725,8 @@ static int virq_debug_show(struct seq_file *m, void *private) data = irq_desc_get_chip_data(desc); seq_printf(m, data ? "0x%p " : " %p ", data); - if (desc->irq_data.domain && desc->irq_data.domain->of_node) - p = desc->irq_data.domain->of_node->full_name; + if (desc->irq_data.domain) + p = of_node_full_name(desc->irq_data.domain->of_node); else p = none; seq_printf(m, "%s\n", p); -- cgit v1.2.3 From 5ca4db61e859526b2dbee3bcea3626d3de49a0b2 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Sun, 3 Jun 2012 22:04:34 -0700 Subject: irqdomain: Simple NUMA awareness. While common irqdesc allocation is node aware, the irqdomain code is not. Presently we observe a number of regressions/inconsistencies on NUMA-capable platforms: - Platforms using irqdomains with legacy mappings, where the irq_descs are allocated node-local and the irqdomain data structure is not. - Drivers implementing irqdomains will lose node locality regardless of the underlying struct device's node id. This plugs in NUMA node id proliferation across the various allocation callsites by way of_node_to_nid() node lookup. While of_node_to_nid() does the right thing for OF-capable platforms it doesn't presently handle the non-DT case. This is trivially dealt with by simply wraping in to numa_node_id() unconditionally. Signed-off-by: Paul Mundt Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Rob Herring Signed-off-by: Grant Likely --- include/linux/of.h | 15 ++++++++++----- kernel/irq/irqdomain.c | 13 ++++++++----- 2 files changed, 18 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/include/linux/of.h b/include/linux/of.h index 1012377cae92..76930ee78db5 100644 --- a/include/linux/of.h +++ b/include/linux/of.h @@ -21,6 +21,7 @@ #include #include #include +#include #include #include @@ -158,11 +159,6 @@ static inline unsigned long of_read_ulong(const __be32 *cell, int size) #define OF_BAD_ADDR ((u64)-1) -#ifndef of_node_to_nid -static inline int of_node_to_nid(struct device_node *np) { return -1; } -#define of_node_to_nid of_node_to_nid -#endif - static inline const char* of_node_full_name(struct device_node *np) { return np ? np->full_name : ""; @@ -412,6 +408,15 @@ static inline int of_machine_is_compatible(const char *compat) while (0) #endif /* CONFIG_OF */ +#ifndef of_node_to_nid +static inline int of_node_to_nid(struct device_node *np) +{ + return numa_node_id(); +} + +#define of_node_to_nid of_node_to_nid +#endif + /** * of_property_read_bool - Findfrom a property * @np: device node from which the property value is to be read. diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 38c5eb839c92..79ae0ebb90ba 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -45,7 +46,8 @@ static struct irq_domain *irq_domain_alloc(struct device_node *of_node, { struct irq_domain *domain; - domain = kzalloc(sizeof(*domain), GFP_KERNEL); + domain = kzalloc_node(sizeof(*domain), GFP_KERNEL, + of_node_to_nid(of_node)); if (WARN_ON(!domain)) return NULL; @@ -229,7 +231,8 @@ struct irq_domain *irq_domain_add_linear(struct device_node *of_node, struct irq_domain *domain; unsigned int *revmap; - revmap = kzalloc(sizeof(*revmap) * size, GFP_KERNEL); + revmap = kzalloc_node(sizeof(*revmap) * size, GFP_KERNEL, + of_node_to_nid(of_node)); if (WARN_ON(!revmap)) return NULL; @@ -367,7 +370,7 @@ unsigned int irq_create_direct_mapping(struct irq_domain *domain) BUG_ON(domain == NULL); WARN_ON(domain->revmap_type != IRQ_DOMAIN_MAP_NOMAP); - virq = irq_alloc_desc_from(1, 0); + virq = irq_alloc_desc_from(1, of_node_to_nid(domain->of_node)); if (!virq) { pr_debug("create_direct virq allocation failed\n"); return 0; @@ -433,9 +436,9 @@ unsigned int irq_create_mapping(struct irq_domain *domain, hint = hwirq % nr_irqs; if (hint == 0) hint++; - virq = irq_alloc_desc_from(hint, 0); + virq = irq_alloc_desc_from(hint, of_node_to_nid(domain->of_node)); if (virq <= 0) - virq = irq_alloc_desc_from(1, 0); + virq = irq_alloc_desc_from(1, of_node_to_nid(domain->of_node)); if (virq <= 0) { pr_debug("-> virq allocation failed\n"); return 0; -- cgit v1.2.3 From 732557047128e5c200c3590efba39f10ac46bcb2 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Sun, 3 Jun 2012 22:04:35 -0700 Subject: irqdomain: Remove unnecessary test for IRQ_DOMAIN_MAP_LEGACY Where irq_domain_associate() is called in irq_create_mapping, there is no need to test for IRQ_DOMAIN_MAP_LEGACY because it is already tested for earlier in the routine. Signed-off-by: Grant Likely Cc: Paul Mundt Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Rob Herring --- kernel/irq/irqdomain.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 79ae0ebb90ba..b1f774cfd089 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -445,8 +445,7 @@ unsigned int irq_create_mapping(struct irq_domain *domain, } if (irq_setup_virq(domain, virq, hwirq)) { - if (domain->revmap_type != IRQ_DOMAIN_MAP_LEGACY) - irq_free_desc(virq); + irq_free_desc(virq); return 0; } -- cgit v1.2.3 From ea131377148cdfe90641b42ae9aa5a6b3a4fa327 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:22 +0200 Subject: uprobes: Valid_vma() should reject VM_HUGETLB __replace_page() obviously can't work with the hugetlbfs mappings, uprobe_register() will likely crash the kernel. Change valid_vma() to check VM_HUGETLB as well. As for PageTransHuge() no need to worry, vma->vm_file != NULL. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120615154322.GA9561@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index b52376d02332..f0d04530af63 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -99,7 +99,8 @@ static bool valid_vma(struct vm_area_struct *vma, bool is_register) if (!is_register) return true; - if ((vma->vm_flags & (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)) == (VM_READ|VM_EXEC)) + if ((vma->vm_flags & (VM_HUGETLB|VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)) + == (VM_READ|VM_EXEC)) return true; return false; -- cgit v1.2.3 From cc359d180fa9c25a4c1819f17e07a422d788353d Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:25 +0200 Subject: uprobes: __copy_insn() should ensure a_ops->readpage != NULL __copy_insn() blindly calls read_mapping_page(), this will crash the kernel if ->readpage == NULL, add the necessary check. For example, hugetlbfs_aops->readpage is NULL. Perhaps we should change read_mapping_page() instead. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120615154325.GA9568@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index f0d04530af63..604930bf9c92 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -610,6 +610,9 @@ __copy_insn(struct address_space *mapping, struct vm_area_struct *vma, char *ins if (!filp) return -EINVAL; + if (!mapping->a_ops->readpage) + return -EIO; + idx = (unsigned long)(offset >> PAGE_CACHE_SHIFT); off1 = offset &= ~PAGE_MASK; -- cgit v1.2.3 From 5323ce71e4b4e1f188ebbc0cc7776885ea6c75fb Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:28 +0200 Subject: uprobes: Write_opcode()->__replace_page() can race with try_to_unmap() write_opcode() gets old_page via get_user_pages() and then calls __replace_page() which assumes that this old_page is still mapped after pte_offset_map_lock(). This is not true if this old_page was already try_to_unmap()'ed, and in this case everything __replace_page() does with old_page is wrong. Just for example, put_page() is not balanced. I think it is possible to teach __replace_page() to handle this unlikely case correctly, but this patch simply changes it to use page_check_address() and return -EAGAIN if it fails. The caller should notice this error code and retry. Note: write_opcode() asks for the cleanups, I'll try to do this in a separate patch. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120615154328.GA9571@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 41 +++++++++++++---------------------------- 1 file changed, 13 insertions(+), 28 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 604930bf9c92..3ccdb29ee8d6 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -129,33 +129,17 @@ static loff_t vma_address(struct vm_area_struct *vma, loff_t offset) static int __replace_page(struct vm_area_struct *vma, struct page *page, struct page *kpage) { struct mm_struct *mm = vma->vm_mm; - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; - pte_t *ptep; - spinlock_t *ptl; unsigned long addr; - int err = -EFAULT; + spinlock_t *ptl; + pte_t *ptep; addr = page_address_in_vma(page, vma); if (addr == -EFAULT) - goto out; - - pgd = pgd_offset(mm, addr); - if (!pgd_present(*pgd)) - goto out; - - pud = pud_offset(pgd, addr); - if (!pud_present(*pud)) - goto out; + return -EFAULT; - pmd = pmd_offset(pud, addr); - if (!pmd_present(*pmd)) - goto out; - - ptep = pte_offset_map_lock(mm, pmd, addr, &ptl); + ptep = page_check_address(page, mm, addr, &ptl, 0); if (!ptep) - goto out; + return -EAGAIN; get_page(kpage); page_add_new_anon_rmap(kpage, vma, addr); @@ -174,10 +158,8 @@ static int __replace_page(struct vm_area_struct *vma, struct page *page, struct try_to_free_swap(page); put_page(page); pte_unmap_unlock(ptep, ptl); - err = 0; -out: - return err; + return 0; } /** @@ -222,9 +204,10 @@ static int write_opcode(struct arch_uprobe *auprobe, struct mm_struct *mm, void *vaddr_old, *vaddr_new; struct vm_area_struct *vma; struct uprobe *uprobe; + unsigned long pgoff; loff_t addr; int ret; - +retry: /* Read the page with vaddr into memory */ ret = get_user_pages(NULL, mm, vaddr, 1, 0, 0, &old_page, &vma); if (ret <= 0) @@ -269,9 +252,9 @@ static int write_opcode(struct arch_uprobe *auprobe, struct mm_struct *mm, memcpy(vaddr_new, vaddr_old, PAGE_SIZE); /* poke the new insn in, ASSUMES we don't cross page boundary */ - vaddr &= ~PAGE_MASK; - BUG_ON(vaddr + UPROBE_SWBP_INSN_SIZE > PAGE_SIZE); - memcpy(vaddr_new + vaddr, &opcode, UPROBE_SWBP_INSN_SIZE); + pgoff = (vaddr & ~PAGE_MASK); + BUG_ON(pgoff + UPROBE_SWBP_INSN_SIZE > PAGE_SIZE); + memcpy(vaddr_new + pgoff, &opcode, UPROBE_SWBP_INSN_SIZE); kunmap_atomic(vaddr_new); kunmap_atomic(vaddr_old); @@ -291,6 +274,8 @@ unlock_out: put_out: put_page(old_page); + if (unlikely(ret == -EAGAIN)) + goto retry; return ret; } -- cgit v1.2.3 From c1914a0936f79ed0236f670122e06e36e4d332ee Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:31 +0200 Subject: uprobes: Install_breakpoint() should fail if is_swbp_insn() == T install_breakpoint() returns -EEXIST if is_swbp_insn(orig_insn) == T, the caller treats this code as success. This is doubly wrong. The successful return should set UPROBE_COPY_INSN, but the real problem is that it shouldn't succeed. If the probed insn is int3 the application should get SIGTRAP, this won't happen with uprobe. Probably we can fix this, we can add the UPROBE_SHARED_BP flag and teach handle_swbp/set_orig_insn to handle this case correctly. But this needs some complications and we have other insns which can't be probed, lets make a simple fix for now. I think this needs a cleanup. UPROBE_COPY_INSN should die, copy_insn() should be called by alloc_uprobe(). arch_uprobe_analyze_insn() depends on ->mm (ia32_compat) but it is called only once. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120615154331.GA9578@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 3ccdb29ee8d6..ec78152e32e9 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -693,7 +693,7 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, return ret; if (is_swbp_insn((uprobe_opcode_t *)uprobe->arch.insn)) - return -EEXIST; + return -ENOTSUPP; ret = arch_uprobe_analyze_insn(&uprobe->arch, mm, addr); if (ret) -- cgit v1.2.3 From 268720903f87e0b84b161626c4447b81671b5d18 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:33 +0200 Subject: uprobes: Rework register_for_each_vma() to make it O(n) Currently register_for_each_vma() is O(n ** 2) + O(n ** 3), every time find_next_vma_info() "restarts" the vma_prio_tree_foreach() loop and each iteration rechecks the whole try_list. This also means that try_list can grow "indefinitely" if register/unregister races with munmap/mmap activity even if the number of mapping is bounded at any time. With this patch register_for_each_vma() builds the list of mm/vaddr structures only once and does install_breakpoint() for each entry. We do not care about the new mappings which can be created after build_map_info() drops mapping->i_mmap_mutex, uprobe_mmap() should do its work. Note that we do not allocate map_info under i_mmap_mutex, this can deadlock with page reclaim (but see the next patch). So we use 2 lists, "curr" which we are going to return, and "prev" which holds the already allocated memory. The main loop deques the entry from "prev" (initially it is empty), and if "prev" becomes empty again it counts the number of entries we need to pre-allocate outside of i_mmap_mutex. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Acked-by: Peter Zijlstra Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Link: http://lkml.kernel.org/r/20120615154333.GA9581@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 199 +++++++++++++++++++++--------------------------- 1 file changed, 86 insertions(+), 113 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index ec78152e32e9..4e0db3496d70 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -60,17 +60,6 @@ static struct mutex uprobes_mmap_mutex[UPROBES_HASH_SZ]; */ static atomic_t uprobe_events = ATOMIC_INIT(0); -/* - * Maintain a temporary per vma info that can be used to search if a vma - * has already been handled. This structure is introduced since extending - * vm_area_struct wasnt recommended. - */ -struct vma_info { - struct list_head probe_list; - struct mm_struct *mm; - loff_t vaddr; -}; - struct uprobe { struct rb_node rb_node; /* node in the rb tree */ atomic_t ref; @@ -742,139 +731,123 @@ static void delete_uprobe(struct uprobe *uprobe) atomic_dec(&uprobe_events); } -static struct vma_info * -__find_next_vma_info(struct address_space *mapping, struct list_head *head, - struct vma_info *vi, loff_t offset, bool is_register) +struct map_info { + struct map_info *next; + struct mm_struct *mm; + loff_t vaddr; +}; + +static inline struct map_info *free_map_info(struct map_info *info) { + struct map_info *next = info->next; + kfree(info); + return next; +} + +static struct map_info * +build_map_info(struct address_space *mapping, loff_t offset, bool is_register) +{ + unsigned long pgoff = offset >> PAGE_SHIFT; struct prio_tree_iter iter; struct vm_area_struct *vma; - struct vma_info *tmpvi; - unsigned long pgoff; - int existing_vma; - loff_t vaddr; - - pgoff = offset >> PAGE_SHIFT; + struct map_info *curr = NULL; + struct map_info *prev = NULL; + struct map_info *info; + int more = 0; + again: + mutex_lock(&mapping->i_mmap_mutex); vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, pgoff, pgoff) { if (!valid_vma(vma, is_register)) continue; - existing_vma = 0; - vaddr = vma_address(vma, offset); - - list_for_each_entry(tmpvi, head, probe_list) { - if (tmpvi->mm == vma->vm_mm && tmpvi->vaddr == vaddr) { - existing_vma = 1; - break; - } - } - - /* - * Another vma needs a probe to be installed. However skip - * installing the probe if the vma is about to be unlinked. - */ - if (!existing_vma && atomic_inc_not_zero(&vma->vm_mm->mm_users)) { - vi->mm = vma->vm_mm; - vi->vaddr = vaddr; - list_add(&vi->probe_list, head); - - return vi; + if (!prev) { + more++; + continue; } - } - - return NULL; -} -/* - * Iterate in the rmap prio tree and find a vma where a probe has not - * yet been inserted. - */ -static struct vma_info * -find_next_vma_info(struct address_space *mapping, struct list_head *head, - loff_t offset, bool is_register) -{ - struct vma_info *vi, *retvi; + if (!atomic_inc_not_zero(&vma->vm_mm->mm_users)) + continue; - vi = kzalloc(sizeof(struct vma_info), GFP_KERNEL); - if (!vi) - return ERR_PTR(-ENOMEM); + info = prev; + prev = prev->next; + info->next = curr; + curr = info; - mutex_lock(&mapping->i_mmap_mutex); - retvi = __find_next_vma_info(mapping, head, vi, offset, is_register); + info->mm = vma->vm_mm; + info->vaddr = vma_address(vma, offset); + } mutex_unlock(&mapping->i_mmap_mutex); - if (!retvi) - kfree(vi); + if (!more) + goto out; + + prev = curr; + while (curr) { + mmput(curr->mm); + curr = curr->next; + } - return retvi; + do { + info = kmalloc(sizeof(struct map_info), GFP_KERNEL); + if (!info) { + curr = ERR_PTR(-ENOMEM); + goto out; + } + info->next = prev; + prev = info; + } while (--more); + + goto again; + out: + while (prev) + prev = free_map_info(prev); + return curr; } static int register_for_each_vma(struct uprobe *uprobe, bool is_register) { - struct list_head try_list; - struct vm_area_struct *vma; - struct address_space *mapping; - struct vma_info *vi, *tmpvi; - struct mm_struct *mm; - loff_t vaddr; - int ret; + struct map_info *info; + int err = 0; - mapping = uprobe->inode->i_mapping; - INIT_LIST_HEAD(&try_list); - - ret = 0; + info = build_map_info(uprobe->inode->i_mapping, + uprobe->offset, is_register); + if (IS_ERR(info)) + return PTR_ERR(info); - for (;;) { - vi = find_next_vma_info(mapping, &try_list, uprobe->offset, is_register); - if (!vi) - break; + while (info) { + struct mm_struct *mm = info->mm; + struct vm_area_struct *vma; + loff_t vaddr; - if (IS_ERR(vi)) { - ret = PTR_ERR(vi); - break; - } + if (err) + goto free; - mm = vi->mm; down_write(&mm->mmap_sem); - vma = find_vma(mm, (unsigned long)vi->vaddr); - if (!vma || !valid_vma(vma, is_register)) { - list_del(&vi->probe_list); - kfree(vi); - up_write(&mm->mmap_sem); - mmput(mm); - continue; - } + vma = find_vma(mm, (unsigned long)info->vaddr); + if (!vma || !valid_vma(vma, is_register)) + goto unlock; + vaddr = vma_address(vma, uprobe->offset); if (vma->vm_file->f_mapping->host != uprobe->inode || - vaddr != vi->vaddr) { - list_del(&vi->probe_list); - kfree(vi); - up_write(&mm->mmap_sem); - mmput(mm); - continue; - } + vaddr != info->vaddr) + goto unlock; - if (is_register) - ret = install_breakpoint(uprobe, mm, vma, vi->vaddr); - else - remove_breakpoint(uprobe, mm, vi->vaddr); - - up_write(&mm->mmap_sem); - mmput(mm); if (is_register) { - if (ret && ret == -EEXIST) - ret = 0; - if (ret) - break; + err = install_breakpoint(uprobe, mm, vma, info->vaddr); + if (err == -EEXIST) + err = 0; + } else { + remove_breakpoint(uprobe, mm, info->vaddr); } + unlock: + up_write(&mm->mmap_sem); + free: + mmput(mm); + info = free_map_info(info); } - list_for_each_entry_safe(vi, tmpvi, &try_list, probe_list) { - list_del(&vi->probe_list); - kfree(vi); - } - - return ret; + return err; } static int __uprobe_register(struct uprobe *uprobe) -- cgit v1.2.3 From 7a5bfb66b07f22d2429db776da7bb8b57bfb5cff Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:36 +0200 Subject: uprobes: Change build_map_info() to try kmalloc(GFP_NOWAIT) first build_map_info() doesn't allocate the memory under i_mmap_mutex to avoid the deadlock with page reclaim. But it can try GFP_NOWAIT first, it should work in the likely case and thus we almost never need the pre-alloc-and-retry path. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Acked-by: Peter Zijlstra Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Link: http://lkml.kernel.org/r/20120615154336.GA9588@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 4e0db3496d70..897417dbca8e 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -761,6 +761,16 @@ build_map_info(struct address_space *mapping, loff_t offset, bool is_register) if (!valid_vma(vma, is_register)) continue; + if (!prev && !more) { + /* + * Needs GFP_NOWAIT to avoid i_mmap_mutex recursion through + * reclaim. This is optimistic, no harm done if it fails. + */ + prev = kmalloc(sizeof(struct map_info), + GFP_NOWAIT | __GFP_NOMEMALLOC | __GFP_NOWARN); + if (prev) + prev->next = NULL; + } if (!prev) { more++; continue; -- cgit v1.2.3 From c5784de2b351fe871bb57487878f7fc7ec5b075c Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 15 Jun 2012 17:43:39 +0200 Subject: uprobes: Document uprobe_register() vs uprobe_mmap() race Because the mind is treacherous and makes us forget we need to write stuff down. Signed-off-by: Peter Zijlstra Signed-off-by: Oleg Nesterov Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120615154339.GA9591@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 31 ++++++++++++++++++++++++++++--- 1 file changed, 28 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 897417dbca8e..2671d9ad49be 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -44,6 +44,23 @@ static DEFINE_SPINLOCK(uprobes_treelock); /* serialize rbtree access */ #define UPROBES_HASH_SZ 13 +/* + * We need separate register/unregister and mmap/munmap lock hashes because + * of mmap_sem nesting. + * + * uprobe_register() needs to install probes on (potentially) all processes + * and thus needs to acquire multiple mmap_sems (consequtively, not + * concurrently), whereas uprobe_mmap() is called while holding mmap_sem + * for the particular process doing the mmap. + * + * uprobe_register()->register_for_each_vma() needs to drop/acquire mmap_sem + * because of lock order against i_mmap_mutex. This means there's a hole in + * the register vma iteration where a mmap() can happen. + * + * Thus uprobe_register() can race with uprobe_mmap() and we can try and + * install a probe where one is already installed. + */ + /* serialize (un)register */ static struct mutex uprobes_mutex[UPROBES_HASH_SZ]; @@ -339,7 +356,9 @@ out: int __weak set_swbp(struct arch_uprobe *auprobe, struct mm_struct *mm, unsigned long vaddr) { int result; - + /* + * See the comment near uprobes_hash(). + */ result = is_swbp_at_addr(mm, vaddr); if (result == 1) return -EEXIST; @@ -845,6 +864,10 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register) if (is_register) { err = install_breakpoint(uprobe, mm, vma, info->vaddr); + /* + * We can race against uprobe_mmap(), see the + * comment near uprobe_hash(). + */ if (err == -EEXIST) err = 0; } else { @@ -1054,8 +1077,10 @@ int uprobe_mmap(struct vm_area_struct *vma) } ret = install_breakpoint(uprobe, vma->vm_mm, vma, vaddr); - - /* Ignore double add: */ + /* + * We can race against uprobe_register(), see the + * comment near uprobe_hash(). + */ if (ret == -EEXIST) { ret = 0; -- cgit v1.2.3 From d436615e60c386095dac4a9bf72b08868d2a7564 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:42 +0200 Subject: uprobes: Copy_insn() shouldn't depend on mm/vma/vaddr 1. copy_insn() doesn't need "addr", it can use uprobe->offset. Remove this argument. 2. Change copy_insn/__copy_insn to accept "struct file*" instead of vma. copy_insn() is called only once and mm/vma/vaddr are random, it shouldn't depend on them. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Acked-by: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120615154342.GA9598@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 2671d9ad49be..08ef566da763 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -591,10 +591,9 @@ static bool consumer_del(struct uprobe *uprobe, struct uprobe_consumer *uc) } static int -__copy_insn(struct address_space *mapping, struct vm_area_struct *vma, char *insn, +__copy_insn(struct address_space *mapping, struct file *filp, char *insn, unsigned long nbytes, unsigned long offset) { - struct file *filp = vma->vm_file; struct page *page; void *vaddr; unsigned long off1; @@ -625,15 +624,13 @@ __copy_insn(struct address_space *mapping, struct vm_area_struct *vma, char *ins return 0; } -static int -copy_insn(struct uprobe *uprobe, struct vm_area_struct *vma, unsigned long addr) +static int copy_insn(struct uprobe *uprobe, struct file *filp) { struct address_space *mapping; unsigned long nbytes; int bytes; - addr &= ~PAGE_MASK; - nbytes = PAGE_SIZE - addr; + nbytes = PAGE_SIZE - (uprobe->offset & ~PAGE_MASK); mapping = uprobe->inode->i_mapping; /* Instruction at end of binary; copy only available bytes */ @@ -644,13 +641,13 @@ copy_insn(struct uprobe *uprobe, struct vm_area_struct *vma, unsigned long addr) /* Instruction at the page-boundary; copy bytes in second page */ if (nbytes < bytes) { - if (__copy_insn(mapping, vma, uprobe->arch.insn + nbytes, + if (__copy_insn(mapping, filp, uprobe->arch.insn + nbytes, bytes - nbytes, uprobe->offset + nbytes)) return -ENOMEM; bytes = nbytes; } - return __copy_insn(mapping, vma, uprobe->arch.insn, bytes, uprobe->offset); + return __copy_insn(mapping, filp, uprobe->arch.insn, bytes, uprobe->offset); } /* @@ -696,7 +693,7 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, addr = (unsigned long)vaddr; if (!(uprobe->flags & UPROBE_COPY_INSN)) { - ret = copy_insn(uprobe, vma, addr); + ret = copy_insn(uprobe, vma->vm_file); if (ret) return ret; -- cgit v1.2.3 From fc36f59565861af2e897225bc3782479a26c5d5a Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:44 +0200 Subject: uprobes: Copy_insn() should not return -ENOMEM if __copy_insn() fails copy_insn() returns -ENOMEM if the first __copy_insn() fails, it should return the correct error code. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Acked-by: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120615154344.GA9601@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 08ef566da763..2db1d94d7dfc 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -641,10 +641,10 @@ static int copy_insn(struct uprobe *uprobe, struct file *filp) /* Instruction at the page-boundary; copy bytes in second page */ if (nbytes < bytes) { - if (__copy_insn(mapping, filp, uprobe->arch.insn + nbytes, - bytes - nbytes, uprobe->offset + nbytes)) - return -ENOMEM; - + int err = __copy_insn(mapping, filp, uprobe->arch.insn + nbytes, + bytes - nbytes, uprobe->offset + nbytes); + if (err) + return err; bytes = nbytes; } return __copy_insn(mapping, filp, uprobe->arch.insn, bytes, uprobe->offset); -- cgit v1.2.3 From eb2bf57bee42c7565032f93adaa211e2c9fcc52c Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:47 +0200 Subject: uprobes: No need to re-check vma_address() in write_opcode() write_opcode() is called by register_for_each_vma() and uprobe_mmap() paths. In both cases the caller has already verified this vaddr under mmap_sem, no need to re-check. Note also that this check is wrong anyway, we should not truncate loff_t returned by vma_address() if we do not trust this mapping. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Acked-by: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120615154347.GA9604@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 2db1d94d7dfc..14c71a2aadad 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -211,7 +211,6 @@ static int write_opcode(struct arch_uprobe *auprobe, struct mm_struct *mm, struct vm_area_struct *vma; struct uprobe *uprobe; unsigned long pgoff; - loff_t addr; int ret; retry: /* Read the page with vaddr into memory */ @@ -235,10 +234,6 @@ retry: if (mapping != vma->vm_file->f_mapping) goto put_out; - addr = vma_address(vma, uprobe->offset); - if (vaddr != (unsigned long)addr) - goto put_out; - ret = -ENOMEM; new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, vaddr); if (!new_page) -- cgit v1.2.3 From d9c4a30e82614d43b55893a73f31e7284007ce82 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:50 +0200 Subject: uprobes: Move BUG_ON(UPROBE_SWBP_INSN_SIZE) from write_opcode() to install_breakpoint() write_opcode() ensures that UPROBE_SWBP_INSN doesn't cross the page boundary. This looks a bit confusing, the check does not depend on vaddr and it is enough to do it only once right after install_breakpoint()->arch_uprobe_analyze_insn(). Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Acked-by: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120615154350.GA9611@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 14c71a2aadad..b9c61bda9029 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -210,7 +210,6 @@ static int write_opcode(struct arch_uprobe *auprobe, struct mm_struct *mm, void *vaddr_old, *vaddr_new; struct vm_area_struct *vma; struct uprobe *uprobe; - unsigned long pgoff; int ret; retry: /* Read the page with vaddr into memory */ @@ -251,11 +250,7 @@ retry: vaddr_new = kmap_atomic(new_page); memcpy(vaddr_new, vaddr_old, PAGE_SIZE); - - /* poke the new insn in, ASSUMES we don't cross page boundary */ - pgoff = (vaddr & ~PAGE_MASK); - BUG_ON(pgoff + UPROBE_SWBP_INSN_SIZE > PAGE_SIZE); - memcpy(vaddr_new + pgoff, &opcode, UPROBE_SWBP_INSN_SIZE); + memcpy(vaddr_new + (vaddr & ~PAGE_MASK), &opcode, UPROBE_SWBP_INSN_SIZE); kunmap_atomic(vaddr_new); kunmap_atomic(vaddr_old); @@ -699,6 +694,10 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, if (ret) return ret; + /* write_opcode() assumes we don't cross page boundary */ + BUG_ON((uprobe->offset & ~PAGE_MASK) + + UPROBE_SWBP_INSN_SIZE > PAGE_SIZE); + uprobe->flags |= UPROBE_COPY_INSN; } -- cgit v1.2.3 From 449d0d7c9fb87277175db34c009c70cb348004a8 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:53 +0200 Subject: uprobes: Simplify the usage of uprobe->pending_list uprobe->pending_list is only used to create the temporary list, it has no meaning after we drop uprobes_mmap_hash(inode). No need to initialize this node or remove it from tmp_list, and we can use list_for_each_entry(). Signed-off-by: Oleg Nesterov Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120615154353.GA9614@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index b9c61bda9029..7d5c78f063ae 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -513,7 +513,6 @@ static struct uprobe *alloc_uprobe(struct inode *inode, loff_t offset) uprobe->inode = igrab(inode); uprobe->offset = offset; init_rwsem(&uprobe->consumer_rwsem); - INIT_LIST_HEAD(&uprobe->pending_list); /* add to uprobes_tree, sorted on inode:offset */ cur_uprobe = insert_uprobe(uprobe); @@ -1037,7 +1036,7 @@ static void build_probe_list(struct inode *inode, struct list_head *head) int uprobe_mmap(struct vm_area_struct *vma) { struct list_head tmp_list; - struct uprobe *uprobe, *u; + struct uprobe *uprobe; struct inode *inode; int ret, count; @@ -1055,10 +1054,9 @@ int uprobe_mmap(struct vm_area_struct *vma) ret = 0; count = 0; - list_for_each_entry_safe(uprobe, u, &tmp_list, pending_list) { + list_for_each_entry(uprobe, &tmp_list, pending_list) { loff_t vaddr; - list_del(&uprobe->pending_list); if (!ret) { vaddr = vma_address(vma, uprobe->offset); @@ -1106,7 +1104,7 @@ int uprobe_mmap(struct vm_area_struct *vma) void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned long end) { struct list_head tmp_list; - struct uprobe *uprobe, *u; + struct uprobe *uprobe; struct inode *inode; if (!atomic_read(&uprobe_events) || !valid_vma(vma, false)) @@ -1123,12 +1121,10 @@ void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned lon mutex_lock(uprobes_mmap_hash(inode)); build_probe_list(inode, &tmp_list); - list_for_each_entry_safe(uprobe, u, &tmp_list, pending_list) { + list_for_each_entry(uprobe, &tmp_list, pending_list) { loff_t vaddr; - list_del(&uprobe->pending_list); vaddr = vma_address(vma, uprobe->offset); - if (vaddr >= start && vaddr < end) { /* * An unregister could have removed the probe before -- cgit v1.2.3 From 816c03fbabe64fa09f66fbb64e932081af381415 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:55 +0200 Subject: uprobes: Don't use loff_t for the valid virtual address loff_t looks confusing when it is used for the virtual address. Change map_info and install_breakpoint/remove_breakpoint paths to use "unsigned long". The patch doesn't change vma_address(), it can't return "long" because it is used to verify the mapping. But probably this needs some cleanups too. Signed-off-by: Oleg Nesterov Signed-off-by: Anton Arapov Acked-by: Srikar Dronamraju Acked-by: Ananth N Mavinakayanahalli Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120615154355.GA9622@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 26 +++++++++----------------- 1 file changed, 9 insertions(+), 17 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 7d5c78f063ae..4df84b76dd48 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -664,9 +664,8 @@ static int copy_insn(struct uprobe *uprobe, struct file *filp) */ static int install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, - struct vm_area_struct *vma, loff_t vaddr) + struct vm_area_struct *vma, unsigned long vaddr) { - unsigned long addr; int ret; /* @@ -679,8 +678,6 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, if (!uprobe->consumers) return -EEXIST; - addr = (unsigned long)vaddr; - if (!(uprobe->flags & UPROBE_COPY_INSN)) { ret = copy_insn(uprobe, vma->vm_file); if (ret) @@ -689,7 +686,7 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, if (is_swbp_insn((uprobe_opcode_t *)uprobe->arch.insn)) return -ENOTSUPP; - ret = arch_uprobe_analyze_insn(&uprobe->arch, mm, addr); + ret = arch_uprobe_analyze_insn(&uprobe->arch, mm, vaddr); if (ret) return ret; @@ -709,7 +706,7 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, * Hence increment before and decrement on failure. */ atomic_inc(&mm->uprobes_state.count); - ret = set_swbp(&uprobe->arch, mm, addr); + ret = set_swbp(&uprobe->arch, mm, vaddr); if (ret) atomic_dec(&mm->uprobes_state.count); @@ -717,9 +714,9 @@ install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, } static void -remove_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, loff_t vaddr) +remove_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, unsigned long vaddr) { - if (!set_orig_insn(&uprobe->arch, mm, (unsigned long)vaddr, true)) + if (!set_orig_insn(&uprobe->arch, mm, vaddr, true)) atomic_dec(&mm->uprobes_state.count); } @@ -743,7 +740,7 @@ static void delete_uprobe(struct uprobe *uprobe) struct map_info { struct map_info *next; struct mm_struct *mm; - loff_t vaddr; + unsigned long vaddr; }; static inline struct map_info *free_map_info(struct map_info *info) @@ -837,7 +834,6 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register) while (info) { struct mm_struct *mm = info->mm; struct vm_area_struct *vma; - loff_t vaddr; if (err) goto free; @@ -847,9 +843,8 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register) if (!vma || !valid_vma(vma, is_register)) goto unlock; - vaddr = vma_address(vma, uprobe->offset); if (vma->vm_file->f_mapping->host != uprobe->inode || - vaddr != info->vaddr) + vma_address(vma, uprobe->offset) != info->vaddr) goto unlock; if (is_register) { @@ -1055,10 +1050,8 @@ int uprobe_mmap(struct vm_area_struct *vma) count = 0; list_for_each_entry(uprobe, &tmp_list, pending_list) { - loff_t vaddr; - if (!ret) { - vaddr = vma_address(vma, uprobe->offset); + loff_t vaddr = vma_address(vma, uprobe->offset); if (vaddr < vma->vm_start || vaddr >= vma->vm_end) { put_uprobe(uprobe); @@ -1122,9 +1115,8 @@ void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned lon build_probe_list(inode, &tmp_list); list_for_each_entry(uprobe, &tmp_list, pending_list) { - loff_t vaddr; + loff_t vaddr = vma_address(vma, uprobe->offset); - vaddr = vma_address(vma, uprobe->offset); if (vaddr >= start && vaddr < end) { /* * An unregister could have removed the probe before -- cgit v1.2.3 From 593609a59600c8377f311b300f14deacb155b9a4 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:43:59 +0200 Subject: uprobes: __copy_insn() needs "loff_t offset" 1. __copy_insn() needs "loff_t offset", not "unsigned long", to read the file. 2. use pgoff_t for "idx" and remove the unnecessary typecast. 3. fix the typo, "&=" is not what we want 4. can't resist, rename off1 to off. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Acked-by: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120615154359.GA9625@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 4df84b76dd48..d1b2eeb80837 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -581,12 +581,12 @@ static bool consumer_del(struct uprobe *uprobe, struct uprobe_consumer *uc) static int __copy_insn(struct address_space *mapping, struct file *filp, char *insn, - unsigned long nbytes, unsigned long offset) + unsigned long nbytes, loff_t offset) { struct page *page; void *vaddr; - unsigned long off1; - unsigned long idx; + unsigned long off; + pgoff_t idx; if (!filp) return -EINVAL; @@ -594,8 +594,8 @@ __copy_insn(struct address_space *mapping, struct file *filp, char *insn, if (!mapping->a_ops->readpage) return -EIO; - idx = (unsigned long)(offset >> PAGE_CACHE_SHIFT); - off1 = offset &= ~PAGE_MASK; + idx = offset >> PAGE_CACHE_SHIFT; + off = offset & ~PAGE_MASK; /* * Ensure that the page that has the original instruction is @@ -606,7 +606,7 @@ __copy_insn(struct address_space *mapping, struct file *filp, char *insn, return PTR_ERR(page); vaddr = kmap_atomic(page); - memcpy(insn, vaddr + off1, nbytes); + memcpy(insn, vaddr + off, nbytes); kunmap_atomic(vaddr); page_cache_release(page); -- cgit v1.2.3 From e227051b13956b8f71c0abecc41ad351e64671c8 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 15 Jun 2012 17:44:01 +0200 Subject: uprobes: Remove the unnecessary initialization in add_utask() Trivial cleanup. No need to nullify ->active_uprobe after kzalloc(). Signed-off-by: Oleg Nesterov Cc: Ananth N Mavinakayanahalli Cc: Anton Arapov Cc: Peter Zijlstra Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120615154401.GA9633@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 1 - 1 file changed, 1 deletion(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index d1b2eeb80837..f93532748bca 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -1392,7 +1392,6 @@ static struct uprobe_task *add_utask(void) if (unlikely(!utask)) return NULL; - utask->active_uprobe = NULL; current->utask = utask; return utask; } -- cgit v1.2.3 From aed98048bd1c83469d96932c1901e867d9ba519a Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Sun, 3 Jun 2012 22:04:39 -0700 Subject: irqdomain: Make ops->map hook optional There isn't a really compelling reason to force ->map to be populated, so allow it to be left unset. Signed-off-by: Grant Likely Acked-by: Benjamin Herrenschmidt Cc: Paul Mundt Cc: Thomas Gleixner Cc: Rob Herring --- kernel/irq/irqdomain.c | 14 ++++---------- 1 file changed, 4 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index b1f774cfd089..d3968e91bfd2 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -205,7 +205,8 @@ struct irq_domain *irq_domain_add_legacy(struct device_node *of_node, * one can then use irq_create_mapping() to * explicitly change them */ - ops->map(domain, irq, hwirq); + if (ops->map) + ops->map(domain, irq, hwirq); /* Clear norequest flags */ irq_clear_status_flags(irq, IRQ_NOREQUEST); @@ -340,8 +341,8 @@ static int irq_setup_virq(struct irq_domain *domain, unsigned int virq, irq_data->hwirq = hwirq; irq_data->domain = domain; - if (domain->ops->map(domain, virq, hwirq)) { - pr_debug("irq-%i==>hwirq-0x%lx mapping failed\n", virq, hwirq); + if (domain->ops->map && domain->ops->map(domain, virq, hwirq)) { + pr_err("irq-%i==>hwirq-0x%lx mapping failed\n", virq, hwirq); irq_data->domain = NULL; irq_data->hwirq = 0; return -1; @@ -763,12 +764,6 @@ static int __init irq_debugfs_init(void) __initcall(irq_debugfs_init); #endif /* CONFIG_IRQ_DOMAIN_DEBUG */ -static int irq_domain_simple_map(struct irq_domain *d, unsigned int irq, - irq_hw_number_t hwirq) -{ - return 0; -} - /** * irq_domain_xlate_onecell() - Generic xlate for direct one cell bindings * @@ -831,7 +826,6 @@ int irq_domain_xlate_onetwocell(struct irq_domain *d, EXPORT_SYMBOL_GPL(irq_domain_xlate_onetwocell); const struct irq_domain_ops irq_domain_simple_ops = { - .map = irq_domain_simple_map, .xlate = irq_domain_xlate_onetwocell, }; EXPORT_SYMBOL_GPL(irq_domain_simple_ops); -- cgit v1.2.3 From fbfc623f8231c8d8c78aab5841e9c6e5811ab638 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Fri, 15 Jun 2012 14:31:31 +0800 Subject: perf: Avoid race between cpu hotplug and installing event perf_event_open() requires the cpu on which to install event is online, but the cpu can go offline after perf_event_open checks that. Add a get_online_cpus()/put_online_cpus() pair to avoid the race. Signed-off-by: Zheng Yan Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1339741902-8449-3-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar --- kernel/events/core.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index d7d71d6ec972..31d182e01549 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -6252,6 +6252,8 @@ SYSCALL_DEFINE5(perf_event_open, } } + get_online_cpus(); + event = perf_event_alloc(&attr, cpu, task, group_leader, NULL, NULL, NULL); if (IS_ERR(event)) { @@ -6391,6 +6393,8 @@ SYSCALL_DEFINE5(perf_event_open, perf_unpin_context(ctx); mutex_unlock(&ctx->mutex); + put_online_cpus(); + event->owner = current; mutex_lock(¤t->perf_event_mutex); @@ -6419,6 +6423,7 @@ err_context: err_alloc: free_event(event); err_task: + put_online_cpus(); if (task) put_task_struct(task); err_group_fd: -- cgit v1.2.3 From e2d37cd213dcc0aeb3db4b37b9bd1710fe36fbf7 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Fri, 15 Jun 2012 14:31:32 +0800 Subject: perf: Allow the PMU driver to choose the CPU on which to install events Allow the pmu->event_init callback to change event->cpu, so the PMU driver can choose the CPU on which to install events. Signed-off-by: Zheng Yan Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1339741902-8449-4-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar --- kernel/events/core.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/events/core.c b/kernel/events/core.c index 31d182e01549..fa36a39e8bb7 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -6306,7 +6306,7 @@ SYSCALL_DEFINE5(perf_event_open, /* * Get the target context (task or percpu): */ - ctx = find_get_context(pmu, task, cpu); + ctx = find_get_context(pmu, task, event->cpu); if (IS_ERR(ctx)) { err = PTR_ERR(ctx); goto err_alloc; @@ -6379,16 +6379,16 @@ SYSCALL_DEFINE5(perf_event_open, mutex_lock(&ctx->mutex); if (move_group) { - perf_install_in_context(ctx, group_leader, cpu); + perf_install_in_context(ctx, group_leader, event->cpu); get_ctx(ctx); list_for_each_entry(sibling, &group_leader->sibling_list, group_entry) { - perf_install_in_context(ctx, sibling, cpu); + perf_install_in_context(ctx, sibling, event->cpu); get_ctx(ctx); } } - perf_install_in_context(ctx, event, cpu); + perf_install_in_context(ctx, event, event->cpu); ++ctx->generation; perf_unpin_context(ctx); mutex_unlock(&ctx->mutex); -- cgit v1.2.3 From 0cda4c023132aa93f2dd94811061f812e88daf4c Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Fri, 15 Jun 2012 14:31:33 +0800 Subject: perf: Introduce perf_pmu_migrate_context() Originally from Peter Zijlstra. The helper migrates perf events from one cpu to another cpu. Signed-off-by: Zheng Yan Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1339741902-8449-5-git-send-email-zheng.z.yan@intel.com Signed-off-by: Ingo Molnar --- include/linux/perf_event.h | 2 ++ kernel/events/core.c | 36 ++++++++++++++++++++++++++++++++++++ 2 files changed, 38 insertions(+) (limited to 'kernel') diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 1ce887abcc5c..76c5c8b724a7 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1107,6 +1107,8 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, struct task_struct *task, perf_overflow_handler_t callback, void *context); +extern void perf_pmu_migrate_context(struct pmu *pmu, + int src_cpu, int dst_cpu); extern u64 perf_event_read_value(struct perf_event *event, u64 *enabled, u64 *running); diff --git a/kernel/events/core.c b/kernel/events/core.c index fa36a39e8bb7..f1cf0edeb39a 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1645,6 +1645,8 @@ perf_install_in_context(struct perf_event_context *ctx, lockdep_assert_held(&ctx->mutex); event->ctx = ctx; + if (event->cpu != -1) + event->cpu = cpu; if (!task) { /* @@ -6379,6 +6381,7 @@ SYSCALL_DEFINE5(perf_event_open, mutex_lock(&ctx->mutex); if (move_group) { + synchronize_rcu(); perf_install_in_context(ctx, group_leader, event->cpu); get_ctx(ctx); list_for_each_entry(sibling, &group_leader->sibling_list, @@ -6484,6 +6487,39 @@ err: } EXPORT_SYMBOL_GPL(perf_event_create_kernel_counter); +void perf_pmu_migrate_context(struct pmu *pmu, int src_cpu, int dst_cpu) +{ + struct perf_event_context *src_ctx; + struct perf_event_context *dst_ctx; + struct perf_event *event, *tmp; + LIST_HEAD(events); + + src_ctx = &per_cpu_ptr(pmu->pmu_cpu_context, src_cpu)->ctx; + dst_ctx = &per_cpu_ptr(pmu->pmu_cpu_context, dst_cpu)->ctx; + + mutex_lock(&src_ctx->mutex); + list_for_each_entry_safe(event, tmp, &src_ctx->event_list, + event_entry) { + perf_remove_from_context(event); + put_ctx(src_ctx); + list_add(&event->event_entry, &events); + } + mutex_unlock(&src_ctx->mutex); + + synchronize_rcu(); + + mutex_lock(&dst_ctx->mutex); + list_for_each_entry_safe(event, tmp, &events, event_entry) { + list_del(&event->event_entry); + if (event->state >= PERF_EVENT_STATE_OFF) + event->state = PERF_EVENT_STATE_INACTIVE; + perf_install_in_context(dst_ctx, event, dst_cpu); + get_ctx(dst_ctx); + } + mutex_unlock(&dst_ctx->mutex); +} +EXPORT_SYMBOL_GPL(perf_pmu_migrate_context); + static void sync_child_event(struct perf_event *child_event, struct task_struct *child) { -- cgit v1.2.3 From 6648bd7e0e62c0c8c03b15e00c9e7015e232feff Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 21 Jun 2012 13:58:31 +0000 Subject: ipv4: Add sysctl knob to control early socket demux This change is meant to add a control for disabling early socket demux. The main motivation behind this patch is to provide an option to disable the feature as it adds an additional cost to routing that reduces overall throughput by up to 5%. For example one of my systems went from 12.1Mpps to 11.6 after the early socket demux was added. It looks like the reason for the regression is that we are now having to perform two lookups, first the one for an established socket, and then the one for the routing table. By adding this patch and toggling the value for ip_early_demux to 0 I am able to get back to the 12.1Mpps I was previously seeing. [ Move local variables in ip_rcv_finish() down into the basic block in which they are actually used. -DaveM ] Signed-off-by: Alexander Duyck Signed-off-by: David S. Miller --- include/linux/sysctl.h | 1 + include/net/ip.h | 3 +++ kernel/sysctl_binary.c | 2 ++ net/ipv4/ip_input.c | 22 +++++++++++++--------- net/ipv4/sysctl_net_ipv4.c | 7 +++++++ 5 files changed, 26 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index c34b4c82b0dc..20825e5f433f 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -425,6 +425,7 @@ enum NET_TCP_ALLOWED_CONG_CONTROL=123, NET_TCP_MAX_SSTHRESH=124, NET_TCP_FRTO_RESPONSE=125, + NET_IPV4_EARLY_DEMUX=126, }; enum { diff --git a/include/net/ip.h b/include/net/ip.h index 83e0619f59d0..50841bd6f10e 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -210,6 +210,9 @@ extern int inet_peer_threshold; extern int inet_peer_minttl; extern int inet_peer_maxttl; +/* From ip_input.c */ +extern int sysctl_ip_early_demux; + /* From ip_output.c */ extern int sysctl_ip_dynaddr; diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index a650694883a1..6a3cf8253aae 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -415,6 +415,8 @@ static const struct bin_table bin_net_ipv4_table[] = { { CTL_INT, NET_IPV4_IPFRAG_SECRET_INTERVAL, "ipfrag_secret_interval" }, /* NET_IPV4_IPFRAG_MAX_DIST "ipfrag_max_dist" no longer used */ + { CTL_INT, NET_IPV4_EARLY_DEMUX, "ip_early_demux" }, + { CTL_INT, 2088 /* NET_IPQ_QMAX */, "ip_queue_maxlen" }, /* NET_TCP_DEFAULT_WIN_SCALE unused */ diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 93b092c9a394..bca25179cdb9 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -313,6 +313,8 @@ drop: return true; } +int sysctl_ip_early_demux __read_mostly = 1; + static int ip_rcv_finish(struct sk_buff *skb) { const struct iphdr *iph = ip_hdr(skb); @@ -323,16 +325,18 @@ static int ip_rcv_finish(struct sk_buff *skb) * how the packet travels inside Linux networking. */ if (skb_dst(skb) == NULL) { - const struct net_protocol *ipprot; - int protocol = iph->protocol; - int err; + int err = -ENOENT; - rcu_read_lock(); - ipprot = rcu_dereference(inet_protos[protocol]); - err = -ENOENT; - if (ipprot && ipprot->early_demux) - err = ipprot->early_demux(skb); - rcu_read_unlock(); + if (sysctl_ip_early_demux) { + const struct net_protocol *ipprot; + int protocol = iph->protocol; + + rcu_read_lock(); + ipprot = rcu_dereference(inet_protos[protocol]); + if (ipprot && ipprot->early_demux) + err = ipprot->early_demux(skb); + rcu_read_unlock(); + } if (err) { err = ip_route_input_noref(skb, iph->daddr, iph->saddr, diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index ef32956ed655..12aa0c5867c4 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -300,6 +300,13 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "ip_early_demux", + .data = &sysctl_ip_early_demux, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { .procname = "ip_dynaddr", .data = &sysctl_ip_dynaddr, -- cgit v1.2.3 From dfbce08c19cba2ba4faaf8c0dd6d7678f46c78dd Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 22 Jun 2012 23:02:22 -0700 Subject: ipv4: Don't add deprecated new binary sysctl value. Reported-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/sysctl.h | 1 - kernel/sysctl_binary.c | 2 -- 2 files changed, 3 deletions(-) (limited to 'kernel') diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 20825e5f433f..c34b4c82b0dc 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -425,7 +425,6 @@ enum NET_TCP_ALLOWED_CONG_CONTROL=123, NET_TCP_MAX_SSTHRESH=124, NET_TCP_FRTO_RESPONSE=125, - NET_IPV4_EARLY_DEMUX=126, }; enum { diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index 6a3cf8253aae..a650694883a1 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -415,8 +415,6 @@ static const struct bin_table bin_net_ipv4_table[] = { { CTL_INT, NET_IPV4_IPFRAG_SECRET_INTERVAL, "ipfrag_secret_interval" }, /* NET_IPV4_IPFRAG_MAX_DIST "ipfrag_max_dist" no longer used */ - { CTL_INT, NET_IPV4_EARLY_DEMUX, "ip_early_demux" }, - { CTL_INT, 2088 /* NET_IPQ_QMAX */, "ip_queue_maxlen" }, /* NET_TCP_DEFAULT_WIN_SCALE unused */ -- cgit v1.2.3 From c64e66c67b574f25a048886807c2007d17d50d0a Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 26 Jun 2012 21:45:21 -0700 Subject: audit: netlink: Move away from NLMSG_NEW(). And use nlmsg_data() while we're here too. Signed-off-by: David S. Miller --- kernel/audit.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/kernel/audit.c b/kernel/audit.c index 1c7f2c61416b..30b252a1fb61 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -384,7 +384,7 @@ static void audit_hold_skb(struct sk_buff *skb) static void audit_printk_skb(struct sk_buff *skb) { struct nlmsghdr *nlh = nlmsg_hdr(skb); - char *data = NLMSG_DATA(nlh); + char *data = nlmsg_data(nlh); if (nlh->nlmsg_type != AUDIT_EOE) { if (printk_ratelimit()) @@ -516,14 +516,15 @@ struct sk_buff *audit_make_reply(int pid, int seq, int type, int done, if (!skb) return NULL; - nlh = NLMSG_NEW(skb, pid, seq, t, size, flags); - data = NLMSG_DATA(nlh); + nlh = nlmsg_put(skb, pid, seq, t, size, flags); + if (!nlh) + goto out_kfree_skb; + data = nlmsg_data(nlh); memcpy(data, payload, size); return skb; -nlmsg_failure: /* Used by NLMSG_NEW */ - if (skb) - kfree_skb(skb); +out_kfree_skb: + kfree_skb(skb); return NULL; } @@ -680,7 +681,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) sessionid = audit_get_sessionid(current); security_task_getsecid(current, &sid); seq = nlh->nlmsg_seq; - data = NLMSG_DATA(nlh); + data = nlmsg_data(nlh); switch (msg_type) { case AUDIT_GET: @@ -1060,13 +1061,15 @@ static struct audit_buffer * audit_buffer_alloc(struct audit_context *ctx, ab->skb = nlmsg_new(AUDIT_BUFSIZ, gfp_mask); if (!ab->skb) - goto nlmsg_failure; + goto err; - nlh = NLMSG_NEW(ab->skb, 0, 0, type, 0, 0); + nlh = nlmsg_put(ab->skb, 0, 0, type, 0, 0); + if (!nlh) + goto out_kfree_skb; return ab; -nlmsg_failure: /* Used by NLMSG_NEW */ +out_kfree_skb: kfree_skb(ab->skb); ab->skb = NULL; err: -- cgit v1.2.3 From 0be61ebc18b919dddbdbcd1c4f42513c310ecf59 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 18 Jun 2012 09:28:16 -0400 Subject: tracing/selftest: Add a WARN_ON() if a tracer test fails Add a WARN_ON() output on test failures so that they are easier to detect in automated tests. Although, the WARN_ON() will not print if the test causes the system to crash, obviously. Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 49249c28690d..748f6401edf6 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -830,6 +830,8 @@ int register_tracer(struct tracer *type) current_trace = saved_tracer; if (ret) { printk(KERN_CONT "FAILED!\n"); + /* Add the warning after printing 'FAILED' */ + WARN_ON(1); goto out; } /* Only reset on passing, to avoid touching corrupted buffers */ -- cgit v1.2.3 From 6d158a813efcd09661c23f16ddf7e2ff834cb20c Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 27 Jun 2012 20:46:14 -0400 Subject: tracing: Remove NR_CPUS array from trace_iterator Replace the NR_CPUS array of buffer_iter from the trace_iterator with an allocated array. This will just create an array of possible CPUS instead of the max number specified. The use of NR_CPUS in that array caused allocation failures for machines that were tight on memory. This did not cause any failures to the system itself (no crashes), but caused unnecessary failures for reading the trace files. Added a helper function called 'trace_buffer_iter()' that returns the buffer_iter item or NULL if it is not defined or the array was not allocated. Some routines do not require the array (tracing_open_pipe() for one). Reported-by: Dave Jones Signed-off-by: Steven Rostedt --- include/linux/ftrace_event.h | 2 +- kernel/trace/trace.c | 27 ++++++++++++++++++--------- kernel/trace/trace.h | 8 ++++++++ kernel/trace/trace_functions_graph.c | 2 +- 4 files changed, 28 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index 1aff18346c71..af961d6f7ab1 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -65,7 +65,7 @@ struct trace_iterator { void *private; int cpu_file; struct mutex mutex; - struct ring_buffer_iter *buffer_iter[NR_CPUS]; + struct ring_buffer_iter **buffer_iter; unsigned long iter_flags; /* trace_seq for __print_flags() and __print_symbolic() etc. */ diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 748f6401edf6..b2af14e94c28 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1710,9 +1710,11 @@ EXPORT_SYMBOL_GPL(trace_vprintk); static void trace_iterator_increment(struct trace_iterator *iter) { + struct ring_buffer_iter *buf_iter = trace_buffer_iter(iter, iter->cpu); + iter->idx++; - if (iter->buffer_iter[iter->cpu]) - ring_buffer_read(iter->buffer_iter[iter->cpu], NULL); + if (buf_iter) + ring_buffer_read(buf_iter, NULL); } static struct trace_entry * @@ -1720,7 +1722,7 @@ peek_next_entry(struct trace_iterator *iter, int cpu, u64 *ts, unsigned long *lost_events) { struct ring_buffer_event *event; - struct ring_buffer_iter *buf_iter = iter->buffer_iter[cpu]; + struct ring_buffer_iter *buf_iter = trace_buffer_iter(iter, cpu); if (buf_iter) event = ring_buffer_iter_peek(buf_iter, ts); @@ -1858,10 +1860,10 @@ void tracing_iter_reset(struct trace_iterator *iter, int cpu) tr->data[cpu]->skipped_entries = 0; - if (!iter->buffer_iter[cpu]) + buf_iter = trace_buffer_iter(iter, cpu); + if (!buf_iter) return; - buf_iter = iter->buffer_iter[cpu]; ring_buffer_iter_reset(buf_iter); /* @@ -2207,13 +2209,15 @@ static enum print_line_t print_bin_fmt(struct trace_iterator *iter) int trace_empty(struct trace_iterator *iter) { + struct ring_buffer_iter *buf_iter; int cpu; /* If we are looking at one CPU buffer, only check that one */ if (iter->cpu_file != TRACE_PIPE_ALL_CPU) { cpu = iter->cpu_file; - if (iter->buffer_iter[cpu]) { - if (!ring_buffer_iter_empty(iter->buffer_iter[cpu])) + buf_iter = trace_buffer_iter(iter, cpu); + if (buf_iter) { + if (!ring_buffer_iter_empty(buf_iter)) return 0; } else { if (!ring_buffer_empty_cpu(iter->tr->buffer, cpu)) @@ -2223,8 +2227,9 @@ int trace_empty(struct trace_iterator *iter) } for_each_tracing_cpu(cpu) { - if (iter->buffer_iter[cpu]) { - if (!ring_buffer_iter_empty(iter->buffer_iter[cpu])) + buf_iter = trace_buffer_iter(iter, cpu); + if (buf_iter) { + if (!ring_buffer_iter_empty(buf_iter)) return 0; } else { if (!ring_buffer_empty_cpu(iter->tr->buffer, cpu)) @@ -2383,6 +2388,8 @@ __tracing_open(struct inode *inode, struct file *file) if (!iter) return ERR_PTR(-ENOMEM); + iter->buffer_iter = kzalloc(sizeof(*iter->buffer_iter) * num_possible_cpus(), + GFP_KERNEL); /* * We make a copy of the current tracer to avoid concurrent * changes on it while we are reading. @@ -2443,6 +2450,7 @@ __tracing_open(struct inode *inode, struct file *file) fail: mutex_unlock(&trace_types_lock); kfree(iter->trace); + kfree(iter->buffer_iter); seq_release_private(inode, file); return ERR_PTR(-ENOMEM); } @@ -2483,6 +2491,7 @@ static int tracing_release(struct inode *inode, struct file *file) mutex_destroy(&iter->mutex); free_cpumask_var(iter->started); kfree(iter->trace); + kfree(iter->buffer_iter); seq_release_private(inode, file); return 0; } diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 5aec220d2de0..55e1f7f0db12 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -317,6 +317,14 @@ struct tracer { #define TRACE_PIPE_ALL_CPU -1 +static inline struct ring_buffer_iter * +trace_buffer_iter(struct trace_iterator *iter, int cpu) +{ + if (iter->buffer_iter && iter->buffer_iter[cpu]) + return iter->buffer_iter[cpu]; + return NULL; +} + int tracer_init(struct tracer *t, struct trace_array *tr); int tracing_is_enabled(void); void trace_wake_up(void); diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index a7d2a4c653d8..ce27c8ba8d31 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -538,7 +538,7 @@ get_return_for_leaf(struct trace_iterator *iter, next = &data->ret; } else { - ring_iter = iter->buffer_iter[iter->cpu]; + ring_iter = trace_buffer_iter(iter, iter->cpu); /* First peek to compare current entry and the next one */ if (ring_iter) -- cgit v1.2.3 From a5fb833172eca69136e9ee1ada778e404086ab8a Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Thu, 28 Jun 2012 13:35:04 -0400 Subject: ring-buffer: Fix uninitialized read_stamp The ring buffer reader page is used to swap a page from the writable ring buffer. If the writer happens to be on that page, it ends up on the reader page, but will simply move off of it, back into the writable ring buffer as writes are added. The time stamp passed back to the readers is stored in the cpu_buffer per CPU descriptor. This stamp is updated when a swap of the reader page takes place, and it reads the current stamp from the page taken from the writable ring buffer. Everytime a writer goes to a new page, it updates the time stamp of that page. The problem happens if a reader reads a page from an empty per CPU ring buffer. If the buffer is empty, the swap still takes place, placing the writer at the start of the reader page. If at a later time, a write happens, it updates the page's time stamp and continues. But the problem is that the read_stamp does not get updated, because the page was already swapped. The solution to this was to not swap the page if the ring buffer happens to be empty. This also removes the side effect that the writes on the reader page will not get updated because the writer never gets back on the reader page without a swap. That is, if a read happens on an empty buffer, but then no reads happen for a while. If a swap took place, and the writer were to start writing a lot of data (function tracer), it will start overflowing the ring buffer and overwrite the older data. But because the writer never goes back onto the reader page, the data left on the reader page never gets overwritten. This causes the reader to see really old data, followed by a jump to newer data. Link: http://lkml.kernel.org/r/1340060577-9112-1-git-send-email-dhsharp@google.com Google-Bug-Id: 6410455 Reported-by: David Sharp tested-by: David Sharp Signed-off-by: Steven Rostedt --- kernel/trace/ring_buffer.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 1d0f6a8a0e5e..82a3e0c56b1d 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -3239,6 +3239,10 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) if (cpu_buffer->commit_page == cpu_buffer->reader_page) goto out; + /* Don't bother swapping if the ring buffer is empty */ + if (rb_num_of_entries(cpu_buffer) == 0) + goto out; + /* * Reset the reader page to size zero. */ -- cgit v1.2.3 From a31f2d17b331db970259e875b7223d3aba7e3821 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Fri, 29 Jun 2012 06:15:21 +0000 Subject: netlink: add netlink_kernel_cfg parameter to netlink_kernel_create This patch adds the following structure: struct netlink_kernel_cfg { unsigned int groups; void (*input)(struct sk_buff *skb); struct mutex *cb_mutex; }; That can be passed to netlink_kernel_create to set optional configurations for netlink kernel sockets. I've populated this structure by looking for NULL and zero parameters at the existing code. The remaining parameters that always need to be set are still left in the original interface. That includes optional parameters for the netlink socket creation. This allows easy extensibility of this interface in the future. This patch also adapts all callers to use this new interface. Signed-off-by: Pablo Neira Ayuso Signed-off-by: David S. Miller --- crypto/crypto_user.c | 7 +++++-- drivers/connector/connector.c | 13 +++++++++---- drivers/infiniband/core/netlink.c | 7 +++++-- drivers/scsi/scsi_netlink.c | 7 +++++-- drivers/scsi/scsi_transport_iscsi.c | 9 ++++++--- drivers/staging/gdm72xx/netlink_k.c | 6 ++++-- include/linux/netlink.h | 15 ++++++++++----- kernel/audit.c | 7 +++++-- lib/kobject_uevent.c | 5 ++++- net/bridge/netfilter/ebt_ulog.c | 6 ++++-- net/core/rtnetlink.c | 9 +++++++-- net/core/sock_diag.c | 8 ++++++-- net/decnet/netfilter/dn_rtmsg.c | 8 +++++--- net/ipv4/fib_frontend.c | 7 +++++-- net/ipv4/netfilter/ipt_ULOG.c | 8 +++++--- net/netfilter/nfnetlink.c | 7 +++++-- net/netlink/af_netlink.c | 16 ++++++++++------ net/netlink/genetlink.c | 10 +++++++--- net/xfrm/xfrm_user.c | 7 +++++-- security/selinux/netlink.c | 6 +++++- 20 files changed, 117 insertions(+), 51 deletions(-) (limited to 'kernel') diff --git a/crypto/crypto_user.c b/crypto/crypto_user.c index 5a37eadb4e56..ba2c611154af 100644 --- a/crypto/crypto_user.c +++ b/crypto/crypto_user.c @@ -496,9 +496,12 @@ static void crypto_netlink_rcv(struct sk_buff *skb) static int __init crypto_user_init(void) { + struct netlink_kernel_cfg cfg = { + .input = crypto_netlink_rcv, + }; + crypto_nlsk = netlink_kernel_create(&init_net, NETLINK_CRYPTO, - 0, crypto_netlink_rcv, - NULL, THIS_MODULE); + THIS_MODULE, &cfg); if (!crypto_nlsk) return -ENOMEM; diff --git a/drivers/connector/connector.c b/drivers/connector/connector.c index 34e0e9e4d913..116cf8d02834 100644 --- a/drivers/connector/connector.c +++ b/drivers/connector/connector.c @@ -251,15 +251,20 @@ static const struct file_operations cn_file_ops = { .release = single_release }; +static struct cn_dev cdev = { + .input = cn_rx_skb, +}; + static int __devinit cn_init(void) { struct cn_dev *dev = &cdev; - - dev->input = cn_rx_skb; + struct netlink_kernel_cfg cfg = { + .groups = CN_NETLINK_USERS + 0xf, + .input = dev->input, + }; dev->nls = netlink_kernel_create(&init_net, NETLINK_CONNECTOR, - CN_NETLINK_USERS + 0xf, - dev->input, NULL, THIS_MODULE); + THIS_MODULE, &cfg); if (!dev->nls) return -EIO; diff --git a/drivers/infiniband/core/netlink.c b/drivers/infiniband/core/netlink.c index 1e691dca1820..3ae2bfd31015 100644 --- a/drivers/infiniband/core/netlink.c +++ b/drivers/infiniband/core/netlink.c @@ -173,8 +173,11 @@ static void ibnl_rcv(struct sk_buff *skb) int __init ibnl_init(void) { - nls = netlink_kernel_create(&init_net, NETLINK_RDMA, 0, ibnl_rcv, - NULL, THIS_MODULE); + struct netlink_kernel_cfg cfg = { + .input = ibnl_rcv, + }; + + nls = netlink_kernel_create(&init_net, NETLINK_RDMA, THIS_MODULE, &cfg); if (!nls) { pr_warn("Failed to create netlink socket\n"); return -ENOMEM; diff --git a/drivers/scsi/scsi_netlink.c b/drivers/scsi/scsi_netlink.c index c77628afbf9f..8818dd681c19 100644 --- a/drivers/scsi/scsi_netlink.c +++ b/drivers/scsi/scsi_netlink.c @@ -486,6 +486,10 @@ void scsi_netlink_init(void) { int error; + struct netlink_kernel_cfg cfg = { + .input = scsi_nl_rcv_msg, + .groups = SCSI_NL_GRP_CNT, + }; INIT_LIST_HEAD(&scsi_nl_drivers); @@ -497,8 +501,7 @@ scsi_netlink_init(void) } scsi_nl_sock = netlink_kernel_create(&init_net, NETLINK_SCSITRANSPORT, - SCSI_NL_GRP_CNT, scsi_nl_rcv_msg, NULL, - THIS_MODULE); + THIS_MODULE, &cfg); if (!scsi_nl_sock) { printk(KERN_ERR "%s: register of receive handler failed\n", __func__); diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 1cf640e575da..6042954d8f3b 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -2936,7 +2936,10 @@ EXPORT_SYMBOL_GPL(iscsi_unregister_transport); static __init int iscsi_transport_init(void) { int err; - + struct netlink_kernel_cfg cfg = { + .groups = 1, + .input = iscsi_if_rx, + }; printk(KERN_INFO "Loading iSCSI transport class v%s.\n", ISCSI_TRANSPORT_VERSION); @@ -2966,8 +2969,8 @@ static __init int iscsi_transport_init(void) if (err) goto unregister_conn_class; - nls = netlink_kernel_create(&init_net, NETLINK_ISCSI, 1, iscsi_if_rx, - NULL, THIS_MODULE); + nls = netlink_kernel_create(&init_net, NETLINK_ISCSI, + THIS_MODULE, &cfg); if (!nls) { err = -ENOBUFS; goto unregister_session_class; diff --git a/drivers/staging/gdm72xx/netlink_k.c b/drivers/staging/gdm72xx/netlink_k.c index 2489bb5597ca..87c3a07ed80e 100644 --- a/drivers/staging/gdm72xx/netlink_k.c +++ b/drivers/staging/gdm72xx/netlink_k.c @@ -88,13 +88,15 @@ struct sock *netlink_init(int unit, void (*cb)(struct net_device *dev, u16 type, void *msg, int len)) { struct sock *sock; + struct netlink_kernel_cfg cfg = { + .input = netlink_rcv, + }; #if !defined(DEFINE_MUTEX) init_MUTEX(&netlink_mutex); #endif - sock = netlink_kernel_create(&init_net, unit, 0, netlink_rcv, NULL, - THIS_MODULE); + sock = netlink_kernel_create(&init_net, unit, THIS_MODULE, &cfg); if (sock) rcv_cb = cb; diff --git a/include/linux/netlink.h b/include/linux/netlink.h index ed33f0901bc2..6085e4919cb3 100644 --- a/include/linux/netlink.h +++ b/include/linux/netlink.h @@ -174,11 +174,16 @@ struct netlink_skb_parms { extern void netlink_table_grab(void); extern void netlink_table_ungrab(void); -extern struct sock *netlink_kernel_create(struct net *net, - int unit,unsigned int groups, - void (*input)(struct sk_buff *skb), - struct mutex *cb_mutex, - struct module *module); +/* optional Netlink kernel configuration parameters */ +struct netlink_kernel_cfg { + unsigned int groups; + void (*input)(struct sk_buff *skb); + struct mutex *cb_mutex; +}; + +extern struct sock *netlink_kernel_create(struct net *net, int unit, + struct module *module, + struct netlink_kernel_cfg *cfg); extern void netlink_kernel_release(struct sock *sk); extern int __netlink_change_ngroups(struct sock *sk, unsigned int groups); extern int netlink_change_ngroups(struct sock *sk, unsigned int groups); diff --git a/kernel/audit.c b/kernel/audit.c index 30b252a1fb61..4a3f28d2ca65 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -962,14 +962,17 @@ static void audit_receive(struct sk_buff *skb) static int __init audit_init(void) { int i; + struct netlink_kernel_cfg cfg = { + .input = audit_receive, + }; if (audit_initialized == AUDIT_DISABLED) return 0; printk(KERN_INFO "audit: initializing netlink socket (%s)\n", audit_default ? "enabled" : "disabled"); - audit_sock = netlink_kernel_create(&init_net, NETLINK_AUDIT, 0, - audit_receive, NULL, THIS_MODULE); + audit_sock = netlink_kernel_create(&init_net, NETLINK_AUDIT, + THIS_MODULE, &cfg); if (!audit_sock) audit_panic("cannot initialize netlink socket"); else diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c index 1a91efa6d121..0401d2916d9f 100644 --- a/lib/kobject_uevent.c +++ b/lib/kobject_uevent.c @@ -373,13 +373,16 @@ EXPORT_SYMBOL_GPL(add_uevent_var); static int uevent_net_init(struct net *net) { struct uevent_sock *ue_sk; + struct netlink_kernel_cfg cfg = { + .groups = 1, + }; ue_sk = kzalloc(sizeof(*ue_sk), GFP_KERNEL); if (!ue_sk) return -ENOMEM; ue_sk->sk = netlink_kernel_create(net, NETLINK_KOBJECT_UEVENT, - 1, NULL, NULL, THIS_MODULE); + THIS_MODULE, &cfg); if (!ue_sk->sk) { printk(KERN_ERR "kobject_uevent: unable to create netlink socket!\n"); diff --git a/net/bridge/netfilter/ebt_ulog.c b/net/bridge/netfilter/ebt_ulog.c index 1bd173218f7b..374bdcd77039 100644 --- a/net/bridge/netfilter/ebt_ulog.c +++ b/net/bridge/netfilter/ebt_ulog.c @@ -282,6 +282,9 @@ static int __init ebt_ulog_init(void) { int ret; int i; + struct netlink_kernel_cfg cfg = { + .groups = EBT_ULOG_MAXNLGROUPS, + }; if (nlbufsiz >= 128*1024) { pr_warning("Netlink buffer has to be <= 128kB," @@ -296,8 +299,7 @@ static int __init ebt_ulog_init(void) } ebtulognl = netlink_kernel_create(&init_net, NETLINK_NFLOG, - EBT_ULOG_MAXNLGROUPS, NULL, NULL, - THIS_MODULE); + THIS_MODULE, &cfg); if (!ebtulognl) ret = -ENOMEM; else if ((ret = xt_register_target(&ebt_ulog_tg_reg)) != 0) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index bc8a1cdaac98..2b325c340b44 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -2353,8 +2353,13 @@ static struct notifier_block rtnetlink_dev_notifier = { static int __net_init rtnetlink_net_init(struct net *net) { struct sock *sk; - sk = netlink_kernel_create(net, NETLINK_ROUTE, RTNLGRP_MAX, - rtnetlink_rcv, &rtnl_mutex, THIS_MODULE); + struct netlink_kernel_cfg cfg = { + .groups = RTNLGRP_MAX, + .input = rtnetlink_rcv, + .cb_mutex = &rtnl_mutex, + }; + + sk = netlink_kernel_create(net, NETLINK_ROUTE, THIS_MODULE, &cfg); if (!sk) return -ENOMEM; net->rtnl = sk; diff --git a/net/core/sock_diag.c b/net/core/sock_diag.c index ff2967acbfae..07a29eb34a41 100644 --- a/net/core/sock_diag.c +++ b/net/core/sock_diag.c @@ -171,8 +171,12 @@ EXPORT_SYMBOL_GPL(sock_diag_nlsk); static int __init sock_diag_init(void) { - sock_diag_nlsk = netlink_kernel_create(&init_net, NETLINK_SOCK_DIAG, 0, - sock_diag_rcv, NULL, THIS_MODULE); + struct netlink_kernel_cfg cfg = { + .input = sock_diag_rcv, + }; + + sock_diag_nlsk = netlink_kernel_create(&init_net, NETLINK_SOCK_DIAG, + THIS_MODULE, &cfg); return sock_diag_nlsk == NULL ? -ENOMEM : 0; } diff --git a/net/decnet/netfilter/dn_rtmsg.c b/net/decnet/netfilter/dn_rtmsg.c index b8f7f5b8c350..11db0ecf342f 100644 --- a/net/decnet/netfilter/dn_rtmsg.c +++ b/net/decnet/netfilter/dn_rtmsg.c @@ -125,11 +125,13 @@ static struct nf_hook_ops dnrmg_ops __read_mostly = { static int __init dn_rtmsg_init(void) { int rv = 0; + struct netlink_kernel_cfg cfg = { + .groups = DNRNG_NLGRP_MAX, + .input = dnrmg_receive_user_skb, + }; dnrmg = netlink_kernel_create(&init_net, - NETLINK_DNRTMSG, DNRNG_NLGRP_MAX, - dnrmg_receive_user_skb, - NULL, THIS_MODULE); + NETLINK_DNRTMSG, THIS_MODULE, &cfg); if (dnrmg == NULL) { printk(KERN_ERR "dn_rtmsg: Cannot create netlink socket"); return -ENOMEM; diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index ae528d1b293a..3e11ea225dad 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -976,8 +976,11 @@ static void nl_fib_input(struct sk_buff *skb) static int __net_init nl_fib_lookup_init(struct net *net) { struct sock *sk; - sk = netlink_kernel_create(net, NETLINK_FIB_LOOKUP, 0, - nl_fib_input, NULL, THIS_MODULE); + struct netlink_kernel_cfg cfg = { + .input = nl_fib_input, + }; + + sk = netlink_kernel_create(net, NETLINK_FIB_LOOKUP, THIS_MODULE, &cfg); if (sk == NULL) return -EAFNOSUPPORT; net->ipv4.fibnl = sk; diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c index 99b3f53f16a7..1109f7f6c254 100644 --- a/net/ipv4/netfilter/ipt_ULOG.c +++ b/net/ipv4/netfilter/ipt_ULOG.c @@ -381,6 +381,9 @@ static struct nf_logger ipt_ulog_logger __read_mostly = { static int __init ulog_tg_init(void) { int ret, i; + struct netlink_kernel_cfg cfg = { + .groups = ULOG_MAXNLGROUPS, + }; pr_debug("init module\n"); @@ -393,9 +396,8 @@ static int __init ulog_tg_init(void) for (i = 0; i < ULOG_MAXNLGROUPS; i++) setup_timer(&ulog_buffers[i].timer, ulog_timer, i); - nflognl = netlink_kernel_create(&init_net, - NETLINK_NFLOG, ULOG_MAXNLGROUPS, NULL, - NULL, THIS_MODULE); + nflognl = netlink_kernel_create(&init_net, NETLINK_NFLOG, + THIS_MODULE, &cfg); if (!nflognl) return -ENOMEM; diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 3e797d1fcb94..700e4616a098 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -203,9 +203,12 @@ static void nfnetlink_rcv(struct sk_buff *skb) static int __net_init nfnetlink_net_init(struct net *net) { struct sock *nfnl; + struct netlink_kernel_cfg cfg = { + .groups = NFNLGRP_MAX, + .input = nfnetlink_rcv, + }; - nfnl = netlink_kernel_create(net, NETLINK_NETFILTER, NFNLGRP_MAX, - nfnetlink_rcv, NULL, THIS_MODULE); + nfnl = netlink_kernel_create(net, NETLINK_NETFILTER, THIS_MODULE, &cfg); if (!nfnl) return -ENOMEM; net->nfnl_stash = nfnl; diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index b3025a603d56..43a124feaad8 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1503,14 +1503,16 @@ static void netlink_data_ready(struct sock *sk, int len) */ struct sock * -netlink_kernel_create(struct net *net, int unit, unsigned int groups, - void (*input)(struct sk_buff *skb), - struct mutex *cb_mutex, struct module *module) +netlink_kernel_create(struct net *net, int unit, + struct module *module, + struct netlink_kernel_cfg *cfg) { struct socket *sock; struct sock *sk; struct netlink_sock *nlk; struct listeners *listeners = NULL; + struct mutex *cb_mutex = cfg ? cfg->cb_mutex : NULL; + unsigned int groups; BUG_ON(!nl_table); @@ -1532,16 +1534,18 @@ netlink_kernel_create(struct net *net, int unit, unsigned int groups, sk = sock->sk; sk_change_net(sk, net); - if (groups < 32) + if (!cfg || cfg->groups < 32) groups = 32; + else + groups = cfg->groups; listeners = kzalloc(sizeof(*listeners) + NLGRPSZ(groups), GFP_KERNEL); if (!listeners) goto out_sock_release; sk->sk_data_ready = netlink_data_ready; - if (input) - nlk_sk(sk)->netlink_rcv = input; + if (cfg && cfg->input) + nlk_sk(sk)->netlink_rcv = cfg->input; if (netlink_insert(sk, net, 0)) goto out_sock_release; diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c index 2cc7c1ee7690..32761b53015e 100644 --- a/net/netlink/genetlink.c +++ b/net/netlink/genetlink.c @@ -915,10 +915,14 @@ static struct genl_multicast_group notify_grp = { static int __net_init genl_pernet_init(struct net *net) { + struct netlink_kernel_cfg cfg = { + .input = genl_rcv, + .cb_mutex = &genl_mutex, + }; + /* we'll bump the group number right afterwards */ - net->genl_sock = netlink_kernel_create(net, NETLINK_GENERIC, 0, - genl_rcv, &genl_mutex, - THIS_MODULE); + net->genl_sock = netlink_kernel_create(net, NETLINK_GENERIC, + THIS_MODULE, &cfg); if (!net->genl_sock && net_eq(net, &init_net)) panic("GENL: Cannot initialize generic netlink\n"); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 540762726aaf..e75d8e47f35c 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -2959,9 +2959,12 @@ static struct xfrm_mgr netlink_mgr = { static int __net_init xfrm_user_net_init(struct net *net) { struct sock *nlsk; + struct netlink_kernel_cfg cfg = { + .groups = XFRMNLGRP_MAX, + .input = xfrm_netlink_rcv, + }; - nlsk = netlink_kernel_create(net, NETLINK_XFRM, XFRMNLGRP_MAX, - xfrm_netlink_rcv, NULL, THIS_MODULE); + nlsk = netlink_kernel_create(net, NETLINK_XFRM, THIS_MODULE, &cfg); if (nlsk == NULL) return -ENOMEM; net->xfrm.nlsk_stash = nlsk; /* Don't set to NULL */ diff --git a/security/selinux/netlink.c b/security/selinux/netlink.c index 8a23a35b9c5b..8a77725423e0 100644 --- a/security/selinux/netlink.c +++ b/security/selinux/netlink.c @@ -111,8 +111,12 @@ void selnl_notify_policyload(u32 seqno) static int __init selnl_init(void) { + struct netlink_kernel_cfg cfg = { + .groups = SELNLGRP_MAX, + }; + selnl = netlink_kernel_create(&init_net, NETLINK_SELINUX, - SELNLGRP_MAX, NULL, NULL, THIS_MODULE); + THIS_MODULE, &cfg); if (selnl == NULL) panic("SELinux: Cannot create netlink socket."); netlink_set_nonroot(NETLINK_SELINUX, NL_NONROOT_RECV); -- cgit v1.2.3 From 62c552ccc3eda1198632a4f344aa32623d226bab Mon Sep 17 00:00:00 2001 From: Bojan Smojver Date: Sat, 16 Jun 2012 00:09:58 +0200 Subject: PM / Hibernate: Enable suspend to both for in-kernel hibernation. It is often useful to suspend to memory after hibernation image has been written to disk. If the battery runs out or power is otherwise lost, the computer will resume from the hibernated image. If not, it will resume from memory and hibernation image will be discarded. Signed-off-by: Bojan Smojver Signed-off-by: Rafael J. Wysocki --- Documentation/power/swsusp.txt | 5 +++++ kernel/power/hibernate.c | 36 ++++++++++++++++++++++++++++++++++++ kernel/power/power.h | 3 +++ kernel/power/swap.c | 28 ++++++++++++++++++++++++++++ 4 files changed, 72 insertions(+) (limited to 'kernel') diff --git a/Documentation/power/swsusp.txt b/Documentation/power/swsusp.txt index ac190cf1963e..92341b84250d 100644 --- a/Documentation/power/swsusp.txt +++ b/Documentation/power/swsusp.txt @@ -33,6 +33,11 @@ echo shutdown > /sys/power/disk; echo disk > /sys/power/state echo platform > /sys/power/disk; echo disk > /sys/power/state +. If you would like to write hibernation image to swap and then suspend +to RAM (provided your platform supports it), you can try + +echo suspend > /sys/power/disk; echo disk > /sys/power/state + . If you have SATA disks, you'll need recent kernels with SATA suspend support. For suspend and resume to work, make sure your disk drivers are built into kernel -- not modules. [There's way to make diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 8b53db38a279..21ad3fe3164f 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -5,6 +5,7 @@ * Copyright (c) 2003 Open Source Development Lab * Copyright (c) 2004 Pavel Machek * Copyright (c) 2009 Rafael J. Wysocki, Novell Inc. + * Copyright (C) 2012 Bojan Smojver * * This file is released under the GPLv2. */ @@ -46,6 +47,9 @@ enum { HIBERNATION_PLATFORM, HIBERNATION_SHUTDOWN, HIBERNATION_REBOOT, +#ifdef CONFIG_SUSPEND + HIBERNATION_SUSPEND, +#endif /* keep last */ __HIBERNATION_AFTER_LAST }; @@ -574,6 +578,10 @@ int hibernation_platform_enter(void) */ static void power_down(void) { +#ifdef CONFIG_SUSPEND + int error; +#endif + switch (hibernation_mode) { case HIBERNATION_REBOOT: kernel_restart(NULL); @@ -583,6 +591,25 @@ static void power_down(void) case HIBERNATION_SHUTDOWN: kernel_power_off(); break; +#ifdef CONFIG_SUSPEND + case HIBERNATION_SUSPEND: + error = suspend_devices_and_enter(PM_SUSPEND_MEM); + if (error) { + if (hibernation_ops) + hibernation_mode = HIBERNATION_PLATFORM; + else + hibernation_mode = HIBERNATION_SHUTDOWN; + power_down(); + } + /* + * Restore swap signature. + */ + error = swsusp_unmark(); + if (error) + printk(KERN_ERR "PM: Swap will be unusable! " + "Try swapon -a.\n"); + return; +#endif } kernel_halt(); /* @@ -827,6 +854,9 @@ static const char * const hibernation_modes[] = { [HIBERNATION_PLATFORM] = "platform", [HIBERNATION_SHUTDOWN] = "shutdown", [HIBERNATION_REBOOT] = "reboot", +#ifdef CONFIG_SUSPEND + [HIBERNATION_SUSPEND] = "suspend", +#endif }; /* @@ -867,6 +897,9 @@ static ssize_t disk_show(struct kobject *kobj, struct kobj_attribute *attr, switch (i) { case HIBERNATION_SHUTDOWN: case HIBERNATION_REBOOT: +#ifdef CONFIG_SUSPEND + case HIBERNATION_SUSPEND: +#endif break; case HIBERNATION_PLATFORM: if (hibernation_ops) @@ -907,6 +940,9 @@ static ssize_t disk_store(struct kobject *kobj, struct kobj_attribute *attr, switch (mode) { case HIBERNATION_SHUTDOWN: case HIBERNATION_REBOOT: +#ifdef CONFIG_SUSPEND + case HIBERNATION_SUSPEND: +#endif hibernation_mode = mode; break; case HIBERNATION_PLATFORM: diff --git a/kernel/power/power.h b/kernel/power/power.h index b0bd4beaebfe..7d4b7ffb3c1d 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -156,6 +156,9 @@ extern void swsusp_free(void); extern int swsusp_read(unsigned int *flags_p); extern int swsusp_write(unsigned int flags); extern void swsusp_close(fmode_t); +#ifdef CONFIG_SUSPEND +extern int swsusp_unmark(void); +#endif /* kernel/power/block_io.c */ extern struct block_device *hib_resume_bdev; diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 11e22c068e8b..83d505142b00 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -1472,6 +1472,34 @@ void swsusp_close(fmode_t mode) blkdev_put(hib_resume_bdev, mode); } +/** + * swsusp_unmark - Unmark swsusp signature in the resume device + */ + +#ifdef CONFIG_SUSPEND +int swsusp_unmark(void) +{ + int error; + + hib_bio_read_page(swsusp_resume_block, swsusp_header, NULL); + if (!memcmp(HIBERNATE_SIG,swsusp_header->sig, 10)) { + memcpy(swsusp_header->sig,swsusp_header->orig_sig, 10); + error = hib_bio_write_page(swsusp_resume_block, + swsusp_header, NULL); + } else { + printk(KERN_ERR "PM: Cannot find swsusp signature!\n"); + error = -ENODEV; + } + + /* + * We just returned from suspend, we don't need the image any more. + */ + free_all_swap_pages(root_swap); + + return error; +} +#endif + static int swsusp_header_init(void) { swsusp_header = (struct swsusp_header*) __get_free_page(GFP_KERNEL); -- cgit v1.2.3 From 443772d408a25af62498793f6f805ce3c559309a Mon Sep 17 00:00:00 2001 From: "Srivatsa S. Bhat" Date: Sat, 16 Jun 2012 15:30:45 +0200 Subject: ftrace: Disable function tracing during suspend/resume and hibernation, again If function tracing is enabled for some of the low-level suspend/resume functions, it leads to triple fault during resume from suspend, ultimately ending up in a reboot instead of a resume (or a total refusal to come out of suspended state, on some machines). This issue was explained in more detail in commit f42ac38c59e0a03d (ftrace: disable tracing for suspend to ram). However, the changes made by that commit got reverted by commit cbe2f5a6e84eebb (tracing: allow tracing of suspend/resume & hibernation code again). So, unfortunately since things are not yet robust enough to allow tracing of low-level suspend/resume functions, suspend/resume is still broken when ftrace is enabled. So fix this by disabling function tracing during suspend/resume & hibernation. Signed-off-by: Srivatsa S. Bhat Cc: stable@vger.kernel.org Signed-off-by: Rafael J. Wysocki --- kernel/power/hibernate.c | 6 ++++++ kernel/power/suspend.c | 3 +++ 2 files changed, 9 insertions(+) (limited to 'kernel') diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 21ad3fe3164f..0d4b354bc1be 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -358,6 +358,7 @@ int hibernation_snapshot(int platform_mode) } suspend_console(); + ftrace_stop(); pm_restrict_gfp_mask(); error = dpm_suspend(PMSG_FREEZE); @@ -383,6 +384,7 @@ int hibernation_snapshot(int platform_mode) if (error || !in_suspend) pm_restore_gfp_mask(); + ftrace_start(); resume_console(); dpm_complete(msg); @@ -485,6 +487,7 @@ int hibernation_restore(int platform_mode) pm_prepare_console(); suspend_console(); + ftrace_stop(); pm_restrict_gfp_mask(); error = dpm_suspend_start(PMSG_QUIESCE); if (!error) { @@ -492,6 +495,7 @@ int hibernation_restore(int platform_mode) dpm_resume_end(PMSG_RECOVER); } pm_restore_gfp_mask(); + ftrace_start(); resume_console(); pm_restore_console(); return error; @@ -518,6 +522,7 @@ int hibernation_platform_enter(void) entering_platform_hibernation = true; suspend_console(); + ftrace_stop(); error = dpm_suspend_start(PMSG_HIBERNATE); if (error) { if (hibernation_ops->recover) @@ -561,6 +566,7 @@ int hibernation_platform_enter(void) Resume_devices: entering_platform_hibernation = false; dpm_resume_end(PMSG_RESTORE); + ftrace_start(); resume_console(); Close: diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 396d262b8fd0..c8b7446b27df 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include "power.h" @@ -212,6 +213,7 @@ int suspend_devices_and_enter(suspend_state_t state) goto Close; } suspend_console(); + ftrace_stop(); suspend_test_start(); error = dpm_suspend_start(PMSG_SUSPEND); if (error) { @@ -231,6 +233,7 @@ int suspend_devices_and_enter(suspend_state_t state) suspend_test_start(); dpm_resume_end(PMSG_RESUME); suspend_test_finish("resume devices"); + ftrace_start(); resume_console(); Close: if (suspend_ops->end) -- cgit v1.2.3 From 4b7760ba0dd3319f66886ab2335a0fbecdbc808a Mon Sep 17 00:00:00 2001 From: Sameer Nanda Date: Tue, 19 Jun 2012 22:23:33 +0200 Subject: PM / Sleep: add knob for printing device resume times Added a new knob called /sys/power/pm_print_times. Setting it to 1 enables printing of time taken by devices to suspend and resume. Setting it to 0 disables this printing (unless overridden by initcall_debug kernel command line option). Signed-off-by: Sameer Nanda Acked-by: Greg KH Signed-off-by: Rafael J. Wysocki --- Documentation/ABI/testing/sysfs-power | 13 +++++++++++++ drivers/base/power/main.c | 4 ++-- drivers/base/power/power.h | 11 +++++++++++ kernel/power/main.c | 33 +++++++++++++++++++++++++++++++++ 4 files changed, 59 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/Documentation/ABI/testing/sysfs-power b/Documentation/ABI/testing/sysfs-power index 31725ffeeb3a..217772615d02 100644 --- a/Documentation/ABI/testing/sysfs-power +++ b/Documentation/ABI/testing/sysfs-power @@ -231,3 +231,16 @@ Description: Reads from this file return a string consisting of the names of wakeup sources created with the help of /sys/power/wake_lock that are inactive at the moment, separated with spaces. + +What: /sys/power/pm_print_times +Date: May 2012 +Contact: Sameer Nanda +Description: + The /sys/power/pm_print_times file allows user space to + control whether the time taken by devices to suspend and + resume is printed. These prints are useful for hunting down + devices that take too long to suspend or resume. + + Writing a "1" enables this printing while writing a "0" + disables it. The default value is "0". Reading from this file + will display the current value. diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 9cb845e49334..6e4db96958d1 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -166,7 +166,7 @@ static ktime_t initcall_debug_start(struct device *dev) { ktime_t calltime = ktime_set(0, 0); - if (initcall_debug) { + if (pm_print_times) { pr_info("calling %s+ @ %i, parent: %s\n", dev_name(dev), task_pid_nr(current), dev->parent ? dev_name(dev->parent) : "none"); @@ -181,7 +181,7 @@ static void initcall_debug_report(struct device *dev, ktime_t calltime, { ktime_t delta, rettime; - if (initcall_debug) { + if (pm_print_times) { rettime = ktime_get(); delta = ktime_sub(rettime, calltime); pr_info("call %s+ returned %d after %Ld usecs\n", dev_name(dev), diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h index eeb4bff9505c..12c77b7ff8e8 100644 --- a/drivers/base/power/power.h +++ b/drivers/base/power/power.h @@ -85,3 +85,14 @@ static inline int pm_qos_sysfs_add(struct device *dev) { return 0; } static inline void pm_qos_sysfs_remove(struct device *dev) {} #endif + +#ifdef CONFIG_PM_DEBUG + +extern int pm_print_times_enabled; +#define pm_print_times (initcall_debug || pm_print_times_enabled) + +#else /* CONFIG_PM_DEBUG */ + +#define pm_print_times initcall_debug + +#endif /* CONFIG_PM_DEBUG */ diff --git a/kernel/power/main.c b/kernel/power/main.c index 428f8a034e96..7beb3fb3670b 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -132,6 +132,38 @@ static ssize_t pm_test_store(struct kobject *kobj, struct kobj_attribute *attr, } power_attr(pm_test); + +/* + * pm_print_times: print time taken by devices to suspend and resume. + * + * show() returns whether printing of suspend and resume times is enabled. + * store() accepts 0 or 1. 0 disables printing and 1 enables it. + */ +int pm_print_times_enabled; + +static ssize_t pm_print_times_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return sprintf(buf, "%d\n", pm_print_times_enabled); +} + +static ssize_t pm_print_times_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t n) +{ + unsigned long val; + + if (kstrtoul(buf, 10, &val)) + return -EINVAL; + + if (val > 1) + return -EINVAL; + + pm_print_times_enabled = val; + return n; +} + +power_attr(pm_print_times); #endif /* CONFIG_PM_DEBUG */ #ifdef CONFIG_DEBUG_FS @@ -530,6 +562,7 @@ static struct attribute * g[] = { #endif #ifdef CONFIG_PM_DEBUG &pm_test_attr.attr, + &pm_print_times_attr.attr, #endif #endif NULL, -- cgit v1.2.3 From b2df1d4f8b95d9d1e3f064cef02fc5c5116b05cf Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 21 Jun 2012 00:19:33 +0200 Subject: PM / Sleep: Separate printing suspend times from initcall_debug Change the behavior of the newly introduced /sys/power/pm_print_times attribute so that its initial value depends on initcall_debug, but setting it to 0 will cause device suspend/resume times not to be printed, even if initcall_debug has been set. This way, the people who use initcall_debug for reasons other than PM debugging will be able to switch the suspend/resume times printing off, if need be. Signed-off-by: Rafael J. Wysocki Reviewed-by: Srivatsa S. Bhat Acked-by: Greg Kroah-Hartman --- drivers/base/power/main.c | 4 +-- drivers/base/power/power.h | 11 ------- include/linux/suspend.h | 6 ++++ kernel/power/Kconfig | 4 +-- kernel/power/main.c | 76 +++++++++++++++++++++++++++------------------- 5 files changed, 54 insertions(+), 47 deletions(-) (limited to 'kernel') diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 6e4db96958d1..df5f41d2ec95 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -166,7 +166,7 @@ static ktime_t initcall_debug_start(struct device *dev) { ktime_t calltime = ktime_set(0, 0); - if (pm_print_times) { + if (pm_print_times_enabled) { pr_info("calling %s+ @ %i, parent: %s\n", dev_name(dev), task_pid_nr(current), dev->parent ? dev_name(dev->parent) : "none"); @@ -181,7 +181,7 @@ static void initcall_debug_report(struct device *dev, ktime_t calltime, { ktime_t delta, rettime; - if (pm_print_times) { + if (pm_print_times_enabled) { rettime = ktime_get(); delta = ktime_sub(rettime, calltime); pr_info("call %s+ returned %d after %Ld usecs\n", dev_name(dev), diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h index 12c77b7ff8e8..eeb4bff9505c 100644 --- a/drivers/base/power/power.h +++ b/drivers/base/power/power.h @@ -85,14 +85,3 @@ static inline int pm_qos_sysfs_add(struct device *dev) { return 0; } static inline void pm_qos_sysfs_remove(struct device *dev) {} #endif - -#ifdef CONFIG_PM_DEBUG - -extern int pm_print_times_enabled; -#define pm_print_times (initcall_debug || pm_print_times_enabled) - -#else /* CONFIG_PM_DEBUG */ - -#define pm_print_times initcall_debug - -#endif /* CONFIG_PM_DEBUG */ diff --git a/include/linux/suspend.h b/include/linux/suspend.h index cd83059fb592..0c808d7fa579 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -408,6 +408,12 @@ static inline void unlock_system_sleep(void) {} #endif /* !CONFIG_PM_SLEEP */ +#ifdef CONFIG_PM_SLEEP_DEBUG +extern bool pm_print_times_enabled; +#else +#define pm_print_times_enabled (false) +#endif + #ifdef CONFIG_PM_AUTOSLEEP /* kernel/power/autosleep.c */ diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 8f9b4eb974e0..a70518c9d82f 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -175,7 +175,7 @@ config PM_TEST_SUSPEND You probably want to have your system's RTC driver statically linked, ensuring that it's available when this test runs. -config CAN_PM_TRACE +config PM_SLEEP_DEBUG def_bool y depends on PM_DEBUG && PM_SLEEP @@ -196,7 +196,7 @@ config PM_TRACE config PM_TRACE_RTC bool "Suspend/resume event tracing" - depends on CAN_PM_TRACE + depends on PM_SLEEP_DEBUG depends on X86 select PM_TRACE ---help--- diff --git a/kernel/power/main.c b/kernel/power/main.c index 7beb3fb3670b..f458238109cc 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -132,38 +132,6 @@ static ssize_t pm_test_store(struct kobject *kobj, struct kobj_attribute *attr, } power_attr(pm_test); - -/* - * pm_print_times: print time taken by devices to suspend and resume. - * - * show() returns whether printing of suspend and resume times is enabled. - * store() accepts 0 or 1. 0 disables printing and 1 enables it. - */ -int pm_print_times_enabled; - -static ssize_t pm_print_times_show(struct kobject *kobj, - struct kobj_attribute *attr, char *buf) -{ - return sprintf(buf, "%d\n", pm_print_times_enabled); -} - -static ssize_t pm_print_times_store(struct kobject *kobj, - struct kobj_attribute *attr, - const char *buf, size_t n) -{ - unsigned long val; - - if (kstrtoul(buf, 10, &val)) - return -EINVAL; - - if (val > 1) - return -EINVAL; - - pm_print_times_enabled = val; - return n; -} - -power_attr(pm_print_times); #endif /* CONFIG_PM_DEBUG */ #ifdef CONFIG_DEBUG_FS @@ -267,6 +235,47 @@ late_initcall(pm_debugfs_init); #endif /* CONFIG_PM_SLEEP */ +#ifdef CONFIG_PM_SLEEP_DEBUG +/* + * pm_print_times: print time taken by devices to suspend and resume. + * + * show() returns whether printing of suspend and resume times is enabled. + * store() accepts 0 or 1. 0 disables printing and 1 enables it. + */ +bool pm_print_times_enabled; + +static ssize_t pm_print_times_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + return sprintf(buf, "%d\n", pm_print_times_enabled); +} + +static ssize_t pm_print_times_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t n) +{ + unsigned long val; + + if (kstrtoul(buf, 10, &val)) + return -EINVAL; + + if (val > 1) + return -EINVAL; + + pm_print_times_enabled = !!val; + return n; +} + +power_attr(pm_print_times); + +static inline void pm_print_times_init(void) +{ + pm_print_times_enabled = !!initcall_debug; +} +#else /* !CONFIG_PP_SLEEP_DEBUG */ +static inline void pm_print_times_init(void) {} +#endif /* CONFIG_PM_SLEEP_DEBUG */ + struct kobject *power_kobj; /** @@ -562,6 +571,8 @@ static struct attribute * g[] = { #endif #ifdef CONFIG_PM_DEBUG &pm_test_attr.attr, +#endif +#ifdef CONFIG_PM_SLEEP_DEBUG &pm_print_times_attr.attr, #endif #endif @@ -599,6 +610,7 @@ static int __init pm_init(void) error = sysfs_create_group(power_kobj, &attr_group); if (error) return error; + pm_print_times_init(); return pm_autosleep_init(); } -- cgit v1.2.3 From d8150d350408de6fb2b9ee7b7625ae8e2bb7aa4a Mon Sep 17 00:00:00 2001 From: Bojan Smojver Date: Thu, 21 Jun 2012 22:27:24 +0200 Subject: PM / Hibernate: Print hibernation/thaw progress indicator one line at a time. With the introduction of suspend to both into in-kernel hibernation code, dmesg was getting polluted with backspace characters printed as part of image saving progress indicator. This patch introduces printing of progress indicator on image save/load every 10% and one line at a time. As an additional benefit, all other messages emitted by the kernel during hibernation/thaw should now print cleanly as well. Signed-off-by: Bojan Smojver Signed-off-by: Rafael J. Wysocki --- kernel/power/swap.c | 54 ++++++++++++++++++++++++++--------------------------- 1 file changed, 27 insertions(+), 27 deletions(-) (limited to 'kernel') diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 83d505142b00..3c9d764eb0d8 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -448,9 +448,9 @@ static int save_image(struct swap_map_handle *handle, struct timeval start; struct timeval stop; - printk(KERN_INFO "PM: Saving image data pages (%u pages) ... ", + printk(KERN_INFO "PM: Saving image data pages (%u pages)...\n", nr_to_write); - m = nr_to_write / 100; + m = nr_to_write / 10; if (!m) m = 1; nr_pages = 0; @@ -464,7 +464,8 @@ static int save_image(struct swap_map_handle *handle, if (ret) break; if (!(nr_pages % m)) - printk(KERN_CONT "\b\b\b\b%3d%%", nr_pages / m); + printk(KERN_INFO "PM: Image saving progress: %3d%%\n", + nr_pages / m * 10); nr_pages++; } err2 = hib_wait_on_bio_chain(&bio); @@ -472,9 +473,7 @@ static int save_image(struct swap_map_handle *handle, if (!ret) ret = err2; if (!ret) - printk(KERN_CONT "\b\b\b\bdone\n"); - else - printk(KERN_CONT "\n"); + printk(KERN_INFO "PM: Image saving done.\n"); swsusp_show_speed(&start, &stop, nr_to_write, "Wrote"); return ret; } @@ -668,9 +667,9 @@ static int save_image_lzo(struct swap_map_handle *handle, printk(KERN_INFO "PM: Using %u thread(s) for compression.\n" - "PM: Compressing and saving image data (%u pages) ... ", + "PM: Compressing and saving image data (%u pages)...\n", nr_threads, nr_to_write); - m = nr_to_write / 100; + m = nr_to_write / 10; if (!m) m = 1; nr_pages = 0; @@ -690,8 +689,10 @@ static int save_image_lzo(struct swap_map_handle *handle, data_of(*snapshot), PAGE_SIZE); if (!(nr_pages % m)) - printk(KERN_CONT "\b\b\b\b%3d%%", - nr_pages / m); + printk(KERN_INFO + "PM: Image saving progress: " + "%3d%%\n", + nr_pages / m * 10); nr_pages++; } if (!off) @@ -761,11 +762,8 @@ out_finish: do_gettimeofday(&stop); if (!ret) ret = err2; - if (!ret) { - printk(KERN_CONT "\b\b\b\bdone\n"); - } else { - printk(KERN_CONT "\n"); - } + if (!ret) + printk(KERN_INFO "PM: Image saving done.\n"); swsusp_show_speed(&start, &stop, nr_to_write, "Wrote"); out_clean: if (crc) { @@ -973,9 +971,9 @@ static int load_image(struct swap_map_handle *handle, int err2; unsigned nr_pages; - printk(KERN_INFO "PM: Loading image data pages (%u pages) ... ", + printk(KERN_INFO "PM: Loading image data pages (%u pages)...\n", nr_to_read); - m = nr_to_read / 100; + m = nr_to_read / 10; if (!m) m = 1; nr_pages = 0; @@ -993,7 +991,8 @@ static int load_image(struct swap_map_handle *handle, if (ret) break; if (!(nr_pages % m)) - printk("\b\b\b\b%3d%%", nr_pages / m); + printk(KERN_INFO "PM: Image loading progress: %3d%%\n", + nr_pages / m * 10); nr_pages++; } err2 = hib_wait_on_bio_chain(&bio); @@ -1001,12 +1000,11 @@ static int load_image(struct swap_map_handle *handle, if (!ret) ret = err2; if (!ret) { - printk("\b\b\b\bdone\n"); + printk(KERN_INFO "PM: Image loading done.\n"); snapshot_write_finalize(snapshot); if (!snapshot_image_loaded(snapshot)) ret = -ENODATA; - } else - printk("\n"); + } swsusp_show_speed(&start, &stop, nr_to_read, "Read"); return ret; } @@ -1185,9 +1183,9 @@ static int load_image_lzo(struct swap_map_handle *handle, printk(KERN_INFO "PM: Using %u thread(s) for decompression.\n" - "PM: Loading and decompressing image data (%u pages) ... ", + "PM: Loading and decompressing image data (%u pages)...\n", nr_threads, nr_to_read); - m = nr_to_read / 100; + m = nr_to_read / 10; if (!m) m = 1; nr_pages = 0; @@ -1319,7 +1317,10 @@ static int load_image_lzo(struct swap_map_handle *handle, data[thr].unc + off, PAGE_SIZE); if (!(nr_pages % m)) - printk("\b\b\b\b%3d%%", nr_pages / m); + printk(KERN_INFO + "PM: Image loading progress: " + "%3d%%\n", + nr_pages / m * 10); nr_pages++; ret = snapshot_write_next(snapshot); @@ -1344,7 +1345,7 @@ out_finish: } do_gettimeofday(&stop); if (!ret) { - printk("\b\b\b\bdone\n"); + printk(KERN_INFO "PM: Image loading done.\n"); snapshot_write_finalize(snapshot); if (!snapshot_image_loaded(snapshot)) ret = -ENODATA; @@ -1357,8 +1358,7 @@ out_finish: } } } - } else - printk("\n"); + } swsusp_show_speed(&start, &stop, nr_to_read, "Read"); out_clean: for (i = 0; i < ring_size; i++) -- cgit v1.2.3 From f885b7f2b2de70be266d2cecc476f773a1e2ca5d Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 23 Apr 2012 15:52:53 -0700 Subject: rcu: Control RCU_FANOUT_LEAF from boot-time parameter Although making RCU_FANOUT_LEAF a kernel configuration parameter rather than a fixed constant makes it easier for people to decrease cache-miss overhead for large systems, it is of little help for people who must run a single pre-built kernel binary. This commit therefore allows the value of RCU_FANOUT_LEAF to be increased (but not decreased!) via a boot-time parameter named rcutree.rcu_fanout_leaf. Reported-by: Mike Galbraith Signed-off-by: Paul E. McKenney --- Documentation/kernel-parameters.txt | 5 ++ kernel/rcutree.c | 97 +++++++++++++++++++++++++++++++------ kernel/rcutree.h | 23 +++++---- kernel/rcutree_plugin.h | 2 + kernel/rcutree_trace.c | 2 +- 5 files changed, 104 insertions(+), 25 deletions(-) (limited to 'kernel') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index a92c5ebf373e..12783fa833c3 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2367,6 +2367,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted. Set maximum number of finished RCU callbacks to process in one batch. + rcutree.fanout_leaf= [KNL,BOOT] + Increase the number of CPUs assigned to each + leaf rcu_node structure. Useful for very large + systems. + rcutree.qhimark= [KNL,BOOT] Set threshold of queued RCU callbacks over which batch limiting is disabled. diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 4b97bba7396e..a4c592b66e10 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -60,17 +60,10 @@ /* Data structures. */ -static struct lock_class_key rcu_node_class[NUM_RCU_LVLS]; +static struct lock_class_key rcu_node_class[RCU_NUM_LVLS]; #define RCU_STATE_INITIALIZER(structname) { \ .level = { &structname##_state.node[0] }, \ - .levelcnt = { \ - NUM_RCU_LVL_0, /* root of hierarchy. */ \ - NUM_RCU_LVL_1, \ - NUM_RCU_LVL_2, \ - NUM_RCU_LVL_3, \ - NUM_RCU_LVL_4, /* == MAX_RCU_LVLS */ \ - }, \ .fqs_state = RCU_GP_IDLE, \ .gpnum = -300, \ .completed = -300, \ @@ -91,6 +84,19 @@ DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); static struct rcu_state *rcu_state; +/* Increase (but not decrease) the CONFIG_RCU_FANOUT_LEAF at boot time. */ +static int rcu_fanout_leaf = CONFIG_RCU_FANOUT_LEAF; +module_param(rcu_fanout_leaf, int, 0); +int rcu_num_lvls __read_mostly = RCU_NUM_LVLS; +static int num_rcu_lvl[] = { /* Number of rcu_nodes at specified level. */ + NUM_RCU_LVL_0, + NUM_RCU_LVL_1, + NUM_RCU_LVL_2, + NUM_RCU_LVL_3, + NUM_RCU_LVL_4, +}; +int rcu_num_nodes __read_mostly = NUM_RCU_NODES; /* Total # rcu_nodes in use. */ + /* * The rcu_scheduler_active variable transitions from zero to one just * before the first task is spawned. So when this variable is zero, RCU @@ -2574,9 +2580,9 @@ static void __init rcu_init_levelspread(struct rcu_state *rsp) { int i; - for (i = NUM_RCU_LVLS - 1; i > 0; i--) + for (i = rcu_num_lvls - 1; i > 0; i--) rsp->levelspread[i] = CONFIG_RCU_FANOUT; - rsp->levelspread[0] = CONFIG_RCU_FANOUT_LEAF; + rsp->levelspread[0] = rcu_fanout_leaf; } #else /* #ifdef CONFIG_RCU_FANOUT_EXACT */ static void __init rcu_init_levelspread(struct rcu_state *rsp) @@ -2586,7 +2592,7 @@ static void __init rcu_init_levelspread(struct rcu_state *rsp) int i; cprv = NR_CPUS; - for (i = NUM_RCU_LVLS - 1; i >= 0; i--) { + for (i = rcu_num_lvls - 1; i >= 0; i--) { ccur = rsp->levelcnt[i]; rsp->levelspread[i] = (cprv + ccur - 1) / ccur; cprv = ccur; @@ -2613,13 +2619,15 @@ static void __init rcu_init_one(struct rcu_state *rsp, /* Initialize the level-tracking arrays. */ - for (i = 1; i < NUM_RCU_LVLS; i++) + for (i = 0; i < rcu_num_lvls; i++) + rsp->levelcnt[i] = num_rcu_lvl[i]; + for (i = 1; i < rcu_num_lvls; i++) rsp->level[i] = rsp->level[i - 1] + rsp->levelcnt[i - 1]; rcu_init_levelspread(rsp); /* Initialize the elements themselves, starting from the leaves. */ - for (i = NUM_RCU_LVLS - 1; i >= 0; i--) { + for (i = rcu_num_lvls - 1; i >= 0; i--) { cpustride *= rsp->levelspread[i]; rnp = rsp->level[i]; for (j = 0; j < rsp->levelcnt[i]; j++, rnp++) { @@ -2649,7 +2657,7 @@ static void __init rcu_init_one(struct rcu_state *rsp, } rsp->rda = rda; - rnp = rsp->level[NUM_RCU_LVLS - 1]; + rnp = rsp->level[rcu_num_lvls - 1]; for_each_possible_cpu(i) { while (i > rnp->grphi) rnp++; @@ -2658,11 +2666,72 @@ static void __init rcu_init_one(struct rcu_state *rsp, } } +/* + * Compute the rcu_node tree geometry from kernel parameters. This cannot + * replace the definitions in rcutree.h because those are needed to size + * the ->node array in the rcu_state structure. + */ +static void __init rcu_init_geometry(void) +{ + int i; + int j; + int n = NR_CPUS; + int rcu_capacity[MAX_RCU_LVLS + 1]; + + /* If the compile-time values are accurate, just leave. */ + if (rcu_fanout_leaf == CONFIG_RCU_FANOUT_LEAF) + return; + + /* + * Compute number of nodes that can be handled an rcu_node tree + * with the given number of levels. Setting rcu_capacity[0] makes + * some of the arithmetic easier. + */ + rcu_capacity[0] = 1; + rcu_capacity[1] = rcu_fanout_leaf; + for (i = 2; i <= MAX_RCU_LVLS; i++) + rcu_capacity[i] = rcu_capacity[i - 1] * CONFIG_RCU_FANOUT; + + /* + * The boot-time rcu_fanout_leaf parameter is only permitted + * to increase the leaf-level fanout, not decrease it. Of course, + * the leaf-level fanout cannot exceed the number of bits in + * the rcu_node masks. Finally, the tree must be able to accommodate + * the configured number of CPUs. Complain and fall back to the + * compile-time values if these limits are exceeded. + */ + if (rcu_fanout_leaf < CONFIG_RCU_FANOUT_LEAF || + rcu_fanout_leaf > sizeof(unsigned long) * 8 || + n > rcu_capacity[MAX_RCU_LVLS]) { + WARN_ON(1); + return; + } + + /* Calculate the number of rcu_nodes at each level of the tree. */ + for (i = 1; i <= MAX_RCU_LVLS; i++) + if (n <= rcu_capacity[i]) { + for (j = 0; j <= i; j++) + num_rcu_lvl[j] = + DIV_ROUND_UP(n, rcu_capacity[i - j]); + rcu_num_lvls = i; + for (j = i + 1; j <= MAX_RCU_LVLS; j++) + num_rcu_lvl[j] = 0; + break; + } + + /* Calculate the total number of rcu_node structures. */ + rcu_num_nodes = 0; + for (i = 0; i <= MAX_RCU_LVLS; i++) + rcu_num_nodes += num_rcu_lvl[i]; + rcu_num_nodes -= n; +} + void __init rcu_init(void) { int cpu; rcu_bootup_announce(); + rcu_init_geometry(); rcu_init_one(&rcu_sched_state, &rcu_sched_data); rcu_init_one(&rcu_bh_state, &rcu_bh_data); __rcu_init_preempt(); diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 19b61ac1079f..780a0195d35a 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -42,28 +42,28 @@ #define RCU_FANOUT_4 (RCU_FANOUT_3 * CONFIG_RCU_FANOUT) #if NR_CPUS <= RCU_FANOUT_1 -# define NUM_RCU_LVLS 1 +# define RCU_NUM_LVLS 1 # define NUM_RCU_LVL_0 1 # define NUM_RCU_LVL_1 (NR_CPUS) # define NUM_RCU_LVL_2 0 # define NUM_RCU_LVL_3 0 # define NUM_RCU_LVL_4 0 #elif NR_CPUS <= RCU_FANOUT_2 -# define NUM_RCU_LVLS 2 +# define RCU_NUM_LVLS 2 # define NUM_RCU_LVL_0 1 # define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) # define NUM_RCU_LVL_2 (NR_CPUS) # define NUM_RCU_LVL_3 0 # define NUM_RCU_LVL_4 0 #elif NR_CPUS <= RCU_FANOUT_3 -# define NUM_RCU_LVLS 3 +# define RCU_NUM_LVLS 3 # define NUM_RCU_LVL_0 1 # define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2) # define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_1) # define NUM_RCU_LVL_3 (NR_CPUS) # define NUM_RCU_LVL_4 0 #elif NR_CPUS <= RCU_FANOUT_4 -# define NUM_RCU_LVLS 4 +# define RCU_NUM_LVLS 4 # define NUM_RCU_LVL_0 1 # define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_3) # define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_2) @@ -76,6 +76,9 @@ #define RCU_SUM (NUM_RCU_LVL_0 + NUM_RCU_LVL_1 + NUM_RCU_LVL_2 + NUM_RCU_LVL_3 + NUM_RCU_LVL_4) #define NUM_RCU_NODES (RCU_SUM - NR_CPUS) +extern int rcu_num_lvls; +extern int rcu_num_nodes; + /* * Dynticks per-CPU state. */ @@ -206,7 +209,7 @@ struct rcu_node { */ #define rcu_for_each_node_breadth_first(rsp, rnp) \ for ((rnp) = &(rsp)->node[0]; \ - (rnp) < &(rsp)->node[NUM_RCU_NODES]; (rnp)++) + (rnp) < &(rsp)->node[rcu_num_nodes]; (rnp)++) /* * Do a breadth-first scan of the non-leaf rcu_node structures for the @@ -215,7 +218,7 @@ struct rcu_node { */ #define rcu_for_each_nonleaf_node_breadth_first(rsp, rnp) \ for ((rnp) = &(rsp)->node[0]; \ - (rnp) < (rsp)->level[NUM_RCU_LVLS - 1]; (rnp)++) + (rnp) < (rsp)->level[rcu_num_lvls - 1]; (rnp)++) /* * Scan the leaves of the rcu_node hierarchy for the specified rcu_state @@ -224,8 +227,8 @@ struct rcu_node { * It is still a leaf node, even if it is also the root node. */ #define rcu_for_each_leaf_node(rsp, rnp) \ - for ((rnp) = (rsp)->level[NUM_RCU_LVLS - 1]; \ - (rnp) < &(rsp)->node[NUM_RCU_NODES]; (rnp)++) + for ((rnp) = (rsp)->level[rcu_num_lvls - 1]; \ + (rnp) < &(rsp)->node[rcu_num_nodes]; (rnp)++) /* Index values for nxttail array in struct rcu_data. */ #define RCU_DONE_TAIL 0 /* Also RCU_WAIT head. */ @@ -357,9 +360,9 @@ do { \ */ struct rcu_state { struct rcu_node node[NUM_RCU_NODES]; /* Hierarchy. */ - struct rcu_node *level[NUM_RCU_LVLS]; /* Hierarchy levels. */ + struct rcu_node *level[RCU_NUM_LVLS]; /* Hierarchy levels. */ u32 levelcnt[MAX_RCU_LVLS + 1]; /* # nodes in each level. */ - u8 levelspread[NUM_RCU_LVLS]; /* kids/node in each level. */ + u8 levelspread[RCU_NUM_LVLS]; /* kids/node in each level. */ struct rcu_data __percpu *rda; /* pointer of percu rcu_data. */ /* The following fields are guarded by the root rcu_node's lock. */ diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 3e4899459f3d..fb92b6dd9980 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -70,6 +70,8 @@ static void __init rcu_bootup_announce_oddness(void) #if NUM_RCU_LVL_4 != 0 printk(KERN_INFO "\tExperimental four-level hierarchy is enabled.\n"); #endif + if (rcu_fanout_leaf != CONFIG_RCU_FANOUT_LEAF) + printk(KERN_INFO "\tExperimental boot-time adjustment of leaf fanout to %d.\n", rcu_fanout_leaf); } #ifdef CONFIG_TREE_PREEMPT_RCU diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index d4bc16ddd1d4..a3556a2d842c 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -278,7 +278,7 @@ static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) rsp->n_force_qs, rsp->n_force_qs_ngp, rsp->n_force_qs - rsp->n_force_qs_ngp, rsp->n_force_qs_lh, rsp->qlen_lazy, rsp->qlen); - for (rnp = &rsp->node[0]; rnp - &rsp->node[0] < NUM_RCU_NODES; rnp++) { + for (rnp = &rsp->node[0]; rnp - &rsp->node[0] < rcu_num_nodes; rnp++) { if (rnp->level != level) { seq_puts(m, "\n"); level = rnp->level; -- cgit v1.2.3 From cc5df65b0370fc6aa2bfe3bb19e0451d5cafb99f Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 15 Jun 2012 18:16:00 -0700 Subject: rcu: Four-level hierarchy is no longer experimental Time to make the four-level-hierarchy setting less scary, so this commit removes "Experimental" from the boot-time message. Leave the message in order to get a heads-up on any possible need to expand to a five-level hierarchy. Signed-off-by: Paul E. McKenney --- kernel/rcutree_plugin.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index fb92b6dd9980..70e0fd256cc6 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -68,7 +68,7 @@ static void __init rcu_bootup_announce_oddness(void) printk(KERN_INFO "\tAdditional per-CPU info printed with stalls.\n"); #endif #if NUM_RCU_LVL_4 != 0 - printk(KERN_INFO "\tExperimental four-level hierarchy is enabled.\n"); + printk(KERN_INFO "\tFour-level hierarchy is enabled.\n"); #endif if (rcu_fanout_leaf != CONFIG_RCU_FANOUT_LEAF) printk(KERN_INFO "\tExperimental boot-time adjustment of leaf fanout to %d.\n", rcu_fanout_leaf); -- cgit v1.2.3 From cca6f3931920a7547d02e68adc2ca635bea5600c Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 8 May 2012 21:00:28 -0700 Subject: rcu: Size rcu_node tree from nr_cpu_ids rather than NR_CPUS The rcu_node tree array is sized based on compile-time constants, including NR_CPUS. Although this approach has worked well in the past, the recent trend by many distros to define NR_CPUS=4096 results in excessive grace-period-initialization latencies. This commit therefore substitutes the run-time computed nr_cpu_ids for the compile-time NR_CPUS when building the tree. This can result in much of the compile-time-allocated rcu_node array being unused. If this is a major problem, you are in a specialized situation anyway, so you can manually adjust the NR_CPUS, RCU_FANOUT, and RCU_FANOUT_LEAF kernel config parameters. Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 2 +- kernel/rcutree_plugin.h | 2 ++ 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index a4c592b66e10..0fdbc5e07302 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -2675,7 +2675,7 @@ static void __init rcu_init_geometry(void) { int i; int j; - int n = NR_CPUS; + int n = nr_cpu_ids; int rcu_capacity[MAX_RCU_LVLS + 1]; /* If the compile-time values are accurate, just leave. */ diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 70e0fd256cc6..ef2b5231afa4 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -72,6 +72,8 @@ static void __init rcu_bootup_announce_oddness(void) #endif if (rcu_fanout_leaf != CONFIG_RCU_FANOUT_LEAF) printk(KERN_INFO "\tExperimental boot-time adjustment of leaf fanout to %d.\n", rcu_fanout_leaf); + if (nr_cpu_ids != NR_CPUS) + printk(KERN_INFO "\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%d.\n", NR_CPUS, nr_cpu_ids); } #ifdef CONFIG_TREE_PREEMPT_RCU -- cgit v1.2.3 From 6c90cc7bf077f28144013e949ee0c122012d194a Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 28 May 2012 19:41:41 -0700 Subject: rcu: Prevent excessive line length in RCU_STATE_INITIALIZER() Upcoming rcu_barrier() concurrency commits will result in line lengths greater than 80 characters in the RCU_STATE_INITIALIZER(), so this commit shortens the name of the macro's argument to prevent this. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 0fdbc5e07302..dd7fd96c90c5 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -62,18 +62,18 @@ static struct lock_class_key rcu_node_class[RCU_NUM_LVLS]; -#define RCU_STATE_INITIALIZER(structname) { \ - .level = { &structname##_state.node[0] }, \ +#define RCU_STATE_INITIALIZER(sname) { \ + .level = { &sname##_state.node[0] }, \ .fqs_state = RCU_GP_IDLE, \ .gpnum = -300, \ .completed = -300, \ - .onofflock = __RAW_SPIN_LOCK_UNLOCKED(&structname##_state.onofflock), \ - .orphan_nxttail = &structname##_state.orphan_nxtlist, \ - .orphan_donetail = &structname##_state.orphan_donelist, \ - .fqslock = __RAW_SPIN_LOCK_UNLOCKED(&structname##_state.fqslock), \ + .onofflock = __RAW_SPIN_LOCK_UNLOCKED(&sname##_state.onofflock), \ + .orphan_nxttail = &sname##_state.orphan_nxtlist, \ + .orphan_donetail = &sname##_state.orphan_donelist, \ + .fqslock = __RAW_SPIN_LOCK_UNLOCKED(&sname##_state.fqslock), \ .n_force_qs = 0, \ .n_force_qs_ngp = 0, \ - .name = #structname, \ + .name = #sname, \ } struct rcu_state rcu_sched_state = RCU_STATE_INITIALIZER(rcu_sched); -- cgit v1.2.3 From 037b64ed0bf2405a1a01542164d3418564b44fff Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 28 May 2012 23:26:01 -0700 Subject: rcu: Place pointer to call_rcu() in rcu_data structure This is a preparatory commit for increasing rcu_barrier()'s concurrency. It adds a pointer in the rcu_data structure to the corresponding call_rcu() function. This allows a pointer to the rcu_data structure to imply the function pointer, which allows _rcu_barrier() state to be placed in the rcu_state structure. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 27 ++++++++++++--------------- kernel/rcutree.h | 2 ++ kernel/rcutree_plugin.h | 5 +++-- 3 files changed, 17 insertions(+), 17 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index dd7fd96c90c5..00c518fa34bb 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -62,8 +62,9 @@ static struct lock_class_key rcu_node_class[RCU_NUM_LVLS]; -#define RCU_STATE_INITIALIZER(sname) { \ +#define RCU_STATE_INITIALIZER(sname, cr) { \ .level = { &sname##_state.node[0] }, \ + .call = cr, \ .fqs_state = RCU_GP_IDLE, \ .gpnum = -300, \ .completed = -300, \ @@ -76,10 +77,11 @@ static struct lock_class_key rcu_node_class[RCU_NUM_LVLS]; .name = #sname, \ } -struct rcu_state rcu_sched_state = RCU_STATE_INITIALIZER(rcu_sched); +struct rcu_state rcu_sched_state = + RCU_STATE_INITIALIZER(rcu_sched, call_rcu_sched); DEFINE_PER_CPU(struct rcu_data, rcu_sched_data); -struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh); +struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh, call_rcu_bh); DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); static struct rcu_state *rcu_state; @@ -2282,21 +2284,17 @@ static void rcu_barrier_func(void *type) { int cpu = smp_processor_id(); struct rcu_head *head = &per_cpu(rcu_barrier_head, cpu); - void (*call_rcu_func)(struct rcu_head *head, - void (*func)(struct rcu_head *head)); + struct rcu_state *rsp = type; atomic_inc(&rcu_barrier_cpu_count); - call_rcu_func = type; - call_rcu_func(head, rcu_barrier_callback); + rsp->call(head, rcu_barrier_callback); } /* * Orchestrate the specified type of RCU barrier, waiting for all * RCU callbacks of the specified type to complete. */ -static void _rcu_barrier(struct rcu_state *rsp, - void (*call_rcu_func)(struct rcu_head *head, - void (*func)(struct rcu_head *head))) +static void _rcu_barrier(struct rcu_state *rsp) { int cpu; unsigned long flags; @@ -2348,8 +2346,7 @@ static void _rcu_barrier(struct rcu_state *rsp, while (cpu_is_offline(cpu) && ACCESS_ONCE(rdp->qlen)) schedule_timeout_interruptible(1); } else if (ACCESS_ONCE(rdp->qlen)) { - smp_call_function_single(cpu, rcu_barrier_func, - (void *)call_rcu_func, 1); + smp_call_function_single(cpu, rcu_barrier_func, rsp, 1); preempt_enable(); } else { preempt_enable(); @@ -2370,7 +2367,7 @@ static void _rcu_barrier(struct rcu_state *rsp, raw_spin_unlock_irqrestore(&rsp->onofflock, flags); atomic_inc(&rcu_barrier_cpu_count); smp_mb__after_atomic_inc(); /* Ensure atomic_inc() before callback. */ - call_rcu_func(&rh, rcu_barrier_callback); + rsp->call(&rh, rcu_barrier_callback); /* * Now that we have an rcu_barrier_callback() callback on each @@ -2393,7 +2390,7 @@ static void _rcu_barrier(struct rcu_state *rsp, */ void rcu_barrier_bh(void) { - _rcu_barrier(&rcu_bh_state, call_rcu_bh); + _rcu_barrier(&rcu_bh_state); } EXPORT_SYMBOL_GPL(rcu_barrier_bh); @@ -2402,7 +2399,7 @@ EXPORT_SYMBOL_GPL(rcu_barrier_bh); */ void rcu_barrier_sched(void) { - _rcu_barrier(&rcu_sched_state, call_rcu_sched); + _rcu_barrier(&rcu_sched_state); } EXPORT_SYMBOL_GPL(rcu_barrier_sched); diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 780a0195d35a..049896a835d9 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -364,6 +364,8 @@ struct rcu_state { u32 levelcnt[MAX_RCU_LVLS + 1]; /* # nodes in each level. */ u8 levelspread[RCU_NUM_LVLS]; /* kids/node in each level. */ struct rcu_data __percpu *rda; /* pointer of percu rcu_data. */ + void (*call)(struct rcu_head *head, /* call_rcu() flavor. */ + void (*func)(struct rcu_head *head)); /* The following fields are guarded by the root rcu_node's lock. */ diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index ef2b5231afa4..9cb3a68819fa 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -78,7 +78,8 @@ static void __init rcu_bootup_announce_oddness(void) #ifdef CONFIG_TREE_PREEMPT_RCU -struct rcu_state rcu_preempt_state = RCU_STATE_INITIALIZER(rcu_preempt); +struct rcu_state rcu_preempt_state = + RCU_STATE_INITIALIZER(rcu_preempt, call_rcu); DEFINE_PER_CPU(struct rcu_data, rcu_preempt_data); static struct rcu_state *rcu_state = &rcu_preempt_state; @@ -944,7 +945,7 @@ static int rcu_preempt_cpu_has_callbacks(int cpu) */ void rcu_barrier(void) { - _rcu_barrier(&rcu_preempt_state, call_rcu); + _rcu_barrier(&rcu_preempt_state); } EXPORT_SYMBOL_GPL(rcu_barrier); -- cgit v1.2.3 From 06668efa9180f4824fe846a8ff96338c18646bc7 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 28 May 2012 23:57:46 -0700 Subject: rcu: Move _rcu_barrier()'s rcu_head structures to rcu_data structures In order for multiple flavors of RCU to each concurrently run one rcu_barrier(), each flavor needs its own per-CPU set of rcu_head structures. This commit therefore moves _rcu_barrier()'s set of per-CPU rcu_head structures from per-CPU variables to the existing per-CPU and per-RCU-flavor rcu_data structures. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 6 ++---- kernel/rcutree.h | 3 +++ 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 00c518fa34bb..1e552598b55d 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -157,7 +157,6 @@ unsigned long rcutorture_vernum; /* State information for rcu_barrier() and friends. */ -static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL}; static atomic_t rcu_barrier_cpu_count; static DEFINE_MUTEX(rcu_barrier_mutex); static struct completion rcu_barrier_completion; @@ -2282,12 +2281,11 @@ static void rcu_barrier_callback(struct rcu_head *notused) */ static void rcu_barrier_func(void *type) { - int cpu = smp_processor_id(); - struct rcu_head *head = &per_cpu(rcu_barrier_head, cpu); struct rcu_state *rsp = type; + struct rcu_data *rdp = __this_cpu_ptr(rsp->rda); atomic_inc(&rcu_barrier_cpu_count); - rsp->call(head, rcu_barrier_callback); + rsp->call(&rdp->barrier_head, rcu_barrier_callback); } /* diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 049896a835d9..586d93c978f2 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -314,6 +314,9 @@ struct rcu_data { unsigned long n_rp_need_fqs; unsigned long n_rp_need_nothing; + /* 6) _rcu_barrier() callback. */ + struct rcu_head barrier_head; + int cpu; struct rcu_state *rsp; }; -- cgit v1.2.3 From 24ebbca8ecdd5129d7f829a7cb5146aaeb531f77 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 29 May 2012 00:34:56 -0700 Subject: rcu: Move rcu_barrier_cpu_count to rcu_state structure In order to allow each RCU flavor to concurrently execute its rcu_barrier() function, it is necessary to move the relevant state to the rcu_state structure. This commit therefore moves the rcu_barrier_cpu_count global variable to a new ->barrier_cpu_count field in the rcu_state structure. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 25 ++++++++++++++----------- kernel/rcutree.h | 1 + 2 files changed, 15 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 1e552598b55d..5929b021666d 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -157,7 +157,6 @@ unsigned long rcutorture_vernum; /* State information for rcu_barrier() and friends. */ -static atomic_t rcu_barrier_cpu_count; static DEFINE_MUTEX(rcu_barrier_mutex); static struct completion rcu_barrier_completion; @@ -2270,9 +2269,12 @@ static int rcu_cpu_has_callbacks(int cpu) * RCU callback function for _rcu_barrier(). If we are last, wake * up the task executing _rcu_barrier(). */ -static void rcu_barrier_callback(struct rcu_head *notused) +static void rcu_barrier_callback(struct rcu_head *rhp) { - if (atomic_dec_and_test(&rcu_barrier_cpu_count)) + struct rcu_data *rdp = container_of(rhp, struct rcu_data, barrier_head); + struct rcu_state *rsp = rdp->rsp; + + if (atomic_dec_and_test(&rsp->barrier_cpu_count)) complete(&rcu_barrier_completion); } @@ -2284,7 +2286,7 @@ static void rcu_barrier_func(void *type) struct rcu_state *rsp = type; struct rcu_data *rdp = __this_cpu_ptr(rsp->rda); - atomic_inc(&rcu_barrier_cpu_count); + atomic_inc(&rsp->barrier_cpu_count); rsp->call(&rdp->barrier_head, rcu_barrier_callback); } @@ -2297,9 +2299,9 @@ static void _rcu_barrier(struct rcu_state *rsp) int cpu; unsigned long flags; struct rcu_data *rdp; - struct rcu_head rh; + struct rcu_data rd; - init_rcu_head_on_stack(&rh); + init_rcu_head_on_stack(&rd.barrier_head); /* Take mutex to serialize concurrent rcu_barrier() requests. */ mutex_lock(&rcu_barrier_mutex); @@ -2324,7 +2326,7 @@ static void _rcu_barrier(struct rcu_state *rsp) * us -- but before CPU 1's orphaned callbacks are invoked!!! */ init_completion(&rcu_barrier_completion); - atomic_set(&rcu_barrier_cpu_count, 1); + atomic_set(&rsp->barrier_cpu_count, 1); raw_spin_lock_irqsave(&rsp->onofflock, flags); rsp->rcu_barrier_in_progress = current; raw_spin_unlock_irqrestore(&rsp->onofflock, flags); @@ -2363,15 +2365,16 @@ static void _rcu_barrier(struct rcu_state *rsp) rcu_adopt_orphan_cbs(rsp); rsp->rcu_barrier_in_progress = NULL; raw_spin_unlock_irqrestore(&rsp->onofflock, flags); - atomic_inc(&rcu_barrier_cpu_count); + atomic_inc(&rsp->barrier_cpu_count); smp_mb__after_atomic_inc(); /* Ensure atomic_inc() before callback. */ - rsp->call(&rh, rcu_barrier_callback); + rd.rsp = rsp; + rsp->call(&rd.barrier_head, rcu_barrier_callback); /* * Now that we have an rcu_barrier_callback() callback on each * CPU, and thus each counted, remove the initial count. */ - if (atomic_dec_and_test(&rcu_barrier_cpu_count)) + if (atomic_dec_and_test(&rsp->barrier_cpu_count)) complete(&rcu_barrier_completion); /* Wait for all rcu_barrier_callback() callbacks to be invoked. */ @@ -2380,7 +2383,7 @@ static void _rcu_barrier(struct rcu_state *rsp) /* Other rcu_barrier() invocations can now safely proceed. */ mutex_unlock(&rcu_barrier_mutex); - destroy_rcu_head_on_stack(&rh); + destroy_rcu_head_on_stack(&rd.barrier_head); } /** diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 586d93c978f2..c57ef0b7f097 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -400,6 +400,7 @@ struct rcu_state { struct task_struct *rcu_barrier_in_progress; /* Task doing rcu_barrier(), */ /* or NULL if no barrier. */ + atomic_t barrier_cpu_count; /* # CPUs waiting on. */ raw_spinlock_t fqslock; /* Only one task forcing */ /* quiescent states. */ unsigned long jiffies_force_qs; /* Time at which to invoke */ -- cgit v1.2.3 From 7db74df88b52844f4e966901e2972bba725e6766 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 29 May 2012 03:03:37 -0700 Subject: rcu: Move rcu_barrier_completion to rcu_state structure In order to allow each RCU flavor to concurrently execute its rcu_barrier() function, it is necessary to move the relevant state to the rcu_state structure. This commit therefore moves the rcu_barrier_completion global variable to a new ->barrier_completion field in the rcu_state structure. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 9 ++++----- kernel/rcutree.h | 1 + 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 5929b021666d..ca7d1678ac79 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -158,7 +158,6 @@ unsigned long rcutorture_vernum; /* State information for rcu_barrier() and friends. */ static DEFINE_MUTEX(rcu_barrier_mutex); -static struct completion rcu_barrier_completion; /* * Return true if an RCU grace period is in progress. The ACCESS_ONCE()s @@ -2275,7 +2274,7 @@ static void rcu_barrier_callback(struct rcu_head *rhp) struct rcu_state *rsp = rdp->rsp; if (atomic_dec_and_test(&rsp->barrier_cpu_count)) - complete(&rcu_barrier_completion); + complete(&rsp->barrier_completion); } /* @@ -2325,7 +2324,7 @@ static void _rcu_barrier(struct rcu_state *rsp) * 6. Both rcu_barrier_callback() callbacks are invoked, awakening * us -- but before CPU 1's orphaned callbacks are invoked!!! */ - init_completion(&rcu_barrier_completion); + init_completion(&rsp->barrier_completion); atomic_set(&rsp->barrier_cpu_count, 1); raw_spin_lock_irqsave(&rsp->onofflock, flags); rsp->rcu_barrier_in_progress = current; @@ -2375,10 +2374,10 @@ static void _rcu_barrier(struct rcu_state *rsp) * CPU, and thus each counted, remove the initial count. */ if (atomic_dec_and_test(&rsp->barrier_cpu_count)) - complete(&rcu_barrier_completion); + complete(&rsp->barrier_completion); /* Wait for all rcu_barrier_callback() callbacks to be invoked. */ - wait_for_completion(&rcu_barrier_completion); + wait_for_completion(&rsp->barrier_completion); /* Other rcu_barrier() invocations can now safely proceed. */ mutex_unlock(&rcu_barrier_mutex); diff --git a/kernel/rcutree.h b/kernel/rcutree.h index c57ef0b7f097..d1ca4424122b 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -401,6 +401,7 @@ struct rcu_state { /* Task doing rcu_barrier(), */ /* or NULL if no barrier. */ atomic_t barrier_cpu_count; /* # CPUs waiting on. */ + struct completion barrier_completion; /* Wake at barrier end. */ raw_spinlock_t fqslock; /* Only one task forcing */ /* quiescent states. */ unsigned long jiffies_force_qs; /* Time at which to invoke */ -- cgit v1.2.3 From 7be7f0be907224445acc62b3884c892f38b7ff40 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 29 May 2012 05:18:53 -0700 Subject: rcu: Move rcu_barrier_mutex to rcu_state structure In order to allow each RCU flavor to concurrently execute its rcu_barrier() function, it is necessary to move the relevant state to the rcu_state structure. This commit therefore moves the rcu_barrier_mutex global variable to a new ->barrier_mutex field in the rcu_state structure. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 9 +++------ kernel/rcutree.h | 1 + 2 files changed, 4 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index ca7d1678ac79..ff992ac24e73 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -71,6 +71,7 @@ static struct lock_class_key rcu_node_class[RCU_NUM_LVLS]; .onofflock = __RAW_SPIN_LOCK_UNLOCKED(&sname##_state.onofflock), \ .orphan_nxttail = &sname##_state.orphan_nxtlist, \ .orphan_donetail = &sname##_state.orphan_donelist, \ + .barrier_mutex = __MUTEX_INITIALIZER(sname##_state.barrier_mutex), \ .fqslock = __RAW_SPIN_LOCK_UNLOCKED(&sname##_state.fqslock), \ .n_force_qs = 0, \ .n_force_qs_ngp = 0, \ @@ -155,10 +156,6 @@ static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp); unsigned long rcutorture_testseq; unsigned long rcutorture_vernum; -/* State information for rcu_barrier() and friends. */ - -static DEFINE_MUTEX(rcu_barrier_mutex); - /* * Return true if an RCU grace period is in progress. The ACCESS_ONCE()s * permit this function to be invoked without holding the root rcu_node @@ -2303,7 +2300,7 @@ static void _rcu_barrier(struct rcu_state *rsp) init_rcu_head_on_stack(&rd.barrier_head); /* Take mutex to serialize concurrent rcu_barrier() requests. */ - mutex_lock(&rcu_barrier_mutex); + mutex_lock(&rsp->barrier_mutex); smp_mb(); /* Prevent any prior operations from leaking in. */ @@ -2380,7 +2377,7 @@ static void _rcu_barrier(struct rcu_state *rsp) wait_for_completion(&rsp->barrier_completion); /* Other rcu_barrier() invocations can now safely proceed. */ - mutex_unlock(&rcu_barrier_mutex); + mutex_unlock(&rsp->barrier_mutex); destroy_rcu_head_on_stack(&rd.barrier_head); } diff --git a/kernel/rcutree.h b/kernel/rcutree.h index d1ca4424122b..7641aec3e59c 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -400,6 +400,7 @@ struct rcu_state { struct task_struct *rcu_barrier_in_progress; /* Task doing rcu_barrier(), */ /* or NULL if no barrier. */ + struct mutex barrier_mutex; /* Guards barrier fields. */ atomic_t barrier_cpu_count; /* # CPUs waiting on. */ struct completion barrier_completion; /* Wake at barrier end. */ raw_spinlock_t fqslock; /* Only one task forcing */ -- cgit v1.2.3 From cfed0a85dad921c683e9c0d25b072bcc5745ede0 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 15 Jun 2012 18:22:05 -0700 Subject: rcu: Remove needless initialization For global variables, C defaults all fields to zero. The initialization of the rcu_state structure's ->n_force_qs and ->n_force_qs_ngp fields is therefore redundant, so this commit removes these initializations. Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index ff992ac24e73..44a8fda9be86 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -73,8 +73,6 @@ static struct lock_class_key rcu_node_class[RCU_NUM_LVLS]; .orphan_donetail = &sname##_state.orphan_donelist, \ .barrier_mutex = __MUTEX_INITIALIZER(sname##_state.barrier_mutex), \ .fqslock = __RAW_SPIN_LOCK_UNLOCKED(&sname##_state.fqslock), \ - .n_force_qs = 0, \ - .n_force_qs_ngp = 0, \ .name = #sname, \ } -- cgit v1.2.3 From cf3a9c4842b1e097dbe0854933c471d43dd24f69 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 29 May 2012 14:56:46 -0700 Subject: rcu: Increase rcu_barrier() concurrency The traditional rcu_barrier() implementation has serialized all requests, regardless of RCU flavor, and also does not coalesce concurrent requests. In the past, this has been good and sufficient. However, systems are getting larger and use of rcu_barrier() has been increasing. This commit therefore introduces a counter-based scheme that allows _rcu_barrier() calls for the same flavor of RCU to take advantage of each others' work. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 36 +++++++++++++++++++++++++++++++++++- kernel/rcutree.h | 2 ++ 2 files changed, 37 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 44a8fda9be86..6bb5d562253f 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -2294,13 +2294,41 @@ static void _rcu_barrier(struct rcu_state *rsp) unsigned long flags; struct rcu_data *rdp; struct rcu_data rd; + unsigned long snap = ACCESS_ONCE(rsp->n_barrier_done); + unsigned long snap_done; init_rcu_head_on_stack(&rd.barrier_head); /* Take mutex to serialize concurrent rcu_barrier() requests. */ mutex_lock(&rsp->barrier_mutex); - smp_mb(); /* Prevent any prior operations from leaking in. */ + /* + * Ensure that all prior references, including to ->n_barrier_done, + * are ordered before the _rcu_barrier() machinery. + */ + smp_mb(); /* See above block comment. */ + + /* + * Recheck ->n_barrier_done to see if others did our work for us. + * This means checking ->n_barrier_done for an even-to-odd-to-even + * transition. The "if" expression below therefore rounds the old + * value up to the next even number and adds two before comparing. + */ + snap_done = ACCESS_ONCE(rsp->n_barrier_done); + if (ULONG_CMP_GE(snap_done, ((snap + 1) & ~0x1) + 2)) { + smp_mb(); /* caller's subsequent code after above check. */ + mutex_unlock(&rsp->barrier_mutex); + return; + } + + /* + * Increment ->n_barrier_done to avoid duplicate work. Use + * ACCESS_ONCE() to prevent the compiler from speculating + * the increment to precede the early-exit check. + */ + ACCESS_ONCE(rsp->n_barrier_done)++; + WARN_ON_ONCE((rsp->n_barrier_done & 0x1) != 1); + smp_mb(); /* Order ->n_barrier_done increment with below mechanism. */ /* * Initialize the count to one rather than to zero in order to @@ -2371,6 +2399,12 @@ static void _rcu_barrier(struct rcu_state *rsp) if (atomic_dec_and_test(&rsp->barrier_cpu_count)) complete(&rsp->barrier_completion); + /* Increment ->n_barrier_done to prevent duplicate work. */ + smp_mb(); /* Keep increment after above mechanism. */ + ACCESS_ONCE(rsp->n_barrier_done)++; + WARN_ON_ONCE((rsp->n_barrier_done & 0x1) != 0); + smp_mb(); /* Keep increment before caller's subsequent code. */ + /* Wait for all rcu_barrier_callback() callbacks to be invoked. */ wait_for_completion(&rsp->barrier_completion); diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 7641aec3e59c..be10286ad380 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -403,6 +403,8 @@ struct rcu_state { struct mutex barrier_mutex; /* Guards barrier fields. */ atomic_t barrier_cpu_count; /* # CPUs waiting on. */ struct completion barrier_completion; /* Wake at barrier end. */ + unsigned long n_barrier_done; /* ++ at start and end of */ + /* _rcu_barrier(). */ raw_spinlock_t fqslock; /* Only one task forcing */ /* quiescent states. */ unsigned long jiffies_force_qs; /* Time at which to invoke */ -- cgit v1.2.3 From a83eff0a82a7f3f14fea477fd41e6c082e7fc96a Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 23 May 2012 18:47:05 -0700 Subject: rcu: Add tracing for _rcu_barrier() This commit adds event tracing for _rcu_barrier() execution. This is defined only if RCU_TRACE=y. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- include/trace/events/rcu.h | 45 +++++++++++++++++++++++++++++++++++++++++++++ kernel/rcutree.c | 29 ++++++++++++++++++++++++++++- 2 files changed, 73 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h index d274734b2aa4..5bde94d8585b 100644 --- a/include/trace/events/rcu.h +++ b/include/trace/events/rcu.h @@ -541,6 +541,50 @@ TRACE_EVENT(rcu_torture_read, __entry->rcutorturename, __entry->rhp) ); +/* + * Tracepoint for _rcu_barrier() execution. The string "s" describes + * the _rcu_barrier phase: + * "Begin": rcu_barrier_callback() started. + * "Check": rcu_barrier_callback() checking for piggybacking. + * "EarlyExit": rcu_barrier_callback() piggybacked, thus early exit. + * "Inc1": rcu_barrier_callback() piggyback check counter incremented. + * "Offline": rcu_barrier_callback() found offline CPU + * "OnlineQ": rcu_barrier_callback() found online CPU with callbacks. + * "OnlineNQ": rcu_barrier_callback() found online CPU, no callbacks. + * "IRQ": An rcu_barrier_callback() callback posted on remote CPU. + * "CB": An rcu_barrier_callback() invoked a callback, not the last. + * "LastCB": An rcu_barrier_callback() invoked the last callback. + * "Inc2": rcu_barrier_callback() piggyback check counter incremented. + * The "cpu" argument is the CPU or -1 if meaningless, the "cnt" argument + * is the count of remaining callbacks, and "done" is the piggybacking count. + */ +TRACE_EVENT(rcu_barrier, + + TP_PROTO(char *rcuname, char *s, int cpu, int cnt, unsigned long done), + + TP_ARGS(rcuname, s, cpu, cnt, done), + + TP_STRUCT__entry( + __field(char *, rcuname) + __field(char *, s) + __field(int, cpu) + __field(int, cnt) + __field(unsigned long, done) + ), + + TP_fast_assign( + __entry->rcuname = rcuname; + __entry->s = s; + __entry->cpu = cpu; + __entry->cnt = cnt; + __entry->done = done; + ), + + TP_printk("%s %s cpu %d remaining %d # %lu", + __entry->rcuname, __entry->s, __entry->cpu, __entry->cnt, + __entry->done) +); + #else /* #ifdef CONFIG_RCU_TRACE */ #define trace_rcu_grace_period(rcuname, gpnum, gpevent) do { } while (0) @@ -564,6 +608,7 @@ TRACE_EVENT(rcu_torture_read, #define trace_rcu_batch_end(rcuname, callbacks_invoked, cb, nr, iit, risk) \ do { } while (0) #define trace_rcu_torture_read(rcutorturename, rhp) do { } while (0) +#define trace_rcu_barrier(name, s, cpu, cnt, done) do { } while (0) #endif /* #else #ifdef CONFIG_RCU_TRACE */ diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 6bb5d562253f..dda43d826504 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -2259,6 +2259,17 @@ static int rcu_cpu_has_callbacks(int cpu) rcu_preempt_cpu_has_callbacks(cpu); } +/* + * Helper function for _rcu_barrier() tracing. If tracing is disabled, + * the compiler is expected to optimize this away. + */ +static void _rcu_barrier_trace(struct rcu_state *rsp, char *s, + int cpu, unsigned long done) +{ + trace_rcu_barrier(rsp->name, s, cpu, + atomic_read(&rsp->barrier_cpu_count), done); +} + /* * RCU callback function for _rcu_barrier(). If we are last, wake * up the task executing _rcu_barrier(). @@ -2268,8 +2279,12 @@ static void rcu_barrier_callback(struct rcu_head *rhp) struct rcu_data *rdp = container_of(rhp, struct rcu_data, barrier_head); struct rcu_state *rsp = rdp->rsp; - if (atomic_dec_and_test(&rsp->barrier_cpu_count)) + if (atomic_dec_and_test(&rsp->barrier_cpu_count)) { + _rcu_barrier_trace(rsp, "LastCB", -1, rsp->n_barrier_done); complete(&rsp->barrier_completion); + } else { + _rcu_barrier_trace(rsp, "CB", -1, rsp->n_barrier_done); + } } /* @@ -2280,6 +2295,7 @@ static void rcu_barrier_func(void *type) struct rcu_state *rsp = type; struct rcu_data *rdp = __this_cpu_ptr(rsp->rda); + _rcu_barrier_trace(rsp, "IRQ", -1, rsp->n_barrier_done); atomic_inc(&rsp->barrier_cpu_count); rsp->call(&rdp->barrier_head, rcu_barrier_callback); } @@ -2298,6 +2314,7 @@ static void _rcu_barrier(struct rcu_state *rsp) unsigned long snap_done; init_rcu_head_on_stack(&rd.barrier_head); + _rcu_barrier_trace(rsp, "Begin", -1, snap); /* Take mutex to serialize concurrent rcu_barrier() requests. */ mutex_lock(&rsp->barrier_mutex); @@ -2315,7 +2332,9 @@ static void _rcu_barrier(struct rcu_state *rsp) * value up to the next even number and adds two before comparing. */ snap_done = ACCESS_ONCE(rsp->n_barrier_done); + _rcu_barrier_trace(rsp, "Check", -1, snap_done); if (ULONG_CMP_GE(snap_done, ((snap + 1) & ~0x1) + 2)) { + _rcu_barrier_trace(rsp, "EarlyExit", -1, snap_done); smp_mb(); /* caller's subsequent code after above check. */ mutex_unlock(&rsp->barrier_mutex); return; @@ -2328,6 +2347,7 @@ static void _rcu_barrier(struct rcu_state *rsp) */ ACCESS_ONCE(rsp->n_barrier_done)++; WARN_ON_ONCE((rsp->n_barrier_done & 0x1) != 1); + _rcu_barrier_trace(rsp, "Inc1", -1, rsp->n_barrier_done); smp_mb(); /* Order ->n_barrier_done increment with below mechanism. */ /* @@ -2364,13 +2384,19 @@ static void _rcu_barrier(struct rcu_state *rsp) preempt_disable(); rdp = per_cpu_ptr(rsp->rda, cpu); if (cpu_is_offline(cpu)) { + _rcu_barrier_trace(rsp, "Offline", cpu, + rsp->n_barrier_done); preempt_enable(); while (cpu_is_offline(cpu) && ACCESS_ONCE(rdp->qlen)) schedule_timeout_interruptible(1); } else if (ACCESS_ONCE(rdp->qlen)) { + _rcu_barrier_trace(rsp, "OnlineQ", cpu, + rsp->n_barrier_done); smp_call_function_single(cpu, rcu_barrier_func, rsp, 1); preempt_enable(); } else { + _rcu_barrier_trace(rsp, "OnlineNQ", cpu, + rsp->n_barrier_done); preempt_enable(); } } @@ -2403,6 +2429,7 @@ static void _rcu_barrier(struct rcu_state *rsp) smp_mb(); /* Keep increment after above mechanism. */ ACCESS_ONCE(rsp->n_barrier_done)++; WARN_ON_ONCE((rsp->n_barrier_done & 0x1) != 0); + _rcu_barrier_trace(rsp, "Inc2", -1, rsp->n_barrier_done); smp_mb(); /* Keep increment before caller's subsequent code. */ /* Wait for all rcu_barrier_callback() callbacks to be invoked. */ -- cgit v1.2.3 From d7e187c8e9f30543f9cadfed094896ff414acb8f Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 25 May 2012 11:49:15 -0700 Subject: rcu: Add rcu_barrier() statistics to debugfs tracing This commit adds an rcubarrier file to RCU's debugfs statistical tracing directory, providing diagnostic information on rcu_barrier(). Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree_trace.c | 39 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+) (limited to 'kernel') diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index a3556a2d842c..057408be6c3b 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -46,6 +46,40 @@ #define RCU_TREE_NONCORE #include "rcutree.h" +static void print_rcubarrier(struct seq_file *m, struct rcu_state *rsp) +{ + seq_printf(m, "%c bcc: %d nbd: %lu\n", + rsp->rcu_barrier_in_progress ? 'B' : '.', + atomic_read(&rsp->barrier_cpu_count), + rsp->n_barrier_done); +} + +static int show_rcubarrier(struct seq_file *m, void *unused) +{ +#ifdef CONFIG_TREE_PREEMPT_RCU + seq_puts(m, "rcu_preempt: "); + print_rcubarrier(m, &rcu_preempt_state); +#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ + seq_puts(m, "rcu_sched: "); + print_rcubarrier(m, &rcu_sched_state); + seq_puts(m, "rcu_bh: "); + print_rcubarrier(m, &rcu_bh_state); + return 0; +} + +static int rcubarrier_open(struct inode *inode, struct file *file) +{ + return single_open(file, show_rcubarrier, NULL); +} + +static const struct file_operations rcubarrier_fops = { + .owner = THIS_MODULE, + .open = rcubarrier_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + #ifdef CONFIG_RCU_BOOST static char convert_kthread_status(unsigned int kthread_status) @@ -453,6 +487,11 @@ static int __init rcutree_trace_init(void) if (!rcudir) goto free_out; + retval = debugfs_create_file("rcubarrier", 0444, rcudir, + NULL, &rcubarrier_fops); + if (!retval) + goto free_out; + retval = debugfs_create_file("rcudata", 0444, rcudir, NULL, &rcudata_fops); if (!retval) -- cgit v1.2.3 From 1bca8cf1a2c3c6683b12ad28a3e826ca7a834978 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 12 Jun 2012 09:40:38 -0700 Subject: rcu: Remove unneeded __rcu_process_callbacks() argument With the advent of __this_cpu_ptr(), it is no longer necessary to pass both the rcu_state and rcu_data structures into __rcu_process_callbacks(). This commit therefore computes the rcu_data pointer from the rcu_state pointer within __rcu_process_callbacks() so that callers can pass in only the pointer to the rcu_state structure. This paves the way for linking the rcu_state structures together and iterating over them. Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 8 ++++---- kernel/rcutree_plugin.h | 3 +-- 2 files changed, 5 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index dda43d826504..5376a156be8a 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1788,9 +1788,10 @@ unlock_fqs_ret: * whom the rdp belongs. */ static void -__rcu_process_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) +__rcu_process_callbacks(struct rcu_state *rsp) { unsigned long flags; + struct rcu_data *rdp = __this_cpu_ptr(rsp->rda); WARN_ON_ONCE(rdp->beenonline == 0); @@ -1827,9 +1828,8 @@ __rcu_process_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) static void rcu_process_callbacks(struct softirq_action *unused) { trace_rcu_utilization("Start RCU core"); - __rcu_process_callbacks(&rcu_sched_state, - &__get_cpu_var(rcu_sched_data)); - __rcu_process_callbacks(&rcu_bh_state, &__get_cpu_var(rcu_bh_data)); + __rcu_process_callbacks(&rcu_sched_state); + __rcu_process_callbacks(&rcu_bh_state); rcu_preempt_process_callbacks(); trace_rcu_utilization("End RCU core"); } diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 9cb3a68819fa..5a80cdd9a0a3 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -687,8 +687,7 @@ static void rcu_preempt_check_callbacks(int cpu) */ static void rcu_preempt_process_callbacks(void) { - __rcu_process_callbacks(&rcu_preempt_state, - &__get_cpu_var(rcu_preempt_data)); + __rcu_process_callbacks(&rcu_preempt_state); } #ifdef CONFIG_RCU_BOOST -- cgit v1.2.3 From 6ce75a2326e6f8b3bdfb60e1de7934b89858e87b Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 12 Jun 2012 11:01:13 -0700 Subject: rcu: Introduce for_each_rcu_flavor() and use it The arrival of TREE_PREEMPT_RCU some years back included some ugly code involving either #ifdef or #ifdef'ed wrapper functions to iterate over all non-SRCU flavors of RCU. This commit therefore introduces a for_each_rcu_flavor() iterator over the rcu_state structures for each flavor of RCU to clean up a bit of the ugliness. Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 53 +++++++++++++--------- kernel/rcutree.h | 12 +++-- kernel/rcutree_plugin.h | 116 ------------------------------------------------ 3 files changed, 37 insertions(+), 144 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 5376a156be8a..b61c3ffc80e9 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -84,6 +84,7 @@ struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh, call_rcu_bh); DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); static struct rcu_state *rcu_state; +LIST_HEAD(rcu_struct_flavors); /* Increase (but not decrease) the CONFIG_RCU_FANOUT_LEAF at boot time. */ static int rcu_fanout_leaf = CONFIG_RCU_FANOUT_LEAF; @@ -860,9 +861,10 @@ static int rcu_panic(struct notifier_block *this, unsigned long ev, void *ptr) */ void rcu_cpu_stall_reset(void) { - rcu_sched_state.jiffies_stall = jiffies + ULONG_MAX / 2; - rcu_bh_state.jiffies_stall = jiffies + ULONG_MAX / 2; - rcu_preempt_stall_reset(); + struct rcu_state *rsp; + + for_each_rcu_flavor(rsp) + rsp->jiffies_stall = jiffies + ULONG_MAX / 2; } static struct notifier_block rcu_panic_block = { @@ -1827,10 +1829,11 @@ __rcu_process_callbacks(struct rcu_state *rsp) */ static void rcu_process_callbacks(struct softirq_action *unused) { + struct rcu_state *rsp; + trace_rcu_utilization("Start RCU core"); - __rcu_process_callbacks(&rcu_sched_state); - __rcu_process_callbacks(&rcu_bh_state); - rcu_preempt_process_callbacks(); + for_each_rcu_flavor(rsp) + __rcu_process_callbacks(rsp); trace_rcu_utilization("End RCU core"); } @@ -2241,9 +2244,12 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) */ static int rcu_pending(int cpu) { - return __rcu_pending(&rcu_sched_state, &per_cpu(rcu_sched_data, cpu)) || - __rcu_pending(&rcu_bh_state, &per_cpu(rcu_bh_data, cpu)) || - rcu_preempt_pending(cpu); + struct rcu_state *rsp; + + for_each_rcu_flavor(rsp) + if (__rcu_pending(rsp, per_cpu_ptr(rsp->rda, cpu))) + return 1; + return 0; } /* @@ -2253,10 +2259,13 @@ static int rcu_pending(int cpu) */ static int rcu_cpu_has_callbacks(int cpu) { + struct rcu_state *rsp; + /* RCU callbacks either ready or pending? */ - return per_cpu(rcu_sched_data, cpu).nxtlist || - per_cpu(rcu_bh_data, cpu).nxtlist || - rcu_preempt_cpu_has_callbacks(cpu); + for_each_rcu_flavor(rsp) + if (per_cpu_ptr(rsp->rda, cpu)->nxtlist) + return 1; + return 0; } /* @@ -2551,9 +2560,11 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptible) static void __cpuinit rcu_prepare_cpu(int cpu) { - rcu_init_percpu_data(cpu, &rcu_sched_state, 0); - rcu_init_percpu_data(cpu, &rcu_bh_state, 0); - rcu_preempt_init_percpu_data(cpu); + struct rcu_state *rsp; + + for_each_rcu_flavor(rsp) + rcu_init_percpu_data(cpu, rsp, + strcmp(rsp->name, "rcu_preempt") == 0); } /* @@ -2565,6 +2576,7 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, long cpu = (long)hcpu; struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, cpu); struct rcu_node *rnp = rdp->mynode; + struct rcu_state *rsp; trace_rcu_utilization("Start CPU hotplug"); switch (action) { @@ -2589,18 +2601,16 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, * touch any data without introducing corruption. We send the * dying CPU's callbacks to an arbitrarily chosen online CPU. */ - rcu_cleanup_dying_cpu(&rcu_bh_state); - rcu_cleanup_dying_cpu(&rcu_sched_state); - rcu_preempt_cleanup_dying_cpu(); + for_each_rcu_flavor(rsp) + rcu_cleanup_dying_cpu(rsp); rcu_cleanup_after_idle(cpu); break; case CPU_DEAD: case CPU_DEAD_FROZEN: case CPU_UP_CANCELED: case CPU_UP_CANCELED_FROZEN: - rcu_cleanup_dead_cpu(cpu, &rcu_bh_state); - rcu_cleanup_dead_cpu(cpu, &rcu_sched_state); - rcu_preempt_cleanup_dead_cpu(cpu); + for_each_rcu_flavor(rsp) + rcu_cleanup_dead_cpu(cpu, rsp); break; default: break; @@ -2717,6 +2727,7 @@ static void __init rcu_init_one(struct rcu_state *rsp, per_cpu_ptr(rsp->rda, i)->mynode = rnp; rcu_boot_init_percpu_data(i, rsp); } + list_add(&rsp->flavors, &rcu_struct_flavors); } /* diff --git a/kernel/rcutree.h b/kernel/rcutree.h index be10286ad380..b92c4550a6e6 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -422,8 +422,13 @@ struct rcu_state { unsigned long gp_max; /* Maximum GP duration in */ /* jiffies. */ char *name; /* Name of structure. */ + struct list_head flavors; /* List of RCU flavors. */ }; +extern struct list_head rcu_struct_flavors; +#define for_each_rcu_flavor(rsp) \ + list_for_each_entry((rsp), &rcu_struct_flavors, flavors) + /* Return values for rcu_preempt_offline_tasks(). */ #define RCU_OFL_TASKS_NORM_GP 0x1 /* Tasks blocking normal */ @@ -466,25 +471,18 @@ static void rcu_stop_cpu_kthread(int cpu); #endif /* #ifdef CONFIG_HOTPLUG_CPU */ static void rcu_print_detail_task_stall(struct rcu_state *rsp); static int rcu_print_task_stall(struct rcu_node *rnp); -static void rcu_preempt_stall_reset(void); static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp); #ifdef CONFIG_HOTPLUG_CPU static int rcu_preempt_offline_tasks(struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp); #endif /* #ifdef CONFIG_HOTPLUG_CPU */ -static void rcu_preempt_cleanup_dead_cpu(int cpu); static void rcu_preempt_check_callbacks(int cpu); -static void rcu_preempt_process_callbacks(void); void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)); #if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_TREE_PREEMPT_RCU) static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp, bool wake); #endif /* #if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_TREE_PREEMPT_RCU) */ -static int rcu_preempt_pending(int cpu); -static int rcu_preempt_cpu_has_callbacks(int cpu); -static void __cpuinit rcu_preempt_init_percpu_data(int cpu); -static void rcu_preempt_cleanup_dying_cpu(void); static void __init __rcu_init_preempt(void); static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags); static void rcu_preempt_boost_start_gp(struct rcu_node *rnp); diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 5a80cdd9a0a3..d18b4d383afe 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -544,16 +544,6 @@ static int rcu_print_task_stall(struct rcu_node *rnp) return ndetected; } -/* - * Suppress preemptible RCU's CPU stall warnings by pushing the - * time of the next stall-warning message comfortably far into the - * future. - */ -static void rcu_preempt_stall_reset(void) -{ - rcu_preempt_state.jiffies_stall = jiffies + ULONG_MAX / 2; -} - /* * Check that the list of blocked tasks for the newly completed grace * period is in fact empty. It is a serious bug to complete a grace @@ -654,14 +644,6 @@ static int rcu_preempt_offline_tasks(struct rcu_state *rsp, #endif /* #ifdef CONFIG_HOTPLUG_CPU */ -/* - * Do CPU-offline processing for preemptible RCU. - */ -static void rcu_preempt_cleanup_dead_cpu(int cpu) -{ - rcu_cleanup_dead_cpu(cpu, &rcu_preempt_state); -} - /* * Check for a quiescent state from the current CPU. When a task blocks, * the task is recorded in the corresponding CPU's rcu_node structure, @@ -682,14 +664,6 @@ static void rcu_preempt_check_callbacks(int cpu) t->rcu_read_unlock_special |= RCU_READ_UNLOCK_NEED_QS; } -/* - * Process callbacks for preemptible RCU. - */ -static void rcu_preempt_process_callbacks(void) -{ - __rcu_process_callbacks(&rcu_preempt_state); -} - #ifdef CONFIG_RCU_BOOST static void rcu_preempt_do_callbacks(void) @@ -921,24 +895,6 @@ mb_ret: } EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); -/* - * Check to see if there is any immediate preemptible-RCU-related work - * to be done. - */ -static int rcu_preempt_pending(int cpu) -{ - return __rcu_pending(&rcu_preempt_state, - &per_cpu(rcu_preempt_data, cpu)); -} - -/* - * Does preemptible RCU have callbacks on this CPU? - */ -static int rcu_preempt_cpu_has_callbacks(int cpu) -{ - return !!per_cpu(rcu_preempt_data, cpu).nxtlist; -} - /** * rcu_barrier - Wait until all in-flight call_rcu() callbacks complete. */ @@ -948,23 +904,6 @@ void rcu_barrier(void) } EXPORT_SYMBOL_GPL(rcu_barrier); -/* - * Initialize preemptible RCU's per-CPU data. - */ -static void __cpuinit rcu_preempt_init_percpu_data(int cpu) -{ - rcu_init_percpu_data(cpu, &rcu_preempt_state, 1); -} - -/* - * Move preemptible RCU's callbacks from dying CPU to other online CPU - * and record a quiescent state. - */ -static void rcu_preempt_cleanup_dying_cpu(void) -{ - rcu_cleanup_dying_cpu(&rcu_preempt_state); -} - /* * Initialize preemptible RCU's state structures. */ @@ -1049,14 +988,6 @@ static int rcu_print_task_stall(struct rcu_node *rnp) return 0; } -/* - * Because preemptible RCU does not exist, there is no need to suppress - * its CPU stall warnings. - */ -static void rcu_preempt_stall_reset(void) -{ -} - /* * Because there is no preemptible RCU, there can be no readers blocked, * so there is no need to check for blocked tasks. So check only for @@ -1084,14 +1015,6 @@ static int rcu_preempt_offline_tasks(struct rcu_state *rsp, #endif /* #ifdef CONFIG_HOTPLUG_CPU */ -/* - * Because preemptible RCU does not exist, it never needs CPU-offline - * processing. - */ -static void rcu_preempt_cleanup_dead_cpu(int cpu) -{ -} - /* * Because preemptible RCU does not exist, it never has any callbacks * to check. @@ -1100,14 +1023,6 @@ static void rcu_preempt_check_callbacks(int cpu) { } -/* - * Because preemptible RCU does not exist, it never has any callbacks - * to process. - */ -static void rcu_preempt_process_callbacks(void) -{ -} - /* * Queue an RCU callback for lazy invocation after a grace period. * This will likely be later named something like "call_rcu_lazy()", @@ -1148,22 +1063,6 @@ static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp, #endif /* #ifdef CONFIG_HOTPLUG_CPU */ -/* - * Because preemptible RCU does not exist, it never has any work to do. - */ -static int rcu_preempt_pending(int cpu) -{ - return 0; -} - -/* - * Because preemptible RCU does not exist, it never has callbacks - */ -static int rcu_preempt_cpu_has_callbacks(int cpu) -{ - return 0; -} - /* * Because preemptible RCU does not exist, rcu_barrier() is just * another name for rcu_barrier_sched(). @@ -1174,21 +1073,6 @@ void rcu_barrier(void) } EXPORT_SYMBOL_GPL(rcu_barrier); -/* - * Because preemptible RCU does not exist, there is no per-CPU - * data to initialize. - */ -static void __cpuinit rcu_preempt_init_percpu_data(int cpu) -{ -} - -/* - * Because there is no preemptible RCU, there is no cleanup to do. - */ -static void rcu_preempt_cleanup_dying_cpu(void) -{ -} - /* * Because preemptible RCU does not exist, it need not be initialized. */ -- cgit v1.2.3 From c0cc962da3e7770feb3665f087ea3e23d8c15479 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 12 Jun 2012 12:25:39 -0700 Subject: rcu: Use for_each_rcu_flavor() in TREE_RCU tracing This commit applies the new for_each_rcu_flavor() macro to the kernel/rcutree_trace.c file. Signed-off-by: Paul E. McKenney --- kernel/rcutree_trace.c | 116 ++++++++++++++++++------------------------------- 1 file changed, 43 insertions(+), 73 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index 057408be6c3b..a16ddbd6fdc4 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -46,24 +46,15 @@ #define RCU_TREE_NONCORE #include "rcutree.h" -static void print_rcubarrier(struct seq_file *m, struct rcu_state *rsp) -{ - seq_printf(m, "%c bcc: %d nbd: %lu\n", - rsp->rcu_barrier_in_progress ? 'B' : '.', - atomic_read(&rsp->barrier_cpu_count), - rsp->n_barrier_done); -} - static int show_rcubarrier(struct seq_file *m, void *unused) { -#ifdef CONFIG_TREE_PREEMPT_RCU - seq_puts(m, "rcu_preempt: "); - print_rcubarrier(m, &rcu_preempt_state); -#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ - seq_puts(m, "rcu_sched: "); - print_rcubarrier(m, &rcu_sched_state); - seq_puts(m, "rcu_bh: "); - print_rcubarrier(m, &rcu_bh_state); + struct rcu_state *rsp; + + for_each_rcu_flavor(rsp) + seq_printf(m, "%s: %c bcc: %d nbd: %lu\n", + rsp->name, rsp->rcu_barrier_in_progress ? 'B' : '.', + atomic_read(&rsp->barrier_cpu_count), + rsp->n_barrier_done); return 0; } @@ -129,24 +120,16 @@ static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp) rdp->n_cbs_invoked, rdp->n_cbs_orphaned, rdp->n_cbs_adopted); } -#define PRINT_RCU_DATA(name, func, m) \ - do { \ - int _p_r_d_i; \ - \ - for_each_possible_cpu(_p_r_d_i) \ - func(m, &per_cpu(name, _p_r_d_i)); \ - } while (0) - static int show_rcudata(struct seq_file *m, void *unused) { -#ifdef CONFIG_TREE_PREEMPT_RCU - seq_puts(m, "rcu_preempt:\n"); - PRINT_RCU_DATA(rcu_preempt_data, print_one_rcu_data, m); -#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ - seq_puts(m, "rcu_sched:\n"); - PRINT_RCU_DATA(rcu_sched_data, print_one_rcu_data, m); - seq_puts(m, "rcu_bh:\n"); - PRINT_RCU_DATA(rcu_bh_data, print_one_rcu_data, m); + int cpu; + struct rcu_state *rsp; + + for_each_rcu_flavor(rsp) { + seq_printf(m, "%s:\n", rsp->name); + for_each_possible_cpu(cpu) + print_one_rcu_data(m, per_cpu_ptr(rsp->rda, cpu)); + } return 0; } @@ -200,6 +183,9 @@ static void print_one_rcu_data_csv(struct seq_file *m, struct rcu_data *rdp) static int show_rcudata_csv(struct seq_file *m, void *unused) { + int cpu; + struct rcu_state *rsp; + seq_puts(m, "\"CPU\",\"Online?\",\"c\",\"g\",\"pq\",\"pgp\",\"pq\","); seq_puts(m, "\"dt\",\"dt nesting\",\"dt NMI nesting\",\"df\","); seq_puts(m, "\"of\",\"qll\",\"ql\",\"qs\""); @@ -207,14 +193,11 @@ static int show_rcudata_csv(struct seq_file *m, void *unused) seq_puts(m, "\"kt\",\"ktl\""); #endif /* #ifdef CONFIG_RCU_BOOST */ seq_puts(m, ",\"b\",\"ci\",\"co\",\"ca\"\n"); -#ifdef CONFIG_TREE_PREEMPT_RCU - seq_puts(m, "\"rcu_preempt:\"\n"); - PRINT_RCU_DATA(rcu_preempt_data, print_one_rcu_data_csv, m); -#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ - seq_puts(m, "\"rcu_sched:\"\n"); - PRINT_RCU_DATA(rcu_sched_data, print_one_rcu_data_csv, m); - seq_puts(m, "\"rcu_bh:\"\n"); - PRINT_RCU_DATA(rcu_bh_data, print_one_rcu_data_csv, m); + for_each_rcu_flavor(rsp) { + seq_printf(m, "\"%s:\"\n", rsp->name); + for_each_possible_cpu(cpu) + print_one_rcu_data_csv(m, per_cpu_ptr(rsp->rda, cpu)); + } return 0; } @@ -304,9 +287,9 @@ static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) struct rcu_node *rnp; gpnum = rsp->gpnum; - seq_printf(m, "c=%lu g=%lu s=%d jfq=%ld j=%x " + seq_printf(m, "%s: c=%lu g=%lu s=%d jfq=%ld j=%x " "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu oqlen=%ld/%ld\n", - rsp->completed, gpnum, rsp->fqs_state, + rsp->name, rsp->completed, gpnum, rsp->fqs_state, (long)(rsp->jiffies_force_qs - jiffies), (int)(jiffies & 0xffff), rsp->n_force_qs, rsp->n_force_qs_ngp, @@ -329,14 +312,10 @@ static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) static int show_rcuhier(struct seq_file *m, void *unused) { -#ifdef CONFIG_TREE_PREEMPT_RCU - seq_puts(m, "rcu_preempt:\n"); - print_one_rcu_state(m, &rcu_preempt_state); -#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ - seq_puts(m, "rcu_sched:\n"); - print_one_rcu_state(m, &rcu_sched_state); - seq_puts(m, "rcu_bh:\n"); - print_one_rcu_state(m, &rcu_bh_state); + struct rcu_state *rsp; + + for_each_rcu_flavor(rsp) + print_one_rcu_state(m, rsp); return 0; } @@ -377,11 +356,10 @@ static void show_one_rcugp(struct seq_file *m, struct rcu_state *rsp) static int show_rcugp(struct seq_file *m, void *unused) { -#ifdef CONFIG_TREE_PREEMPT_RCU - show_one_rcugp(m, &rcu_preempt_state); -#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ - show_one_rcugp(m, &rcu_sched_state); - show_one_rcugp(m, &rcu_bh_state); + struct rcu_state *rsp; + + for_each_rcu_flavor(rsp) + show_one_rcugp(m, rsp); return 0; } @@ -416,28 +394,20 @@ static void print_one_rcu_pending(struct seq_file *m, struct rcu_data *rdp) rdp->n_rp_need_nothing); } -static void print_rcu_pendings(struct seq_file *m, struct rcu_state *rsp) +static int show_rcu_pending(struct seq_file *m, void *unused) { int cpu; struct rcu_data *rdp; - - for_each_possible_cpu(cpu) { - rdp = per_cpu_ptr(rsp->rda, cpu); - if (rdp->beenonline) - print_one_rcu_pending(m, rdp); + struct rcu_state *rsp; + + for_each_rcu_flavor(rsp) { + seq_printf(m, "%s:\n", rsp->name); + for_each_possible_cpu(cpu) { + rdp = per_cpu_ptr(rsp->rda, cpu); + if (rdp->beenonline) + print_one_rcu_pending(m, rdp); + } } -} - -static int show_rcu_pending(struct seq_file *m, void *unused) -{ -#ifdef CONFIG_TREE_PREEMPT_RCU - seq_puts(m, "rcu_preempt:\n"); - print_rcu_pendings(m, &rcu_preempt_state); -#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ - seq_puts(m, "rcu_sched:\n"); - print_rcu_pendings(m, &rcu_sched_state); - seq_puts(m, "rcu_bh:\n"); - print_rcu_pendings(m, &rcu_bh_state); return 0; } -- cgit v1.2.3 From ff015030c939f0bec68fa9b8898da3aaa7fe55ea Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 12 Jun 2012 13:16:58 -0700 Subject: rcu: RCU_SAVE_DYNTICK code no longer ever dead Before RCU had unified idle, the RCU_SAVE_DYNTICK leg of the switch statement in force_quiescent_state() was dead code for CONFIG_NO_HZ=n kernel builds. With unified idle, the code is never dead. This commit therefore removes the "if" statement designed to make gcc aware of when the code was and was not dead. Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index b61c3ffc80e9..967b4bed2cf3 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1747,8 +1747,6 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) break; /* grace period idle or initializing, ignore. */ case RCU_SAVE_DYNTICK: - if (RCU_SIGNAL_INIT != RCU_SAVE_DYNTICK) - break; /* So gcc recognizes the dead code. */ raw_spin_unlock(&rnp->lock); /* irqs remain disabled */ -- cgit v1.2.3 From 751a68b2e46fe11a42450cb55a16e8065eddec7e Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 7 May 2012 13:44:44 -0700 Subject: rcu: Rationalize ordering of torture_ops list Move the raw SRCU interfaces out of the middle of the normal SRCU interfaces. Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutorture.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index e66b34ab7555..9850479f319d 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -1908,8 +1908,8 @@ rcu_torture_init(void) static struct rcu_torture_ops *torture_ops[] = { &rcu_ops, &rcu_sync_ops, &rcu_expedited_ops, &rcu_bh_ops, &rcu_bh_sync_ops, &rcu_bh_expedited_ops, - &srcu_ops, &srcu_sync_ops, &srcu_raw_ops, - &srcu_raw_sync_ops, &srcu_expedited_ops, + &srcu_ops, &srcu_sync_ops, &srcu_expedited_ops, + &srcu_raw_ops, &srcu_raw_sync_ops, &sched_ops, &sched_sync_ops, &sched_expedited_ops, }; mutex_lock(&fullstop_mutex); -- cgit v1.2.3 From e3f8d3788ed7cf55946030dc9b76e73edb111602 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 8 May 2012 10:21:50 -0700 Subject: rcu: Test srcu_barrier() from rcutorture test suite SRCU now has a call_srcu() and an srcu_barrier(), but rcutorture does not test them. This commit adds the machinery to allow rcutorture's existing tests for call_rcu() and rcu_barrier() to apply to the SRCU equivalents. Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutorture.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 9850479f319d..7b6935e0cee3 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -635,6 +635,17 @@ static void srcu_torture_synchronize(void) synchronize_srcu(&srcu_ctl); } +static void srcu_torture_call(struct rcu_head *head, + void (*func)(struct rcu_head *head)) +{ + call_srcu(&srcu_ctl, head, func); +} + +static void srcu_torture_barrier(void) +{ + srcu_barrier(&srcu_ctl); +} + static int srcu_torture_stats(char *page) { int cnt = 0; @@ -661,8 +672,8 @@ static struct rcu_torture_ops srcu_ops = { .completed = srcu_torture_completed, .deferred_free = srcu_torture_deferred_free, .sync = srcu_torture_synchronize, - .call = NULL, - .cb_barrier = NULL, + .call = srcu_torture_call, + .cb_barrier = srcu_torture_barrier, .stats = srcu_torture_stats, .name = "srcu" }; -- cgit v1.2.3 From c6ebcbb60c8c68a88160fe54302e851700d1362c Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 28 May 2012 19:21:41 -0700 Subject: rcu: Fix bug in rcu_barrier() torture test The child threads in the rcu_torture_barrier_cbs() are improperly synchronized, which can cause the rcu_barrier() tests to hang. The failure mode is as follows: 1. CPU 0 running in rcu_torture_barrier() sets barrier_cbs_count to n_barrier_cbs. 2. CPU 1 running in rcu_torture_barrier_cbs() wakes up, posts its RCU callback, and atomically decrements barrier_cbs_count. Because barrier_cbs_count is not zero, it does not do the wake_up(). 3. CPU 2 running in rcu_torture_barrier_cbs() wakes up, but finds that barrier_cbs_count is not equal to n_barrier_cbs, and so returns to sleep. 4. The value of barrier_cbs_count therefore never reaches zero, which causes the test to hang. This commit therefore uses a phase variable to coordinate the test, preventing this scenario from occurring. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutorture.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 7b6935e0cee3..f7fe73e59c9f 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -206,6 +206,7 @@ static unsigned long boost_starttime; /* jiffies of next boost test start. */ DEFINE_MUTEX(boost_mutex); /* protect setting boost_starttime */ /* and boost task create/destroy. */ static atomic_t barrier_cbs_count; /* Barrier callbacks registered. */ +static bool barrier_phase; /* Test phase. */ static atomic_t barrier_cbs_invoked; /* Barrier callbacks invoked. */ static wait_queue_head_t *barrier_cbs_wq; /* Coordinate barrier testing. */ static DECLARE_WAIT_QUEUE_HEAD(barrier_wq); @@ -1642,6 +1643,7 @@ void rcu_torture_barrier_cbf(struct rcu_head *rcu) static int rcu_torture_barrier_cbs(void *arg) { long myid = (long)arg; + bool lastphase = 0; struct rcu_head rcu; init_rcu_head_on_stack(&rcu); @@ -1649,9 +1651,11 @@ static int rcu_torture_barrier_cbs(void *arg) set_user_nice(current, 19); do { wait_event(barrier_cbs_wq[myid], - atomic_read(&barrier_cbs_count) == n_barrier_cbs || + barrier_phase != lastphase || kthread_should_stop() || fullstop != FULLSTOP_DONTSTOP); + lastphase = barrier_phase; + smp_mb(); /* ensure barrier_phase load before ->call(). */ if (kthread_should_stop() || fullstop != FULLSTOP_DONTSTOP) break; cur_ops->call(&rcu, rcu_torture_barrier_cbf); @@ -1676,7 +1680,8 @@ static int rcu_torture_barrier(void *arg) do { atomic_set(&barrier_cbs_invoked, 0); atomic_set(&barrier_cbs_count, n_barrier_cbs); - /* wake_up() path contains the required barriers. */ + smp_mb(); /* Ensure barrier_phase after prior assignments. */ + barrier_phase = !barrier_phase; for (i = 0; i < n_barrier_cbs; i++) wake_up(&barrier_cbs_wq[i]); wait_event(barrier_wq, -- cgit v1.2.3 From 143aa672f4fc643420c8325ad09c379ed33a27cf Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 24 May 2012 18:17:48 -0700 Subject: rcu: Fix diagnostic-printk typo in rcutorture The rcu_torture_barrier() function has a copy-and-paste typo in the string passed to rcutorture_shutdown_absorb(), which this commit fixes. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutorture.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index f7fe73e59c9f..045a3dc233ea 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -1700,7 +1700,7 @@ static int rcu_torture_barrier(void *arg) schedule_timeout_interruptible(HZ / 10); } while (!kthread_should_stop() && fullstop == FULLSTOP_DONTSTOP); VERBOSE_PRINTK_STRING("rcu_torture_barrier task stopping"); - rcutorture_shutdown_absorb("rcu_torture_barrier_cbs"); + rcutorture_shutdown_absorb("rcu_torture_barrier"); while (!kthread_should_stop()) schedule_timeout_interruptible(1); return 0; -- cgit v1.2.3 From 72472a02a9c4507ef54d03d71bb253c26015f52c Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 29 May 2012 17:50:51 -0700 Subject: rcu: Make rcutorture fakewriters invoke rcu_barrier() The current rcutorture rcu_barrier() testing never intentionally runs more than one instance of rcu_barrier() at a given time. This fails to test the the shiny new concurrency features of rcu_barrier(). This commit therefore modifies the rcutorture fakewriter kthread to randomly invoke rcu_barrier() rather than the usual synchronize_rcu(). Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutorture.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 045a3dc233ea..c279ee920947 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -1025,7 +1025,11 @@ rcu_torture_fakewriter(void *arg) do { schedule_timeout_uninterruptible(1 + rcu_random(&rand)%10); udelay(rcu_random(&rand) & 0x3ff); - cur_ops->sync(); + if (cur_ops->cb_barrier != NULL && + rcu_random(&rand) % (nfakewriters * 8) == 0) + cur_ops->cb_barrier(); + else + cur_ops->sync(); rcu_stutter_wait("rcu_torture_fakewriter"); } while (!kthread_should_stop() && fullstop == FULLSTOP_DONTSTOP); -- cgit v1.2.3 From 285fe29481d865ae381ad3924c80894e6968c2d8 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 9 May 2012 08:45:12 -0700 Subject: rcu: Fix detection of abruptly-ending stall The code that attempts to identify stalls that end just as we detect them is broken by both flavors of initialization failure. This commit therefore properly initializes and computes the count of the number of reasons why the RCU grace period is stalled. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 4b97bba7396e..dc8c5284fe06 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -733,7 +733,7 @@ static void print_other_cpu_stall(struct rcu_state *rsp) int cpu; long delta; unsigned long flags; - int ndetected; + int ndetected = 0; struct rcu_node *rnp = rcu_get_root(rsp); /* Only let one CPU complain about others per time interval. */ @@ -774,7 +774,7 @@ static void print_other_cpu_stall(struct rcu_state *rsp) */ rnp = rcu_get_root(rsp); raw_spin_lock_irqsave(&rnp->lock, flags); - ndetected = rcu_print_task_stall(rnp); + ndetected += rcu_print_task_stall(rnp); raw_spin_unlock_irqrestore(&rnp->lock, flags); print_cpu_stall_info_end(); -- cgit v1.2.3 From 3f5d3ea64f1783f0d4ea0d35890ae3297f045a8b Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 9 May 2012 15:39:56 -0700 Subject: rcu: Consolidate duplicate callback-list initialization There are a couple of open-coded initializations of the rcu_data structure's RCU callback list. This commit therefore consolidates them into a new init_callback_list() function. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index dc8c5284fe06..81e0394e46af 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -936,6 +936,18 @@ check_for_new_grace_period(struct rcu_state *rsp, struct rcu_data *rdp) return ret; } +/* + * Initialize the specified rcu_data structure's callback list to empty. + */ +static void init_callback_list(struct rcu_data *rdp) +{ + int i; + + rdp->nxtlist = NULL; + for (i = 0; i < RCU_NEXT_SIZE; i++) + rdp->nxttail[i] = &rdp->nxtlist; +} + /* * Advance this CPU's callbacks, but only if the current grace period * has ended. This may be called only from the CPU to whom the rdp @@ -1328,8 +1340,6 @@ static void rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) { - int i; - /* * Orphan the callbacks. First adjust the counts. This is safe * because ->onofflock excludes _rcu_barrier()'s adoption of @@ -1369,9 +1379,7 @@ rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp, } /* Finally, initialize the rcu_data structure's list to empty. */ - rdp->nxtlist = NULL; - for (i = 0; i < RCU_NEXT_SIZE; i++) - rdp->nxttail[i] = &rdp->nxtlist; + init_callback_list(rdp); } /* @@ -2407,16 +2415,13 @@ static void __init rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) { unsigned long flags; - int i; struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp = rcu_get_root(rsp); /* Set up local state, ensuring consistent view of global state. */ raw_spin_lock_irqsave(&rnp->lock, flags); rdp->grpmask = 1UL << (cpu - rdp->mynode->grplo); - rdp->nxtlist = NULL; - for (i = 0; i < RCU_NEXT_SIZE; i++) - rdp->nxttail[i] = &rdp->nxtlist; + init_callback_list(rdp); rdp->qlen_lazy = 0; rdp->qlen = 0; rdp->dynticks = &per_cpu(rcu_dynticks, cpu); -- cgit v1.2.3 From 1d1fb395f6dbc07b36285bbedcf01a73b57f7cb5 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 9 May 2012 15:44:42 -0700 Subject: rcu: Add ACCESS_ONCE() to ->qlen accesses The _rcu_barrier() function accesses other CPUs' rcu_data structure's ->qlen field without benefit of locking. This commit therefore adds the required ACCESS_ONCE() wrappers around accesses and updates that need it. ACCESS_ONCE() is not needed when a CPU accesses its own ->qlen, or in code that cannot run while _rcu_barrier() is sampling ->qlen fields. Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 81e0394e46af..89addada3e3a 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1350,7 +1350,7 @@ rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp, rsp->qlen += rdp->qlen; rdp->n_cbs_orphaned += rdp->qlen; rdp->qlen_lazy = 0; - rdp->qlen = 0; + ACCESS_ONCE(rdp->qlen) = 0; } /* @@ -1600,7 +1600,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) } smp_mb(); /* List handling before counting for rcu_barrier(). */ rdp->qlen_lazy -= count_lazy; - rdp->qlen -= count; + ACCESS_ONCE(rdp->qlen) -= count; rdp->n_cbs_invoked += count; /* Reinstate batch limit if we have worked down the excess. */ @@ -1889,7 +1889,7 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), rdp = this_cpu_ptr(rsp->rda); /* Add the callback to our list. */ - rdp->qlen++; + ACCESS_ONCE(rdp->qlen)++; if (lazy) rdp->qlen_lazy++; else @@ -2423,7 +2423,7 @@ rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) rdp->grpmask = 1UL << (cpu - rdp->mynode->grplo); init_callback_list(rdp); rdp->qlen_lazy = 0; - rdp->qlen = 0; + ACCESS_ONCE(rdp->qlen) = 0; rdp->dynticks = &per_cpu(rcu_dynticks, cpu); WARN_ON_ONCE(rdp->dynticks->dynticks_nesting != DYNTICK_TASK_EXIT_IDLE); WARN_ON_ONCE(atomic_read(&rdp->dynticks->dynticks) != 1); -- cgit v1.2.3 From 2a3fa843b555d202e682bf08c65ee1a4a93c79cf Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 21 May 2012 11:58:36 -0700 Subject: rcu: Consolidate tree/tiny __rcu_read_{,un}lock() implementations The CONFIG_TREE_PREEMPT_RCU and CONFIG_TINY_PREEMPT_RCU versions of __rcu_read_lock() and __rcu_read_unlock() are identical, so this commit consolidates them into kernel/rcupdate.h. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- include/linux/rcupdate.h | 1 + kernel/rcupdate.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ kernel/rcutiny_plugin.h | 47 +---------------------------------------------- kernel/rcutree_plugin.h | 47 +---------------------------------------------- 4 files changed, 47 insertions(+), 92 deletions(-) (limited to 'kernel') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index fb8e6db511d7..698555ebf49b 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -147,6 +147,7 @@ extern void synchronize_sched(void); extern void __rcu_read_lock(void); extern void __rcu_read_unlock(void); +extern void rcu_read_unlock_special(struct task_struct *t); void synchronize_rcu(void); /* diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index 95cba41ce1e9..4e6a61b15e86 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -53,6 +53,50 @@ #ifdef CONFIG_PREEMPT_RCU +/* + * Preemptible RCU implementation for rcu_read_lock(). + * Just increment ->rcu_read_lock_nesting, shared state will be updated + * if we block. + */ +void __rcu_read_lock(void) +{ + current->rcu_read_lock_nesting++; + barrier(); /* critical section after entry code. */ +} +EXPORT_SYMBOL_GPL(__rcu_read_lock); + +/* + * Preemptible RCU implementation for rcu_read_unlock(). + * Decrement ->rcu_read_lock_nesting. If the result is zero (outermost + * rcu_read_unlock()) and ->rcu_read_unlock_special is non-zero, then + * invoke rcu_read_unlock_special() to clean up after a context switch + * in an RCU read-side critical section and other special cases. + */ +void __rcu_read_unlock(void) +{ + struct task_struct *t = current; + + if (t->rcu_read_lock_nesting != 1) { + --t->rcu_read_lock_nesting; + } else { + barrier(); /* critical section before exit code. */ + t->rcu_read_lock_nesting = INT_MIN; + barrier(); /* assign before ->rcu_read_unlock_special load */ + if (unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) + rcu_read_unlock_special(t); + barrier(); /* ->rcu_read_unlock_special load before assign */ + t->rcu_read_lock_nesting = 0; + } +#ifdef CONFIG_PROVE_LOCKING + { + int rrln = ACCESS_ONCE(t->rcu_read_lock_nesting); + + WARN_ON_ONCE(rrln < 0 && rrln > INT_MIN / 2); + } +#endif /* #ifdef CONFIG_PROVE_LOCKING */ +} +EXPORT_SYMBOL_GPL(__rcu_read_unlock); + /* * Check for a task exiting while in a preemptible-RCU read-side * critical section, clean up if so. No need to issue warnings, diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h index fc31a2d65100..a269b0da0eb6 100644 --- a/kernel/rcutiny_plugin.h +++ b/kernel/rcutiny_plugin.h @@ -132,7 +132,6 @@ static struct rcu_preempt_ctrlblk rcu_preempt_ctrlblk = { RCU_TRACE(.rcb.name = "rcu_preempt") }; -static void rcu_read_unlock_special(struct task_struct *t); static int rcu_preempted_readers_exp(void); static void rcu_report_exp_done(void); @@ -526,24 +525,12 @@ void rcu_preempt_note_context_switch(void) local_irq_restore(flags); } -/* - * Tiny-preemptible RCU implementation for rcu_read_lock(). - * Just increment ->rcu_read_lock_nesting, shared state will be updated - * if we block. - */ -void __rcu_read_lock(void) -{ - current->rcu_read_lock_nesting++; - barrier(); /* needed if we ever invoke rcu_read_lock in rcutiny.c */ -} -EXPORT_SYMBOL_GPL(__rcu_read_lock); - /* * Handle special cases during rcu_read_unlock(), such as needing to * notify RCU core processing or task having blocked during the RCU * read-side critical section. */ -static noinline void rcu_read_unlock_special(struct task_struct *t) +void rcu_read_unlock_special(struct task_struct *t) { int empty; int empty_exp; @@ -626,38 +613,6 @@ static noinline void rcu_read_unlock_special(struct task_struct *t) local_irq_restore(flags); } -/* - * Tiny-preemptible RCU implementation for rcu_read_unlock(). - * Decrement ->rcu_read_lock_nesting. If the result is zero (outermost - * rcu_read_unlock()) and ->rcu_read_unlock_special is non-zero, then - * invoke rcu_read_unlock_special() to clean up after a context switch - * in an RCU read-side critical section and other special cases. - */ -void __rcu_read_unlock(void) -{ - struct task_struct *t = current; - - barrier(); /* needed if we ever invoke rcu_read_unlock in rcutiny.c */ - if (t->rcu_read_lock_nesting != 1) - --t->rcu_read_lock_nesting; - else { - t->rcu_read_lock_nesting = INT_MIN; - barrier(); /* assign before ->rcu_read_unlock_special load */ - if (unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) - rcu_read_unlock_special(t); - barrier(); /* ->rcu_read_unlock_special load before assign */ - t->rcu_read_lock_nesting = 0; - } -#ifdef CONFIG_PROVE_LOCKING - { - int rrln = ACCESS_ONCE(t->rcu_read_lock_nesting); - - WARN_ON_ONCE(rrln < 0 && rrln > INT_MIN / 2); - } -#endif /* #ifdef CONFIG_PROVE_LOCKING */ -} -EXPORT_SYMBOL_GPL(__rcu_read_unlock); - /* * Check for a quiescent state from the current CPU. When a task blocks, * the task is recorded in the rcu_preempt_ctrlblk structure, which is diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 3e4899459f3d..4b6b17cdf66b 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -78,7 +78,6 @@ struct rcu_state rcu_preempt_state = RCU_STATE_INITIALIZER(rcu_preempt); DEFINE_PER_CPU(struct rcu_data, rcu_preempt_data); static struct rcu_state *rcu_state = &rcu_preempt_state; -static void rcu_read_unlock_special(struct task_struct *t); static int rcu_preempted_readers_exp(struct rcu_node *rnp); /* @@ -232,18 +231,6 @@ static void rcu_preempt_note_context_switch(int cpu) local_irq_restore(flags); } -/* - * Tree-preemptible RCU implementation for rcu_read_lock(). - * Just increment ->rcu_read_lock_nesting, shared state will be updated - * if we block. - */ -void __rcu_read_lock(void) -{ - current->rcu_read_lock_nesting++; - barrier(); /* needed if we ever invoke rcu_read_lock in rcutree.c */ -} -EXPORT_SYMBOL_GPL(__rcu_read_lock); - /* * Check for preempted RCU readers blocking the current grace period * for the specified rcu_node structure. If the caller needs a reliable @@ -310,7 +297,7 @@ static struct list_head *rcu_next_node_entry(struct task_struct *t, * notify RCU core processing or task having blocked during the RCU * read-side critical section. */ -static noinline void rcu_read_unlock_special(struct task_struct *t) +void rcu_read_unlock_special(struct task_struct *t) { int empty; int empty_exp; @@ -418,38 +405,6 @@ static noinline void rcu_read_unlock_special(struct task_struct *t) } } -/* - * Tree-preemptible RCU implementation for rcu_read_unlock(). - * Decrement ->rcu_read_lock_nesting. If the result is zero (outermost - * rcu_read_unlock()) and ->rcu_read_unlock_special is non-zero, then - * invoke rcu_read_unlock_special() to clean up after a context switch - * in an RCU read-side critical section and other special cases. - */ -void __rcu_read_unlock(void) -{ - struct task_struct *t = current; - - if (t->rcu_read_lock_nesting != 1) - --t->rcu_read_lock_nesting; - else { - barrier(); /* critical section before exit code. */ - t->rcu_read_lock_nesting = INT_MIN; - barrier(); /* assign before ->rcu_read_unlock_special load */ - if (unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) - rcu_read_unlock_special(t); - barrier(); /* ->rcu_read_unlock_special load before assign */ - t->rcu_read_lock_nesting = 0; - } -#ifdef CONFIG_PROVE_LOCKING - { - int rrln = ACCESS_ONCE(t->rcu_read_lock_nesting); - - WARN_ON_ONCE(rrln < 0 && rrln > INT_MIN / 2); - } -#endif /* #ifdef CONFIG_PROVE_LOCKING */ -} -EXPORT_SYMBOL_GPL(__rcu_read_unlock); - #ifdef CONFIG_RCU_CPU_STALL_VERBOSE /* -- cgit v1.2.3 From 62fde6edf12b60fddb13a3f0a779c8be0bb7447e Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 22 May 2012 22:10:24 -0700 Subject: rcu: Make __call_rcu() handle invocation from idle Although __call_rcu() is handled correctly when called from a momentary non-idle period, if it is called on a CPU that RCU believes to be idle on RCU_FAST_NO_HZ kernels, the callback might be indefinitely postponed. This commit therefore ensures that RCU is aware of the new callback and has a chance to force the CPU out of dyntick-idle mode when a new callback is posted. Reported-by: Frederic Weisbecker Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- include/linux/rcupdate.h | 13 ++++--------- kernel/rcutree.c | 15 +++++++++------ 2 files changed, 13 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 31568c734525..26f6417f0264 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -256,6 +256,10 @@ static inline void destroy_rcu_head_on_stack(struct rcu_head *head) } #endif /* #else !CONFIG_DEBUG_OBJECTS_RCU_HEAD */ +#if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_SMP) +extern int rcu_is_cpu_idle(void); +#endif /* #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_SMP) */ + #if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PROVE_RCU) bool rcu_lockdep_current_cpu_online(void); #else /* #if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PROVE_RCU) */ @@ -267,15 +271,6 @@ static inline bool rcu_lockdep_current_cpu_online(void) #ifdef CONFIG_DEBUG_LOCK_ALLOC -#ifdef CONFIG_PROVE_RCU -extern int rcu_is_cpu_idle(void); -#else /* !CONFIG_PROVE_RCU */ -static inline int rcu_is_cpu_idle(void) -{ - return 0; -} -#endif /* else !CONFIG_PROVE_RCU */ - static inline void rcu_lock_acquire(struct lockdep_map *map) { lock_acquire(map, 0, 0, 2, 1, NULL, _THIS_IP_); diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 89addada3e3a..a4a9c916ad36 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -585,8 +585,6 @@ void rcu_nmi_exit(void) WARN_ON_ONCE(atomic_read(&rdtp->dynticks) & 0x1); } -#ifdef CONFIG_PROVE_RCU - /** * rcu_is_cpu_idle - see if RCU thinks that the current CPU is idle * @@ -604,7 +602,7 @@ int rcu_is_cpu_idle(void) } EXPORT_SYMBOL(rcu_is_cpu_idle); -#ifdef CONFIG_HOTPLUG_CPU +#if defined(CONFIG_PROVE_RCU) && defined(CONFIG_HOTPLUG_CPU) /* * Is the current CPU online? Disable preemption to avoid false positives @@ -645,9 +643,7 @@ bool rcu_lockdep_current_cpu_online(void) } EXPORT_SYMBOL_GPL(rcu_lockdep_current_cpu_online); -#endif /* #ifdef CONFIG_HOTPLUG_CPU */ - -#endif /* #ifdef CONFIG_PROVE_RCU */ +#endif /* #if defined(CONFIG_PROVE_RCU) && defined(CONFIG_HOTPLUG_CPU) */ /** * rcu_is_cpu_rrupt_from_idle - see if idle or immediately interrupted from idle @@ -1904,6 +1900,13 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), else trace_rcu_callback(rsp->name, head, rdp->qlen_lazy, rdp->qlen); + /* + * If called from an extended quiescent state, invoke the RCU + * core in order to force a re-evaluation of RCU's idleness. + */ + if (rcu_is_cpu_idle()) + invoke_rcu_core(); + /* If interrupts were disabled, don't dive into RCU core. */ if (irqs_disabled_flags(flags)) { local_irq_restore(flags); -- cgit v1.2.3 From a16b7a693430406dc229ab0c6b154f669a2031c5 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 26 May 2012 08:56:01 -0700 Subject: rcu: Prevent __call_rcu() from invoking RCU core on offline CPUs The __call_rcu() function will invoke the RCU core, for example, if it detects that the current CPU has too many callbacks. However, this can happen on an offline CPU that is on its way to the idle loop, in which case it is an error to invoke the RCU core, and the excess callbacks will be adopted in any case. This commit therefore adds checks to __call_rcu() for running on an offline CPU, refraining from invoking the RCU core in this case. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index a4a9c916ad36..ceaa95923a87 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1904,11 +1904,11 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), * If called from an extended quiescent state, invoke the RCU * core in order to force a re-evaluation of RCU's idleness. */ - if (rcu_is_cpu_idle()) + if (rcu_is_cpu_idle() && cpu_online(smp_processor_id())) invoke_rcu_core(); - /* If interrupts were disabled, don't dive into RCU core. */ - if (irqs_disabled_flags(flags)) { + /* If interrupts were disabled or CPU offline, don't invoke RCU core. */ + if (irqs_disabled_flags(flags) || cpu_is_offline(smp_processor_id())) { local_irq_restore(flags); return; } -- cgit v1.2.3 From 29154c57e35a191c83b19c61b1935c9f21957662 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 30 May 2012 03:21:48 -0700 Subject: rcu: Split RCU core processing out of __call_rcu() The __call_rcu() function is a bit overweight, so this commit splits it into actual enqueuing of and accounting for the callback (__call_rcu()) and associated RCU-core processing (__call_rcu_core()). Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 90 ++++++++++++++++++++++++++++++-------------------------- 1 file changed, 49 insertions(+), 41 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index ceaa95923a87..70c4da7d2a97 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1861,45 +1861,12 @@ static void invoke_rcu_core(void) raise_softirq(RCU_SOFTIRQ); } -static void -__call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), - struct rcu_state *rsp, bool lazy) +/* + * Handle any core-RCU processing required by a call_rcu() invocation. + */ +static void __call_rcu_core(struct rcu_state *rsp, struct rcu_data *rdp, + struct rcu_head *head, unsigned long flags) { - unsigned long flags; - struct rcu_data *rdp; - - WARN_ON_ONCE((unsigned long)head & 0x3); /* Misaligned rcu_head! */ - debug_rcu_head_queue(head); - head->func = func; - head->next = NULL; - - smp_mb(); /* Ensure RCU update seen before callback registry. */ - - /* - * Opportunistically note grace-period endings and beginnings. - * Note that we might see a beginning right after we see an - * end, but never vice versa, since this CPU has to pass through - * a quiescent state betweentimes. - */ - local_irq_save(flags); - rdp = this_cpu_ptr(rsp->rda); - - /* Add the callback to our list. */ - ACCESS_ONCE(rdp->qlen)++; - if (lazy) - rdp->qlen_lazy++; - else - rcu_idle_count_callbacks_posted(); - smp_mb(); /* Count before adding callback for rcu_barrier(). */ - *rdp->nxttail[RCU_NEXT_TAIL] = head; - rdp->nxttail[RCU_NEXT_TAIL] = &head->next; - - if (__is_kfree_rcu_offset((unsigned long)func)) - trace_rcu_kfree_callback(rsp->name, head, (unsigned long)func, - rdp->qlen_lazy, rdp->qlen); - else - trace_rcu_callback(rsp->name, head, rdp->qlen_lazy, rdp->qlen); - /* * If called from an extended quiescent state, invoke the RCU * core in order to force a re-evaluation of RCU's idleness. @@ -1908,10 +1875,8 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), invoke_rcu_core(); /* If interrupts were disabled or CPU offline, don't invoke RCU core. */ - if (irqs_disabled_flags(flags) || cpu_is_offline(smp_processor_id())) { - local_irq_restore(flags); + if (irqs_disabled_flags(flags) || cpu_is_offline(smp_processor_id())) return; - } /* * Force the grace period if too many callbacks or too long waiting. @@ -1944,6 +1909,49 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), } } else if (ULONG_CMP_LT(ACCESS_ONCE(rsp->jiffies_force_qs), jiffies)) force_quiescent_state(rsp, 1); +} + +static void +__call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), + struct rcu_state *rsp, bool lazy) +{ + unsigned long flags; + struct rcu_data *rdp; + + WARN_ON_ONCE((unsigned long)head & 0x3); /* Misaligned rcu_head! */ + debug_rcu_head_queue(head); + head->func = func; + head->next = NULL; + + smp_mb(); /* Ensure RCU update seen before callback registry. */ + + /* + * Opportunistically note grace-period endings and beginnings. + * Note that we might see a beginning right after we see an + * end, but never vice versa, since this CPU has to pass through + * a quiescent state betweentimes. + */ + local_irq_save(flags); + rdp = this_cpu_ptr(rsp->rda); + + /* Add the callback to our list. */ + ACCESS_ONCE(rdp->qlen)++; + if (lazy) + rdp->qlen_lazy++; + else + rcu_idle_count_callbacks_posted(); + smp_mb(); /* Count before adding callback for rcu_barrier(). */ + *rdp->nxttail[RCU_NEXT_TAIL] = head; + rdp->nxttail[RCU_NEXT_TAIL] = &head->next; + + if (__is_kfree_rcu_offset((unsigned long)func)) + trace_rcu_kfree_callback(rsp->name, head, (unsigned long)func, + rdp->qlen_lazy, rdp->qlen); + else + trace_rcu_callback(rsp->name, head, rdp->qlen_lazy, rdp->qlen); + + /* Go handle any RCU core processing required. */ + __call_rcu_core(rsp, rdp, head, flags); local_irq_restore(flags); } -- cgit v1.2.3 From d7118175cce7e76b3292b60832813ef1f28b6523 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 6 Jun 2012 07:12:43 -0700 Subject: rcu: Fix rcu_is_cpu_idle() #ifdef in TINY_RCU The rcu_is_cpu_idle() function is used if CONFIG_DEBUG_LOCK_ALLOC, but TINY_RCU defines it only when CONFIG_PROVE_RCU. This causes build failures when CONFIG_DEBUG_LOCK_ALLOC=y but CONFIG_PROVE_RCU=n. This commit therefore adjusts the #ifdefs for rcu_is_cpu_idle() so that it is defined when CONFIG_DEBUG_LOCK_ALLOC=y. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutiny.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c index 37a5444204d2..547b1fe5b052 100644 --- a/kernel/rcutiny.c +++ b/kernel/rcutiny.c @@ -172,7 +172,7 @@ void rcu_irq_enter(void) local_irq_restore(flags); } -#ifdef CONFIG_PROVE_RCU +#ifdef CONFIG_DEBUG_LOCK_ALLOC /* * Test whether RCU thinks that the current CPU is idle. @@ -183,7 +183,7 @@ int rcu_is_cpu_idle(void) } EXPORT_SYMBOL(rcu_is_cpu_idle); -#endif /* #ifdef CONFIG_PROVE_RCU */ +#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ /* * Test whether the current CPU was interrupted from idle. Nested -- cgit v1.2.3 From 1c17e4d4437d8045a596d9f06c1558dc09e2b372 Mon Sep 17 00:00:00 2001 From: Carsten Emde Date: Tue, 19 Jun 2012 10:43:16 +0200 Subject: rcu: Prevent uninitialized string in RCU CPU stall info An uninitialized string may be displayed at the end of the rcu_preempt detected stall info such as 0: (1 GPs behind) idle=075/140000000000000/0 =8?^D=8?^D ^^^^^^^^^^ if CONFIG_RCU_FAST_NO_HZ is not defined. This trivial patch clears the string in this case. Signed-off-by: Carsten Emde Signed-off-by: Paul E. McKenney --- kernel/rcutree_plugin.h | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 4b6b17cdf66b..395cdd1e0634 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -2224,6 +2224,7 @@ static void print_cpu_stall_fast_no_hz(char *cp, int cpu) static void print_cpu_stall_fast_no_hz(char *cp, int cpu) { + *cp = '\0'; } #endif /* #else #ifdef CONFIG_RCU_FAST_NO_HZ */ -- cgit v1.2.3 From 95f0c1de3e6ed4383cc4b5f52ce4ecfb21026b49 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 19 Jun 2012 11:58:27 -0700 Subject: rcu: Disable preemption in rcu_blocking_is_gp() It is time to optimize CONFIG_TREE_PREEMPT_RCU's synchronize_rcu() for uniprocessor optimization, which means that rcu_blocking_is_gp() can no longer rely on RCU read-side critical sections having disabled preemption. This commit therefore disables preemption across rcu_blocking_is_gp()'s scan of the cpu_online_mask. (Updated from previous version to fix embarrassing bug spotted by Wu Fengguang.) Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 24 ++++++------------------ 1 file changed, 6 insertions(+), 18 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 70c4da7d2a97..e000a623e635 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1981,28 +1981,16 @@ EXPORT_SYMBOL_GPL(call_rcu_bh); * occasionally incorrectly indicate that there are multiple CPUs online * when there was in fact only one the whole time, as this just adds * some overhead: RCU still operates correctly. - * - * Of course, sampling num_online_cpus() with preemption enabled can - * give erroneous results if there are concurrent CPU-hotplug operations. - * For example, given a demonic sequence of preemptions in num_online_cpus() - * and CPU-hotplug operations, there could be two or more CPUs online at - * all times, but num_online_cpus() might well return one (or even zero). - * - * However, all such demonic sequences require at least one CPU-offline - * operation. Furthermore, rcu_blocking_is_gp() giving the wrong answer - * is only a problem if there is an RCU read-side critical section executing - * throughout. But RCU-sched and RCU-bh read-side critical sections - * disable either preemption or bh, which prevents a CPU from going offline. - * Therefore, the only way that rcu_blocking_is_gp() can incorrectly return - * that there is only one CPU when in fact there was more than one throughout - * is when there were no RCU readers in the system. If there are no - * RCU readers, the grace period by definition can be of zero length, - * regardless of the number of online CPUs. */ static inline int rcu_blocking_is_gp(void) { + int ret; + might_sleep(); /* Check for RCU read-side critical section. */ - return num_online_cpus() <= 1; + preempt_disable(); + ret = num_online_cpus() <= 1; + preempt_enable(); + return ret; } /** -- cgit v1.2.3 From cf01537ecf192d9ff15c32a355db5d5af22bea4d Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 21 Jun 2012 11:26:42 -0700 Subject: rcu: Add check for CPUs going offline with callbacks queued If a CPU goes offline with callbacks queued, those callbacks might be indefinitely postponed, which can result in a system hang. This commit therefore inserts warnings for this condition. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index e000a623e635..95c7b61e77e4 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1509,6 +1509,9 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp) raw_spin_unlock_irqrestore(&rnp->lock, flags); if (need_report & RCU_OFL_TASKS_EXP_GP) rcu_report_exp_rnp(rsp, rnp, true); + WARN_ONCE(rdp->qlen != 0 || rdp->nxtlist != NULL, + "rcu_cleanup_dead_cpu: Callbacks on offline CPU %d: qlen=%lu, nxtlist=%p\n", + cpu, rdp->qlen, rdp->nxtlist); } #else /* #ifdef CONFIG_HOTPLUG_CPU */ -- cgit v1.2.3 From bf1304e9cd755be7814ac8365834b7a1d0f06b58 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 9 May 2012 15:55:39 -0700 Subject: rcu: Dump only the current CPU's buffers for idle-entry/exit warnings Problems in RCU idle entry and exit are almost always confined to the offending CPU. This commit therefore switches ftrace_dump() from DUMP_ALL to DUMP_ORIG. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Tested-by: Heiko Carstens Tested-by: Pascal Chapperon --- kernel/rcutree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 4b97bba7396e..6eb48f13eeeb 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -358,7 +358,7 @@ static void rcu_idle_enter_common(struct rcu_dynticks *rdtp, long long oldval) struct task_struct *idle = idle_task(smp_processor_id()); trace_rcu_dyntick("Error on entry: not idle task", oldval, 0); - ftrace_dump(DUMP_ALL); + ftrace_dump(DUMP_ORIG); WARN_ONCE(1, "Current pid: %d comm: %s / Idle pid: %d comm: %s", current->pid, current->comm, idle->pid, idle->comm); /* must be idle task! */ @@ -468,7 +468,7 @@ static void rcu_idle_exit_common(struct rcu_dynticks *rdtp, long long oldval) trace_rcu_dyntick("Error on exit: not idle task", oldval, rdtp->dynticks_nesting); - ftrace_dump(DUMP_ALL); + ftrace_dump(DUMP_ORIG); WARN_ONCE(1, "Current pid: %d comm: %s / Idle pid: %d comm: %s", current->pid, current->comm, idle->pid, idle->comm); /* must be idle task! */ -- cgit v1.2.3 From 28f8555364ef5b54b21251c5c8022109a70626e9 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 10 May 2012 15:37:48 -0700 Subject: rcu: The rcu_needs_cpu() function is not a quiescent state The TINY_PREEMPT_RCU() function rcu_preempt_needs_cpu(), which is called from rcu_needs_cpu(), assumes that it is in a quiescent state with respect to the CPU. This is no longer the case. This commit therefore updates rcu_preempt_needs_cpu() to make it aware that it is not running in a quiescent state. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Tested-by: Heiko Carstens Tested-by: Pascal Chapperon --- kernel/rcutiny_plugin.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h index fc31a2d65100..849ede51c12d 100644 --- a/kernel/rcutiny_plugin.h +++ b/kernel/rcutiny_plugin.h @@ -846,8 +846,6 @@ EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); */ int rcu_preempt_needs_cpu(void) { - if (!rcu_preempt_running_reader()) - rcu_preempt_cpu_qs(); return rcu_preempt_ctrlblk.rcb.rcucblist != NULL; } -- cgit v1.2.3 From e84c48ae3024ac2f14ed1c3671e5ea37c60fb838 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 4 Jun 2012 20:45:10 -0700 Subject: rcu: Round FAST_NO_HZ lazy timeout to nearest second Currently, if several CPUs in the same package have all lazy RCU callbacks, their wakeups will be uncorrelated. If all the CPUs are in the same power domain (as is often the case), this will result in unnecessary power-ups of the package. This commit therefore uses round_jiffies() to round the timeouts to a second boundary, increasing the odds that they can be coalesced with each other or with other timeouts. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree_plugin.h | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 3e4899459f3d..c28d25522a81 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -1968,7 +1968,7 @@ static void rcu_idle_count_callbacks_posted(void) */ #define RCU_IDLE_FLUSHES 5 /* Number of dyntick-idle tries. */ #define RCU_IDLE_OPT_FLUSHES 3 /* Optional dyntick-idle tries. */ -#define RCU_IDLE_GP_DELAY 6 /* Roughly one grace period. */ +#define RCU_IDLE_GP_DELAY 4 /* Roughly one grace period. */ #define RCU_IDLE_LAZY_GP_DELAY (6 * HZ) /* Roughly six seconds. */ /* @@ -2047,10 +2047,13 @@ int rcu_needs_cpu(int cpu, unsigned long *delta_jiffies) return 1; } /* Set up for the possibility that RCU will post a timer. */ - if (rcu_cpu_has_nonlazy_callbacks(cpu)) - *delta_jiffies = RCU_IDLE_GP_DELAY; - else - *delta_jiffies = RCU_IDLE_LAZY_GP_DELAY; + if (rcu_cpu_has_nonlazy_callbacks(cpu)) { + *delta_jiffies = round_up(RCU_IDLE_GP_DELAY + jiffies, + RCU_IDLE_GP_DELAY) - jiffies; + } else { + *delta_jiffies = jiffies + RCU_IDLE_LAZY_GP_DELAY; + *delta_jiffies = round_jiffies(*delta_jiffies) - jiffies; + } return 0; } @@ -2187,10 +2190,11 @@ static void rcu_prepare_for_idle(int cpu) if (rcu_cpu_has_nonlazy_callbacks(cpu)) { trace_rcu_prep_idle("Dyntick with callbacks"); rdtp->idle_gp_timer_expires = - jiffies + RCU_IDLE_GP_DELAY; + round_up(jiffies + RCU_IDLE_GP_DELAY, + RCU_IDLE_GP_DELAY); } else { rdtp->idle_gp_timer_expires = - jiffies + RCU_IDLE_LAZY_GP_DELAY; + round_jiffies(jiffies + RCU_IDLE_LAZY_GP_DELAY); trace_rcu_prep_idle("Dyntick with lazy callbacks"); } tp = &rdtp->idle_gp_timer; -- cgit v1.2.3 From 9d2ad24306f2fafc3612e5a216aab31f9e56e879 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 24 Jun 2012 10:15:02 -0700 Subject: rcu: Make RCU_FAST_NO_HZ respect nohz= boot parameter If the nohz= boot parameter disables nohz, then RCU_FAST_NO_HZ needs to also disable itself. This commit therefore checks for tick_nohz_enabled being zero, disabling rcu_prepare_for_idle() if so. This commit assumes that tick_nohz_enabled can change at runtime: If this is not the case, then a simpler approach suffices. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- kernel/rcutree.h | 1 + kernel/rcutree_plugin.h | 15 +++++++++++++++ kernel/time/tick-sched.c | 2 +- 3 files changed, 17 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 19b61ac1079f..e978845c6800 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -97,6 +97,7 @@ struct rcu_dynticks { /* # times non-lazy CBs posted to CPU. */ unsigned long nonlazy_posted_snap; /* idle-period nonlazy_posted snapshot. */ + int tick_nohz_enabled_snap; /* Previously seen value from sysfs. */ #endif /* #ifdef CONFIG_RCU_FAST_NO_HZ */ }; diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index c28d25522a81..3508000e077d 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -1971,6 +1971,8 @@ static void rcu_idle_count_callbacks_posted(void) #define RCU_IDLE_GP_DELAY 4 /* Roughly one grace period. */ #define RCU_IDLE_LAZY_GP_DELAY (6 * HZ) /* Roughly six seconds. */ +extern int tick_nohz_enabled; + /* * Does the specified flavor of RCU have non-lazy callbacks pending on * the specified CPU? Both RCU flavor and CPU are specified by the @@ -2112,6 +2114,7 @@ static void rcu_cleanup_after_idle(int cpu) del_timer(&rdtp->idle_gp_timer); trace_rcu_prep_idle("Cleanup after idle"); + rdtp->tick_nohz_enabled_snap = ACCESS_ONCE(tick_nohz_enabled); } /* @@ -2137,6 +2140,18 @@ static void rcu_prepare_for_idle(int cpu) { struct timer_list *tp; struct rcu_dynticks *rdtp = &per_cpu(rcu_dynticks, cpu); + int tne; + + /* Handle nohz enablement switches conservatively. */ + tne = ACCESS_ONCE(tick_nohz_enabled); + if (tne != rdtp->tick_nohz_enabled_snap) { + if (rcu_cpu_has_callbacks(cpu)) + invoke_rcu_core(); /* force nohz to see update. */ + rdtp->tick_nohz_enabled_snap = tne; + return; + } + if (!tne) + return; /* * If this is an idle re-entry, for example, due to use of diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 869997833928..66ff07f6184c 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -105,7 +105,7 @@ static ktime_t tick_init_jiffy_update(void) /* * NO HZ enabled ? */ -static int tick_nohz_enabled __read_mostly = 1; +int tick_nohz_enabled __read_mostly = 1; /* * Enable / Disable tickless mode -- cgit v1.2.3 From 74a7f08448adea6cb47cd9b260c98ff168117e92 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Fri, 15 Jun 2012 11:50:25 -0600 Subject: devicetree: add helper inline for retrieving a node's full name The pattern (np ? np->full_name : "") is rather common in the kernel, but can also make for quite long lines. This patch adds a new inline function, of_node_full_name() so that the test for a valid node pointer doesn't need to be open coded at all call sites. Signed-off-by: Grant Likely Cc: Paul Mundt Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Signed-off-by: Rob Herring --- arch/microblaze/pci/pci-common.c | 6 ++---- arch/powerpc/kernel/pci-common.c | 6 ++---- arch/powerpc/kernel/vio.c | 5 ++--- arch/powerpc/platforms/cell/iommu.c | 3 +-- arch/powerpc/platforms/pseries/iommu.c | 2 +- arch/sparc/kernel/of_device_64.c | 2 +- drivers/of/base.c | 2 +- drivers/of/irq.c | 2 +- include/linux/of.h | 10 ++++++++++ kernel/irq/irqdomain.c | 8 ++++---- 10 files changed, 25 insertions(+), 21 deletions(-) (limited to 'kernel') diff --git a/arch/microblaze/pci/pci-common.c b/arch/microblaze/pci/pci-common.c index ed22bfc5db14..ca8f6e769960 100644 --- a/arch/microblaze/pci/pci-common.c +++ b/arch/microblaze/pci/pci-common.c @@ -249,8 +249,7 @@ int pci_read_irq_line(struct pci_dev *pci_dev) } else { pr_debug(" Got one, spec %d cells (0x%08x 0x%08x...) on %s\n", oirq.size, oirq.specifier[0], oirq.specifier[1], - oirq.controller ? oirq.controller->full_name : - ""); + of_node_full_name(oirq.controller)); virq = irq_create_of_mapping(oirq.controller, oirq.specifier, oirq.size); @@ -1493,8 +1492,7 @@ static void __devinit pcibios_scan_phb(struct pci_controller *hose) struct pci_bus *bus; struct device_node *node = hose->dn; - pr_debug("PCI: Scanning PHB %s\n", - node ? node->full_name : ""); + pr_debug("PCI: Scanning PHB %s\n", of_node_full_name(node)); pcibios_setup_phb_resources(hose, &resources); diff --git a/arch/powerpc/kernel/pci-common.c b/arch/powerpc/kernel/pci-common.c index 8e78e93c8185..886c254fd565 100644 --- a/arch/powerpc/kernel/pci-common.c +++ b/arch/powerpc/kernel/pci-common.c @@ -248,8 +248,7 @@ static int pci_read_irq_line(struct pci_dev *pci_dev) } else { pr_debug(" Got one, spec %d cells (0x%08x 0x%08x...) on %s\n", oirq.size, oirq.specifier[0], oirq.specifier[1], - oirq.controller ? oirq.controller->full_name : - ""); + of_node_full_name(oirq.controller)); virq = irq_create_of_mapping(oirq.controller, oirq.specifier, oirq.size); @@ -1628,8 +1627,7 @@ void __devinit pcibios_scan_phb(struct pci_controller *hose) struct device_node *node = hose->dn; int mode; - pr_debug("PCI: Scanning PHB %s\n", - node ? node->full_name : ""); + pr_debug("PCI: Scanning PHB %s\n", of_node_full_name(node)); /* Get some IO space for the new PHB */ pcibios_setup_phb_io_space(hose); diff --git a/arch/powerpc/kernel/vio.c b/arch/powerpc/kernel/vio.c index cb87301ccd55..63f72ede4341 100644 --- a/arch/powerpc/kernel/vio.c +++ b/arch/powerpc/kernel/vio.c @@ -1296,8 +1296,7 @@ static void __devinit vio_dev_release(struct device *dev) struct iommu_table *tbl = get_iommu_table_base(dev); if (tbl) - iommu_free_table(tbl, dev->of_node ? - dev->of_node->full_name : dev_name(dev)); + iommu_free_table(tbl, of_node_full_name(dev->of_node)); of_node_put(dev->of_node); kfree(to_vio_dev(dev)); } @@ -1509,7 +1508,7 @@ static ssize_t devspec_show(struct device *dev, { struct device_node *of_node = dev->of_node; - return sprintf(buf, "%s\n", of_node ? of_node->full_name : "none"); + return sprintf(buf, "%s\n", of_node_full_name(of_node)); } static ssize_t modalias_show(struct device *dev, struct device_attribute *attr, diff --git a/arch/powerpc/platforms/cell/iommu.c b/arch/powerpc/platforms/cell/iommu.c index b9f509a34c01..b6732004c882 100644 --- a/arch/powerpc/platforms/cell/iommu.c +++ b/arch/powerpc/platforms/cell/iommu.c @@ -552,8 +552,7 @@ static struct iommu_table *cell_get_iommu_table(struct device *dev) iommu = cell_iommu_for_node(dev_to_node(dev)); if (iommu == NULL || list_empty(&iommu->windows)) { printk(KERN_ERR "iommu: missing iommu for %s (node %d)\n", - dev->of_node ? dev->of_node->full_name : "?", - dev_to_node(dev)); + of_node_full_name(dev->of_node), dev_to_node(dev)); return NULL; } window = list_entry(iommu->windows.next, struct iommu_window, list); diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c index 2d311c0caf8e..6b58a395dff6 100644 --- a/arch/powerpc/platforms/pseries/iommu.c +++ b/arch/powerpc/platforms/pseries/iommu.c @@ -1051,7 +1051,7 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev) if (!pdn || !PCI_DN(pdn)) { printk(KERN_WARNING "pci_dma_dev_setup_pSeriesLP: " "no DMA window found for pci dev=%s dn=%s\n", - pci_name(dev), dn? dn->full_name : ""); + pci_name(dev), of_node_full_name(dn)); return; } pr_debug(" parent is %s\n", pdn->full_name); diff --git a/arch/sparc/kernel/of_device_64.c b/arch/sparc/kernel/of_device_64.c index 7a3be6f6737a..7bbdc26d9512 100644 --- a/arch/sparc/kernel/of_device_64.c +++ b/arch/sparc/kernel/of_device_64.c @@ -580,7 +580,7 @@ static unsigned int __init build_one_device_irq(struct platform_device *op, printk("%s: Apply [%s:%x] imap --> [%s:%x]\n", op->dev.of_node->full_name, pp->full_name, this_orig_irq, - (iret ? iret->full_name : "NULL"), irq); + of_node_full_name(iret), irq); if (!iret) break; diff --git a/drivers/of/base.c b/drivers/of/base.c index 85757952f12d..9ec0a2f1b028 100644 --- a/drivers/of/base.c +++ b/drivers/of/base.c @@ -1173,7 +1173,7 @@ static void of_alias_add(struct alias_prop *ap, struct device_node *np, ap->stem[stem_len] = 0; list_add_tail(&ap->link, &aliases_lookup); pr_debug("adding DT alias:%s: stem=%s id=%i node=%s\n", - ap->alias, ap->stem, ap->id, np ? np->full_name : NULL); + ap->alias, ap->stem, ap->id, of_node_full_name(np)); } /** diff --git a/drivers/of/irq.c b/drivers/of/irq.c index 9cf00602f566..ff8ab7b27373 100644 --- a/drivers/of/irq.c +++ b/drivers/of/irq.c @@ -255,7 +255,7 @@ int of_irq_map_raw(struct device_node *parent, const __be32 *intspec, skiplevel: /* Iterate again with new parent */ - pr_debug(" -> new parent: %s\n", newpar ? newpar->full_name : "<>"); + pr_debug(" -> new parent: %s\n", of_node_full_name(newpar)); of_node_put(ipar); ipar = newpar; newpar = NULL; diff --git a/include/linux/of.h b/include/linux/of.h index 2ec1083af7ff..1012377cae92 100644 --- a/include/linux/of.h +++ b/include/linux/of.h @@ -163,6 +163,11 @@ static inline int of_node_to_nid(struct device_node *np) { return -1; } #define of_node_to_nid of_node_to_nid #endif +static inline const char* of_node_full_name(struct device_node *np) +{ + return np ? np->full_name : ""; +} + extern struct device_node *of_find_node_by_name(struct device_node *from, const char *name); #define for_each_node_by_name(dn, name) \ @@ -303,6 +308,11 @@ const char *of_prop_next_string(struct property *prop, const char *cur); #else /* CONFIG_OF */ +static inline const char* of_node_full_name(struct device_node *np) +{ + return ""; +} + static inline bool of_have_populated_dt(void) { return false; diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 41c1564103f1..38c5eb839c92 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -448,7 +448,7 @@ unsigned int irq_create_mapping(struct irq_domain *domain, } pr_debug("irq %lu on domain %s mapped to virtual irq %u\n", - hwirq, domain->of_node ? domain->of_node->full_name : "null", virq); + hwirq, of_node_full_name(domain->of_node), virq); return virq; } @@ -477,7 +477,7 @@ unsigned int irq_create_of_mapping(struct device_node *controller, return intspec[0]; #endif pr_warning("no irq domain found for %s !\n", - controller->full_name); + of_node_full_name(controller)); return 0; } @@ -725,8 +725,8 @@ static int virq_debug_show(struct seq_file *m, void *private) data = irq_desc_get_chip_data(desc); seq_printf(m, data ? "0x%p " : " %p ", data); - if (desc->irq_data.domain && desc->irq_data.domain->of_node) - p = desc->irq_data.domain->of_node->full_name; + if (desc->irq_data.domain) + p = of_node_full_name(desc->irq_data.domain->of_node); else p = none; seq_printf(m, "%s\n", p); -- cgit v1.2.3 From cfca927972e31a5b3da49bf641c525732ff3c357 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 25 Jun 2012 12:54:17 -0700 Subject: rcu: Introduce check for callback list/count mismatch The recent bug that introduced the RCU callback list/count mismatch showed the need for a diagnostic to check for this, which this commit adds. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutree.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 95c7b61e77e4..4154c9567a6d 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1612,6 +1612,7 @@ static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) rdp->n_force_qs_snap = rsp->n_force_qs; } else if (rdp->qlen < rdp->qlen_last_fqs_check - qhimark) rdp->qlen_last_fqs_check = rdp->qlen; + WARN_ON_ONCE((rdp->nxtlist == NULL) != (rdp->qlen == 0)); local_irq_restore(flags); -- cgit v1.2.3 From c701d5d9b384ff03ceb232ef21236364d784a411 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 28 Jun 2012 08:08:25 -0700 Subject: rcu: Fix code-style issues involving "else" The Linux kernel coding style says that single-statement blocks should omit curly braces unless the other leg of the "if" statement has multiple statements, in which case the curly braces should be included. This commit fixes RCU's violations of this rule. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/rcutiny_plugin.h | 7 ++++--- kernel/rcutorture.c | 3 ++- kernel/rcutree.c | 7 ++++--- kernel/rcutree_plugin.h | 14 ++++++++------ 4 files changed, 18 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/rcutiny_plugin.h b/kernel/rcutiny_plugin.h index 116725b5edfb..918fd1e8509c 100644 --- a/kernel/rcutiny_plugin.h +++ b/kernel/rcutiny_plugin.h @@ -350,8 +350,9 @@ static int rcu_initiate_boost(void) rcu_preempt_ctrlblk.boost_tasks = rcu_preempt_ctrlblk.gp_tasks; invoke_rcu_callbacks(); - } else + } else { RCU_TRACE(rcu_initiate_boost_trace()); + } return 1; } @@ -778,9 +779,9 @@ void synchronize_rcu_expedited(void) rpcp->exp_tasks = NULL; /* Wait for tail of ->blkd_tasks list to drain. */ - if (!rcu_preempted_readers_exp()) + if (!rcu_preempted_readers_exp()) { local_irq_restore(flags); - else { + } else { rcu_initiate_boost(); local_irq_restore(flags); wait_event(sync_rcu_preempt_exp_wq, diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index c279ee920947..155fb129b641 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -408,8 +408,9 @@ rcu_torture_cb(struct rcu_head *p) if (++rp->rtort_pipe_count >= RCU_TORTURE_PIPE_LEN) { rp->rtort_mbtest = 0; rcu_torture_free(rp); - } else + } else { cur_ops->deferred_free(rp); + } } static int rcu_no_completed(void) diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 117218a43724..f280e542e3e9 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -892,8 +892,9 @@ static void __note_new_gpnum(struct rcu_state *rsp, struct rcu_node *rnp, struct if (rnp->qsmask & rdp->grpmask) { rdp->qs_pending = 1; rdp->passed_quiesce = 0; - } else + } else { rdp->qs_pending = 0; + } zero_cpu_stall_ticks(rdp); } } @@ -2130,9 +2131,9 @@ void synchronize_sched_expedited(void) put_online_cpus(); /* No joy, try again later. Or just synchronize_sched(). */ - if (trycount++ < 10) + if (trycount++ < 10) { udelay(trycount * num_online_cpus()); - else { + } else { synchronize_sched(); return; } diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index a9194d5606c4..7f3244c0df01 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -390,8 +390,9 @@ void rcu_read_unlock_special(struct task_struct *t) rnp->grphi, !!rnp->gp_tasks); rcu_report_unblock_qs_rnp(rnp, flags); - } else + } else { raw_spin_unlock_irqrestore(&rnp->lock, flags); + } #ifdef CONFIG_RCU_BOOST /* Unboost if we were boosted. */ @@ -757,9 +758,9 @@ sync_rcu_preempt_exp_init(struct rcu_state *rsp, struct rcu_node *rnp) int must_wait = 0; raw_spin_lock_irqsave(&rnp->lock, flags); - if (list_empty(&rnp->blkd_tasks)) + if (list_empty(&rnp->blkd_tasks)) { raw_spin_unlock_irqrestore(&rnp->lock, flags); - else { + } else { rnp->exp_tasks = rnp->blkd_tasks.next; rcu_initiate_boost(rnp, flags); /* releases rnp->lock */ must_wait = 1; @@ -803,9 +804,9 @@ void synchronize_rcu_expedited(void) * expedited grace period for us, just leave. */ while (!mutex_trylock(&sync_rcu_preempt_exp_mutex)) { - if (trycount++ < 10) + if (trycount++ < 10) { udelay(trycount * num_online_cpus()); - else { + } else { synchronize_rcu(); return; } @@ -2093,8 +2094,9 @@ static void rcu_prepare_for_idle(int cpu) if (rcu_cpu_has_callbacks(cpu)) { trace_rcu_prep_idle("More callbacks"); invoke_rcu_core(); - } else + } else { trace_rcu_prep_idle("Callbacks drained"); + } } /* -- cgit v1.2.3 From 5cf05ad758c30d17ff23c2be346b5de982bc2121 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 17 May 2012 15:12:45 -0700 Subject: rcu: Fix broken strings in RCU's source code. Although the C language allows you to break strings across lines, doing this makes it hard for people to find the Linux kernel code corresponding to a given console message. This commit therefore fixes broken strings throughout RCU's source code. Suggested-by: Josh Triplett Suggested-by: Ingo Molnar Signed-off-by: Paul E. McKenney --- include/linux/rcupdate.h | 3 +-- kernel/rcutorture.c | 33 ++++++++++++++------------------- kernel/rcutree_trace.c | 25 ++++++++++++------------- lib/list_debug.c | 6 ++---- 4 files changed, 29 insertions(+), 38 deletions(-) (limited to 'kernel') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index c2c0d86dd3ac..115ead2b5155 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -427,8 +427,7 @@ extern int rcu_my_thread_group_empty(void); static inline void rcu_preempt_sleep_check(void) { rcu_lockdep_assert(!lock_is_held(&rcu_lock_map), - "Illegal context switch in RCU read-side " - "critical section"); + "Illegal context switch in RCU read-side critical section"); } #else /* #ifdef CONFIG_PROVE_RCU */ static inline void rcu_preempt_sleep_check(void) diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 155fb129b641..25b15033c61f 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -49,8 +49,7 @@ #include MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Paul E. McKenney and " - "Josh Triplett "); +MODULE_AUTHOR("Paul E. McKenney and Josh Triplett "); static int nreaders = -1; /* # reader threads, defaults to 2*ncpus */ static int nfakewriters = 4; /* # fake writer threads */ @@ -1200,27 +1199,27 @@ rcu_torture_printk(char *page) } cnt += sprintf(&page[cnt], "%s%s ", torture_type, TORTURE_FLAG); cnt += sprintf(&page[cnt], - "rtc: %p ver: %lu tfle: %d rta: %d rtaf: %d rtf: %d " - "rtmbe: %d rtbke: %ld rtbre: %ld " - "rtbf: %ld rtb: %ld nt: %ld " - "onoff: %ld/%ld:%ld/%ld " - "barrier: %ld/%ld:%ld", + "rtc: %p ver: %lu tfle: %d rta: %d rtaf: %d rtf: %d ", rcu_torture_current, rcu_torture_current_version, list_empty(&rcu_torture_freelist), atomic_read(&n_rcu_torture_alloc), atomic_read(&n_rcu_torture_alloc_fail), - atomic_read(&n_rcu_torture_free), + atomic_read(&n_rcu_torture_free)); + cnt += sprintf(&page[cnt], "rtmbe: %d rtbke: %ld rtbre: %ld ", atomic_read(&n_rcu_torture_mberror), n_rcu_torture_boost_ktrerror, - n_rcu_torture_boost_rterror, + n_rcu_torture_boost_rterror); + cnt += sprintf(&page[cnt], "rtbf: %ld rtb: %ld nt: %ld ", n_rcu_torture_boost_failure, n_rcu_torture_boosts, - n_rcu_torture_timers, + n_rcu_torture_timers); + cnt += sprintf(&page[cnt], "onoff: %ld/%ld:%ld/%ld ", n_online_successes, n_online_attempts, n_offline_successes, - n_offline_attempts, + n_offline_attempts); + cnt += sprintf(&page[cnt], "barrier: %ld/%ld:%ld", n_barrier_successes, n_barrier_attempts, n_rcu_torture_barrier_error); @@ -1462,8 +1461,7 @@ rcu_torture_shutdown(void *arg) delta = shutdown_time - jiffies_snap; if (verbose) printk(KERN_ALERT "%s" TORTURE_FLAG - "rcu_torture_shutdown task: %lu " - "jiffies remaining\n", + "rcu_torture_shutdown task: %lu jiffies remaining\n", torture_type, delta); schedule_timeout_interruptible(delta); jiffies_snap = ACCESS_ONCE(jiffies); @@ -1515,8 +1513,7 @@ rcu_torture_onoff(void *arg) if (cpu_down(cpu) == 0) { if (verbose) printk(KERN_ALERT "%s" TORTURE_FLAG - "rcu_torture_onoff task: " - "offlined %d\n", + "rcu_torture_onoff task: offlined %d\n", torture_type, cpu); n_offline_successes++; } @@ -1529,8 +1526,7 @@ rcu_torture_onoff(void *arg) if (cpu_up(cpu) == 0) { if (verbose) printk(KERN_ALERT "%s" TORTURE_FLAG - "rcu_torture_onoff task: " - "onlined %d\n", + "rcu_torture_onoff task: onlined %d\n", torture_type, cpu); n_online_successes++; } @@ -1952,8 +1948,7 @@ rcu_torture_init(void) return -EINVAL; } if (cur_ops->fqs == NULL && fqs_duration != 0) { - printk(KERN_ALERT "rcu-torture: ->fqs NULL and non-zero " - "fqs_duration, fqs disabled.\n"); + printk(KERN_ALERT "rcu-torture: ->fqs NULL and non-zero fqs_duration, fqs disabled.\n"); fqs_duration = 0; } if (cur_ops->init) diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index a16ddbd6fdc4..abffb486e94e 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -218,8 +218,7 @@ static const struct file_operations rcudata_csv_fops = { static void print_one_rcu_node_boost(struct seq_file *m, struct rcu_node *rnp) { - seq_printf(m, "%d:%d tasks=%c%c%c%c kt=%c ntb=%lu neb=%lu nnb=%lu " - "j=%04x bt=%04x\n", + seq_printf(m, "%d:%d tasks=%c%c%c%c kt=%c ntb=%lu neb=%lu nnb=%lu ", rnp->grplo, rnp->grphi, "T."[list_empty(&rnp->blkd_tasks)], "N."[!rnp->gp_tasks], @@ -227,11 +226,11 @@ static void print_one_rcu_node_boost(struct seq_file *m, struct rcu_node *rnp) "B."[!rnp->boost_tasks], convert_kthread_status(rnp->boost_kthread_status), rnp->n_tasks_boosted, rnp->n_exp_boosts, - rnp->n_normal_boosts, + rnp->n_normal_boosts); + seq_printf(m, "j=%04x bt=%04x\n", (int)(jiffies & 0xffff), (int)(rnp->boost_time & 0xffff)); - seq_printf(m, "%s: nt=%lu egt=%lu bt=%lu nb=%lu ny=%lu nos=%lu\n", - " balk", + seq_printf(m, " balk: nt=%lu egt=%lu bt=%lu nb=%lu ny=%lu nos=%lu\n", rnp->n_balk_blkd_tasks, rnp->n_balk_exp_gp_tasks, rnp->n_balk_boost_tasks, @@ -287,11 +286,11 @@ static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) struct rcu_node *rnp; gpnum = rsp->gpnum; - seq_printf(m, "%s: c=%lu g=%lu s=%d jfq=%ld j=%x " - "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu oqlen=%ld/%ld\n", + seq_printf(m, "%s: c=%lu g=%lu s=%d jfq=%ld j=%x ", rsp->name, rsp->completed, gpnum, rsp->fqs_state, (long)(rsp->jiffies_force_qs - jiffies), - (int)(jiffies & 0xffff), + (int)(jiffies & 0xffff)); + seq_printf(m, "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu oqlen=%ld/%ld\n", rsp->n_force_qs, rsp->n_force_qs_ngp, rsp->n_force_qs - rsp->n_force_qs_ngp, rsp->n_force_qs_lh, rsp->qlen_lazy, rsp->qlen); @@ -378,16 +377,16 @@ static const struct file_operations rcugp_fops = { static void print_one_rcu_pending(struct seq_file *m, struct rcu_data *rdp) { - seq_printf(m, "%3d%cnp=%ld " - "qsp=%ld rpq=%ld cbr=%ld cng=%ld " - "gpc=%ld gps=%ld nf=%ld nn=%ld\n", + seq_printf(m, "%3d%cnp=%ld ", rdp->cpu, cpu_is_offline(rdp->cpu) ? '!' : ' ', - rdp->n_rcu_pending, + rdp->n_rcu_pending); + seq_printf(m, "qsp=%ld rpq=%ld cbr=%ld cng=%ld ", rdp->n_rp_qs_pending, rdp->n_rp_report_qs, rdp->n_rp_cb_ready, - rdp->n_rp_cpu_needs_gp, + rdp->n_rp_cpu_needs_gp); + seq_printf(m, "gpc=%ld gps=%ld nf=%ld nn=%ld\n", rdp->n_rp_gp_completed, rdp->n_rp_gp_started, rdp->n_rp_need_fqs, diff --git a/lib/list_debug.c b/lib/list_debug.c index 23a5e031cd8b..c24c2f7e296f 100644 --- a/lib/list_debug.c +++ b/lib/list_debug.c @@ -87,12 +87,10 @@ void __list_add_rcu(struct list_head *new, struct list_head *prev, struct list_head *next) { WARN(next->prev != prev, - "list_add_rcu corruption. next->prev should be " - "prev (%p), but was %p. (next=%p).\n", + "list_add_rcu corruption. next->prev should be prev (%p), but was %p. (next=%p).\n", prev, next->prev, next); WARN(prev->next != next, - "list_add_rcu corruption. prev->next should be " - "next (%p), but was %p. (prev=%p).\n", + "list_add_rcu corruption. prev->next should be next (%p), but was %p. (prev=%p).\n", next, prev->next, prev); new->next = next; new->prev = prev; -- cgit v1.2.3 From ce27e317ba22b359bde02216afab934dac3af095 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 3 Jul 2012 10:38:06 -0700 Subject: cgroup: cgroup_rm_files() was calling simple_unlink() with the wrong inode While refactoring cgroup file removal path, 05ef1d7c4a "cgroup: introduce struct cfent" incorrectly changed the @dir argument of simple_unlink() to the inode of the file being deleted instead of that of the containing directory. The effect of this bug is minor - ctime and mtime of the parent weren't properly updated on file deletion. Fix it by using @cgrp->dentry->d_inode instead. Signed-off-by: Tejun Heo Reported-by: Al Viro Acked-by: Li Zefan Cc: stable@vger.kernel.org --- kernel/cgroup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index dec62f5936ef..a56805aa0f1b 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -959,7 +959,7 @@ static int cgroup_rm_file(struct cgroup *cgrp, const struct cftype *cft) dget(d); d_delete(d); - simple_unlink(d->d_inode, d); + simple_unlink(cgrp->dentry->d_inode, d); list_del_init(&cfe->node); dput(d); -- cgit v1.2.3 From 781d0f46d81e2c26c70649903b503bcfe817efc8 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Thu, 5 Jul 2012 12:19:19 +0100 Subject: irq_domain: Standardise legacy/linear domain selection A large proportion of interrupt controllers that support legacy mappings do so because non-DT systems need to use fixed IRQ numbers when registering devices via buses but can otherwise use a linear mapping. The interrupt controller itself typically is not affected by the mapping used and best practice is to use a linear mapping where possible so drivers frequently select at runtime depending on if a legacy range has been allocated to them. Standardise this behaviour by providing irq_domain_register_simple() which will allocate a linear mapping unless a positive first_irq is provided in which case it will fall back to a legacy mapping. This helps make best practice for irq_domain adoption clearer. Signed-off-by: Mark Brown Signed-off-by: Grant Likely --- Documentation/IRQ-domain.txt | 5 +++++ include/linux/irqdomain.h | 5 +++++ kernel/irq/irqdomain.c | 30 ++++++++++++++++++++++++++++++ 3 files changed, 40 insertions(+) (limited to 'kernel') diff --git a/Documentation/IRQ-domain.txt b/Documentation/IRQ-domain.txt index 27dcaabfb4db..1401cece745a 100644 --- a/Documentation/IRQ-domain.txt +++ b/Documentation/IRQ-domain.txt @@ -93,6 +93,7 @@ Linux IRQ number into the hardware. Most drivers cannot use this mapping. ==== Legacy ==== +irq_domain_add_simple() irq_domain_add_legacy() irq_domain_add_legacy_isa() @@ -115,3 +116,7 @@ The legacy map should only be used if fixed IRQ mappings must be supported. For example, ISA controllers would use the legacy map for mapping Linux IRQs 0-15 so that existing ISA drivers get the correct IRQ numbers. + +Most users of legacy mappings should use irq_domain_add_simple() which +will use a legacy domain only if an IRQ range is supplied by the +system and will otherwise use a linear domain mapping. diff --git a/include/linux/irqdomain.h b/include/linux/irqdomain.h index 5abb533eb8eb..17b60be30fff 100644 --- a/include/linux/irqdomain.h +++ b/include/linux/irqdomain.h @@ -112,6 +112,11 @@ struct irq_domain { }; #ifdef CONFIG_IRQ_DOMAIN +struct irq_domain *irq_domain_add_simple(struct device_node *of_node, + unsigned int size, + unsigned int first_irq, + const struct irq_domain_ops *ops, + void *host_data); struct irq_domain *irq_domain_add_legacy(struct device_node *of_node, unsigned int size, unsigned int first_irq, diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index d3968e91bfd2..0c51958c6335 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -139,6 +139,36 @@ static unsigned int irq_domain_legacy_revmap(struct irq_domain *domain, return hwirq - first_hwirq + domain->revmap_data.legacy.first_irq; } +/** + * irq_domain_add_simple() - Allocate and register a simple irq_domain. + * @of_node: pointer to interrupt controller's device tree node. + * @size: total number of irqs in mapping + * @first_irq: first number of irq block assigned to the domain + * @ops: map/unmap domain callbacks + * @host_data: Controller private data pointer + * + * Allocates a legacy irq_domain if irq_base is positive or a linear + * domain otherwise. + * + * This is intended to implement the expected behaviour for most + * interrupt controllers which is that a linear mapping should + * normally be used unless the system requires a legacy mapping in + * order to support supplying interrupt numbers during non-DT + * registration of devices. + */ +struct irq_domain *irq_domain_add_simple(struct device_node *of_node, + unsigned int size, + unsigned int first_irq, + const struct irq_domain_ops *ops, + void *host_data) +{ + if (first_irq > 0) + return irq_domain_add_legacy(of_node, size, first_irq, 0, + ops, host_data); + else + return irq_domain_add_linear(of_node, size, ops, host_data); +} + /** * irq_domain_add_legacy() - Allocate and register a legacy revmap irq_domain. * @of_node: pointer to interrupt controller's device tree node. -- cgit v1.2.3 From 22076c7712be29a602de45b1c573f31adbd428a9 Mon Sep 17 00:00:00 2001 From: Dong Aisheng Date: Wed, 20 Jun 2012 17:00:30 +0800 Subject: irq_domain: correct a minor wrong comment for linear revmap The revmap type should be linear for irq_domain_add_linear function. Signed-off-by: Dong Aisheng Signed-off-by: Grant Likely --- kernel/irq/irqdomain.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 0c51958c6335..622fdf4b82b6 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -248,7 +248,7 @@ struct irq_domain *irq_domain_add_legacy(struct device_node *of_node, EXPORT_SYMBOL_GPL(irq_domain_add_legacy); /** - * irq_domain_add_linear() - Allocate and register a legacy revmap irq_domain. + * irq_domain_add_linear() - Allocate and register a linear revmap irq_domain. * @of_node: pointer to interrupt controller's device tree node. * @size: Number of interrupts in the domain. * @ops: map/unmap domain callbacks -- cgit v1.2.3 From 913af2070731bfc1bd39bb35c5cd2fd66f5eff12 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Sun, 3 Jun 2012 22:04:35 -0700 Subject: irqdomain: Split disassociating code into separate function This patch moves the irq disassociation code out into a separate function in preparation to extend irq_setup_virq to handle multiple irqs and rename it for use by interrupt controller drivers. The new function will be used by irq_setup_virq() in its error path. Signed-off-by: Grant Likely Acked-by: Benjamin Herrenschmidt Cc: Paul Mundt Cc: Thomas Gleixner Cc: Rob Herring --- kernel/irq/irqdomain.c | 75 +++++++++++++++++++++++++++++++------------------- 1 file changed, 47 insertions(+), 28 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 622fdf4b82b6..17634f2c9f6c 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -364,6 +364,52 @@ void irq_set_default_host(struct irq_domain *domain) } EXPORT_SYMBOL_GPL(irq_set_default_host); +static void irq_domain_disassociate_many(struct irq_domain *domain, + unsigned int irq_base, int count) +{ + /* + * disassociate in reverse order; + * not strictly necessary, but nice for unwinding + */ + while (count--) { + int irq = irq_base + count; + struct irq_data *irq_data = irq_get_irq_data(irq); + irq_hw_number_t hwirq = irq_data->hwirq; + + if (WARN_ON(!irq_data || irq_data->domain != domain)) + continue; + + irq_set_status_flags(irq, IRQ_NOREQUEST); + + /* remove chip and handler */ + irq_set_chip_and_handler(irq, NULL, NULL); + + /* Make sure it's completed */ + synchronize_irq(irq); + + /* Tell the PIC about it */ + if (domain->ops->unmap) + domain->ops->unmap(domain, irq); + smp_mb(); + + irq_data->domain = NULL; + irq_data->hwirq = 0; + + /* Clear reverse map */ + switch(domain->revmap_type) { + case IRQ_DOMAIN_MAP_LINEAR: + if (hwirq < domain->revmap_data.linear.size) + domain->revmap_data.linear.revmap[hwirq] = 0; + break; + case IRQ_DOMAIN_MAP_TREE: + mutex_lock(&revmap_trees_mutex); + radix_tree_delete(&domain->revmap_data.tree, hwirq); + mutex_unlock(&revmap_trees_mutex); + break; + } + } +} + static int irq_setup_virq(struct irq_domain *domain, unsigned int virq, irq_hw_number_t hwirq) { @@ -544,7 +590,6 @@ void irq_dispose_mapping(unsigned int virq) { struct irq_data *irq_data = irq_get_irq_data(virq); struct irq_domain *domain; - irq_hw_number_t hwirq; if (!virq || !irq_data) return; @@ -557,33 +602,7 @@ void irq_dispose_mapping(unsigned int virq) if (domain->revmap_type == IRQ_DOMAIN_MAP_LEGACY) return; - irq_set_status_flags(virq, IRQ_NOREQUEST); - - /* remove chip and handler */ - irq_set_chip_and_handler(virq, NULL, NULL); - - /* Make sure it's completed */ - synchronize_irq(virq); - - /* Tell the PIC about it */ - if (domain->ops->unmap) - domain->ops->unmap(domain, virq); - smp_mb(); - - /* Clear reverse map */ - hwirq = irq_data->hwirq; - switch(domain->revmap_type) { - case IRQ_DOMAIN_MAP_LINEAR: - if (hwirq < domain->revmap_data.linear.size) - domain->revmap_data.linear.revmap[hwirq] = 0; - break; - case IRQ_DOMAIN_MAP_TREE: - mutex_lock(&revmap_trees_mutex); - radix_tree_delete(&domain->revmap_data.tree, hwirq); - mutex_unlock(&revmap_trees_mutex); - break; - } - + irq_domain_disassociate_many(domain, virq, 1); irq_free_desc(virq); } EXPORT_SYMBOL_GPL(irq_dispose_mapping); -- cgit v1.2.3 From 2a71a1a9da40dfbd5b23d4312aa1641385581f4a Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Sun, 3 Jun 2012 22:04:36 -0700 Subject: irqdomain: Always update revmap when setting up a virq At irq_setup_virq() time all of the data needed to update the reverse map is available, but the current code ignores it and relies upon the slow path to insert revmap records. This patch adds revmap updating to the setup path so the slow path will no longer be necessary. Signed-off-by: Grant Likely Cc: Paul Mundt Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Rob Herring --- arch/powerpc/sysdev/xics/xics-common.c | 3 --- kernel/irq/irqdomain.c | 12 ++++++++++++ 2 files changed, 12 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/arch/powerpc/sysdev/xics/xics-common.c b/arch/powerpc/sysdev/xics/xics-common.c index cd1d18db92c6..9049d9f44485 100644 --- a/arch/powerpc/sysdev/xics/xics-common.c +++ b/arch/powerpc/sysdev/xics/xics-common.c @@ -329,9 +329,6 @@ static int xics_host_map(struct irq_domain *h, unsigned int virq, pr_devel("xics: map virq %d, hwirq 0x%lx\n", virq, hw); - /* Insert the interrupt mapping into the radix tree for fast lookup */ - irq_radix_revmap_insert(xics_host, virq, hw); - /* They aren't all level sensitive but we just don't really know */ irq_set_status_flags(virq, IRQ_LEVEL); diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 17634f2c9f6c..091732c9dbdc 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -424,6 +424,18 @@ static int irq_setup_virq(struct irq_domain *domain, unsigned int virq, return -1; } + switch (domain->revmap_type) { + case IRQ_DOMAIN_MAP_LINEAR: + if (hwirq < domain->revmap_data.linear.size) + domain->revmap_data.linear.revmap[hwirq] = virq; + break; + case IRQ_DOMAIN_MAP_TREE: + mutex_lock(&revmap_trees_mutex); + irq_radix_revmap_insert(domain, virq, hwirq); + mutex_unlock(&revmap_trees_mutex); + break; + } + irq_clear_status_flags(virq, IRQ_NOREQUEST); return 0; -- cgit v1.2.3 From 98aa468e045a0091a7c34d9f5205a629634fabf4 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Sun, 17 Jun 2012 16:17:04 -0600 Subject: irqdomain: Support for static IRQ mapping and association. This adds a new strict mapping API for supporting creation of linux IRQs at existing positions within the domain. The new routines are as follows: For dynamic allocation and insertion to specified ranges: - irq_create_identity_mapping() - irq_create_strict_mappings() These will allocate and associate a range of linux IRQs at the specified location. This can be used by controllers that have their own static linux IRQ definitions to map a hwirq range to, as well as for platforms that wish to establish 1:1 identity mapping between linux and hwirq space. For insertion to specified ranges by platforms that do their own irq_desc management: - irq_domain_associate() - irq_domain_associate_many() These in turn call back in to the domain's ->map() routine, for further processing by the platform. Disassociation of IRQs get handled through irq_dispose_mapping() as normal. With these in place it should be possible to begin migration of legacy IRQ domains to linear ones, without requiring special handling for static vs dynamic IRQ definitions in DT vs non-DT paths. This also makes it possible for domains with static mappings to adopt whichever tree model best fits their needs, rather than simply restricting them to linear revmaps. Signed-off-by: Paul Mundt [grant.likely: Reorganized irq_domain_associate{,_many} to have all logic in one place] [grant.likely: Add error checking for unallocated irq_descs at associate time] Signed-off-by: Grant Likely Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Rob Herring --- include/linux/irqdomain.h | 19 ++++++++ kernel/irq/irqdomain.c | 113 ++++++++++++++++++++++++++++++++++++---------- 2 files changed, 107 insertions(+), 25 deletions(-) (limited to 'kernel') diff --git a/include/linux/irqdomain.h b/include/linux/irqdomain.h index 17b60be30fff..eab8a0e60b8e 100644 --- a/include/linux/irqdomain.h +++ b/include/linux/irqdomain.h @@ -149,12 +149,31 @@ static inline struct irq_domain *irq_domain_add_legacy_isa( extern void irq_domain_remove(struct irq_domain *host); +extern int irq_domain_associate_many(struct irq_domain *domain, + unsigned int irq_base, + irq_hw_number_t hwirq_base, int count); +static inline int irq_domain_associate(struct irq_domain *domain, unsigned int irq, + irq_hw_number_t hwirq) +{ + return irq_domain_associate_many(domain, irq, hwirq, 1); +} + extern unsigned int irq_create_mapping(struct irq_domain *host, irq_hw_number_t hwirq); extern void irq_dispose_mapping(unsigned int virq); extern unsigned int irq_find_mapping(struct irq_domain *host, irq_hw_number_t hwirq); extern unsigned int irq_create_direct_mapping(struct irq_domain *host); +extern int irq_create_strict_mappings(struct irq_domain *domain, + unsigned int irq_base, + irq_hw_number_t hwirq_base, int count); + +static inline int irq_create_identity_mapping(struct irq_domain *host, + irq_hw_number_t hwirq) +{ + return irq_create_strict_mappings(host, hwirq, hwirq, 1); +} + extern void irq_radix_revmap_insert(struct irq_domain *host, unsigned int virq, irq_hw_number_t hwirq); extern unsigned int irq_radix_revmap_lookup(struct irq_domain *host, diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 091732c9dbdc..a07d92446b66 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -410,36 +410,61 @@ static void irq_domain_disassociate_many(struct irq_domain *domain, } } -static int irq_setup_virq(struct irq_domain *domain, unsigned int virq, - irq_hw_number_t hwirq) +int irq_domain_associate_many(struct irq_domain *domain, unsigned int irq_base, + irq_hw_number_t hwirq_base, int count) { - struct irq_data *irq_data = irq_get_irq_data(virq); + unsigned int virq = irq_base; + irq_hw_number_t hwirq = hwirq_base; + int i; - irq_data->hwirq = hwirq; - irq_data->domain = domain; - if (domain->ops->map && domain->ops->map(domain, virq, hwirq)) { - pr_err("irq-%i==>hwirq-0x%lx mapping failed\n", virq, hwirq); - irq_data->domain = NULL; - irq_data->hwirq = 0; - return -1; - } + pr_debug("%s(%s, irqbase=%i, hwbase=%i, count=%i)\n", __func__, + of_node_full_name(domain->of_node), irq_base, (int)hwirq_base, count); - switch (domain->revmap_type) { - case IRQ_DOMAIN_MAP_LINEAR: - if (hwirq < domain->revmap_data.linear.size) - domain->revmap_data.linear.revmap[hwirq] = virq; - break; - case IRQ_DOMAIN_MAP_TREE: - mutex_lock(&revmap_trees_mutex); - irq_radix_revmap_insert(domain, virq, hwirq); - mutex_unlock(&revmap_trees_mutex); - break; - } + for (i = 0; i < count; i++) { + struct irq_data *irq_data = irq_get_irq_data(virq + i); + + if (WARN(!irq_data, "error: irq_desc not allocated; " + "irq=%i hwirq=0x%x\n", virq + i, (int)hwirq + i)) + return -EINVAL; + if (WARN(irq_data->domain, "error: irq_desc already associated; " + "irq=%i hwirq=0x%x\n", virq + i, (int)hwirq + i)) + return -EINVAL; + }; + + for (i = 0; i < count; i++, virq++, hwirq++) { + struct irq_data *irq_data = irq_get_irq_data(virq); + + irq_data->hwirq = hwirq; + irq_data->domain = domain; + if (domain->ops->map && domain->ops->map(domain, virq, hwirq)) { + pr_err("irq-%i==>hwirq-0x%lx mapping failed\n", virq, hwirq); + irq_data->domain = NULL; + irq_data->hwirq = 0; + goto err_unmap; + } + + switch (domain->revmap_type) { + case IRQ_DOMAIN_MAP_LINEAR: + if (hwirq < domain->revmap_data.linear.size) + domain->revmap_data.linear.revmap[hwirq] = virq; + break; + case IRQ_DOMAIN_MAP_TREE: + mutex_lock(&revmap_trees_mutex); + irq_radix_revmap_insert(domain, virq, hwirq); + mutex_unlock(&revmap_trees_mutex); + break; + } - irq_clear_status_flags(virq, IRQ_NOREQUEST); + irq_clear_status_flags(virq, IRQ_NOREQUEST); + } return 0; + + err_unmap: + irq_domain_disassociate_many(domain, irq_base, i); + return -EINVAL; } +EXPORT_SYMBOL_GPL(irq_domain_associate_many); /** * irq_create_direct_mapping() - Allocate an irq for direct mapping @@ -472,7 +497,7 @@ unsigned int irq_create_direct_mapping(struct irq_domain *domain) } pr_debug("create_direct obtained virq %d\n", virq); - if (irq_setup_virq(domain, virq, virq)) { + if (irq_domain_associate(domain, virq, virq)) { irq_free_desc(virq); return 0; } @@ -533,7 +558,7 @@ unsigned int irq_create_mapping(struct irq_domain *domain, return 0; } - if (irq_setup_virq(domain, virq, hwirq)) { + if (irq_domain_associate(domain, virq, hwirq)) { irq_free_desc(virq); return 0; } @@ -545,6 +570,44 @@ unsigned int irq_create_mapping(struct irq_domain *domain, } EXPORT_SYMBOL_GPL(irq_create_mapping); +/** + * irq_create_strict_mappings() - Map a range of hw irqs to fixed linux irqs + * @domain: domain owning the interrupt range + * @irq_base: beginning of linux IRQ range + * @hwirq_base: beginning of hardware IRQ range + * @count: Number of interrupts to map + * + * This routine is used for allocating and mapping a range of hardware + * irqs to linux irqs where the linux irq numbers are at pre-defined + * locations. For use by controllers that already have static mappings + * to insert in to the domain. + * + * Non-linear users can use irq_create_identity_mapping() for IRQ-at-a-time + * domain insertion. + * + * 0 is returned upon success, while any failure to establish a static + * mapping is treated as an error. + */ +int irq_create_strict_mappings(struct irq_domain *domain, unsigned int irq_base, + irq_hw_number_t hwirq_base, int count) +{ + int ret; + + ret = irq_alloc_descs(irq_base, irq_base, count, + of_node_to_nid(domain->of_node)); + if (unlikely(ret < 0)) + return ret; + + ret = irq_domain_associate_many(domain, irq_base, hwirq_base, count); + if (unlikely(ret < 0)) { + irq_free_descs(irq_base, count); + return ret; + } + + return 0; +} +EXPORT_SYMBOL_GPL(irq_create_strict_mappings); + unsigned int irq_create_of_mapping(struct device_node *controller, const u32 *intspec, unsigned int intsize) { -- cgit v1.2.3 From d6b0d1f7058f7cf818138cd7fd116dca3f3576d9 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Sun, 3 Jun 2012 22:04:37 -0700 Subject: irqdomain: Eliminate dedicated radix lookup functions In preparation to remove the slow revmap path, eliminate the public radix revmap lookup functions. This simplifies the code and makes the slowpath removal patch a lot simpler. Signed-off-by: Grant Likely Cc: Paul Mundt Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Rob Herring --- arch/powerpc/sysdev/xics/icp-hv.c | 2 +- arch/powerpc/sysdev/xics/icp-native.c | 2 +- include/linux/irqdomain.h | 4 --- kernel/irq/irqdomain.c | 60 +---------------------------------- 4 files changed, 3 insertions(+), 65 deletions(-) (limited to 'kernel') diff --git a/arch/powerpc/sysdev/xics/icp-hv.c b/arch/powerpc/sysdev/xics/icp-hv.c index 253dce98c16e..14469cf9df68 100644 --- a/arch/powerpc/sysdev/xics/icp-hv.c +++ b/arch/powerpc/sysdev/xics/icp-hv.c @@ -111,7 +111,7 @@ static unsigned int icp_hv_get_irq(void) if (vec == XICS_IRQ_SPURIOUS) return NO_IRQ; - irq = irq_radix_revmap_lookup(xics_host, vec); + irq = irq_find_mapping(xics_host, vec); if (likely(irq != NO_IRQ)) { xics_push_cppr(vec); return irq; diff --git a/arch/powerpc/sysdev/xics/icp-native.c b/arch/powerpc/sysdev/xics/icp-native.c index 4c79b6fbee1c..48861d3fcd07 100644 --- a/arch/powerpc/sysdev/xics/icp-native.c +++ b/arch/powerpc/sysdev/xics/icp-native.c @@ -119,7 +119,7 @@ static unsigned int icp_native_get_irq(void) if (vec == XICS_IRQ_SPURIOUS) return NO_IRQ; - irq = irq_radix_revmap_lookup(xics_host, vec); + irq = irq_find_mapping(xics_host, vec); if (likely(irq != NO_IRQ)) { xics_push_cppr(vec); return irq; diff --git a/include/linux/irqdomain.h b/include/linux/irqdomain.h index eab8a0e60b8e..0d5b17bf5e51 100644 --- a/include/linux/irqdomain.h +++ b/include/linux/irqdomain.h @@ -174,10 +174,6 @@ static inline int irq_create_identity_mapping(struct irq_domain *host, return irq_create_strict_mappings(host, hwirq, hwirq, 1); } -extern void irq_radix_revmap_insert(struct irq_domain *host, unsigned int virq, - irq_hw_number_t hwirq); -extern unsigned int irq_radix_revmap_lookup(struct irq_domain *host, - irq_hw_number_t hwirq); extern unsigned int irq_linear_revmap(struct irq_domain *host, irq_hw_number_t hwirq); diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index a07d92446b66..f540bb1eff84 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -450,7 +450,7 @@ int irq_domain_associate_many(struct irq_domain *domain, unsigned int irq_base, break; case IRQ_DOMAIN_MAP_TREE: mutex_lock(&revmap_trees_mutex); - irq_radix_revmap_insert(domain, virq, hwirq); + radix_tree_insert(&domain->revmap_data.tree, hwirq, irq_data); mutex_unlock(&revmap_trees_mutex); break; } @@ -723,64 +723,6 @@ unsigned int irq_find_mapping(struct irq_domain *domain, } EXPORT_SYMBOL_GPL(irq_find_mapping); -/** - * irq_radix_revmap_lookup() - Find a linux irq from a hw irq number. - * @domain: domain owning this hardware interrupt - * @hwirq: hardware irq number in that domain space - * - * This is a fast path, for use by irq controller code that uses radix tree - * revmaps - */ -unsigned int irq_radix_revmap_lookup(struct irq_domain *domain, - irq_hw_number_t hwirq) -{ - struct irq_data *irq_data; - - if (WARN_ON_ONCE(domain->revmap_type != IRQ_DOMAIN_MAP_TREE)) - return irq_find_mapping(domain, hwirq); - - /* - * Freeing an irq can delete nodes along the path to - * do the lookup via call_rcu. - */ - rcu_read_lock(); - irq_data = radix_tree_lookup(&domain->revmap_data.tree, hwirq); - rcu_read_unlock(); - - /* - * If found in radix tree, then fine. - * Else fallback to linear lookup - this should not happen in practice - * as it means that we failed to insert the node in the radix tree. - */ - return irq_data ? irq_data->irq : irq_find_mapping(domain, hwirq); -} -EXPORT_SYMBOL_GPL(irq_radix_revmap_lookup); - -/** - * irq_radix_revmap_insert() - Insert a hw irq to linux irq number mapping. - * @domain: domain owning this hardware interrupt - * @virq: linux irq number - * @hwirq: hardware irq number in that domain space - * - * This is for use by irq controllers that use a radix tree reverse - * mapping for fast lookup. - */ -void irq_radix_revmap_insert(struct irq_domain *domain, unsigned int virq, - irq_hw_number_t hwirq) -{ - struct irq_data *irq_data = irq_get_irq_data(virq); - - if (WARN_ON(domain->revmap_type != IRQ_DOMAIN_MAP_TREE)) - return; - - if (virq) { - mutex_lock(&revmap_trees_mutex); - radix_tree_insert(&domain->revmap_data.tree, hwirq, irq_data); - mutex_unlock(&revmap_trees_mutex); - } -} -EXPORT_SYMBOL_GPL(irq_radix_revmap_insert); - /** * irq_linear_revmap() - Find a linux irq from a hw irq number. * @domain: domain owning this hardware interrupt -- cgit v1.2.3 From 9844a5524ec532aee826c35e3031637c7fc8287b Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Sun, 3 Jun 2012 22:04:38 -0700 Subject: irqdomain: Fix irq_create_direct_mapping() to test irq_domain type. irq_create_direct_mapping can only be used with the NOMAP type. Make the function test to ensure it is passed the correct type of irq_domain. Signed-off-by: Grant Likely Cc: Paul Mundt Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Rob Herring --- kernel/irq/irqdomain.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index f540bb1eff84..c0e638ba9c2a 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -481,8 +481,8 @@ unsigned int irq_create_direct_mapping(struct irq_domain *domain) if (domain == NULL) domain = irq_default_domain; - BUG_ON(domain == NULL); - WARN_ON(domain->revmap_type != IRQ_DOMAIN_MAP_NOMAP); + if (WARN_ON(!domain || domain->revmap_type != IRQ_DOMAIN_MAP_NOMAP)) + return 0; virq = irq_alloc_desc_from(1, of_node_to_nid(domain->of_node)); if (!virq) { -- cgit v1.2.3 From 93574fcc5b50cc7b8834698acb2ce947e5b6a5dc Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 11 Jul 2012 09:35:08 +0300 Subject: tracing: Check for allocation failure in __tracing_open() Clean up and return -ENOMEM on if the kzalloc() fails. This also prevents a potential crash, as the pointer that failed to allocate would be later used. Link: http://lkml.kernel.org/r/20120711063507.GF11812@elgon.mountain Cc: Frederic Weisbecker Cc: Ingo Molnar Signed-off-by: Dan Carpenter Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 814ff306ae74..a120f98c4112 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2390,6 +2390,9 @@ __tracing_open(struct inode *inode, struct file *file) iter->buffer_iter = kzalloc(sizeof(*iter->buffer_iter) * num_possible_cpus(), GFP_KERNEL); + if (!iter->buffer_iter) + goto release; + /* * We make a copy of the current tracer to avoid concurrent * changes on it while we are reading. @@ -2451,6 +2454,7 @@ __tracing_open(struct inode *inode, struct file *file) mutex_unlock(&trace_types_lock); kfree(iter->trace); kfree(iter->buffer_iter); +release: seq_release_private(inode, file); return ERR_PTR(-ENOMEM); } -- cgit v1.2.3 From 974271c485a4d8bb801decc616748f90aafb07ec Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 12 Jul 2012 14:46:37 -0700 Subject: workqueue: don't use WQ_HIGHPRI for unbound workqueues Unbound wqs aren't concurrency-managed and try to execute work items as soon as possible. This is currently achieved by implicitly setting %WQ_HIGHPRI on all unbound workqueues; however, WQ_HIGHPRI implementation is about to be restructured and this usage won't be valid anymore. Add an explicit chain-wakeup path for unbound workqueues in process_one_work() instead of piggy backing on %WQ_HIGHPRI. Signed-off-by: Tejun Heo --- kernel/workqueue.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 9a3128dc67df..27637c284cb9 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -580,6 +580,10 @@ static bool __need_more_worker(struct global_cwq *gcwq) /* * Need to wake up a worker? Called from anything but currently * running workers. + * + * Note that, because unbound workers never contribute to nr_running, this + * function will always return %true for unbound gcwq as long as the + * worklist isn't empty. */ static bool need_more_worker(struct global_cwq *gcwq) { @@ -1867,6 +1871,13 @@ __acquires(&gcwq->lock) if (unlikely(cpu_intensive)) worker_set_flags(worker, WORKER_CPU_INTENSIVE, true); + /* + * Unbound gcwq isn't concurrency managed and work items should be + * executed ASAP. Wake up another worker if necessary. + */ + if ((worker->flags & WORKER_UNBOUND) && need_more_worker(gcwq)) + wake_up_worker(gcwq); + spin_unlock_irq(&gcwq->lock); work_clear_pending(work); @@ -2984,13 +2995,6 @@ struct workqueue_struct *__alloc_workqueue_key(const char *fmt, if (flags & WQ_MEM_RECLAIM) flags |= WQ_RESCUER; - /* - * Unbound workqueues aren't concurrency managed and should be - * dispatched to workers immediately. - */ - if (flags & WQ_UNBOUND) - flags |= WQ_HIGHPRI; - max_active = max_active ?: WQ_DFL_ACTIVE; max_active = wq_clamp_max_active(max_active, flags, wq->name); -- cgit v1.2.3 From bd7bdd43dcb81bb08240b9401b36a104f77dc135 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 12 Jul 2012 14:46:37 -0700 Subject: workqueue: factor out worker_pool from global_cwq Move worklist and all worker management fields from global_cwq into the new struct worker_pool. worker_pool points back to the containing gcwq. worker and cpu_workqueue_struct are updated to point to worker_pool instead of gcwq too. This change is mechanical and doesn't introduce any functional difference other than rearranging of fields and an added level of indirection in some places. This is to prepare for multiple pools per gcwq. v2: Comment typo fixes as suggested by Namhyung. Signed-off-by: Tejun Heo Cc: Namhyung Kim --- include/trace/events/workqueue.h | 2 +- kernel/workqueue.c | 216 +++++++++++++++++++++------------------ 2 files changed, 118 insertions(+), 100 deletions(-) (limited to 'kernel') diff --git a/include/trace/events/workqueue.h b/include/trace/events/workqueue.h index 4018f5058f27..f28d1b65f178 100644 --- a/include/trace/events/workqueue.h +++ b/include/trace/events/workqueue.h @@ -54,7 +54,7 @@ TRACE_EVENT(workqueue_queue_work, __entry->function = work->func; __entry->workqueue = cwq->wq; __entry->req_cpu = req_cpu; - __entry->cpu = cwq->gcwq->cpu; + __entry->cpu = cwq->pool->gcwq->cpu; ), TP_printk("work struct=%p function=%pf workqueue=%p req_cpu=%u cpu=%u", diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 27637c284cb9..61f154467026 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -115,6 +115,7 @@ enum { */ struct global_cwq; +struct worker_pool; /* * The poor guys doing the actual heavy lifting. All on-duty workers @@ -131,7 +132,7 @@ struct worker { struct cpu_workqueue_struct *current_cwq; /* L: current_work's cwq */ struct list_head scheduled; /* L: scheduled works */ struct task_struct *task; /* I: worker task */ - struct global_cwq *gcwq; /* I: the associated gcwq */ + struct worker_pool *pool; /* I: the associated pool */ /* 64 bytes boundary on 64bit, 32 on 32bit */ unsigned long last_active; /* L: last active timestamp */ unsigned int flags; /* X: flags */ @@ -139,6 +140,21 @@ struct worker { struct work_struct rebind_work; /* L: rebind worker to cpu */ }; +struct worker_pool { + struct global_cwq *gcwq; /* I: the owning gcwq */ + + struct list_head worklist; /* L: list of pending works */ + int nr_workers; /* L: total number of workers */ + int nr_idle; /* L: currently idle ones */ + + struct list_head idle_list; /* X: list of idle workers */ + struct timer_list idle_timer; /* L: worker idle timeout */ + struct timer_list mayday_timer; /* L: SOS timer for workers */ + + struct ida worker_ida; /* L: for worker IDs */ + struct worker *first_idle; /* L: first idle worker */ +}; + /* * Global per-cpu workqueue. There's one and only one for each cpu * and all works are queued and processed here regardless of their @@ -146,27 +162,18 @@ struct worker { */ struct global_cwq { spinlock_t lock; /* the gcwq lock */ - struct list_head worklist; /* L: list of pending works */ unsigned int cpu; /* I: the associated cpu */ unsigned int flags; /* L: GCWQ_* flags */ - int nr_workers; /* L: total number of workers */ - int nr_idle; /* L: currently idle ones */ - - /* workers are chained either in the idle_list or busy_hash */ - struct list_head idle_list; /* X: list of idle workers */ + /* workers are chained either in busy_hash or pool idle_list */ struct hlist_head busy_hash[BUSY_WORKER_HASH_SIZE]; /* L: hash of busy workers */ - struct timer_list idle_timer; /* L: worker idle timeout */ - struct timer_list mayday_timer; /* L: SOS timer for dworkers */ - - struct ida worker_ida; /* L: for worker IDs */ + struct worker_pool pool; /* the worker pools */ struct task_struct *trustee; /* L: for gcwq shutdown */ unsigned int trustee_state; /* L: trustee state */ wait_queue_head_t trustee_wait; /* trustee wait */ - struct worker *first_idle; /* L: first idle worker */ } ____cacheline_aligned_in_smp; /* @@ -175,7 +182,7 @@ struct global_cwq { * aligned at two's power of the number of flag bits. */ struct cpu_workqueue_struct { - struct global_cwq *gcwq; /* I: the associated gcwq */ + struct worker_pool *pool; /* I: the associated pool */ struct workqueue_struct *wq; /* I: the owning workqueue */ int work_color; /* L: current color */ int flush_color; /* L: flushing color */ @@ -555,7 +562,7 @@ static struct global_cwq *get_work_gcwq(struct work_struct *work) if (data & WORK_STRUCT_CWQ) return ((struct cpu_workqueue_struct *) - (data & WORK_STRUCT_WQ_DATA_MASK))->gcwq; + (data & WORK_STRUCT_WQ_DATA_MASK))->pool->gcwq; cpu = data >> WORK_STRUCT_FLAG_BITS; if (cpu == WORK_CPU_NONE) @@ -587,13 +594,13 @@ static bool __need_more_worker(struct global_cwq *gcwq) */ static bool need_more_worker(struct global_cwq *gcwq) { - return !list_empty(&gcwq->worklist) && __need_more_worker(gcwq); + return !list_empty(&gcwq->pool.worklist) && __need_more_worker(gcwq); } /* Can I start working? Called from busy but !running workers. */ static bool may_start_working(struct global_cwq *gcwq) { - return gcwq->nr_idle; + return gcwq->pool.nr_idle; } /* Do I need to keep working? Called from currently running workers. */ @@ -601,7 +608,7 @@ static bool keep_working(struct global_cwq *gcwq) { atomic_t *nr_running = get_gcwq_nr_running(gcwq->cpu); - return !list_empty(&gcwq->worklist) && + return !list_empty(&gcwq->pool.worklist) && (atomic_read(nr_running) <= 1 || gcwq->flags & GCWQ_HIGHPRI_PENDING); } @@ -622,8 +629,8 @@ static bool need_to_manage_workers(struct global_cwq *gcwq) static bool too_many_workers(struct global_cwq *gcwq) { bool managing = gcwq->flags & GCWQ_MANAGING_WORKERS; - int nr_idle = gcwq->nr_idle + managing; /* manager is considered idle */ - int nr_busy = gcwq->nr_workers - nr_idle; + int nr_idle = gcwq->pool.nr_idle + managing; /* manager is considered idle */ + int nr_busy = gcwq->pool.nr_workers - nr_idle; return nr_idle > 2 && (nr_idle - 2) * MAX_IDLE_WORKERS_RATIO >= nr_busy; } @@ -635,10 +642,10 @@ static bool too_many_workers(struct global_cwq *gcwq) /* Return the first worker. Safe with preemption disabled */ static struct worker *first_worker(struct global_cwq *gcwq) { - if (unlikely(list_empty(&gcwq->idle_list))) + if (unlikely(list_empty(&gcwq->pool.idle_list))) return NULL; - return list_first_entry(&gcwq->idle_list, struct worker, entry); + return list_first_entry(&gcwq->pool.idle_list, struct worker, entry); } /** @@ -696,7 +703,8 @@ struct task_struct *wq_worker_sleeping(struct task_struct *task, unsigned int cpu) { struct worker *worker = kthread_data(task), *to_wakeup = NULL; - struct global_cwq *gcwq = get_gcwq(cpu); + struct worker_pool *pool = worker->pool; + struct global_cwq *gcwq = pool->gcwq; atomic_t *nr_running = get_gcwq_nr_running(cpu); if (worker->flags & WORKER_NOT_RUNNING) @@ -716,7 +724,7 @@ struct task_struct *wq_worker_sleeping(struct task_struct *task, * could be manipulating idle_list, so dereferencing idle_list * without gcwq lock is safe. */ - if (atomic_dec_and_test(nr_running) && !list_empty(&gcwq->worklist)) + if (atomic_dec_and_test(nr_running) && !list_empty(&pool->worklist)) to_wakeup = first_worker(gcwq); return to_wakeup ? to_wakeup->task : NULL; } @@ -737,7 +745,8 @@ struct task_struct *wq_worker_sleeping(struct task_struct *task, static inline void worker_set_flags(struct worker *worker, unsigned int flags, bool wakeup) { - struct global_cwq *gcwq = worker->gcwq; + struct worker_pool *pool = worker->pool; + struct global_cwq *gcwq = pool->gcwq; WARN_ON_ONCE(worker->task != current); @@ -752,7 +761,7 @@ static inline void worker_set_flags(struct worker *worker, unsigned int flags, if (wakeup) { if (atomic_dec_and_test(nr_running) && - !list_empty(&gcwq->worklist)) + !list_empty(&pool->worklist)) wake_up_worker(gcwq); } else atomic_dec(nr_running); @@ -773,7 +782,7 @@ static inline void worker_set_flags(struct worker *worker, unsigned int flags, */ static inline void worker_clr_flags(struct worker *worker, unsigned int flags) { - struct global_cwq *gcwq = worker->gcwq; + struct global_cwq *gcwq = worker->pool->gcwq; unsigned int oflags = worker->flags; WARN_ON_ONCE(worker->task != current); @@ -894,9 +903,9 @@ static inline struct list_head *gcwq_determine_ins_pos(struct global_cwq *gcwq, struct work_struct *twork; if (likely(!(cwq->wq->flags & WQ_HIGHPRI))) - return &gcwq->worklist; + return &gcwq->pool.worklist; - list_for_each_entry(twork, &gcwq->worklist, entry) { + list_for_each_entry(twork, &gcwq->pool.worklist, entry) { struct cpu_workqueue_struct *tcwq = get_work_cwq(twork); if (!(tcwq->wq->flags & WQ_HIGHPRI)) @@ -924,7 +933,7 @@ static void insert_work(struct cpu_workqueue_struct *cwq, struct work_struct *work, struct list_head *head, unsigned int extra_flags) { - struct global_cwq *gcwq = cwq->gcwq; + struct global_cwq *gcwq = cwq->pool->gcwq; /* we own @work, set data and link */ set_work_cwq(work, cwq, extra_flags); @@ -1196,7 +1205,8 @@ EXPORT_SYMBOL_GPL(queue_delayed_work_on); */ static void worker_enter_idle(struct worker *worker) { - struct global_cwq *gcwq = worker->gcwq; + struct worker_pool *pool = worker->pool; + struct global_cwq *gcwq = pool->gcwq; BUG_ON(worker->flags & WORKER_IDLE); BUG_ON(!list_empty(&worker->entry) && @@ -1204,15 +1214,15 @@ static void worker_enter_idle(struct worker *worker) /* can't use worker_set_flags(), also called from start_worker() */ worker->flags |= WORKER_IDLE; - gcwq->nr_idle++; + pool->nr_idle++; worker->last_active = jiffies; /* idle_list is LIFO */ - list_add(&worker->entry, &gcwq->idle_list); + list_add(&worker->entry, &pool->idle_list); if (likely(!(worker->flags & WORKER_ROGUE))) { - if (too_many_workers(gcwq) && !timer_pending(&gcwq->idle_timer)) - mod_timer(&gcwq->idle_timer, + if (too_many_workers(gcwq) && !timer_pending(&pool->idle_timer)) + mod_timer(&pool->idle_timer, jiffies + IDLE_WORKER_TIMEOUT); } else wake_up_all(&gcwq->trustee_wait); @@ -1223,7 +1233,7 @@ static void worker_enter_idle(struct worker *worker) * warning may trigger spuriously. Check iff trustee is idle. */ WARN_ON_ONCE(gcwq->trustee_state == TRUSTEE_DONE && - gcwq->nr_workers == gcwq->nr_idle && + pool->nr_workers == pool->nr_idle && atomic_read(get_gcwq_nr_running(gcwq->cpu))); } @@ -1238,11 +1248,11 @@ static void worker_enter_idle(struct worker *worker) */ static void worker_leave_idle(struct worker *worker) { - struct global_cwq *gcwq = worker->gcwq; + struct worker_pool *pool = worker->pool; BUG_ON(!(worker->flags & WORKER_IDLE)); worker_clr_flags(worker, WORKER_IDLE); - gcwq->nr_idle--; + pool->nr_idle--; list_del_init(&worker->entry); } @@ -1279,7 +1289,7 @@ static void worker_leave_idle(struct worker *worker) static bool worker_maybe_bind_and_lock(struct worker *worker) __acquires(&gcwq->lock) { - struct global_cwq *gcwq = worker->gcwq; + struct global_cwq *gcwq = worker->pool->gcwq; struct task_struct *task = worker->task; while (true) { @@ -1321,7 +1331,7 @@ __acquires(&gcwq->lock) static void worker_rebind_fn(struct work_struct *work) { struct worker *worker = container_of(work, struct worker, rebind_work); - struct global_cwq *gcwq = worker->gcwq; + struct global_cwq *gcwq = worker->pool->gcwq; if (worker_maybe_bind_and_lock(worker)) worker_clr_flags(worker, WORKER_REBIND); @@ -1362,13 +1372,14 @@ static struct worker *alloc_worker(void) static struct worker *create_worker(struct global_cwq *gcwq, bool bind) { bool on_unbound_cpu = gcwq->cpu == WORK_CPU_UNBOUND; + struct worker_pool *pool = &gcwq->pool; struct worker *worker = NULL; int id = -1; spin_lock_irq(&gcwq->lock); - while (ida_get_new(&gcwq->worker_ida, &id)) { + while (ida_get_new(&pool->worker_ida, &id)) { spin_unlock_irq(&gcwq->lock); - if (!ida_pre_get(&gcwq->worker_ida, GFP_KERNEL)) + if (!ida_pre_get(&pool->worker_ida, GFP_KERNEL)) goto fail; spin_lock_irq(&gcwq->lock); } @@ -1378,7 +1389,7 @@ static struct worker *create_worker(struct global_cwq *gcwq, bool bind) if (!worker) goto fail; - worker->gcwq = gcwq; + worker->pool = pool; worker->id = id; if (!on_unbound_cpu) @@ -1409,7 +1420,7 @@ static struct worker *create_worker(struct global_cwq *gcwq, bool bind) fail: if (id >= 0) { spin_lock_irq(&gcwq->lock); - ida_remove(&gcwq->worker_ida, id); + ida_remove(&pool->worker_ida, id); spin_unlock_irq(&gcwq->lock); } kfree(worker); @@ -1428,7 +1439,7 @@ fail: static void start_worker(struct worker *worker) { worker->flags |= WORKER_STARTED; - worker->gcwq->nr_workers++; + worker->pool->nr_workers++; worker_enter_idle(worker); wake_up_process(worker->task); } @@ -1444,7 +1455,8 @@ static void start_worker(struct worker *worker) */ static void destroy_worker(struct worker *worker) { - struct global_cwq *gcwq = worker->gcwq; + struct worker_pool *pool = worker->pool; + struct global_cwq *gcwq = pool->gcwq; int id = worker->id; /* sanity check frenzy */ @@ -1452,9 +1464,9 @@ static void destroy_worker(struct worker *worker) BUG_ON(!list_empty(&worker->scheduled)); if (worker->flags & WORKER_STARTED) - gcwq->nr_workers--; + pool->nr_workers--; if (worker->flags & WORKER_IDLE) - gcwq->nr_idle--; + pool->nr_idle--; list_del_init(&worker->entry); worker->flags |= WORKER_DIE; @@ -1465,7 +1477,7 @@ static void destroy_worker(struct worker *worker) kfree(worker); spin_lock_irq(&gcwq->lock); - ida_remove(&gcwq->worker_ida, id); + ida_remove(&pool->worker_ida, id); } static void idle_worker_timeout(unsigned long __gcwq) @@ -1479,11 +1491,12 @@ static void idle_worker_timeout(unsigned long __gcwq) unsigned long expires; /* idle_list is kept in LIFO order, check the last one */ - worker = list_entry(gcwq->idle_list.prev, struct worker, entry); + worker = list_entry(gcwq->pool.idle_list.prev, struct worker, + entry); expires = worker->last_active + IDLE_WORKER_TIMEOUT; if (time_before(jiffies, expires)) - mod_timer(&gcwq->idle_timer, expires); + mod_timer(&gcwq->pool.idle_timer, expires); else { /* it's been idle for too long, wake up manager */ gcwq->flags |= GCWQ_MANAGE_WORKERS; @@ -1504,7 +1517,7 @@ static bool send_mayday(struct work_struct *work) return false; /* mayday mayday mayday */ - cpu = cwq->gcwq->cpu; + cpu = cwq->pool->gcwq->cpu; /* WORK_CPU_UNBOUND can't be set in cpumask, use cpu 0 instead */ if (cpu == WORK_CPU_UNBOUND) cpu = 0; @@ -1527,13 +1540,13 @@ static void gcwq_mayday_timeout(unsigned long __gcwq) * allocation deadlock. Send distress signals to * rescuers. */ - list_for_each_entry(work, &gcwq->worklist, entry) + list_for_each_entry(work, &gcwq->pool.worklist, entry) send_mayday(work); } spin_unlock_irq(&gcwq->lock); - mod_timer(&gcwq->mayday_timer, jiffies + MAYDAY_INTERVAL); + mod_timer(&gcwq->pool.mayday_timer, jiffies + MAYDAY_INTERVAL); } /** @@ -1568,14 +1581,14 @@ restart: spin_unlock_irq(&gcwq->lock); /* if we don't make progress in MAYDAY_INITIAL_TIMEOUT, call for help */ - mod_timer(&gcwq->mayday_timer, jiffies + MAYDAY_INITIAL_TIMEOUT); + mod_timer(&gcwq->pool.mayday_timer, jiffies + MAYDAY_INITIAL_TIMEOUT); while (true) { struct worker *worker; worker = create_worker(gcwq, true); if (worker) { - del_timer_sync(&gcwq->mayday_timer); + del_timer_sync(&gcwq->pool.mayday_timer); spin_lock_irq(&gcwq->lock); start_worker(worker); BUG_ON(need_to_create_worker(gcwq)); @@ -1592,7 +1605,7 @@ restart: break; } - del_timer_sync(&gcwq->mayday_timer); + del_timer_sync(&gcwq->pool.mayday_timer); spin_lock_irq(&gcwq->lock); if (need_to_create_worker(gcwq)) goto restart; @@ -1622,11 +1635,12 @@ static bool maybe_destroy_workers(struct global_cwq *gcwq) struct worker *worker; unsigned long expires; - worker = list_entry(gcwq->idle_list.prev, struct worker, entry); + worker = list_entry(gcwq->pool.idle_list.prev, struct worker, + entry); expires = worker->last_active + IDLE_WORKER_TIMEOUT; if (time_before(jiffies, expires)) { - mod_timer(&gcwq->idle_timer, expires); + mod_timer(&gcwq->pool.idle_timer, expires); break; } @@ -1659,7 +1673,7 @@ static bool maybe_destroy_workers(struct global_cwq *gcwq) */ static bool manage_workers(struct worker *worker) { - struct global_cwq *gcwq = worker->gcwq; + struct global_cwq *gcwq = worker->pool->gcwq; bool ret = false; if (gcwq->flags & GCWQ_MANAGING_WORKERS) @@ -1732,7 +1746,7 @@ static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq) { struct work_struct *work = list_first_entry(&cwq->delayed_works, struct work_struct, entry); - struct list_head *pos = gcwq_determine_ins_pos(cwq->gcwq, cwq); + struct list_head *pos = gcwq_determine_ins_pos(cwq->pool->gcwq, cwq); trace_workqueue_activate_work(work); move_linked_works(work, pos, NULL); @@ -1808,7 +1822,8 @@ __releases(&gcwq->lock) __acquires(&gcwq->lock) { struct cpu_workqueue_struct *cwq = get_work_cwq(work); - struct global_cwq *gcwq = cwq->gcwq; + struct worker_pool *pool = worker->pool; + struct global_cwq *gcwq = pool->gcwq; struct hlist_head *bwh = busy_worker_head(gcwq, work); bool cpu_intensive = cwq->wq->flags & WQ_CPU_INTENSIVE; work_func_t f = work->func; @@ -1854,10 +1869,10 @@ __acquires(&gcwq->lock) * wake up another worker; otherwise, clear HIGHPRI_PENDING. */ if (unlikely(gcwq->flags & GCWQ_HIGHPRI_PENDING)) { - struct work_struct *nwork = list_first_entry(&gcwq->worklist, - struct work_struct, entry); + struct work_struct *nwork = list_first_entry(&pool->worklist, + struct work_struct, entry); - if (!list_empty(&gcwq->worklist) && + if (!list_empty(&pool->worklist) && get_work_cwq(nwork)->wq->flags & WQ_HIGHPRI) wake_up_worker(gcwq); else @@ -1950,7 +1965,8 @@ static void process_scheduled_works(struct worker *worker) static int worker_thread(void *__worker) { struct worker *worker = __worker; - struct global_cwq *gcwq = worker->gcwq; + struct worker_pool *pool = worker->pool; + struct global_cwq *gcwq = pool->gcwq; /* tell the scheduler that this is a workqueue worker */ worker->task->flags |= PF_WQ_WORKER; @@ -1990,7 +2006,7 @@ recheck: do { struct work_struct *work = - list_first_entry(&gcwq->worklist, + list_first_entry(&pool->worklist, struct work_struct, entry); if (likely(!(*work_data_bits(work) & WORK_STRUCT_LINKED))) { @@ -2064,14 +2080,15 @@ repeat: for_each_mayday_cpu(cpu, wq->mayday_mask) { unsigned int tcpu = is_unbound ? WORK_CPU_UNBOUND : cpu; struct cpu_workqueue_struct *cwq = get_cwq(tcpu, wq); - struct global_cwq *gcwq = cwq->gcwq; + struct worker_pool *pool = cwq->pool; + struct global_cwq *gcwq = pool->gcwq; struct work_struct *work, *n; __set_current_state(TASK_RUNNING); mayday_clear_cpu(cpu, wq->mayday_mask); /* migrate to the target cpu if possible */ - rescuer->gcwq = gcwq; + rescuer->pool = pool; worker_maybe_bind_and_lock(rescuer); /* @@ -2079,7 +2096,7 @@ repeat: * process'em. */ BUG_ON(!list_empty(&rescuer->scheduled)); - list_for_each_entry_safe(work, n, &gcwq->worklist, entry) + list_for_each_entry_safe(work, n, &pool->worklist, entry) if (get_work_cwq(work) == cwq) move_linked_works(work, scheduled, &n); @@ -2216,7 +2233,7 @@ static bool flush_workqueue_prep_cwqs(struct workqueue_struct *wq, for_each_cwq_cpu(cpu, wq) { struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); - struct global_cwq *gcwq = cwq->gcwq; + struct global_cwq *gcwq = cwq->pool->gcwq; spin_lock_irq(&gcwq->lock); @@ -2432,9 +2449,9 @@ reflush: struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); bool drained; - spin_lock_irq(&cwq->gcwq->lock); + spin_lock_irq(&cwq->pool->gcwq->lock); drained = !cwq->nr_active && list_empty(&cwq->delayed_works); - spin_unlock_irq(&cwq->gcwq->lock); + spin_unlock_irq(&cwq->pool->gcwq->lock); if (drained) continue; @@ -2474,7 +2491,7 @@ static bool start_flush_work(struct work_struct *work, struct wq_barrier *barr, */ smp_rmb(); cwq = get_work_cwq(work); - if (unlikely(!cwq || gcwq != cwq->gcwq)) + if (unlikely(!cwq || gcwq != cwq->pool->gcwq)) goto already_gone; } else if (wait_executing) { worker = find_worker_executing_work(gcwq, work); @@ -3017,7 +3034,7 @@ struct workqueue_struct *__alloc_workqueue_key(const char *fmt, struct global_cwq *gcwq = get_gcwq(cpu); BUG_ON((unsigned long)cwq & WORK_STRUCT_FLAG_MASK); - cwq->gcwq = gcwq; + cwq->pool = &gcwq->pool; cwq->wq = wq; cwq->flush_color = -1; cwq->max_active = max_active; @@ -3344,7 +3361,7 @@ static int __cpuinit trustee_thread(void *__gcwq) gcwq->flags |= GCWQ_MANAGING_WORKERS; - list_for_each_entry(worker, &gcwq->idle_list, entry) + list_for_each_entry(worker, &gcwq->pool.idle_list, entry) worker->flags |= WORKER_ROGUE; for_each_busy_worker(worker, i, pos, gcwq) @@ -3369,7 +3386,7 @@ static int __cpuinit trustee_thread(void *__gcwq) atomic_set(get_gcwq_nr_running(gcwq->cpu), 0); spin_unlock_irq(&gcwq->lock); - del_timer_sync(&gcwq->idle_timer); + del_timer_sync(&gcwq->pool.idle_timer); spin_lock_irq(&gcwq->lock); /* @@ -3391,17 +3408,17 @@ static int __cpuinit trustee_thread(void *__gcwq) * may be frozen works in freezable cwqs. Don't declare * completion while frozen. */ - while (gcwq->nr_workers != gcwq->nr_idle || + while (gcwq->pool.nr_workers != gcwq->pool.nr_idle || gcwq->flags & GCWQ_FREEZING || gcwq->trustee_state == TRUSTEE_IN_CHARGE) { int nr_works = 0; - list_for_each_entry(work, &gcwq->worklist, entry) { + list_for_each_entry(work, &gcwq->pool.worklist, entry) { send_mayday(work); nr_works++; } - list_for_each_entry(worker, &gcwq->idle_list, entry) { + list_for_each_entry(worker, &gcwq->pool.idle_list, entry) { if (!nr_works--) break; wake_up_process(worker->task); @@ -3428,11 +3445,11 @@ static int __cpuinit trustee_thread(void *__gcwq) * all workers till we're canceled. */ do { - rc = trustee_wait_event(!list_empty(&gcwq->idle_list)); - while (!list_empty(&gcwq->idle_list)) - destroy_worker(list_first_entry(&gcwq->idle_list, + rc = trustee_wait_event(!list_empty(&gcwq->pool.idle_list)); + while (!list_empty(&gcwq->pool.idle_list)) + destroy_worker(list_first_entry(&gcwq->pool.idle_list, struct worker, entry)); - } while (gcwq->nr_workers && rc >= 0); + } while (gcwq->pool.nr_workers && rc >= 0); /* * At this point, either draining has completed and no worker @@ -3441,7 +3458,7 @@ static int __cpuinit trustee_thread(void *__gcwq) * Tell the remaining busy ones to rebind once it finishes the * currently scheduled works by scheduling the rebind_work. */ - WARN_ON(!list_empty(&gcwq->idle_list)); + WARN_ON(!list_empty(&gcwq->pool.idle_list)); for_each_busy_worker(worker, i, pos, gcwq) { struct work_struct *rebind_work = &worker->rebind_work; @@ -3522,7 +3539,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, kthread_bind(new_trustee, cpu); /* fall through */ case CPU_UP_PREPARE: - BUG_ON(gcwq->first_idle); + BUG_ON(gcwq->pool.first_idle); new_worker = create_worker(gcwq, false); if (!new_worker) { if (new_trustee) @@ -3544,8 +3561,8 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, wait_trustee_state(gcwq, TRUSTEE_IN_CHARGE); /* fall through */ case CPU_UP_PREPARE: - BUG_ON(gcwq->first_idle); - gcwq->first_idle = new_worker; + BUG_ON(gcwq->pool.first_idle); + gcwq->pool.first_idle = new_worker; break; case CPU_DYING: @@ -3562,8 +3579,8 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, gcwq->trustee_state = TRUSTEE_BUTCHER; /* fall through */ case CPU_UP_CANCELED: - destroy_worker(gcwq->first_idle); - gcwq->first_idle = NULL; + destroy_worker(gcwq->pool.first_idle); + gcwq->pool.first_idle = NULL; break; case CPU_DOWN_FAILED: @@ -3581,11 +3598,11 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, * take a look. */ spin_unlock_irq(&gcwq->lock); - kthread_bind(gcwq->first_idle->task, cpu); + kthread_bind(gcwq->pool.first_idle->task, cpu); spin_lock_irq(&gcwq->lock); gcwq->flags |= GCWQ_MANAGE_WORKERS; - start_worker(gcwq->first_idle); - gcwq->first_idle = NULL; + start_worker(gcwq->pool.first_idle); + gcwq->pool.first_idle = NULL; break; } @@ -3794,22 +3811,23 @@ static int __init init_workqueues(void) struct global_cwq *gcwq = get_gcwq(cpu); spin_lock_init(&gcwq->lock); - INIT_LIST_HEAD(&gcwq->worklist); + gcwq->pool.gcwq = gcwq; + INIT_LIST_HEAD(&gcwq->pool.worklist); gcwq->cpu = cpu; gcwq->flags |= GCWQ_DISASSOCIATED; - INIT_LIST_HEAD(&gcwq->idle_list); + INIT_LIST_HEAD(&gcwq->pool.idle_list); for (i = 0; i < BUSY_WORKER_HASH_SIZE; i++) INIT_HLIST_HEAD(&gcwq->busy_hash[i]); - init_timer_deferrable(&gcwq->idle_timer); - gcwq->idle_timer.function = idle_worker_timeout; - gcwq->idle_timer.data = (unsigned long)gcwq; + init_timer_deferrable(&gcwq->pool.idle_timer); + gcwq->pool.idle_timer.function = idle_worker_timeout; + gcwq->pool.idle_timer.data = (unsigned long)gcwq; - setup_timer(&gcwq->mayday_timer, gcwq_mayday_timeout, + setup_timer(&gcwq->pool.mayday_timer, gcwq_mayday_timeout, (unsigned long)gcwq); - ida_init(&gcwq->worker_ida); + ida_init(&gcwq->pool.worker_ida); gcwq->trustee_state = TRUSTEE_DONE; init_waitqueue_head(&gcwq->trustee_wait); -- cgit v1.2.3 From 63d95a9150ee3bbd4117fcd609dee40313b454d9 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 12 Jul 2012 14:46:37 -0700 Subject: workqueue: use @pool instead of @gcwq or @cpu where applicable Modify all functions which deal with per-pool properties to pass around @pool instead of @gcwq or @cpu. The changes in this patch are mechanical and don't caues any functional difference. This is to prepare for multiple pools per gcwq. Signed-off-by: Tejun Heo --- kernel/workqueue.c | 218 +++++++++++++++++++++++++++-------------------------- 1 file changed, 111 insertions(+), 107 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 61f154467026..2d82f7b193a0 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -471,8 +471,10 @@ static struct global_cwq *get_gcwq(unsigned int cpu) return &unbound_global_cwq; } -static atomic_t *get_gcwq_nr_running(unsigned int cpu) +static atomic_t *get_pool_nr_running(struct worker_pool *pool) { + int cpu = pool->gcwq->cpu; + if (cpu != WORK_CPU_UNBOUND) return &per_cpu(gcwq_nr_running, cpu); else @@ -578,10 +580,10 @@ static struct global_cwq *get_work_gcwq(struct work_struct *work) * assume that they're being called with gcwq->lock held. */ -static bool __need_more_worker(struct global_cwq *gcwq) +static bool __need_more_worker(struct worker_pool *pool) { - return !atomic_read(get_gcwq_nr_running(gcwq->cpu)) || - gcwq->flags & GCWQ_HIGHPRI_PENDING; + return !atomic_read(get_pool_nr_running(pool)) || + pool->gcwq->flags & GCWQ_HIGHPRI_PENDING; } /* @@ -592,45 +594,46 @@ static bool __need_more_worker(struct global_cwq *gcwq) * function will always return %true for unbound gcwq as long as the * worklist isn't empty. */ -static bool need_more_worker(struct global_cwq *gcwq) +static bool need_more_worker(struct worker_pool *pool) { - return !list_empty(&gcwq->pool.worklist) && __need_more_worker(gcwq); + return !list_empty(&pool->worklist) && __need_more_worker(pool); } /* Can I start working? Called from busy but !running workers. */ -static bool may_start_working(struct global_cwq *gcwq) +static bool may_start_working(struct worker_pool *pool) { - return gcwq->pool.nr_idle; + return pool->nr_idle; } /* Do I need to keep working? Called from currently running workers. */ -static bool keep_working(struct global_cwq *gcwq) +static bool keep_working(struct worker_pool *pool) { - atomic_t *nr_running = get_gcwq_nr_running(gcwq->cpu); + atomic_t *nr_running = get_pool_nr_running(pool); - return !list_empty(&gcwq->pool.worklist) && + return !list_empty(&pool->worklist) && (atomic_read(nr_running) <= 1 || - gcwq->flags & GCWQ_HIGHPRI_PENDING); + pool->gcwq->flags & GCWQ_HIGHPRI_PENDING); } /* Do we need a new worker? Called from manager. */ -static bool need_to_create_worker(struct global_cwq *gcwq) +static bool need_to_create_worker(struct worker_pool *pool) { - return need_more_worker(gcwq) && !may_start_working(gcwq); + return need_more_worker(pool) && !may_start_working(pool); } /* Do I need to be the manager? */ -static bool need_to_manage_workers(struct global_cwq *gcwq) +static bool need_to_manage_workers(struct worker_pool *pool) { - return need_to_create_worker(gcwq) || gcwq->flags & GCWQ_MANAGE_WORKERS; + return need_to_create_worker(pool) || + pool->gcwq->flags & GCWQ_MANAGE_WORKERS; } /* Do we have too many workers and should some go away? */ -static bool too_many_workers(struct global_cwq *gcwq) +static bool too_many_workers(struct worker_pool *pool) { - bool managing = gcwq->flags & GCWQ_MANAGING_WORKERS; - int nr_idle = gcwq->pool.nr_idle + managing; /* manager is considered idle */ - int nr_busy = gcwq->pool.nr_workers - nr_idle; + bool managing = pool->gcwq->flags & GCWQ_MANAGING_WORKERS; + int nr_idle = pool->nr_idle + managing; /* manager is considered idle */ + int nr_busy = pool->nr_workers - nr_idle; return nr_idle > 2 && (nr_idle - 2) * MAX_IDLE_WORKERS_RATIO >= nr_busy; } @@ -640,26 +643,26 @@ static bool too_many_workers(struct global_cwq *gcwq) */ /* Return the first worker. Safe with preemption disabled */ -static struct worker *first_worker(struct global_cwq *gcwq) +static struct worker *first_worker(struct worker_pool *pool) { - if (unlikely(list_empty(&gcwq->pool.idle_list))) + if (unlikely(list_empty(&pool->idle_list))) return NULL; - return list_first_entry(&gcwq->pool.idle_list, struct worker, entry); + return list_first_entry(&pool->idle_list, struct worker, entry); } /** * wake_up_worker - wake up an idle worker - * @gcwq: gcwq to wake worker for + * @pool: worker pool to wake worker from * - * Wake up the first idle worker of @gcwq. + * Wake up the first idle worker of @pool. * * CONTEXT: * spin_lock_irq(gcwq->lock). */ -static void wake_up_worker(struct global_cwq *gcwq) +static void wake_up_worker(struct worker_pool *pool) { - struct worker *worker = first_worker(gcwq); + struct worker *worker = first_worker(pool); if (likely(worker)) wake_up_process(worker->task); @@ -681,7 +684,7 @@ void wq_worker_waking_up(struct task_struct *task, unsigned int cpu) struct worker *worker = kthread_data(task); if (!(worker->flags & WORKER_NOT_RUNNING)) - atomic_inc(get_gcwq_nr_running(cpu)); + atomic_inc(get_pool_nr_running(worker->pool)); } /** @@ -704,8 +707,7 @@ struct task_struct *wq_worker_sleeping(struct task_struct *task, { struct worker *worker = kthread_data(task), *to_wakeup = NULL; struct worker_pool *pool = worker->pool; - struct global_cwq *gcwq = pool->gcwq; - atomic_t *nr_running = get_gcwq_nr_running(cpu); + atomic_t *nr_running = get_pool_nr_running(pool); if (worker->flags & WORKER_NOT_RUNNING) return NULL; @@ -725,7 +727,7 @@ struct task_struct *wq_worker_sleeping(struct task_struct *task, * without gcwq lock is safe. */ if (atomic_dec_and_test(nr_running) && !list_empty(&pool->worklist)) - to_wakeup = first_worker(gcwq); + to_wakeup = first_worker(pool); return to_wakeup ? to_wakeup->task : NULL; } @@ -746,7 +748,6 @@ static inline void worker_set_flags(struct worker *worker, unsigned int flags, bool wakeup) { struct worker_pool *pool = worker->pool; - struct global_cwq *gcwq = pool->gcwq; WARN_ON_ONCE(worker->task != current); @@ -757,12 +758,12 @@ static inline void worker_set_flags(struct worker *worker, unsigned int flags, */ if ((flags & WORKER_NOT_RUNNING) && !(worker->flags & WORKER_NOT_RUNNING)) { - atomic_t *nr_running = get_gcwq_nr_running(gcwq->cpu); + atomic_t *nr_running = get_pool_nr_running(pool); if (wakeup) { if (atomic_dec_and_test(nr_running) && !list_empty(&pool->worklist)) - wake_up_worker(gcwq); + wake_up_worker(pool); } else atomic_dec(nr_running); } @@ -782,7 +783,7 @@ static inline void worker_set_flags(struct worker *worker, unsigned int flags, */ static inline void worker_clr_flags(struct worker *worker, unsigned int flags) { - struct global_cwq *gcwq = worker->pool->gcwq; + struct worker_pool *pool = worker->pool; unsigned int oflags = worker->flags; WARN_ON_ONCE(worker->task != current); @@ -796,7 +797,7 @@ static inline void worker_clr_flags(struct worker *worker, unsigned int flags) */ if ((flags & WORKER_NOT_RUNNING) && (oflags & WORKER_NOT_RUNNING)) if (!(worker->flags & WORKER_NOT_RUNNING)) - atomic_inc(get_gcwq_nr_running(gcwq->cpu)); + atomic_inc(get_pool_nr_running(pool)); } /** @@ -880,15 +881,15 @@ static struct worker *find_worker_executing_work(struct global_cwq *gcwq, } /** - * gcwq_determine_ins_pos - find insertion position - * @gcwq: gcwq of interest + * pool_determine_ins_pos - find insertion position + * @pool: pool of interest * @cwq: cwq a work is being queued for * - * A work for @cwq is about to be queued on @gcwq, determine insertion + * A work for @cwq is about to be queued on @pool, determine insertion * position for the work. If @cwq is for HIGHPRI wq, the work is * queued at the head of the queue but in FIFO order with respect to * other HIGHPRI works; otherwise, at the end of the queue. This - * function also sets GCWQ_HIGHPRI_PENDING flag to hint @gcwq that + * function also sets GCWQ_HIGHPRI_PENDING flag to hint @pool that * there are HIGHPRI works pending. * * CONTEXT: @@ -897,22 +898,22 @@ static struct worker *find_worker_executing_work(struct global_cwq *gcwq, * RETURNS: * Pointer to inserstion position. */ -static inline struct list_head *gcwq_determine_ins_pos(struct global_cwq *gcwq, +static inline struct list_head *pool_determine_ins_pos(struct worker_pool *pool, struct cpu_workqueue_struct *cwq) { struct work_struct *twork; if (likely(!(cwq->wq->flags & WQ_HIGHPRI))) - return &gcwq->pool.worklist; + return &pool->worklist; - list_for_each_entry(twork, &gcwq->pool.worklist, entry) { + list_for_each_entry(twork, &pool->worklist, entry) { struct cpu_workqueue_struct *tcwq = get_work_cwq(twork); if (!(tcwq->wq->flags & WQ_HIGHPRI)) break; } - gcwq->flags |= GCWQ_HIGHPRI_PENDING; + pool->gcwq->flags |= GCWQ_HIGHPRI_PENDING; return &twork->entry; } @@ -933,7 +934,7 @@ static void insert_work(struct cpu_workqueue_struct *cwq, struct work_struct *work, struct list_head *head, unsigned int extra_flags) { - struct global_cwq *gcwq = cwq->pool->gcwq; + struct worker_pool *pool = cwq->pool; /* we own @work, set data and link */ set_work_cwq(work, cwq, extra_flags); @@ -953,8 +954,8 @@ static void insert_work(struct cpu_workqueue_struct *cwq, */ smp_mb(); - if (__need_more_worker(gcwq)) - wake_up_worker(gcwq); + if (__need_more_worker(pool)) + wake_up_worker(pool); } /* @@ -1056,7 +1057,7 @@ static void __queue_work(unsigned int cpu, struct workqueue_struct *wq, if (likely(cwq->nr_active < cwq->max_active)) { trace_workqueue_activate_work(work); cwq->nr_active++; - worklist = gcwq_determine_ins_pos(gcwq, cwq); + worklist = pool_determine_ins_pos(cwq->pool, cwq); } else { work_flags |= WORK_STRUCT_DELAYED; worklist = &cwq->delayed_works; @@ -1221,7 +1222,7 @@ static void worker_enter_idle(struct worker *worker) list_add(&worker->entry, &pool->idle_list); if (likely(!(worker->flags & WORKER_ROGUE))) { - if (too_many_workers(gcwq) && !timer_pending(&pool->idle_timer)) + if (too_many_workers(pool) && !timer_pending(&pool->idle_timer)) mod_timer(&pool->idle_timer, jiffies + IDLE_WORKER_TIMEOUT); } else @@ -1234,7 +1235,7 @@ static void worker_enter_idle(struct worker *worker) */ WARN_ON_ONCE(gcwq->trustee_state == TRUSTEE_DONE && pool->nr_workers == pool->nr_idle && - atomic_read(get_gcwq_nr_running(gcwq->cpu))); + atomic_read(get_pool_nr_running(pool))); } /** @@ -1356,10 +1357,10 @@ static struct worker *alloc_worker(void) /** * create_worker - create a new workqueue worker - * @gcwq: gcwq the new worker will belong to + * @pool: pool the new worker will belong to * @bind: whether to set affinity to @cpu or not * - * Create a new worker which is bound to @gcwq. The returned worker + * Create a new worker which is bound to @pool. The returned worker * can be started by calling start_worker() or destroyed using * destroy_worker(). * @@ -1369,10 +1370,10 @@ static struct worker *alloc_worker(void) * RETURNS: * Pointer to the newly created worker. */ -static struct worker *create_worker(struct global_cwq *gcwq, bool bind) +static struct worker *create_worker(struct worker_pool *pool, bool bind) { + struct global_cwq *gcwq = pool->gcwq; bool on_unbound_cpu = gcwq->cpu == WORK_CPU_UNBOUND; - struct worker_pool *pool = &gcwq->pool; struct worker *worker = NULL; int id = -1; @@ -1480,27 +1481,27 @@ static void destroy_worker(struct worker *worker) ida_remove(&pool->worker_ida, id); } -static void idle_worker_timeout(unsigned long __gcwq) +static void idle_worker_timeout(unsigned long __pool) { - struct global_cwq *gcwq = (void *)__gcwq; + struct worker_pool *pool = (void *)__pool; + struct global_cwq *gcwq = pool->gcwq; spin_lock_irq(&gcwq->lock); - if (too_many_workers(gcwq)) { + if (too_many_workers(pool)) { struct worker *worker; unsigned long expires; /* idle_list is kept in LIFO order, check the last one */ - worker = list_entry(gcwq->pool.idle_list.prev, struct worker, - entry); + worker = list_entry(pool->idle_list.prev, struct worker, entry); expires = worker->last_active + IDLE_WORKER_TIMEOUT; if (time_before(jiffies, expires)) - mod_timer(&gcwq->pool.idle_timer, expires); + mod_timer(&pool->idle_timer, expires); else { /* it's been idle for too long, wake up manager */ gcwq->flags |= GCWQ_MANAGE_WORKERS; - wake_up_worker(gcwq); + wake_up_worker(pool); } } @@ -1526,37 +1527,38 @@ static bool send_mayday(struct work_struct *work) return true; } -static void gcwq_mayday_timeout(unsigned long __gcwq) +static void gcwq_mayday_timeout(unsigned long __pool) { - struct global_cwq *gcwq = (void *)__gcwq; + struct worker_pool *pool = (void *)__pool; + struct global_cwq *gcwq = pool->gcwq; struct work_struct *work; spin_lock_irq(&gcwq->lock); - if (need_to_create_worker(gcwq)) { + if (need_to_create_worker(pool)) { /* * We've been trying to create a new worker but * haven't been successful. We might be hitting an * allocation deadlock. Send distress signals to * rescuers. */ - list_for_each_entry(work, &gcwq->pool.worklist, entry) + list_for_each_entry(work, &pool->worklist, entry) send_mayday(work); } spin_unlock_irq(&gcwq->lock); - mod_timer(&gcwq->pool.mayday_timer, jiffies + MAYDAY_INTERVAL); + mod_timer(&pool->mayday_timer, jiffies + MAYDAY_INTERVAL); } /** * maybe_create_worker - create a new worker if necessary - * @gcwq: gcwq to create a new worker for + * @pool: pool to create a new worker for * - * Create a new worker for @gcwq if necessary. @gcwq is guaranteed to + * Create a new worker for @pool if necessary. @pool is guaranteed to * have at least one idle worker on return from this function. If * creating a new worker takes longer than MAYDAY_INTERVAL, mayday is - * sent to all rescuers with works scheduled on @gcwq to resolve + * sent to all rescuers with works scheduled on @pool to resolve * possible allocation deadlock. * * On return, need_to_create_worker() is guaranteed to be false and @@ -1571,52 +1573,54 @@ static void gcwq_mayday_timeout(unsigned long __gcwq) * false if no action was taken and gcwq->lock stayed locked, true * otherwise. */ -static bool maybe_create_worker(struct global_cwq *gcwq) +static bool maybe_create_worker(struct worker_pool *pool) __releases(&gcwq->lock) __acquires(&gcwq->lock) { - if (!need_to_create_worker(gcwq)) + struct global_cwq *gcwq = pool->gcwq; + + if (!need_to_create_worker(pool)) return false; restart: spin_unlock_irq(&gcwq->lock); /* if we don't make progress in MAYDAY_INITIAL_TIMEOUT, call for help */ - mod_timer(&gcwq->pool.mayday_timer, jiffies + MAYDAY_INITIAL_TIMEOUT); + mod_timer(&pool->mayday_timer, jiffies + MAYDAY_INITIAL_TIMEOUT); while (true) { struct worker *worker; - worker = create_worker(gcwq, true); + worker = create_worker(pool, true); if (worker) { - del_timer_sync(&gcwq->pool.mayday_timer); + del_timer_sync(&pool->mayday_timer); spin_lock_irq(&gcwq->lock); start_worker(worker); - BUG_ON(need_to_create_worker(gcwq)); + BUG_ON(need_to_create_worker(pool)); return true; } - if (!need_to_create_worker(gcwq)) + if (!need_to_create_worker(pool)) break; __set_current_state(TASK_INTERRUPTIBLE); schedule_timeout(CREATE_COOLDOWN); - if (!need_to_create_worker(gcwq)) + if (!need_to_create_worker(pool)) break; } - del_timer_sync(&gcwq->pool.mayday_timer); + del_timer_sync(&pool->mayday_timer); spin_lock_irq(&gcwq->lock); - if (need_to_create_worker(gcwq)) + if (need_to_create_worker(pool)) goto restart; return true; } /** * maybe_destroy_worker - destroy workers which have been idle for a while - * @gcwq: gcwq to destroy workers for + * @pool: pool to destroy workers for * - * Destroy @gcwq workers which have been idle for longer than + * Destroy @pool workers which have been idle for longer than * IDLE_WORKER_TIMEOUT. * * LOCKING: @@ -1627,20 +1631,19 @@ restart: * false if no action was taken and gcwq->lock stayed locked, true * otherwise. */ -static bool maybe_destroy_workers(struct global_cwq *gcwq) +static bool maybe_destroy_workers(struct worker_pool *pool) { bool ret = false; - while (too_many_workers(gcwq)) { + while (too_many_workers(pool)) { struct worker *worker; unsigned long expires; - worker = list_entry(gcwq->pool.idle_list.prev, struct worker, - entry); + worker = list_entry(pool->idle_list.prev, struct worker, entry); expires = worker->last_active + IDLE_WORKER_TIMEOUT; if (time_before(jiffies, expires)) { - mod_timer(&gcwq->pool.idle_timer, expires); + mod_timer(&pool->idle_timer, expires); break; } @@ -1673,7 +1676,8 @@ static bool maybe_destroy_workers(struct global_cwq *gcwq) */ static bool manage_workers(struct worker *worker) { - struct global_cwq *gcwq = worker->pool->gcwq; + struct worker_pool *pool = worker->pool; + struct global_cwq *gcwq = pool->gcwq; bool ret = false; if (gcwq->flags & GCWQ_MANAGING_WORKERS) @@ -1686,8 +1690,8 @@ static bool manage_workers(struct worker *worker) * Destroy and then create so that may_start_working() is true * on return. */ - ret |= maybe_destroy_workers(gcwq); - ret |= maybe_create_worker(gcwq); + ret |= maybe_destroy_workers(pool); + ret |= maybe_create_worker(pool); gcwq->flags &= ~GCWQ_MANAGING_WORKERS; @@ -1746,7 +1750,7 @@ static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq) { struct work_struct *work = list_first_entry(&cwq->delayed_works, struct work_struct, entry); - struct list_head *pos = gcwq_determine_ins_pos(cwq->pool->gcwq, cwq); + struct list_head *pos = pool_determine_ins_pos(cwq->pool, cwq); trace_workqueue_activate_work(work); move_linked_works(work, pos, NULL); @@ -1874,7 +1878,7 @@ __acquires(&gcwq->lock) if (!list_empty(&pool->worklist) && get_work_cwq(nwork)->wq->flags & WQ_HIGHPRI) - wake_up_worker(gcwq); + wake_up_worker(pool); else gcwq->flags &= ~GCWQ_HIGHPRI_PENDING; } @@ -1890,8 +1894,8 @@ __acquires(&gcwq->lock) * Unbound gcwq isn't concurrency managed and work items should be * executed ASAP. Wake up another worker if necessary. */ - if ((worker->flags & WORKER_UNBOUND) && need_more_worker(gcwq)) - wake_up_worker(gcwq); + if ((worker->flags & WORKER_UNBOUND) && need_more_worker(pool)) + wake_up_worker(pool); spin_unlock_irq(&gcwq->lock); @@ -1983,11 +1987,11 @@ woke_up: worker_leave_idle(worker); recheck: /* no more worker necessary? */ - if (!need_more_worker(gcwq)) + if (!need_more_worker(pool)) goto sleep; /* do we need to manage? */ - if (unlikely(!may_start_working(gcwq)) && manage_workers(worker)) + if (unlikely(!may_start_working(pool)) && manage_workers(worker)) goto recheck; /* @@ -2018,11 +2022,11 @@ recheck: move_linked_works(work, &worker->scheduled, NULL); process_scheduled_works(worker); } - } while (keep_working(gcwq)); + } while (keep_working(pool)); worker_set_flags(worker, WORKER_PREP, false); sleep: - if (unlikely(need_to_manage_workers(gcwq)) && manage_workers(worker)) + if (unlikely(need_to_manage_workers(pool)) && manage_workers(worker)) goto recheck; /* @@ -2107,8 +2111,8 @@ repeat: * regular worker; otherwise, we end up with 0 concurrency * and stalling the execution. */ - if (keep_working(gcwq)) - wake_up_worker(gcwq); + if (keep_working(pool)) + wake_up_worker(pool); spin_unlock_irq(&gcwq->lock); } @@ -3383,7 +3387,7 @@ static int __cpuinit trustee_thread(void *__gcwq) * keep_working() are always true as long as the worklist is * not empty. */ - atomic_set(get_gcwq_nr_running(gcwq->cpu), 0); + atomic_set(get_pool_nr_running(&gcwq->pool), 0); spin_unlock_irq(&gcwq->lock); del_timer_sync(&gcwq->pool.idle_timer); @@ -3424,9 +3428,9 @@ static int __cpuinit trustee_thread(void *__gcwq) wake_up_process(worker->task); } - if (need_to_create_worker(gcwq)) { + if (need_to_create_worker(&gcwq->pool)) { spin_unlock_irq(&gcwq->lock); - worker = create_worker(gcwq, false); + worker = create_worker(&gcwq->pool, false); spin_lock_irq(&gcwq->lock); if (worker) { worker->flags |= WORKER_ROGUE; @@ -3540,7 +3544,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, /* fall through */ case CPU_UP_PREPARE: BUG_ON(gcwq->pool.first_idle); - new_worker = create_worker(gcwq, false); + new_worker = create_worker(&gcwq->pool, false); if (!new_worker) { if (new_trustee) kthread_stop(new_trustee); @@ -3788,7 +3792,7 @@ void thaw_workqueues(void) cwq_activate_first_delayed(cwq); } - wake_up_worker(gcwq); + wake_up_worker(&gcwq->pool); spin_unlock_irq(&gcwq->lock); } @@ -3822,10 +3826,10 @@ static int __init init_workqueues(void) init_timer_deferrable(&gcwq->pool.idle_timer); gcwq->pool.idle_timer.function = idle_worker_timeout; - gcwq->pool.idle_timer.data = (unsigned long)gcwq; + gcwq->pool.idle_timer.data = (unsigned long)&gcwq->pool; setup_timer(&gcwq->pool.mayday_timer, gcwq_mayday_timeout, - (unsigned long)gcwq); + (unsigned long)&gcwq->pool); ida_init(&gcwq->pool.worker_ida); @@ -3840,7 +3844,7 @@ static int __init init_workqueues(void) if (cpu != WORK_CPU_UNBOUND) gcwq->flags &= ~GCWQ_DISASSOCIATED; - worker = create_worker(gcwq, true); + worker = create_worker(&gcwq->pool, true); BUG_ON(!worker); spin_lock_irq(&gcwq->lock); start_worker(worker); -- cgit v1.2.3 From 11ebea50dbc1ade5994b2c838a096078d4c02399 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 12 Jul 2012 14:46:37 -0700 Subject: workqueue: separate out worker_pool flags GCWQ_MANAGE_WORKERS, GCWQ_MANAGING_WORKERS and GCWQ_HIGHPRI_PENDING are per-pool properties. Add worker_pool->flags and make the above three flags per-pool flags. The changes in this patch are mechanical and don't caues any functional difference. This is to prepare for multiple pools per gcwq. Signed-off-by: Tejun Heo --- kernel/workqueue.c | 47 +++++++++++++++++++++++++---------------------- 1 file changed, 25 insertions(+), 22 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 2d82f7b193a0..7a98bae635fa 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -46,11 +46,13 @@ enum { /* global_cwq flags */ - GCWQ_MANAGE_WORKERS = 1 << 0, /* need to manage workers */ - GCWQ_MANAGING_WORKERS = 1 << 1, /* managing workers */ - GCWQ_DISASSOCIATED = 1 << 2, /* cpu can't serve workers */ - GCWQ_FREEZING = 1 << 3, /* freeze in progress */ - GCWQ_HIGHPRI_PENDING = 1 << 4, /* highpri works on queue */ + GCWQ_DISASSOCIATED = 1 << 0, /* cpu can't serve workers */ + GCWQ_FREEZING = 1 << 1, /* freeze in progress */ + + /* pool flags */ + POOL_MANAGE_WORKERS = 1 << 0, /* need to manage workers */ + POOL_MANAGING_WORKERS = 1 << 1, /* managing workers */ + POOL_HIGHPRI_PENDING = 1 << 2, /* highpri works on queue */ /* worker flags */ WORKER_STARTED = 1 << 0, /* started */ @@ -142,6 +144,7 @@ struct worker { struct worker_pool { struct global_cwq *gcwq; /* I: the owning gcwq */ + unsigned int flags; /* X: flags */ struct list_head worklist; /* L: list of pending works */ int nr_workers; /* L: total number of workers */ @@ -583,7 +586,7 @@ static struct global_cwq *get_work_gcwq(struct work_struct *work) static bool __need_more_worker(struct worker_pool *pool) { return !atomic_read(get_pool_nr_running(pool)) || - pool->gcwq->flags & GCWQ_HIGHPRI_PENDING; + (pool->flags & POOL_HIGHPRI_PENDING); } /* @@ -612,7 +615,7 @@ static bool keep_working(struct worker_pool *pool) return !list_empty(&pool->worklist) && (atomic_read(nr_running) <= 1 || - pool->gcwq->flags & GCWQ_HIGHPRI_PENDING); + (pool->flags & POOL_HIGHPRI_PENDING)); } /* Do we need a new worker? Called from manager. */ @@ -625,13 +628,13 @@ static bool need_to_create_worker(struct worker_pool *pool) static bool need_to_manage_workers(struct worker_pool *pool) { return need_to_create_worker(pool) || - pool->gcwq->flags & GCWQ_MANAGE_WORKERS; + (pool->flags & POOL_MANAGE_WORKERS); } /* Do we have too many workers and should some go away? */ static bool too_many_workers(struct worker_pool *pool) { - bool managing = pool->gcwq->flags & GCWQ_MANAGING_WORKERS; + bool managing = pool->flags & POOL_MANAGING_WORKERS; int nr_idle = pool->nr_idle + managing; /* manager is considered idle */ int nr_busy = pool->nr_workers - nr_idle; @@ -889,7 +892,7 @@ static struct worker *find_worker_executing_work(struct global_cwq *gcwq, * position for the work. If @cwq is for HIGHPRI wq, the work is * queued at the head of the queue but in FIFO order with respect to * other HIGHPRI works; otherwise, at the end of the queue. This - * function also sets GCWQ_HIGHPRI_PENDING flag to hint @pool that + * function also sets POOL_HIGHPRI_PENDING flag to hint @pool that * there are HIGHPRI works pending. * * CONTEXT: @@ -913,7 +916,7 @@ static inline struct list_head *pool_determine_ins_pos(struct worker_pool *pool, break; } - pool->gcwq->flags |= GCWQ_HIGHPRI_PENDING; + pool->flags |= POOL_HIGHPRI_PENDING; return &twork->entry; } @@ -1500,7 +1503,7 @@ static void idle_worker_timeout(unsigned long __pool) mod_timer(&pool->idle_timer, expires); else { /* it's been idle for too long, wake up manager */ - gcwq->flags |= GCWQ_MANAGE_WORKERS; + pool->flags |= POOL_MANAGE_WORKERS; wake_up_worker(pool); } } @@ -1680,11 +1683,11 @@ static bool manage_workers(struct worker *worker) struct global_cwq *gcwq = pool->gcwq; bool ret = false; - if (gcwq->flags & GCWQ_MANAGING_WORKERS) + if (pool->flags & POOL_MANAGING_WORKERS) return ret; - gcwq->flags &= ~GCWQ_MANAGE_WORKERS; - gcwq->flags |= GCWQ_MANAGING_WORKERS; + pool->flags &= ~POOL_MANAGE_WORKERS; + pool->flags |= POOL_MANAGING_WORKERS; /* * Destroy and then create so that may_start_working() is true @@ -1693,7 +1696,7 @@ static bool manage_workers(struct worker *worker) ret |= maybe_destroy_workers(pool); ret |= maybe_create_worker(pool); - gcwq->flags &= ~GCWQ_MANAGING_WORKERS; + pool->flags &= ~POOL_MANAGING_WORKERS; /* * The trustee might be waiting to take over the manager @@ -1872,7 +1875,7 @@ __acquires(&gcwq->lock) * If HIGHPRI_PENDING, check the next work, and, if HIGHPRI, * wake up another worker; otherwise, clear HIGHPRI_PENDING. */ - if (unlikely(gcwq->flags & GCWQ_HIGHPRI_PENDING)) { + if (unlikely(pool->flags & POOL_HIGHPRI_PENDING)) { struct work_struct *nwork = list_first_entry(&pool->worklist, struct work_struct, entry); @@ -1880,7 +1883,7 @@ __acquires(&gcwq->lock) get_work_cwq(nwork)->wq->flags & WQ_HIGHPRI) wake_up_worker(pool); else - gcwq->flags &= ~GCWQ_HIGHPRI_PENDING; + pool->flags &= ~POOL_HIGHPRI_PENDING; } /* @@ -3360,10 +3363,10 @@ static int __cpuinit trustee_thread(void *__gcwq) * cancelled. */ BUG_ON(gcwq->cpu != smp_processor_id()); - rc = trustee_wait_event(!(gcwq->flags & GCWQ_MANAGING_WORKERS)); + rc = trustee_wait_event(!(gcwq->pool.flags & POOL_MANAGING_WORKERS)); BUG_ON(rc < 0); - gcwq->flags |= GCWQ_MANAGING_WORKERS; + gcwq->pool.flags |= POOL_MANAGING_WORKERS; list_for_each_entry(worker, &gcwq->pool.idle_list, entry) worker->flags |= WORKER_ROGUE; @@ -3487,7 +3490,7 @@ static int __cpuinit trustee_thread(void *__gcwq) } /* relinquish manager role */ - gcwq->flags &= ~GCWQ_MANAGING_WORKERS; + gcwq->pool.flags &= ~POOL_MANAGING_WORKERS; /* notify completion */ gcwq->trustee = NULL; @@ -3604,7 +3607,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, spin_unlock_irq(&gcwq->lock); kthread_bind(gcwq->pool.first_idle->task, cpu); spin_lock_irq(&gcwq->lock); - gcwq->flags |= GCWQ_MANAGE_WORKERS; + gcwq->pool.flags |= POOL_MANAGE_WORKERS; start_worker(gcwq->pool.first_idle); gcwq->pool.first_idle = NULL; break; -- cgit v1.2.3 From 4ce62e9e30cacc26885cab133ad1de358dd79f21 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 13 Jul 2012 22:16:44 -0700 Subject: workqueue: introduce NR_WORKER_POOLS and for_each_worker_pool() Introduce NR_WORKER_POOLS and for_each_worker_pool() and convert code paths which need to manipulate all pools in a gcwq to use them. NR_WORKER_POOLS is currently one and for_each_worker_pool() iterates over only @gcwq->pool. Note that nr_running is per-pool property and converted to an array with NR_WORKER_POOLS elements and renamed to pool_nr_running. Note that get_pool_nr_running() currently assumes 0 index. The next patch will make use of non-zero index. The changes in this patch are mechanical and don't caues any functional difference. This is to prepare for multiple pools per gcwq. v2: nr_running indexing bug in get_pool_nr_running() fixed. v3: Pointer to array is stupid. Don't use it in get_pool_nr_running() as suggested by Linus. Signed-off-by: Tejun Heo Cc: Tony Luck Cc: Fengguang Wu Cc: Linus Torvalds --- kernel/workqueue.c | 223 ++++++++++++++++++++++++++++++++++++----------------- 1 file changed, 153 insertions(+), 70 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 7a98bae635fa..b0daaea44eaa 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -74,6 +74,8 @@ enum { TRUSTEE_RELEASE = 3, /* release workers */ TRUSTEE_DONE = 4, /* trustee is done */ + NR_WORKER_POOLS = 1, /* # worker pools per gcwq */ + BUSY_WORKER_HASH_ORDER = 6, /* 64 pointers */ BUSY_WORKER_HASH_SIZE = 1 << BUSY_WORKER_HASH_ORDER, BUSY_WORKER_HASH_MASK = BUSY_WORKER_HASH_SIZE - 1, @@ -274,6 +276,9 @@ EXPORT_SYMBOL_GPL(system_nrt_freezable_wq); #define CREATE_TRACE_POINTS #include +#define for_each_worker_pool(pool, gcwq) \ + for ((pool) = &(gcwq)->pool; (pool); (pool) = NULL) + #define for_each_busy_worker(worker, i, pos, gcwq) \ for (i = 0; i < BUSY_WORKER_HASH_SIZE; i++) \ hlist_for_each_entry(worker, pos, &gcwq->busy_hash[i], hentry) @@ -454,7 +459,7 @@ static bool workqueue_freezing; /* W: have wqs started freezing? */ * try_to_wake_up(). Put it in a separate cacheline. */ static DEFINE_PER_CPU(struct global_cwq, global_cwq); -static DEFINE_PER_CPU_SHARED_ALIGNED(atomic_t, gcwq_nr_running); +static DEFINE_PER_CPU_SHARED_ALIGNED(atomic_t, pool_nr_running[NR_WORKER_POOLS]); /* * Global cpu workqueue and nr_running counter for unbound gcwq. The @@ -462,7 +467,9 @@ static DEFINE_PER_CPU_SHARED_ALIGNED(atomic_t, gcwq_nr_running); * workers have WORKER_UNBOUND set. */ static struct global_cwq unbound_global_cwq; -static atomic_t unbound_gcwq_nr_running = ATOMIC_INIT(0); /* always 0 */ +static atomic_t unbound_pool_nr_running[NR_WORKER_POOLS] = { + [0 ... NR_WORKER_POOLS - 1] = ATOMIC_INIT(0), /* always 0 */ +}; static int worker_thread(void *__worker); @@ -477,11 +484,12 @@ static struct global_cwq *get_gcwq(unsigned int cpu) static atomic_t *get_pool_nr_running(struct worker_pool *pool) { int cpu = pool->gcwq->cpu; + int idx = 0; if (cpu != WORK_CPU_UNBOUND) - return &per_cpu(gcwq_nr_running, cpu); + return &per_cpu(pool_nr_running, cpu)[idx]; else - return &unbound_gcwq_nr_running; + return &unbound_pool_nr_running[idx]; } static struct cpu_workqueue_struct *get_cwq(unsigned int cpu, @@ -3345,9 +3353,30 @@ EXPORT_SYMBOL_GPL(work_busy); __ret1 < 0 ? -1 : 0; \ }) +static bool gcwq_is_managing_workers(struct global_cwq *gcwq) +{ + struct worker_pool *pool; + + for_each_worker_pool(pool, gcwq) + if (pool->flags & POOL_MANAGING_WORKERS) + return true; + return false; +} + +static bool gcwq_has_idle_workers(struct global_cwq *gcwq) +{ + struct worker_pool *pool; + + for_each_worker_pool(pool, gcwq) + if (!list_empty(&pool->idle_list)) + return true; + return false; +} + static int __cpuinit trustee_thread(void *__gcwq) { struct global_cwq *gcwq = __gcwq; + struct worker_pool *pool; struct worker *worker; struct work_struct *work; struct hlist_node *pos; @@ -3363,13 +3392,15 @@ static int __cpuinit trustee_thread(void *__gcwq) * cancelled. */ BUG_ON(gcwq->cpu != smp_processor_id()); - rc = trustee_wait_event(!(gcwq->pool.flags & POOL_MANAGING_WORKERS)); + rc = trustee_wait_event(!gcwq_is_managing_workers(gcwq)); BUG_ON(rc < 0); - gcwq->pool.flags |= POOL_MANAGING_WORKERS; + for_each_worker_pool(pool, gcwq) { + pool->flags |= POOL_MANAGING_WORKERS; - list_for_each_entry(worker, &gcwq->pool.idle_list, entry) - worker->flags |= WORKER_ROGUE; + list_for_each_entry(worker, &pool->idle_list, entry) + worker->flags |= WORKER_ROGUE; + } for_each_busy_worker(worker, i, pos, gcwq) worker->flags |= WORKER_ROGUE; @@ -3390,10 +3421,12 @@ static int __cpuinit trustee_thread(void *__gcwq) * keep_working() are always true as long as the worklist is * not empty. */ - atomic_set(get_pool_nr_running(&gcwq->pool), 0); + for_each_worker_pool(pool, gcwq) + atomic_set(get_pool_nr_running(pool), 0); spin_unlock_irq(&gcwq->lock); - del_timer_sync(&gcwq->pool.idle_timer); + for_each_worker_pool(pool, gcwq) + del_timer_sync(&pool->idle_timer); spin_lock_irq(&gcwq->lock); /* @@ -3415,29 +3448,38 @@ static int __cpuinit trustee_thread(void *__gcwq) * may be frozen works in freezable cwqs. Don't declare * completion while frozen. */ - while (gcwq->pool.nr_workers != gcwq->pool.nr_idle || - gcwq->flags & GCWQ_FREEZING || - gcwq->trustee_state == TRUSTEE_IN_CHARGE) { - int nr_works = 0; + while (true) { + bool busy = false; - list_for_each_entry(work, &gcwq->pool.worklist, entry) { - send_mayday(work); - nr_works++; - } + for_each_worker_pool(pool, gcwq) + busy |= pool->nr_workers != pool->nr_idle; - list_for_each_entry(worker, &gcwq->pool.idle_list, entry) { - if (!nr_works--) - break; - wake_up_process(worker->task); - } + if (!busy && !(gcwq->flags & GCWQ_FREEZING) && + gcwq->trustee_state != TRUSTEE_IN_CHARGE) + break; - if (need_to_create_worker(&gcwq->pool)) { - spin_unlock_irq(&gcwq->lock); - worker = create_worker(&gcwq->pool, false); - spin_lock_irq(&gcwq->lock); - if (worker) { - worker->flags |= WORKER_ROGUE; - start_worker(worker); + for_each_worker_pool(pool, gcwq) { + int nr_works = 0; + + list_for_each_entry(work, &pool->worklist, entry) { + send_mayday(work); + nr_works++; + } + + list_for_each_entry(worker, &pool->idle_list, entry) { + if (!nr_works--) + break; + wake_up_process(worker->task); + } + + if (need_to_create_worker(pool)) { + spin_unlock_irq(&gcwq->lock); + worker = create_worker(pool, false); + spin_lock_irq(&gcwq->lock); + if (worker) { + worker->flags |= WORKER_ROGUE; + start_worker(worker); + } } } @@ -3452,11 +3494,18 @@ static int __cpuinit trustee_thread(void *__gcwq) * all workers till we're canceled. */ do { - rc = trustee_wait_event(!list_empty(&gcwq->pool.idle_list)); - while (!list_empty(&gcwq->pool.idle_list)) - destroy_worker(list_first_entry(&gcwq->pool.idle_list, - struct worker, entry)); - } while (gcwq->pool.nr_workers && rc >= 0); + rc = trustee_wait_event(gcwq_has_idle_workers(gcwq)); + + i = 0; + for_each_worker_pool(pool, gcwq) { + while (!list_empty(&pool->idle_list)) { + worker = list_first_entry(&pool->idle_list, + struct worker, entry); + destroy_worker(worker); + } + i |= pool->nr_workers; + } + } while (i && rc >= 0); /* * At this point, either draining has completed and no worker @@ -3465,7 +3514,8 @@ static int __cpuinit trustee_thread(void *__gcwq) * Tell the remaining busy ones to rebind once it finishes the * currently scheduled works by scheduling the rebind_work. */ - WARN_ON(!list_empty(&gcwq->pool.idle_list)); + for_each_worker_pool(pool, gcwq) + WARN_ON(!list_empty(&pool->idle_list)); for_each_busy_worker(worker, i, pos, gcwq) { struct work_struct *rebind_work = &worker->rebind_work; @@ -3490,7 +3540,8 @@ static int __cpuinit trustee_thread(void *__gcwq) } /* relinquish manager role */ - gcwq->pool.flags &= ~POOL_MANAGING_WORKERS; + for_each_worker_pool(pool, gcwq) + pool->flags &= ~POOL_MANAGING_WORKERS; /* notify completion */ gcwq->trustee = NULL; @@ -3532,8 +3583,10 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, unsigned int cpu = (unsigned long)hcpu; struct global_cwq *gcwq = get_gcwq(cpu); struct task_struct *new_trustee = NULL; - struct worker *uninitialized_var(new_worker); + struct worker *new_workers[NR_WORKER_POOLS] = { }; + struct worker_pool *pool; unsigned long flags; + int i; action &= ~CPU_TASKS_FROZEN; @@ -3546,12 +3599,12 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, kthread_bind(new_trustee, cpu); /* fall through */ case CPU_UP_PREPARE: - BUG_ON(gcwq->pool.first_idle); - new_worker = create_worker(&gcwq->pool, false); - if (!new_worker) { - if (new_trustee) - kthread_stop(new_trustee); - return NOTIFY_BAD; + i = 0; + for_each_worker_pool(pool, gcwq) { + BUG_ON(pool->first_idle); + new_workers[i] = create_worker(pool, false); + if (!new_workers[i++]) + goto err_destroy; } } @@ -3568,8 +3621,11 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, wait_trustee_state(gcwq, TRUSTEE_IN_CHARGE); /* fall through */ case CPU_UP_PREPARE: - BUG_ON(gcwq->pool.first_idle); - gcwq->pool.first_idle = new_worker; + i = 0; + for_each_worker_pool(pool, gcwq) { + BUG_ON(pool->first_idle); + pool->first_idle = new_workers[i++]; + } break; case CPU_DYING: @@ -3586,8 +3642,10 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, gcwq->trustee_state = TRUSTEE_BUTCHER; /* fall through */ case CPU_UP_CANCELED: - destroy_worker(gcwq->pool.first_idle); - gcwq->pool.first_idle = NULL; + for_each_worker_pool(pool, gcwq) { + destroy_worker(pool->first_idle); + pool->first_idle = NULL; + } break; case CPU_DOWN_FAILED: @@ -3604,18 +3662,32 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, * Put the first_idle in and request a real manager to * take a look. */ - spin_unlock_irq(&gcwq->lock); - kthread_bind(gcwq->pool.first_idle->task, cpu); - spin_lock_irq(&gcwq->lock); - gcwq->pool.flags |= POOL_MANAGE_WORKERS; - start_worker(gcwq->pool.first_idle); - gcwq->pool.first_idle = NULL; + for_each_worker_pool(pool, gcwq) { + spin_unlock_irq(&gcwq->lock); + kthread_bind(pool->first_idle->task, cpu); + spin_lock_irq(&gcwq->lock); + pool->flags |= POOL_MANAGE_WORKERS; + start_worker(pool->first_idle); + pool->first_idle = NULL; + } break; } spin_unlock_irqrestore(&gcwq->lock, flags); return notifier_from_errno(0); + +err_destroy: + if (new_trustee) + kthread_stop(new_trustee); + + spin_lock_irqsave(&gcwq->lock, flags); + for (i = 0; i < NR_WORKER_POOLS; i++) + if (new_workers[i]) + destroy_worker(new_workers[i]); + spin_unlock_irqrestore(&gcwq->lock, flags); + + return NOTIFY_BAD; } #ifdef CONFIG_SMP @@ -3774,6 +3846,7 @@ void thaw_workqueues(void) for_each_gcwq_cpu(cpu) { struct global_cwq *gcwq = get_gcwq(cpu); + struct worker_pool *pool; struct workqueue_struct *wq; spin_lock_irq(&gcwq->lock); @@ -3795,7 +3868,8 @@ void thaw_workqueues(void) cwq_activate_first_delayed(cwq); } - wake_up_worker(&gcwq->pool); + for_each_worker_pool(pool, gcwq) + wake_up_worker(pool); spin_unlock_irq(&gcwq->lock); } @@ -3816,25 +3890,29 @@ static int __init init_workqueues(void) /* initialize gcwqs */ for_each_gcwq_cpu(cpu) { struct global_cwq *gcwq = get_gcwq(cpu); + struct worker_pool *pool; spin_lock_init(&gcwq->lock); - gcwq->pool.gcwq = gcwq; - INIT_LIST_HEAD(&gcwq->pool.worklist); gcwq->cpu = cpu; gcwq->flags |= GCWQ_DISASSOCIATED; - INIT_LIST_HEAD(&gcwq->pool.idle_list); for (i = 0; i < BUSY_WORKER_HASH_SIZE; i++) INIT_HLIST_HEAD(&gcwq->busy_hash[i]); - init_timer_deferrable(&gcwq->pool.idle_timer); - gcwq->pool.idle_timer.function = idle_worker_timeout; - gcwq->pool.idle_timer.data = (unsigned long)&gcwq->pool; + for_each_worker_pool(pool, gcwq) { + pool->gcwq = gcwq; + INIT_LIST_HEAD(&pool->worklist); + INIT_LIST_HEAD(&pool->idle_list); - setup_timer(&gcwq->pool.mayday_timer, gcwq_mayday_timeout, - (unsigned long)&gcwq->pool); + init_timer_deferrable(&pool->idle_timer); + pool->idle_timer.function = idle_worker_timeout; + pool->idle_timer.data = (unsigned long)pool; - ida_init(&gcwq->pool.worker_ida); + setup_timer(&pool->mayday_timer, gcwq_mayday_timeout, + (unsigned long)pool); + + ida_init(&pool->worker_ida); + } gcwq->trustee_state = TRUSTEE_DONE; init_waitqueue_head(&gcwq->trustee_wait); @@ -3843,15 +3921,20 @@ static int __init init_workqueues(void) /* create the initial worker */ for_each_online_gcwq_cpu(cpu) { struct global_cwq *gcwq = get_gcwq(cpu); - struct worker *worker; + struct worker_pool *pool; if (cpu != WORK_CPU_UNBOUND) gcwq->flags &= ~GCWQ_DISASSOCIATED; - worker = create_worker(&gcwq->pool, true); - BUG_ON(!worker); - spin_lock_irq(&gcwq->lock); - start_worker(worker); - spin_unlock_irq(&gcwq->lock); + + for_each_worker_pool(pool, gcwq) { + struct worker *worker; + + worker = create_worker(pool, true); + BUG_ON(!worker); + spin_lock_irq(&gcwq->lock); + start_worker(worker); + spin_unlock_irq(&gcwq->lock); + } } system_wq = alloc_workqueue("events", 0, 0); -- cgit v1.2.3 From 3270476a6c0ce322354df8679652f060d66526dc Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 13 Jul 2012 22:16:45 -0700 Subject: workqueue: reimplement WQ_HIGHPRI using a separate worker_pool WQ_HIGHPRI was implemented by queueing highpri work items at the head of the global worklist. Other than queueing at the head, they weren't handled differently; unfortunately, this could lead to execution latency of a few seconds on heavily loaded systems. Now that workqueue code has been updated to deal with multiple worker_pools per global_cwq, this patch reimplements WQ_HIGHPRI using a separate worker_pool. NR_WORKER_POOLS is bumped to two and gcwq->pools[0] is used for normal pri work items and ->pools[1] for highpri. Highpri workers get -20 nice level and has 'H' suffix in their names. Note that this change increases the number of kworkers per cpu. POOL_HIGHPRI_PENDING, pool_determine_ins_pos() and highpri chain wakeup code in process_one_work() are no longer used and removed. This allows proper prioritization of highpri work items and removes high execution latency of highpri work items. v2: nr_running indexing bug in get_pool_nr_running() fixed. v3: Refreshed for the get_pool_nr_running() update in the previous patch. Signed-off-by: Tejun Heo Reported-by: Josh Hunt LKML-Reference: Cc: Tony Luck Cc: Fengguang Wu --- Documentation/workqueue.txt | 103 ++++++++++++++++---------------------------- kernel/workqueue.c | 100 ++++++++++++------------------------------ 2 files changed, 65 insertions(+), 138 deletions(-) (limited to 'kernel') diff --git a/Documentation/workqueue.txt b/Documentation/workqueue.txt index a0b577de918f..a6ab4b62d926 100644 --- a/Documentation/workqueue.txt +++ b/Documentation/workqueue.txt @@ -89,25 +89,28 @@ called thread-pools. The cmwq design differentiates between the user-facing workqueues that subsystems and drivers queue work items on and the backend mechanism -which manages thread-pool and processes the queued work items. +which manages thread-pools and processes the queued work items. The backend is called gcwq. There is one gcwq for each possible CPU -and one gcwq to serve work items queued on unbound workqueues. +and one gcwq to serve work items queued on unbound workqueues. Each +gcwq has two thread-pools - one for normal work items and the other +for high priority ones. Subsystems and drivers can create and queue work items through special workqueue API functions as they see fit. They can influence some aspects of the way the work items are executed by setting flags on the workqueue they are putting the work item on. These flags include -things like CPU locality, reentrancy, concurrency limits and more. To -get a detailed overview refer to the API description of +things like CPU locality, reentrancy, concurrency limits, priority and +more. To get a detailed overview refer to the API description of alloc_workqueue() below. -When a work item is queued to a workqueue, the target gcwq is -determined according to the queue parameters and workqueue attributes -and appended on the shared worklist of the gcwq. For example, unless -specifically overridden, a work item of a bound workqueue will be -queued on the worklist of exactly that gcwq that is associated to the -CPU the issuer is running on. +When a work item is queued to a workqueue, the target gcwq and +thread-pool is determined according to the queue parameters and +workqueue attributes and appended on the shared worklist of the +thread-pool. For example, unless specifically overridden, a work item +of a bound workqueue will be queued on the worklist of either normal +or highpri thread-pool of the gcwq that is associated to the CPU the +issuer is running on. For any worker pool implementation, managing the concurrency level (how many execution contexts are active) is an important issue. cmwq @@ -115,26 +118,26 @@ tries to keep the concurrency at a minimal but sufficient level. Minimal to save resources and sufficient in that the system is used at its full capacity. -Each gcwq bound to an actual CPU implements concurrency management by -hooking into the scheduler. The gcwq is notified whenever an active -worker wakes up or sleeps and keeps track of the number of the -currently runnable workers. Generally, work items are not expected to -hog a CPU and consume many cycles. That means maintaining just enough -concurrency to prevent work processing from stalling should be -optimal. As long as there are one or more runnable workers on the -CPU, the gcwq doesn't start execution of a new work, but, when the -last running worker goes to sleep, it immediately schedules a new -worker so that the CPU doesn't sit idle while there are pending work -items. This allows using a minimal number of workers without losing -execution bandwidth. +Each thread-pool bound to an actual CPU implements concurrency +management by hooking into the scheduler. The thread-pool is notified +whenever an active worker wakes up or sleeps and keeps track of the +number of the currently runnable workers. Generally, work items are +not expected to hog a CPU and consume many cycles. That means +maintaining just enough concurrency to prevent work processing from +stalling should be optimal. As long as there are one or more runnable +workers on the CPU, the thread-pool doesn't start execution of a new +work, but, when the last running worker goes to sleep, it immediately +schedules a new worker so that the CPU doesn't sit idle while there +are pending work items. This allows using a minimal number of workers +without losing execution bandwidth. Keeping idle workers around doesn't cost other than the memory space for kthreads, so cmwq holds onto idle ones for a while before killing them. For an unbound wq, the above concurrency management doesn't apply and -the gcwq for the pseudo unbound CPU tries to start executing all work -items as soon as possible. The responsibility of regulating +the thread-pools for the pseudo unbound CPU try to start executing all +work items as soon as possible. The responsibility of regulating concurrency level is on the users. There is also a flag to mark a bound wq to ignore the concurrency management. Please refer to the API section for details. @@ -205,31 +208,22 @@ resources, scheduled and executed. WQ_HIGHPRI - Work items of a highpri wq are queued at the head of the - worklist of the target gcwq and start execution regardless of - the current concurrency level. In other words, highpri work - items will always start execution as soon as execution - resource is available. + Work items of a highpri wq are queued to the highpri + thread-pool of the target gcwq. Highpri thread-pools are + served by worker threads with elevated nice level. - Ordering among highpri work items is preserved - a highpri - work item queued after another highpri work item will start - execution after the earlier highpri work item starts. - - Although highpri work items are not held back by other - runnable work items, they still contribute to the concurrency - level. Highpri work items in runnable state will prevent - non-highpri work items from starting execution. - - This flag is meaningless for unbound wq. + Note that normal and highpri thread-pools don't interact with + each other. Each maintain its separate pool of workers and + implements concurrency management among its workers. WQ_CPU_INTENSIVE Work items of a CPU intensive wq do not contribute to the concurrency level. In other words, runnable CPU intensive - work items will not prevent other work items from starting - execution. This is useful for bound work items which are - expected to hog CPU cycles so that their execution is - regulated by the system scheduler. + work items will not prevent other work items in the same + thread-pool from starting execution. This is useful for bound + work items which are expected to hog CPU cycles so that their + execution is regulated by the system scheduler. Although CPU intensive work items don't contribute to the concurrency level, start of their executions is still @@ -239,14 +233,6 @@ resources, scheduled and executed. This flag is meaningless for unbound wq. - WQ_HIGHPRI | WQ_CPU_INTENSIVE - - This combination makes the wq avoid interaction with - concurrency management completely and behave as a simple - per-CPU execution context provider. Work items queued on a - highpri CPU-intensive wq start execution as soon as resources - are available and don't affect execution of other work items. - @max_active: @max_active determines the maximum number of execution contexts per @@ -328,20 +314,7 @@ If @max_active == 2, 35 w2 wakes up and finishes Now, let's assume w1 and w2 are queued to a different wq q1 which has -WQ_HIGHPRI set, - - TIME IN MSECS EVENT - 0 w1 and w2 start and burn CPU - 5 w1 sleeps - 10 w2 sleeps - 10 w0 starts and burns CPU - 15 w0 sleeps - 15 w1 wakes up and finishes - 20 w2 wakes up and finishes - 25 w0 wakes up and burns CPU - 30 w0 finishes - -If q1 has WQ_CPU_INTENSIVE set, +WQ_CPU_INTENSIVE set, TIME IN MSECS EVENT 0 w0 starts and burns CPU diff --git a/kernel/workqueue.c b/kernel/workqueue.c index b0daaea44eaa..4fa9e3552f1e 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -52,7 +52,6 @@ enum { /* pool flags */ POOL_MANAGE_WORKERS = 1 << 0, /* need to manage workers */ POOL_MANAGING_WORKERS = 1 << 1, /* managing workers */ - POOL_HIGHPRI_PENDING = 1 << 2, /* highpri works on queue */ /* worker flags */ WORKER_STARTED = 1 << 0, /* started */ @@ -74,7 +73,7 @@ enum { TRUSTEE_RELEASE = 3, /* release workers */ TRUSTEE_DONE = 4, /* trustee is done */ - NR_WORKER_POOLS = 1, /* # worker pools per gcwq */ + NR_WORKER_POOLS = 2, /* # worker pools per gcwq */ BUSY_WORKER_HASH_ORDER = 6, /* 64 pointers */ BUSY_WORKER_HASH_SIZE = 1 << BUSY_WORKER_HASH_ORDER, @@ -95,6 +94,7 @@ enum { * all cpus. Give -20. */ RESCUER_NICE_LEVEL = -20, + HIGHPRI_NICE_LEVEL = -20, }; /* @@ -174,7 +174,7 @@ struct global_cwq { struct hlist_head busy_hash[BUSY_WORKER_HASH_SIZE]; /* L: hash of busy workers */ - struct worker_pool pool; /* the worker pools */ + struct worker_pool pools[2]; /* normal and highpri pools */ struct task_struct *trustee; /* L: for gcwq shutdown */ unsigned int trustee_state; /* L: trustee state */ @@ -277,7 +277,8 @@ EXPORT_SYMBOL_GPL(system_nrt_freezable_wq); #include #define for_each_worker_pool(pool, gcwq) \ - for ((pool) = &(gcwq)->pool; (pool); (pool) = NULL) + for ((pool) = &(gcwq)->pools[0]; \ + (pool) < &(gcwq)->pools[NR_WORKER_POOLS]; (pool)++) #define for_each_busy_worker(worker, i, pos, gcwq) \ for (i = 0; i < BUSY_WORKER_HASH_SIZE; i++) \ @@ -473,6 +474,11 @@ static atomic_t unbound_pool_nr_running[NR_WORKER_POOLS] = { static int worker_thread(void *__worker); +static int worker_pool_pri(struct worker_pool *pool) +{ + return pool - pool->gcwq->pools; +} + static struct global_cwq *get_gcwq(unsigned int cpu) { if (cpu != WORK_CPU_UNBOUND) @@ -484,7 +490,7 @@ static struct global_cwq *get_gcwq(unsigned int cpu) static atomic_t *get_pool_nr_running(struct worker_pool *pool) { int cpu = pool->gcwq->cpu; - int idx = 0; + int idx = worker_pool_pri(pool); if (cpu != WORK_CPU_UNBOUND) return &per_cpu(pool_nr_running, cpu)[idx]; @@ -586,15 +592,14 @@ static struct global_cwq *get_work_gcwq(struct work_struct *work) } /* - * Policy functions. These define the policies on how the global - * worker pool is managed. Unless noted otherwise, these functions - * assume that they're being called with gcwq->lock held. + * Policy functions. These define the policies on how the global worker + * pools are managed. Unless noted otherwise, these functions assume that + * they're being called with gcwq->lock held. */ static bool __need_more_worker(struct worker_pool *pool) { - return !atomic_read(get_pool_nr_running(pool)) || - (pool->flags & POOL_HIGHPRI_PENDING); + return !atomic_read(get_pool_nr_running(pool)); } /* @@ -621,9 +626,7 @@ static bool keep_working(struct worker_pool *pool) { atomic_t *nr_running = get_pool_nr_running(pool); - return !list_empty(&pool->worklist) && - (atomic_read(nr_running) <= 1 || - (pool->flags & POOL_HIGHPRI_PENDING)); + return !list_empty(&pool->worklist) && atomic_read(nr_running) <= 1; } /* Do we need a new worker? Called from manager. */ @@ -891,43 +894,6 @@ static struct worker *find_worker_executing_work(struct global_cwq *gcwq, work); } -/** - * pool_determine_ins_pos - find insertion position - * @pool: pool of interest - * @cwq: cwq a work is being queued for - * - * A work for @cwq is about to be queued on @pool, determine insertion - * position for the work. If @cwq is for HIGHPRI wq, the work is - * queued at the head of the queue but in FIFO order with respect to - * other HIGHPRI works; otherwise, at the end of the queue. This - * function also sets POOL_HIGHPRI_PENDING flag to hint @pool that - * there are HIGHPRI works pending. - * - * CONTEXT: - * spin_lock_irq(gcwq->lock). - * - * RETURNS: - * Pointer to inserstion position. - */ -static inline struct list_head *pool_determine_ins_pos(struct worker_pool *pool, - struct cpu_workqueue_struct *cwq) -{ - struct work_struct *twork; - - if (likely(!(cwq->wq->flags & WQ_HIGHPRI))) - return &pool->worklist; - - list_for_each_entry(twork, &pool->worklist, entry) { - struct cpu_workqueue_struct *tcwq = get_work_cwq(twork); - - if (!(tcwq->wq->flags & WQ_HIGHPRI)) - break; - } - - pool->flags |= POOL_HIGHPRI_PENDING; - return &twork->entry; -} - /** * insert_work - insert a work into gcwq * @cwq: cwq @work belongs to @@ -1068,7 +1034,7 @@ static void __queue_work(unsigned int cpu, struct workqueue_struct *wq, if (likely(cwq->nr_active < cwq->max_active)) { trace_workqueue_activate_work(work); cwq->nr_active++; - worklist = pool_determine_ins_pos(cwq->pool, cwq); + worklist = &cwq->pool->worklist; } else { work_flags |= WORK_STRUCT_DELAYED; worklist = &cwq->delayed_works; @@ -1385,6 +1351,7 @@ static struct worker *create_worker(struct worker_pool *pool, bool bind) { struct global_cwq *gcwq = pool->gcwq; bool on_unbound_cpu = gcwq->cpu == WORK_CPU_UNBOUND; + const char *pri = worker_pool_pri(pool) ? "H" : ""; struct worker *worker = NULL; int id = -1; @@ -1406,15 +1373,17 @@ static struct worker *create_worker(struct worker_pool *pool, bool bind) if (!on_unbound_cpu) worker->task = kthread_create_on_node(worker_thread, - worker, - cpu_to_node(gcwq->cpu), - "kworker/%u:%d", gcwq->cpu, id); + worker, cpu_to_node(gcwq->cpu), + "kworker/%u:%d%s", gcwq->cpu, id, pri); else worker->task = kthread_create(worker_thread, worker, - "kworker/u:%d", id); + "kworker/u:%d%s", id, pri); if (IS_ERR(worker->task)) goto fail; + if (worker_pool_pri(pool)) + set_user_nice(worker->task, HIGHPRI_NICE_LEVEL); + /* * A rogue worker will become a regular one if CPU comes * online later on. Make sure every worker has @@ -1761,10 +1730,9 @@ static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq) { struct work_struct *work = list_first_entry(&cwq->delayed_works, struct work_struct, entry); - struct list_head *pos = pool_determine_ins_pos(cwq->pool, cwq); trace_workqueue_activate_work(work); - move_linked_works(work, pos, NULL); + move_linked_works(work, &cwq->pool->worklist, NULL); __clear_bit(WORK_STRUCT_DELAYED_BIT, work_data_bits(work)); cwq->nr_active++; } @@ -1879,21 +1847,6 @@ __acquires(&gcwq->lock) set_work_cpu(work, gcwq->cpu); list_del_init(&work->entry); - /* - * If HIGHPRI_PENDING, check the next work, and, if HIGHPRI, - * wake up another worker; otherwise, clear HIGHPRI_PENDING. - */ - if (unlikely(pool->flags & POOL_HIGHPRI_PENDING)) { - struct work_struct *nwork = list_first_entry(&pool->worklist, - struct work_struct, entry); - - if (!list_empty(&pool->worklist) && - get_work_cwq(nwork)->wq->flags & WQ_HIGHPRI) - wake_up_worker(pool); - else - pool->flags &= ~POOL_HIGHPRI_PENDING; - } - /* * CPU intensive works don't participate in concurrency * management. They're the scheduler's responsibility. @@ -3047,9 +3000,10 @@ struct workqueue_struct *__alloc_workqueue_key(const char *fmt, for_each_cwq_cpu(cpu, wq) { struct cpu_workqueue_struct *cwq = get_cwq(cpu, wq); struct global_cwq *gcwq = get_gcwq(cpu); + int pool_idx = (bool)(flags & WQ_HIGHPRI); BUG_ON((unsigned long)cwq & WORK_STRUCT_FLAG_MASK); - cwq->pool = &gcwq->pool; + cwq->pool = &gcwq->pools[pool_idx]; cwq->wq = wq; cwq->flush_color = -1; cwq->max_active = max_active; -- cgit v1.2.3 From 00cd8dd3bf95f2cc8435b4cac01d9995635c6d0b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 10 Jun 2012 17:13:09 -0400 Subject: stop passing nameidata to ->lookup() Just the flags; only NFS cares even about that, but there are legitimate uses for such argument. And getting rid of that completely would require splitting ->lookup() into a couple of methods (at least), so let's leave that alone for now... Signed-off-by: Al Viro --- Documentation/filesystems/Locking | 3 +-- Documentation/filesystems/porting | 4 ++-- Documentation/filesystems/vfs.txt | 2 +- fs/9p/v9fs.h | 2 +- fs/9p/vfs_inode.c | 8 ++++---- fs/9p/vfs_inode_dotl.c | 2 +- fs/adfs/dir.c | 2 +- fs/affs/affs.h | 2 +- fs/affs/namei.c | 2 +- fs/afs/dir.c | 4 ++-- fs/afs/mntpt.c | 4 ++-- fs/autofs4/root.c | 4 ++-- fs/bad_inode.c | 2 +- fs/befs/linuxvfs.c | 4 ++-- fs/bfs/dir.c | 2 +- fs/btrfs/inode.c | 2 +- fs/ceph/dir.c | 6 +++--- fs/cifs/cifsfs.h | 2 +- fs/cifs/dir.c | 4 ++-- fs/coda/dir.c | 4 ++-- fs/configfs/dir.c | 2 +- fs/cramfs/inode.c | 2 +- fs/ecryptfs/inode.c | 2 +- fs/efs/efs.h | 2 +- fs/efs/namei.c | 3 ++- fs/exofs/namei.c | 2 +- fs/ext2/namei.c | 2 +- fs/ext3/namei.c | 2 +- fs/ext4/namei.c | 2 +- fs/fat/namei_msdos.c | 2 +- fs/fat/namei_vfat.c | 2 +- fs/freevxfs/vxfs_lookup.c | 4 ++-- fs/fuse/dir.c | 4 ++-- fs/gfs2/inode.c | 2 +- fs/hfs/dir.c | 2 +- fs/hfs/inode.c | 2 +- fs/hfsplus/dir.c | 2 +- fs/hfsplus/inode.c | 2 +- fs/hostfs/hostfs_kern.c | 2 +- fs/hpfs/dir.c | 2 +- fs/hpfs/hpfs_fn.h | 2 +- fs/hppfs/hppfs.c | 2 +- fs/isofs/isofs.h | 2 +- fs/isofs/namei.c | 2 +- fs/jffs2/dir.c | 4 ++-- fs/jfs/namei.c | 2 +- fs/libfs.c | 2 +- fs/logfs/dir.c | 2 +- fs/minix/namei.c | 2 +- fs/namei.c | 2 +- fs/ncpfs/dir.c | 4 ++-- fs/nfs/dir.c | 8 ++++---- fs/nilfs2/namei.c | 2 +- fs/ntfs/namei.c | 2 +- fs/ocfs2/namei.c | 2 +- fs/omfs/dir.c | 2 +- fs/openpromfs/inode.c | 4 ++-- fs/proc/base.c | 18 ++++++++++-------- fs/proc/generic.c | 2 +- fs/proc/internal.h | 4 ++-- fs/proc/namespaces.c | 2 +- fs/proc/proc_net.c | 2 +- fs/proc/proc_sysctl.c | 2 +- fs/proc/root.c | 7 +++---- fs/qnx4/namei.c | 2 +- fs/qnx4/qnx4.h | 2 +- fs/qnx6/namei.c | 2 +- fs/qnx6/qnx6.h | 2 +- fs/reiserfs/namei.c | 2 +- fs/romfs/super.c | 2 +- fs/squashfs/namei.c | 2 +- fs/sysfs/dir.c | 2 +- fs/sysv/namei.c | 2 +- fs/ubifs/dir.c | 2 +- fs/udf/namei.c | 2 +- fs/ufs/namei.c | 2 +- fs/xfs/xfs_iops.c | 4 ++-- include/linux/fs.h | 4 ++-- kernel/cgroup.c | 4 ++-- 79 files changed, 115 insertions(+), 114 deletions(-) (limited to 'kernel') diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index 52a057367f6f..33f2c8f1db81 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -38,8 +38,7 @@ d_manage: no no yes (ref-walk) maybe --------------------------- inode_operations --------------------------- prototypes: int (*create) (struct inode *,struct dentry *,umode_t, struct nameidata *); - struct dentry * (*lookup) (struct inode *,struct dentry *, struct nameid -ata *); + struct dentry * (*lookup) (struct inode *,struct dentry *, unsigned int); int (*link) (struct dentry *,struct inode *,struct dentry *); int (*unlink) (struct inode *,struct dentry *); int (*symlink) (struct inode *,struct dentry *,const char *); diff --git a/Documentation/filesystems/porting b/Documentation/filesystems/porting index 56750b714d1e..690f573928b9 100644 --- a/Documentation/filesystems/porting +++ b/Documentation/filesystems/porting @@ -434,5 +434,5 @@ d_make_root() drops the reference to inode if dentry allocation fails. -- [mandatory] - The witch is dead! Well, 1/3 of it, anyway. ->d_revalidate() does *not* -take struct nameidata anymore; just the flags. + The witch is dead! Well, 2/3 of it, anyway. ->d_revalidate() and +->lookup() do *not* take struct nameidata anymore; just the flags. diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt index b9a406b2ed0f..ee786354946c 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -342,7 +342,7 @@ filesystem. As of kernel 2.6.22, the following members are defined: struct inode_operations { int (*create) (struct inode *,struct dentry *, umode_t, struct nameidata *); - struct dentry * (*lookup) (struct inode *,struct dentry *, struct nameidata *); + struct dentry * (*lookup) (struct inode *,struct dentry *, unsigned int); int (*link) (struct dentry *,struct inode *,struct dentry *); int (*unlink) (struct inode *,struct dentry *); int (*symlink) (struct inode *,struct dentry *,const char *); diff --git a/fs/9p/v9fs.h b/fs/9p/v9fs.h index e78956cbd702..34c59f14a1c9 100644 --- a/fs/9p/v9fs.h +++ b/fs/9p/v9fs.h @@ -144,7 +144,7 @@ extern void v9fs_session_close(struct v9fs_session_info *v9ses); extern void v9fs_session_cancel(struct v9fs_session_info *v9ses); extern void v9fs_session_begin_cancel(struct v9fs_session_info *v9ses); extern struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nameidata); + unsigned int flags); extern int v9fs_vfs_unlink(struct inode *i, struct dentry *d); extern int v9fs_vfs_rmdir(struct inode *i, struct dentry *d); extern int v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index eae476fb401c..bb0d7627f95b 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -785,7 +785,7 @@ static int v9fs_vfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode */ struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nameidata) + unsigned int flags) { struct dentry *res; struct super_block *sb; @@ -795,8 +795,8 @@ struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, char *name; int result = 0; - p9_debug(P9_DEBUG_VFS, "dir: %p dentry: (%s) %p nameidata: %p\n", - dir, dentry->d_name.name, dentry, nameidata); + p9_debug(P9_DEBUG_VFS, "dir: %p dentry: (%s) %p flags: %x\n", + dir, dentry->d_name.name, dentry, flags); if (dentry->d_name.len > NAME_MAX) return ERR_PTR(-ENAMETOOLONG); @@ -869,7 +869,7 @@ v9fs_vfs_atomic_open(struct inode *dir, struct dentry *dentry, struct dentry *res = NULL; if (d_unhashed(dentry)) { - res = v9fs_vfs_lookup(dir, dentry, NULL); + res = v9fs_vfs_lookup(dir, dentry, 0); if (IS_ERR(res)) return PTR_ERR(res); diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index 1ee10c89df97..b97619fed196 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -259,7 +259,7 @@ v9fs_vfs_atomic_open_dotl(struct inode *dir, struct dentry *dentry, struct dentry *res = NULL; if (d_unhashed(dentry)) { - res = v9fs_vfs_lookup(dir, dentry, NULL); + res = v9fs_vfs_lookup(dir, dentry, 0); if (IS_ERR(res)) return PTR_ERR(res); diff --git a/fs/adfs/dir.c b/fs/adfs/dir.c index 3d83075aaa2e..b3be2e7c5643 100644 --- a/fs/adfs/dir.c +++ b/fs/adfs/dir.c @@ -266,7 +266,7 @@ const struct dentry_operations adfs_dentry_operations = { }; static struct dentry * -adfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +adfs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { struct inode *inode = NULL; struct object_info obj; diff --git a/fs/affs/affs.h b/fs/affs/affs.h index 3a130e27eb15..49e4e3457bfd 100644 --- a/fs/affs/affs.h +++ b/fs/affs/affs.h @@ -153,7 +153,7 @@ extern void affs_free_bitmap(struct super_block *sb); /* namei.c */ extern int affs_hash_name(struct super_block *sb, const u8 *name, unsigned int len); -extern struct dentry *affs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *); +extern struct dentry *affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int); extern int affs_unlink(struct inode *dir, struct dentry *dentry); extern int affs_create(struct inode *dir, struct dentry *dentry, umode_t mode, struct nameidata *); extern int affs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); diff --git a/fs/affs/namei.c b/fs/affs/namei.c index 47806940aac0..7f9721be709f 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -211,7 +211,7 @@ affs_find_entry(struct inode *dir, struct dentry *dentry) } struct dentry * -affs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { struct super_block *sb = dir->i_sb; struct buffer_head *bh; diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 65c54ab04733..ffb33e36ea72 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -20,7 +20,7 @@ #include "internal.h" static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd); + unsigned int flags); static int afs_dir_open(struct inode *inode, struct file *file); static int afs_readdir(struct file *file, void *dirent, filldir_t filldir); static int afs_d_revalidate(struct dentry *dentry, unsigned int flags); @@ -516,7 +516,7 @@ out: * look up an entry in a directory */ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct afs_vnode *vnode; struct afs_fid fid; diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index 298cf8919ec7..9682c33d5daf 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -22,7 +22,7 @@ static struct dentry *afs_mntpt_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd); + unsigned int flags); static int afs_mntpt_open(struct inode *inode, struct file *file); static void afs_mntpt_expiry_timed_out(struct work_struct *work); @@ -104,7 +104,7 @@ out: */ static struct dentry *afs_mntpt_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { _enter("%p,%p{%p{%s},%s}", dir, diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index 75e5f1c8e028..e7396cfdb109 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -32,7 +32,7 @@ static long autofs4_root_ioctl(struct file *,unsigned int,unsigned long); static long autofs4_root_compat_ioctl(struct file *,unsigned int,unsigned long); #endif static int autofs4_dir_open(struct inode *inode, struct file *file); -static struct dentry *autofs4_lookup(struct inode *,struct dentry *, struct nameidata *); +static struct dentry *autofs4_lookup(struct inode *,struct dentry *, unsigned int); static struct vfsmount *autofs4_d_automount(struct path *); static int autofs4_d_manage(struct dentry *, bool); static void autofs4_dentry_release(struct dentry *); @@ -458,7 +458,7 @@ int autofs4_d_manage(struct dentry *dentry, bool rcu_walk) } /* Lookups in the root directory */ -static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { struct autofs_sb_info *sbi; struct autofs_info *ino; diff --git a/fs/bad_inode.c b/fs/bad_inode.c index 1b35d6bd06b0..d27e73c69ba4 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -179,7 +179,7 @@ static int bad_inode_create (struct inode *dir, struct dentry *dentry, } static struct dentry *bad_inode_lookup(struct inode *dir, - struct dentry *dentry, struct nameidata *nd) + struct dentry *dentry, unsigned int flags) { return ERR_PTR(-EIO); } diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index e18da23d42b5..cf7f3c67c8b7 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -34,7 +34,7 @@ static int befs_readdir(struct file *, void *, filldir_t); static int befs_get_block(struct inode *, sector_t, struct buffer_head *, int); static int befs_readpage(struct file *file, struct page *page); static sector_t befs_bmap(struct address_space *mapping, sector_t block); -static struct dentry *befs_lookup(struct inode *, struct dentry *, struct nameidata *); +static struct dentry *befs_lookup(struct inode *, struct dentry *, unsigned int); static struct inode *befs_iget(struct super_block *, unsigned long); static struct inode *befs_alloc_inode(struct super_block *sb); static void befs_destroy_inode(struct inode *inode); @@ -159,7 +159,7 @@ befs_get_block(struct inode *inode, sector_t block, } static struct dentry * -befs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +befs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { struct inode *inode = NULL; struct super_block *sb = dir->i_sb; diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index d12c7966db27..3f1cd3b71681 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -133,7 +133,7 @@ static int bfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, } static struct dentry *bfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct inode *inode = NULL; struct buffer_head *bh; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a101572f1cea..e5f1f81b2d65 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4247,7 +4247,7 @@ static void btrfs_dentry_release(struct dentry *dentry) } static struct dentry *btrfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct dentry *ret; diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 8898eef8bca9..74b2f3c54fe7 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -576,7 +576,7 @@ static int is_root_ceph_dentry(struct inode *inode, struct dentry *dentry) * the MDS so that it gets our 'caps wanted' value in a single op. */ static struct dentry *ceph_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct ceph_fs_client *fsc = ceph_sb_to_client(dir->i_sb); struct ceph_mds_client *mdsc = fsc->mdsc; @@ -653,7 +653,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, } if (d_unhashed(dentry)) { - res = ceph_lookup(dir, dentry, NULL); + res = ceph_lookup(dir, dentry, 0); if (IS_ERR(res)) return PTR_ERR(res); @@ -678,7 +678,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, */ int ceph_handle_notrace_create(struct inode *dir, struct dentry *dentry) { - struct dentry *result = ceph_lookup(dir, dentry, NULL); + struct dentry *result = ceph_lookup(dir, dentry, 0); if (result && !IS_ERR(result)) { /* diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 48bb474ce294..1abd31fd5bf0 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -50,7 +50,7 @@ extern int cifs_atomic_open(struct inode *, struct dentry *, struct file *, unsigned, umode_t, int *); extern struct dentry *cifs_lookup(struct inode *, struct dentry *, - struct nameidata *); + unsigned int); extern int cifs_unlink(struct inode *dir, struct dentry *dentry); extern int cifs_hardlink(struct dentry *, struct inode *, struct dentry *); extern int cifs_mknod(struct inode *, struct dentry *, umode_t, dev_t); diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index b97ff48b7df6..2d732b9276ee 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -401,7 +401,7 @@ cifs_atomic_open(struct inode *inode, struct dentry *direntry, * in network traffic in the other paths. */ if (!(oflags & O_CREAT)) { - struct dentry *res = cifs_lookup(inode, direntry, NULL); + struct dentry *res = cifs_lookup(inode, direntry, 0); if (IS_ERR(res)) return PTR_ERR(res); @@ -621,7 +621,7 @@ mknod_out: struct dentry * cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, - struct nameidata *nd) + unsigned int flags) { int xid; int rc = 0; /* to get around spurious gcc warning, set to zero here */ diff --git a/fs/coda/dir.c b/fs/coda/dir.c index 7f8f1a7c6d87..da35e965861d 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -31,7 +31,7 @@ /* dir inode-ops */ static int coda_create(struct inode *dir, struct dentry *new, umode_t mode, struct nameidata *nd); -static struct dentry *coda_lookup(struct inode *dir, struct dentry *target, struct nameidata *nd); +static struct dentry *coda_lookup(struct inode *dir, struct dentry *target, unsigned int flags); static int coda_link(struct dentry *old_dentry, struct inode *dir_inode, struct dentry *entry); static int coda_unlink(struct inode *dir_inode, struct dentry *entry); @@ -94,7 +94,7 @@ const struct file_operations coda_dir_operations = { /* inode operations for directories */ /* access routines: lookup, readlink, permission */ -static struct dentry *coda_lookup(struct inode *dir, struct dentry *entry, struct nameidata *nd) +static struct dentry *coda_lookup(struct inode *dir, struct dentry *entry, unsigned int flags) { struct super_block *sb = dir->i_sb; const char *name = entry->d_name.name; diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index 7e6c52d8a207..7414ae24a79b 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -442,7 +442,7 @@ static int configfs_attach_attr(struct configfs_dirent * sd, struct dentry * den static struct dentry * configfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct configfs_dirent * parent_sd = dentry->d_parent->d_fsdata; struct configfs_dirent * sd; diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index d013c46402ed..28cca01ca9c9 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -417,7 +417,7 @@ static int cramfs_readdir(struct file *filp, void *dirent, filldir_t filldir) /* * Lookup and fill in the inode data.. */ -static struct dentry * cramfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry * cramfs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { unsigned int offset = 0; struct inode *inode = NULL; diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index a07441a0a878..4ab50c3f5ab2 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -374,7 +374,7 @@ static int ecryptfs_lookup_interpose(struct dentry *dentry, */ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode, struct dentry *ecryptfs_dentry, - struct nameidata *ecryptfs_nd) + unsigned int flags) { char *encrypted_and_encoded_name = NULL; size_t encrypted_and_encoded_name_size; diff --git a/fs/efs/efs.h b/fs/efs/efs.h index d8305b582ab0..5528926ac7f6 100644 --- a/fs/efs/efs.h +++ b/fs/efs/efs.h @@ -129,7 +129,7 @@ extern struct inode *efs_iget(struct super_block *, unsigned long); extern efs_block_t efs_map_block(struct inode *, efs_block_t); extern int efs_get_block(struct inode *, sector_t, struct buffer_head *, int); -extern struct dentry *efs_lookup(struct inode *, struct dentry *, struct nameidata *); +extern struct dentry *efs_lookup(struct inode *, struct dentry *, unsigned int); extern struct dentry *efs_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len, int fh_type); extern struct dentry *efs_fh_to_parent(struct super_block *sb, struct fid *fid, diff --git a/fs/efs/namei.c b/fs/efs/namei.c index 832b10ded82f..96f66d213a19 100644 --- a/fs/efs/namei.c +++ b/fs/efs/namei.c @@ -58,7 +58,8 @@ static efs_ino_t efs_find_entry(struct inode *inode, const char *name, int len) return(0); } -struct dentry *efs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { +struct dentry *efs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) +{ efs_ino_t inodenum; struct inode *inode = NULL; diff --git a/fs/exofs/namei.c b/fs/exofs/namei.c index fc7161d6bf6b..909ed6ea4cf6 100644 --- a/fs/exofs/namei.c +++ b/fs/exofs/namei.c @@ -46,7 +46,7 @@ static inline int exofs_add_nondir(struct dentry *dentry, struct inode *inode) } static struct dentry *exofs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct inode *inode; ino_t ino; diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c index f663a67d7bf0..b3e6778cd1e7 100644 --- a/fs/ext2/namei.c +++ b/fs/ext2/namei.c @@ -55,7 +55,7 @@ static inline int ext2_add_nondir(struct dentry *dentry, struct inode *inode) * Methods themselves. */ -static struct dentry *ext2_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *ext2_lookup(struct inode * dir, struct dentry *dentry, unsigned int flags) { struct inode * inode; ino_t ino; diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index eeb63dfc5d20..86d25f3f6043 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -1011,7 +1011,7 @@ errout: return NULL; } -static struct dentry *ext3_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *ext3_lookup(struct inode * dir, struct dentry *dentry, unsigned int flags) { struct inode * inode; struct ext3_dir_entry_2 * de; diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 5845cd97bf8b..4fba3cd42e2b 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1312,7 +1312,7 @@ errout: return NULL; } -static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { struct inode *inode; struct ext4_dir_entry_2 *de; diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c index c5938c9084b9..47c608b05294 100644 --- a/fs/fat/namei_msdos.c +++ b/fs/fat/namei_msdos.c @@ -201,7 +201,7 @@ static const struct dentry_operations msdos_dentry_operations = { /***** Get inode using directory and name */ static struct dentry *msdos_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct super_block *sb = dir->i_sb; struct fat_slot_info sinfo; diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c index 0bbdf3990060..44152571524e 100644 --- a/fs/fat/namei_vfat.c +++ b/fs/fat/namei_vfat.c @@ -714,7 +714,7 @@ static int vfat_d_anon_disconn(struct dentry *dentry) } static struct dentry *vfat_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct super_block *sb = dir->i_sb; struct fat_slot_info sinfo; diff --git a/fs/freevxfs/vxfs_lookup.c b/fs/freevxfs/vxfs_lookup.c index 3360f1e678ad..bd447e88f208 100644 --- a/fs/freevxfs/vxfs_lookup.c +++ b/fs/freevxfs/vxfs_lookup.c @@ -48,7 +48,7 @@ #define VXFS_BLOCK_PER_PAGE(sbp) ((PAGE_CACHE_SIZE / (sbp)->s_blocksize)) -static struct dentry * vxfs_lookup(struct inode *, struct dentry *, struct nameidata *); +static struct dentry * vxfs_lookup(struct inode *, struct dentry *, unsigned int); static int vxfs_readdir(struct file *, void *, filldir_t); const struct inode_operations vxfs_dir_inode_ops = { @@ -203,7 +203,7 @@ vxfs_inode_by_name(struct inode *dip, struct dentry *dp) * in the return pointer. */ static struct dentry * -vxfs_lookup(struct inode *dip, struct dentry *dp, struct nameidata *nd) +vxfs_lookup(struct inode *dip, struct dentry *dp, unsigned int flags) { struct inode *ip = NULL; ino_t ino; diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index eba30bd9ba2b..385235ac137d 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -316,7 +316,7 @@ int fuse_lookup_name(struct super_block *sb, u64 nodeid, struct qstr *name, } static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, - struct nameidata *nd) + unsigned int flags) { int err; struct fuse_entry_out outarg; @@ -478,7 +478,7 @@ static int fuse_atomic_open(struct inode *dir, struct dentry *entry, struct dentry *res = NULL; if (d_unhashed(entry)) { - res = fuse_lookup(dir, entry, NULL); + res = fuse_lookup(dir, entry, 0); if (IS_ERR(res)) return PTR_ERR(res); diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index a9ba2444e077..19e443b73354 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -775,7 +775,7 @@ static int gfs2_create(struct inode *dir, struct dentry *dentry, */ static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct inode *inode = gfs2_lookupi(dir, &dentry->d_name, 0); if (inode && !IS_ERR(inode)) { diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c index 62fc14ea4b73..617b1ed71f52 100644 --- a/fs/hfs/dir.c +++ b/fs/hfs/dir.c @@ -18,7 +18,7 @@ * hfs_lookup() */ static struct dentry *hfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { hfs_cat_rec rec; struct hfs_find_data fd; diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index 761ec06354b4..451c97281b83 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -489,7 +489,7 @@ out: } static struct dentry *hfs_file_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct inode *inode = NULL; hfs_cat_rec rec; diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 26b53fb09f68..90c2f78b2c79 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -25,7 +25,7 @@ static inline void hfsplus_instantiate(struct dentry *dentry, /* Find the entry inside dir named dentry->d_name */ static struct dentry *hfsplus_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct inode *inode = NULL; struct hfs_find_data fd; diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index 82b69ee4dacc..7009265b746f 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -168,7 +168,7 @@ const struct dentry_operations hfsplus_dentry_operations = { }; static struct dentry *hfsplus_file_lookup(struct inode *dir, - struct dentry *dentry, struct nameidata *nd) + struct dentry *dentry, unsigned int flags) { struct hfs_find_data fd; struct super_block *sb = dir->i_sb; diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 2afa5bbccf9b..0ea005228e1b 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -595,7 +595,7 @@ int hostfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, } struct dentry *hostfs_lookup(struct inode *ino, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct inode *inode; char *name; diff --git a/fs/hpfs/dir.c b/fs/hpfs/dir.c index b8472f803f4e..78e12b2e0ea2 100644 --- a/fs/hpfs/dir.c +++ b/fs/hpfs/dir.c @@ -189,7 +189,7 @@ out: * to tell read_inode to read fnode or not. */ -struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +struct dentry *hpfs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { const unsigned char *name = dentry->d_name.name; unsigned len = dentry->d_name.len; diff --git a/fs/hpfs/hpfs_fn.h b/fs/hpfs/hpfs_fn.h index c07ef1f1ced6..ac1ead194db5 100644 --- a/fs/hpfs/hpfs_fn.h +++ b/fs/hpfs/hpfs_fn.h @@ -220,7 +220,7 @@ extern const struct dentry_operations hpfs_dentry_operations; /* dir.c */ -struct dentry *hpfs_lookup(struct inode *, struct dentry *, struct nameidata *); +struct dentry *hpfs_lookup(struct inode *, struct dentry *, unsigned int); extern const struct file_operations hpfs_dir_ops; /* dnode.c */ diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index d4f93b52cec5..e5c06531dcc4 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -138,7 +138,7 @@ static int file_removed(struct dentry *dentry, const char *file) } static struct dentry *hppfs_lookup(struct inode *ino, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct dentry *proc_dentry, *parent; struct qstr *name = &dentry->d_name; diff --git a/fs/isofs/isofs.h b/fs/isofs/isofs.h index 0e73f63d9274..3620ad1ea9bc 100644 --- a/fs/isofs/isofs.h +++ b/fs/isofs/isofs.h @@ -114,7 +114,7 @@ extern int isofs_name_translate(struct iso_directory_record *, char *, struct in int get_joliet_filename(struct iso_directory_record *, unsigned char *, struct inode *); int get_acorn_filename(struct iso_directory_record *, char *, struct inode *); -extern struct dentry *isofs_lookup(struct inode *, struct dentry *, struct nameidata *); +extern struct dentry *isofs_lookup(struct inode *, struct dentry *, unsigned int flags); extern struct buffer_head *isofs_bread(struct inode *, sector_t); extern int isofs_get_blocks(struct inode *, sector_t, struct buffer_head **, unsigned long); diff --git a/fs/isofs/namei.c b/fs/isofs/namei.c index 1e2946f2a69e..c167028844ed 100644 --- a/fs/isofs/namei.c +++ b/fs/isofs/namei.c @@ -163,7 +163,7 @@ isofs_find_entry(struct inode *dir, struct dentry *dentry, return 0; } -struct dentry *isofs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +struct dentry *isofs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { int found; unsigned long uninitialized_var(block); diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index b56018896d5e..6a601673f89f 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -27,7 +27,7 @@ static int jffs2_readdir (struct file *, void *, filldir_t); static int jffs2_create (struct inode *,struct dentry *,umode_t, struct nameidata *); static struct dentry *jffs2_lookup (struct inode *,struct dentry *, - struct nameidata *); + unsigned int); static int jffs2_link (struct dentry *,struct inode *,struct dentry *); static int jffs2_unlink (struct inode *,struct dentry *); static int jffs2_symlink (struct inode *,struct dentry *,const char *); @@ -74,7 +74,7 @@ const struct inode_operations jffs2_dir_inode_operations = nice and simple */ static struct dentry *jffs2_lookup(struct inode *dir_i, struct dentry *target, - struct nameidata *nd) + unsigned int flags) { struct jffs2_inode_info *dir_f; struct jffs2_full_dirent *fd = NULL, *fd_list; diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index f37977fb0871..34fe85555caf 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -1436,7 +1436,7 @@ static int jfs_mknod(struct inode *dir, struct dentry *dentry, return rc; } -static struct dentry *jfs_lookup(struct inode *dip, struct dentry *dentry, struct nameidata *nd) +static struct dentry *jfs_lookup(struct inode *dip, struct dentry *dentry, unsigned int flags) { struct btstack btstack; ino_t inum; diff --git a/fs/libfs.c b/fs/libfs.c index f86ec27a4230..ebd03f6910d5 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -53,7 +53,7 @@ static int simple_delete_dentry(const struct dentry *dentry) * Lookup the data. This is trivial - if the dentry didn't already * exist, we know it is negative. Set d_op to delete negative dentries. */ -struct dentry *simple_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +struct dentry *simple_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { static const struct dentry_operations simple_dentry_operations = { .d_delete = simple_delete_dentry, diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c index bea5d1b9954b..8a3dcc615b39 100644 --- a/fs/logfs/dir.c +++ b/fs/logfs/dir.c @@ -349,7 +349,7 @@ static void logfs_set_name(struct logfs_disk_dentry *dd, struct qstr *name) } static struct dentry *logfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct page *page; struct logfs_disk_dentry *dd; diff --git a/fs/minix/namei.c b/fs/minix/namei.c index 2d0ee1786305..1f245240ea08 100644 --- a/fs/minix/namei.c +++ b/fs/minix/namei.c @@ -18,7 +18,7 @@ static int add_nondir(struct dentry *dentry, struct inode *inode) return err; } -static struct dentry *minix_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *minix_lookup(struct inode * dir, struct dentry *dentry, unsigned int flags) { struct inode * inode = NULL; ino_t ino; diff --git a/fs/namei.c b/fs/namei.c index 2e943ab04f32..175e81b8f261 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1090,7 +1090,7 @@ static struct dentry *lookup_real(struct inode *dir, struct dentry *dentry, return ERR_PTR(-ENOENT); } - old = dir->i_op->lookup(dir, dentry, nd); + old = dir->i_op->lookup(dir, dentry, nd ? nd->flags : 0); if (unlikely(old)) { dput(dentry); dentry = old; diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c index 32607f749588..a0cff22bfc9b 100644 --- a/fs/ncpfs/dir.c +++ b/fs/ncpfs/dir.c @@ -31,7 +31,7 @@ static void ncp_do_readdir(struct file *, void *, filldir_t, static int ncp_readdir(struct file *, void *, filldir_t); static int ncp_create(struct inode *, struct dentry *, umode_t, struct nameidata *); -static struct dentry *ncp_lookup(struct inode *, struct dentry *, struct nameidata *); +static struct dentry *ncp_lookup(struct inode *, struct dentry *, unsigned int); static int ncp_unlink(struct inode *, struct dentry *); static int ncp_mkdir(struct inode *, struct dentry *, umode_t); static int ncp_rmdir(struct inode *, struct dentry *); @@ -836,7 +836,7 @@ out: return result; } -static struct dentry *ncp_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *ncp_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { struct ncp_server *server = NCP_SERVER(dir); struct inode *inode = NULL; diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 656f52e9aa2e..8f21205c5896 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -46,7 +46,7 @@ static int nfs_opendir(struct inode *, struct file *); static int nfs_closedir(struct inode *, struct file *); static int nfs_readdir(struct file *, void *, filldir_t); -static struct dentry *nfs_lookup(struct inode *, struct dentry *, struct nameidata *); +static struct dentry *nfs_lookup(struct inode *, struct dentry *, unsigned int); static int nfs_create(struct inode *, struct dentry *, umode_t, struct nameidata *); static int nfs_mkdir(struct inode *, struct dentry *, umode_t); static int nfs_rmdir(struct inode *, struct dentry *); @@ -1270,7 +1270,7 @@ const struct dentry_operations nfs_dentry_operations = { .d_release = nfs_d_release, }; -static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd) +static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags) { struct dentry *res; struct dentry *parent; @@ -1291,7 +1291,7 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru * If we're doing an exclusive create, optimize away the lookup * but don't hash the dentry. */ - if (nd && nfs_is_exclusive_create(dir, nd->flags)) { + if (nfs_is_exclusive_create(dir, flags)) { d_instantiate(dentry, NULL); res = NULL; goto out; @@ -1482,7 +1482,7 @@ out: return err; no_open: - res = nfs_lookup(dir, dentry, NULL); + res = nfs_lookup(dir, dentry, 0); err = PTR_ERR(res); if (IS_ERR(res)) goto out; diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c index b72847988b78..5e5f779db76f 100644 --- a/fs/nilfs2/namei.c +++ b/fs/nilfs2/namei.c @@ -63,7 +63,7 @@ static inline int nilfs_add_nondir(struct dentry *dentry, struct inode *inode) */ static struct dentry * -nilfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +nilfs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { struct inode *inode; ino_t ino; diff --git a/fs/ntfs/namei.c b/fs/ntfs/namei.c index 358273e59ade..436f36037e09 100644 --- a/fs/ntfs/namei.c +++ b/fs/ntfs/namei.c @@ -101,7 +101,7 @@ * Locking: Caller must hold i_mutex on the directory. */ static struct dentry *ntfs_lookup(struct inode *dir_ino, struct dentry *dent, - struct nameidata *nd) + unsigned int flags) { ntfs_volume *vol = NTFS_SB(dir_ino->i_sb); struct inode *dent_inode; diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 9f39c640cddf..fd71f6e5841f 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -98,7 +98,7 @@ static int ocfs2_create_symlink_data(struct ocfs2_super *osb, #define OCFS2_ORPHAN_NAMELEN ((int)(2 * sizeof(u64))) static struct dentry *ocfs2_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { int status; u64 blkno; diff --git a/fs/omfs/dir.c b/fs/omfs/dir.c index f00576ec320f..3d254872e641 100644 --- a/fs/omfs/dir.c +++ b/fs/omfs/dir.c @@ -291,7 +291,7 @@ static int omfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, } static struct dentry *omfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct buffer_head *bh; struct inode *inode = NULL; diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index bc49c975d501..4a3477949bca 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -170,13 +170,13 @@ static const struct file_operations openprom_operations = { .llseek = generic_file_llseek, }; -static struct dentry *openpromfs_lookup(struct inode *, struct dentry *, struct nameidata *); +static struct dentry *openpromfs_lookup(struct inode *, struct dentry *, unsigned int); static const struct inode_operations openprom_inode_operations = { .lookup = openpromfs_lookup, }; -static struct dentry *openpromfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *openpromfs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { struct op_inode_info *ent_oi, *oi = OP_I(dir); struct device_node *dp, *child; diff --git a/fs/proc/base.c b/fs/proc/base.c index bf749cca4cc6..8eaa5ea1c613 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1956,7 +1956,7 @@ out_no_task: } static struct dentry *proc_lookupfd(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { return proc_lookupfd_common(dir, dentry, proc_fd_instantiate); } @@ -2145,7 +2145,7 @@ proc_map_files_instantiate(struct inode *dir, struct dentry *dentry, } static struct dentry *proc_map_files_lookup(struct inode *dir, - struct dentry *dentry, struct nameidata *nd) + struct dentry *dentry, unsigned int flags) { unsigned long vm_start, vm_end; struct vm_area_struct *vma; @@ -2380,7 +2380,7 @@ static struct dentry *proc_fdinfo_instantiate(struct inode *dir, static struct dentry *proc_lookupfdinfo(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { return proc_lookupfd_common(dir, dentry, proc_fdinfo_instantiate); } @@ -2630,7 +2630,7 @@ static const struct file_operations proc_attr_dir_operations = { }; static struct dentry *proc_attr_dir_lookup(struct inode *dir, - struct dentry *dentry, struct nameidata *nd) + struct dentry *dentry, unsigned int flags) { return proc_pident_lookup(dir, dentry, attr_dir_stuff, ARRAY_SIZE(attr_dir_stuff)); @@ -3114,7 +3114,8 @@ static const struct file_operations proc_tgid_base_operations = { .llseek = default_llseek, }; -static struct dentry *proc_tgid_base_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd){ +static struct dentry *proc_tgid_base_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) +{ return proc_pident_lookup(dir, dentry, tgid_base_stuff, ARRAY_SIZE(tgid_base_stuff)); } @@ -3243,7 +3244,7 @@ out: return error; } -struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd) +struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags) { struct dentry *result; struct task_struct *task; @@ -3470,7 +3471,8 @@ static int proc_tid_base_readdir(struct file * filp, tid_base_stuff,ARRAY_SIZE(tid_base_stuff)); } -static struct dentry *proc_tid_base_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd){ +static struct dentry *proc_tid_base_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) +{ return proc_pident_lookup(dir, dentry, tid_base_stuff, ARRAY_SIZE(tid_base_stuff)); } @@ -3514,7 +3516,7 @@ out: return error; } -static struct dentry *proc_task_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *nd) +static struct dentry *proc_task_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags) { struct dentry *result = ERR_PTR(-ENOENT); struct task_struct *task; diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 2edf34f2eb61..b3647fe6a608 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -446,7 +446,7 @@ out_unlock: } struct dentry *proc_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { return proc_lookup_de(PDE(dir), dir, dentry); } diff --git a/fs/proc/internal.h b/fs/proc/internal.h index e0c2a48dab73..e1167a1c9126 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -106,7 +106,7 @@ void pde_users_dec(struct proc_dir_entry *pde); extern spinlock_t proc_subdir_lock; -struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, struct nameidata *); +struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, unsigned int); int proc_pid_readdir(struct file * filp, void * dirent, filldir_t filldir); unsigned long task_vsize(struct mm_struct *); unsigned long task_statm(struct mm_struct *, @@ -132,7 +132,7 @@ int proc_remount(struct super_block *sb, int *flags, char *data); * of the /proc/ subdirectories. */ int proc_readdir(struct file *, void *, filldir_t); -struct dentry *proc_lookup(struct inode *, struct dentry *, struct nameidata *); +struct dentry *proc_lookup(struct inode *, struct dentry *, unsigned int); diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c index 40ceb40f9853..b178ed733c36 100644 --- a/fs/proc/namespaces.c +++ b/fs/proc/namespaces.c @@ -140,7 +140,7 @@ const struct file_operations proc_ns_dir_operations = { }; static struct dentry *proc_ns_dir_lookup(struct inode *dir, - struct dentry *dentry, struct nameidata *nd) + struct dentry *dentry, unsigned int flags) { struct dentry *error; struct task_struct *task = get_proc_task(dir); diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 06e1cc17caf6..fe72cd073dea 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -119,7 +119,7 @@ static struct net *get_proc_task_net(struct inode *dir) } static struct dentry *proc_tgid_net_lookup(struct inode *dir, - struct dentry *dentry, struct nameidata *nd) + struct dentry *dentry, unsigned int flags) { struct dentry *de; struct net *net; diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index fda69fa39099..dfafeb2b05a0 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -433,7 +433,7 @@ static struct ctl_table_header *grab_header(struct inode *inode) } static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct ctl_table_header *head = grab_header(dir); struct ctl_table_header *h = NULL; diff --git a/fs/proc/root.c b/fs/proc/root.c index 7c30fce037c0..568b20290c75 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -200,13 +200,12 @@ static int proc_root_getattr(struct vfsmount *mnt, struct dentry *dentry, struct return 0; } -static struct dentry *proc_root_lookup(struct inode * dir, struct dentry * dentry, struct nameidata *nd) +static struct dentry *proc_root_lookup(struct inode * dir, struct dentry * dentry, unsigned int flags) { - if (!proc_lookup(dir, dentry, nd)) { + if (!proc_lookup(dir, dentry, flags)) return NULL; - } - return proc_pid_lookup(dir, dentry, nd); + return proc_pid_lookup(dir, dentry, flags); } static int proc_root_readdir(struct file * filp, diff --git a/fs/qnx4/namei.c b/fs/qnx4/namei.c index a512c0b30e8e..d024505ba007 100644 --- a/fs/qnx4/namei.c +++ b/fs/qnx4/namei.c @@ -95,7 +95,7 @@ static struct buffer_head *qnx4_find_entry(int len, struct inode *dir, return NULL; } -struct dentry * qnx4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +struct dentry * qnx4_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { int ino; struct qnx4_inode_entry *de; diff --git a/fs/qnx4/qnx4.h b/fs/qnx4/qnx4.h index 244d4620189b..34e2d329c97e 100644 --- a/fs/qnx4/qnx4.h +++ b/fs/qnx4/qnx4.h @@ -23,7 +23,7 @@ struct qnx4_inode_info { }; extern struct inode *qnx4_iget(struct super_block *, unsigned long); -extern struct dentry *qnx4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd); +extern struct dentry *qnx4_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags); extern unsigned long qnx4_count_free_blocks(struct super_block *sb); extern unsigned long qnx4_block_map(struct inode *inode, long iblock); diff --git a/fs/qnx6/namei.c b/fs/qnx6/namei.c index 8a97289e04ad..0561326a94f5 100644 --- a/fs/qnx6/namei.c +++ b/fs/qnx6/namei.c @@ -13,7 +13,7 @@ #include "qnx6.h" struct dentry *qnx6_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { unsigned ino; struct page *page; diff --git a/fs/qnx6/qnx6.h b/fs/qnx6/qnx6.h index 6c5e02a0b6a8..b00fcc960d37 100644 --- a/fs/qnx6/qnx6.h +++ b/fs/qnx6/qnx6.h @@ -45,7 +45,7 @@ struct qnx6_inode_info { extern struct inode *qnx6_iget(struct super_block *sb, unsigned ino); extern struct dentry *qnx6_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd); + unsigned int flags); #ifdef CONFIG_QNX6FS_DEBUG extern void qnx6_superblock_debug(struct qnx6_super_block *, diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index 84e8a69cee9d..1d9cf248c471 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -322,7 +322,7 @@ static int reiserfs_find_entry(struct inode *dir, const char *name, int namelen, } static struct dentry *reiserfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { int retval; int lock_depth; diff --git a/fs/romfs/super.c b/fs/romfs/super.c index e64f6b5f7ae5..77c5f2173983 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -210,7 +210,7 @@ out: * look up an entry in a directory */ static struct dentry *romfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { unsigned long offset, maxoff; struct inode *inode; diff --git a/fs/squashfs/namei.c b/fs/squashfs/namei.c index abcc58f3c152..7834a517f7f4 100644 --- a/fs/squashfs/namei.c +++ b/fs/squashfs/namei.c @@ -134,7 +134,7 @@ out: static struct dentry *squashfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { const unsigned char *name = dentry->d_name.name; int len = dentry->d_name.len; diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 038e74b3af87..efd373e3e0aa 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -764,7 +764,7 @@ int sysfs_create_dir(struct kobject * kobj) } static struct dentry * sysfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct dentry *ret = NULL; struct dentry *parent = dentry->d_parent; diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c index d7466e293614..a8c4359cd0e1 100644 --- a/fs/sysv/namei.c +++ b/fs/sysv/namei.c @@ -43,7 +43,7 @@ const struct dentry_operations sysv_dentry_operations = { .d_hash = sysv_hash, }; -static struct dentry *sysv_lookup(struct inode * dir, struct dentry * dentry, struct nameidata *nd) +static struct dentry *sysv_lookup(struct inode * dir, struct dentry * dentry, unsigned int flags) { struct inode * inode = NULL; ino_t ino; diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index a6d42efc76d2..845b2df08317 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -184,7 +184,7 @@ static int dbg_check_name(const struct ubifs_info *c, } static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { int err; union ubifs_key key; diff --git a/fs/udf/namei.c b/fs/udf/namei.c index 18024178ac4c..929cc205985a 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -251,7 +251,7 @@ out_ok: } static struct dentry *udf_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct inode *inode = NULL; struct fileIdentDesc cfi; diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c index a2281cadefa1..bc77fa170b9d 100644 --- a/fs/ufs/namei.c +++ b/fs/ufs/namei.c @@ -46,7 +46,7 @@ static inline int ufs_add_nondir(struct dentry *dentry, struct inode *inode) return err; } -static struct dentry *ufs_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *ufs_lookup(struct inode * dir, struct dentry *dentry, unsigned int flags) { struct inode * inode = NULL; ino_t ino; diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 1a25fd802798..b41cfba14faf 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -197,7 +197,7 @@ STATIC struct dentry * xfs_vn_lookup( struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct xfs_inode *cip; struct xfs_name name; @@ -222,7 +222,7 @@ STATIC struct dentry * xfs_vn_ci_lookup( struct inode *dir, struct dentry *dentry, - struct nameidata *nd) + unsigned int flags) { struct xfs_inode *ip; struct xfs_name xname; diff --git a/include/linux/fs.h b/include/linux/fs.h index 17ee20dba86c..7a71709b7fa7 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1666,7 +1666,7 @@ struct file_operations { }; struct inode_operations { - struct dentry * (*lookup) (struct inode *,struct dentry *, struct nameidata *); + struct dentry * (*lookup) (struct inode *,struct dentry *, unsigned int); void * (*follow_link) (struct dentry *, struct nameidata *); int (*permission) (struct inode *, int); struct posix_acl * (*get_acl)(struct inode *, int); @@ -2571,7 +2571,7 @@ extern int simple_write_end(struct file *file, struct address_space *mapping, loff_t pos, unsigned len, unsigned copied, struct page *page, void *fsdata); -extern struct dentry *simple_lookup(struct inode *, struct dentry *, struct nameidata *); +extern struct dentry *simple_lookup(struct inode *, struct dentry *, unsigned int flags); extern ssize_t generic_read_dir(struct file *, char __user *, size_t, loff_t *); extern const struct file_operations simple_dir_operations; extern const struct inode_operations simple_dir_inode_operations; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index b303dfc7dce0..0cd1314acdaf 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -822,7 +822,7 @@ EXPORT_SYMBOL_GPL(cgroup_unlock); */ static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); -static struct dentry *cgroup_lookup(struct inode *, struct dentry *, struct nameidata *); +static struct dentry *cgroup_lookup(struct inode *, struct dentry *, unsigned int); static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry); static int cgroup_populate_dir(struct cgroup *cgrp); static const struct inode_operations cgroup_dir_inode_operations; @@ -2570,7 +2570,7 @@ static const struct inode_operations cgroup_dir_inode_operations = { .rename = cgroup_rename, }; -static struct dentry *cgroup_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) +static struct dentry *cgroup_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { if (dentry->d_name.len > NAME_MAX) return ERR_PTR(-ENAMETOOLONG); -- cgit v1.2.3 From 79714f72d3b964611997de512cb29198c9f2dbbb Mon Sep 17 00:00:00 2001 From: Al Viro Date: Fri, 15 Jun 2012 03:01:42 +0400 Subject: get rid of kern_path_parent() all callers want the same thing, actually - a kinda-sorta analog of kern_path_create(). I.e. they want parent vfsmount/dentry (with ->i_mutex held, to make sure the child dentry is still their child) + the child dentry. Signed-off-by Al Viro --- drivers/base/devtmpfs.c | 98 +++++++++++++++++++++---------------------------- fs/namei.c | 22 ++++++++++- include/linux/namei.h | 2 +- kernel/audit_watch.c | 25 ++----------- 4 files changed, 65 insertions(+), 82 deletions(-) (limited to 'kernel') diff --git a/drivers/base/devtmpfs.c b/drivers/base/devtmpfs.c index 765c3a28077a..d91a3a0b2325 100644 --- a/drivers/base/devtmpfs.c +++ b/drivers/base/devtmpfs.c @@ -227,33 +227,24 @@ static int handle_create(const char *nodename, umode_t mode, struct device *dev) static int dev_rmdir(const char *name) { - struct nameidata nd; + struct path parent; struct dentry *dentry; int err; - err = kern_path_parent(name, &nd); - if (err) - return err; - - mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); - dentry = lookup_one_len(nd.last.name, nd.path.dentry, nd.last.len); - if (!IS_ERR(dentry)) { - if (dentry->d_inode) { - if (dentry->d_inode->i_private == &thread) - err = vfs_rmdir(nd.path.dentry->d_inode, - dentry); - else - err = -EPERM; - } else { - err = -ENOENT; - } - dput(dentry); + dentry = kern_path_locked(name, &parent); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); + if (dentry->d_inode) { + if (dentry->d_inode->i_private == &thread) + err = vfs_rmdir(parent.dentry->d_inode, dentry); + else + err = -EPERM; } else { - err = PTR_ERR(dentry); + err = -ENOENT; } - - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); - path_put(&nd.path); + dput(dentry); + mutex_unlock(&parent.dentry->d_inode->i_mutex); + path_put(&parent); return err; } @@ -305,50 +296,43 @@ static int dev_mynode(struct device *dev, struct inode *inode, struct kstat *sta static int handle_remove(const char *nodename, struct device *dev) { - struct nameidata nd; + struct path parent; struct dentry *dentry; - struct kstat stat; int deleted = 1; int err; - err = kern_path_parent(nodename, &nd); - if (err) - return err; + dentry = kern_path_locked(nodename, &parent); + if (IS_ERR(dentry)) + return PTR_ERR(dentry); - mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); - dentry = lookup_one_len(nd.last.name, nd.path.dentry, nd.last.len); - if (!IS_ERR(dentry)) { - if (dentry->d_inode) { - err = vfs_getattr(nd.path.mnt, dentry, &stat); - if (!err && dev_mynode(dev, dentry->d_inode, &stat)) { - struct iattr newattrs; - /* - * before unlinking this node, reset permissions - * of possible references like hardlinks - */ - newattrs.ia_uid = 0; - newattrs.ia_gid = 0; - newattrs.ia_mode = stat.mode & ~0777; - newattrs.ia_valid = - ATTR_UID|ATTR_GID|ATTR_MODE; - mutex_lock(&dentry->d_inode->i_mutex); - notify_change(dentry, &newattrs); - mutex_unlock(&dentry->d_inode->i_mutex); - err = vfs_unlink(nd.path.dentry->d_inode, - dentry); - if (!err || err == -ENOENT) - deleted = 1; - } - } else { - err = -ENOENT; + if (dentry->d_inode) { + struct kstat stat; + err = vfs_getattr(parent.mnt, dentry, &stat); + if (!err && dev_mynode(dev, dentry->d_inode, &stat)) { + struct iattr newattrs; + /* + * before unlinking this node, reset permissions + * of possible references like hardlinks + */ + newattrs.ia_uid = 0; + newattrs.ia_gid = 0; + newattrs.ia_mode = stat.mode & ~0777; + newattrs.ia_valid = + ATTR_UID|ATTR_GID|ATTR_MODE; + mutex_lock(&dentry->d_inode->i_mutex); + notify_change(dentry, &newattrs); + mutex_unlock(&dentry->d_inode->i_mutex); + err = vfs_unlink(parent.dentry->d_inode, dentry); + if (!err || err == -ENOENT) + deleted = 1; } - dput(dentry); } else { - err = PTR_ERR(dentry); + err = -ENOENT; } - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + dput(dentry); + mutex_unlock(&parent.dentry->d_inode->i_mutex); - path_put(&nd.path); + path_put(&parent); if (deleted && strchr(nodename, '/')) delete_path(nodename); return err; diff --git a/fs/namei.c b/fs/namei.c index 5abab9176903..6b29a51bef5d 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1814,9 +1814,27 @@ static int do_path_lookup(int dfd, const char *name, return retval; } -int kern_path_parent(const char *name, struct nameidata *nd) +/* does lookup, returns the object with parent locked */ +struct dentry *kern_path_locked(const char *name, struct path *path) { - return do_path_lookup(AT_FDCWD, name, LOOKUP_PARENT, nd); + struct nameidata nd; + struct dentry *d; + int err = do_path_lookup(AT_FDCWD, name, LOOKUP_PARENT, &nd); + if (err) + return ERR_PTR(err); + if (nd.last_type != LAST_NORM) { + path_put(&nd.path); + return ERR_PTR(-EINVAL); + } + mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); + d = lookup_one_len(nd.last.name, nd.path.dentry, nd.last.len); + if (IS_ERR(d)) { + mutex_unlock(&nd.path.dentry->d_inode->i_mutex); + path_put(&nd.path); + return d; + } + *path = nd.path; + return d; } int kern_path(const char *name, unsigned int flags, struct path *path) diff --git a/include/linux/namei.h b/include/linux/namei.h index 23d859879210..f5931489e150 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -67,7 +67,7 @@ extern int kern_path(const char *, unsigned, struct path *); extern struct dentry *kern_path_create(int, const char *, struct path *, int); extern struct dentry *user_path_create(int, const char __user *, struct path *, int); -extern int kern_path_parent(const char *, struct nameidata *); +extern struct dentry *kern_path_locked(const char *, struct path *); extern int vfs_path_lookup(struct dentry *, struct vfsmount *, const char *, unsigned int, struct path *); diff --git a/kernel/audit_watch.c b/kernel/audit_watch.c index e683869365d9..3823281401b5 100644 --- a/kernel/audit_watch.c +++ b/kernel/audit_watch.c @@ -355,34 +355,15 @@ static void audit_remove_parent_watches(struct audit_parent *parent) /* Get path information necessary for adding watches. */ static int audit_get_nd(struct audit_watch *watch, struct path *parent) { - struct nameidata nd; - struct dentry *d; - int err; - - err = kern_path_parent(watch->path, &nd); - if (err) - return err; - - if (nd.last_type != LAST_NORM) { - path_put(&nd.path); - return -EINVAL; - } - - mutex_lock_nested(&nd.path.dentry->d_inode->i_mutex, I_MUTEX_PARENT); - d = lookup_one_len(nd.last.name, nd.path.dentry, nd.last.len); - if (IS_ERR(d)) { - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); - path_put(&nd.path); + struct dentry *d = kern_path_locked(watch->path, parent); + if (IS_ERR(d)) return PTR_ERR(d); - } + mutex_unlock(&parent->dentry->d_inode->i_mutex); if (d->d_inode) { /* update watch filter fields */ watch->dev = d->d_inode->i_sb->s_dev; watch->ino = d->d_inode->i_ino; } - mutex_unlock(&nd.path.dentry->d_inode->i_mutex); - - *parent = nd.path; dput(d); return 0; } -- cgit v1.2.3 From be34d1a3bc4b6f357a49acb55ae870c81337e4f0 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 25 Jun 2012 12:55:18 +0100 Subject: VFS: Make clone_mnt()/copy_tree()/collect_mounts() return errors copy_tree() can theoretically fail in a case other than ENOMEM, but always returns NULL which is interpreted by callers as -ENOMEM. Change it to return an explicit error. Also change clone_mnt() for consistency and because union mounts will add new error cases. Thanks to Andreas Gruenbacher for a bug fix. [AV: folded braino fix by Dan Carpenter] Original-author: Valerie Aurora Signed-off-by: David Howells Cc: Valerie Aurora Cc: Andreas Gruenbacher Signed-off-by: Al Viro --- fs/namespace.c | 120 ++++++++++++++++++++++++++++------------------------ fs/pnode.c | 5 ++- kernel/audit_tree.c | 10 ++--- 3 files changed, 73 insertions(+), 62 deletions(-) (limited to 'kernel') diff --git a/fs/namespace.c b/fs/namespace.c index 8f412abcb67f..be1b07a774f1 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -708,56 +708,60 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root, int flag) { struct super_block *sb = old->mnt.mnt_sb; - struct mount *mnt = alloc_vfsmnt(old->mnt_devname); + struct mount *mnt; + int err; - if (mnt) { - if (flag & (CL_SLAVE | CL_PRIVATE)) - mnt->mnt_group_id = 0; /* not a peer of original */ - else - mnt->mnt_group_id = old->mnt_group_id; - - if ((flag & CL_MAKE_SHARED) && !mnt->mnt_group_id) { - int err = mnt_alloc_group_id(mnt); - if (err) - goto out_free; - } + mnt = alloc_vfsmnt(old->mnt_devname); + if (!mnt) + return ERR_PTR(-ENOMEM); - mnt->mnt.mnt_flags = old->mnt.mnt_flags & ~MNT_WRITE_HOLD; - atomic_inc(&sb->s_active); - mnt->mnt.mnt_sb = sb; - mnt->mnt.mnt_root = dget(root); - mnt->mnt_mountpoint = mnt->mnt.mnt_root; - mnt->mnt_parent = mnt; - br_write_lock(&vfsmount_lock); - list_add_tail(&mnt->mnt_instance, &sb->s_mounts); - br_write_unlock(&vfsmount_lock); + if (flag & (CL_SLAVE | CL_PRIVATE)) + mnt->mnt_group_id = 0; /* not a peer of original */ + else + mnt->mnt_group_id = old->mnt_group_id; - if (flag & CL_SLAVE) { - list_add(&mnt->mnt_slave, &old->mnt_slave_list); - mnt->mnt_master = old; - CLEAR_MNT_SHARED(mnt); - } else if (!(flag & CL_PRIVATE)) { - if ((flag & CL_MAKE_SHARED) || IS_MNT_SHARED(old)) - list_add(&mnt->mnt_share, &old->mnt_share); - if (IS_MNT_SLAVE(old)) - list_add(&mnt->mnt_slave, &old->mnt_slave); - mnt->mnt_master = old->mnt_master; - } - if (flag & CL_MAKE_SHARED) - set_mnt_shared(mnt); - - /* stick the duplicate mount on the same expiry list - * as the original if that was on one */ - if (flag & CL_EXPIRE) { - if (!list_empty(&old->mnt_expire)) - list_add(&mnt->mnt_expire, &old->mnt_expire); - } + if ((flag & CL_MAKE_SHARED) && !mnt->mnt_group_id) { + err = mnt_alloc_group_id(mnt); + if (err) + goto out_free; + } + + mnt->mnt.mnt_flags = old->mnt.mnt_flags & ~MNT_WRITE_HOLD; + atomic_inc(&sb->s_active); + mnt->mnt.mnt_sb = sb; + mnt->mnt.mnt_root = dget(root); + mnt->mnt_mountpoint = mnt->mnt.mnt_root; + mnt->mnt_parent = mnt; + br_write_lock(&vfsmount_lock); + list_add_tail(&mnt->mnt_instance, &sb->s_mounts); + br_write_unlock(&vfsmount_lock); + + if (flag & CL_SLAVE) { + list_add(&mnt->mnt_slave, &old->mnt_slave_list); + mnt->mnt_master = old; + CLEAR_MNT_SHARED(mnt); + } else if (!(flag & CL_PRIVATE)) { + if ((flag & CL_MAKE_SHARED) || IS_MNT_SHARED(old)) + list_add(&mnt->mnt_share, &old->mnt_share); + if (IS_MNT_SLAVE(old)) + list_add(&mnt->mnt_slave, &old->mnt_slave); + mnt->mnt_master = old->mnt_master; + } + if (flag & CL_MAKE_SHARED) + set_mnt_shared(mnt); + + /* stick the duplicate mount on the same expiry list + * as the original if that was on one */ + if (flag & CL_EXPIRE) { + if (!list_empty(&old->mnt_expire)) + list_add(&mnt->mnt_expire, &old->mnt_expire); } + return mnt; out_free: free_vfsmnt(mnt); - return NULL; + return ERR_PTR(err); } static inline void mntfree(struct mount *mnt) @@ -1242,11 +1246,12 @@ struct mount *copy_tree(struct mount *mnt, struct dentry *dentry, struct path path; if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(mnt)) - return NULL; + return ERR_PTR(-EINVAL); res = q = clone_mnt(mnt, dentry, flag); - if (!q) - goto Enomem; + if (IS_ERR(q)) + return q; + q->mnt_mountpoint = mnt->mnt_mountpoint; p = mnt; @@ -1268,8 +1273,8 @@ struct mount *copy_tree(struct mount *mnt, struct dentry *dentry, path.mnt = &q->mnt; path.dentry = p->mnt_mountpoint; q = clone_mnt(p, p->mnt.mnt_root, flag); - if (!q) - goto Enomem; + if (IS_ERR(q)) + goto out; br_write_lock(&vfsmount_lock); list_add_tail(&q->mnt_list, &res->mnt_list); attach_mnt(q, &path); @@ -1277,7 +1282,7 @@ struct mount *copy_tree(struct mount *mnt, struct dentry *dentry, } } return res; -Enomem: +out: if (res) { LIST_HEAD(umount_list); br_write_lock(&vfsmount_lock); @@ -1285,9 +1290,11 @@ Enomem: br_write_unlock(&vfsmount_lock); release_mounts(&umount_list); } - return NULL; + return q; } +/* Caller should check returned pointer for errors */ + struct vfsmount *collect_mounts(struct path *path) { struct mount *tree; @@ -1295,7 +1302,9 @@ struct vfsmount *collect_mounts(struct path *path) tree = copy_tree(real_mount(path->mnt), path->dentry, CL_COPY_ALL | CL_PRIVATE); up_write(&namespace_sem); - return tree ? &tree->mnt : NULL; + if (IS_ERR(tree)) + return NULL; + return &tree->mnt; } void drop_collected_mounts(struct vfsmount *mnt) @@ -1590,14 +1599,15 @@ static int do_loopback(struct path *path, char *old_name, if (!check_mnt(real_mount(path->mnt)) || !check_mnt(old)) goto out2; - err = -ENOMEM; if (recurse) mnt = copy_tree(old, old_path.dentry, 0); else mnt = clone_mnt(old, old_path.dentry, 0); - if (!mnt) - goto out2; + if (IS_ERR(mnt)) { + err = PTR_ERR(mnt); + goto out; + } err = graft_tree(mnt, path); if (err) { @@ -2211,10 +2221,10 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, down_write(&namespace_sem); /* First pass: copy the tree topology */ new = copy_tree(old, old->mnt.mnt_root, CL_COPY_ALL | CL_EXPIRE); - if (!new) { + if (IS_ERR(new)) { up_write(&namespace_sem); kfree(new_ns); - return ERR_PTR(-ENOMEM); + return ERR_CAST(new); } new_ns->root = new; br_write_lock(&vfsmount_lock); diff --git a/fs/pnode.c b/fs/pnode.c index bed378db0758..3e000a51ac0d 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -237,8 +237,9 @@ int propagate_mnt(struct mount *dest_mnt, struct dentry *dest_dentry, source = get_source(m, prev_dest_mnt, prev_src_mnt, &type); - if (!(child = copy_tree(source, source->mnt.mnt_root, type))) { - ret = -ENOMEM; + child = copy_tree(source, source->mnt.mnt_root, type); + if (IS_ERR(child)) { + ret = PTR_ERR(child); list_splice(tree_list, tmp_list.prev); goto out; } diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 5bf0790497e7..3a5ca582ba1e 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -595,7 +595,7 @@ void audit_trim_trees(void) root_mnt = collect_mounts(&path); path_put(&path); - if (!root_mnt) + if (IS_ERR(root_mnt)) goto skip_it; spin_lock(&hash_lock); @@ -669,8 +669,8 @@ int audit_add_tree_rule(struct audit_krule *rule) goto Err; mnt = collect_mounts(&path); path_put(&path); - if (!mnt) { - err = -ENOMEM; + if (IS_ERR(mnt)) { + err = PTR_ERR(mnt); goto Err; } @@ -719,8 +719,8 @@ int audit_tag_tree(char *old, char *new) return err; tagged = collect_mounts(&path2); path_put(&path2); - if (!tagged) - return -ENOMEM; + if (IS_ERR(tagged)) + return PTR_ERR(tagged); err = kern_path(old, 0, &path1); if (err) { -- cgit v1.2.3 From 9249e17fe094d853d1ef7475dd559a2cc7e23d42 Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 25 Jun 2012 12:55:37 +0100 Subject: VFS: Pass mount flags to sget() Pass mount flags to sget() so that it can use them in initialising a new superblock before the set function is called. They could also be passed to the compare function. Signed-off-by: David Howells Signed-off-by: Al Viro --- drivers/mtd/mtdsuper.c | 4 +--- fs/9p/vfs_super.c | 4 ++-- fs/afs/super.c | 3 +-- fs/btrfs/super.c | 4 ++-- fs/ceph/super.c | 2 +- fs/cifs/cifsfs.c | 9 ++++----- fs/devpts/inode.c | 6 +++--- fs/ecryptfs/main.c | 3 +-- fs/gfs2/ops_fstype.c | 5 ++--- fs/libfs.c | 4 ++-- fs/logfs/super.c | 3 +-- fs/nfs/super.c | 2 +- fs/nilfs2/super.c | 4 ++-- fs/proc/root.c | 3 +-- fs/reiserfs/procfs.c | 2 +- fs/super.c | 22 +++++++++++----------- fs/sysfs/mount.c | 3 +-- fs/ubifs/super.c | 3 +-- include/linux/fs.h | 2 +- kernel/cgroup.c | 2 +- 20 files changed, 40 insertions(+), 50 deletions(-) (limited to 'kernel') diff --git a/drivers/mtd/mtdsuper.c b/drivers/mtd/mtdsuper.c index a90bfe79916d..334da5f583c0 100644 --- a/drivers/mtd/mtdsuper.c +++ b/drivers/mtd/mtdsuper.c @@ -63,7 +63,7 @@ static struct dentry *mount_mtd_aux(struct file_system_type *fs_type, int flags, struct super_block *sb; int ret; - sb = sget(fs_type, get_sb_mtd_compare, get_sb_mtd_set, mtd); + sb = sget(fs_type, get_sb_mtd_compare, get_sb_mtd_set, flags, mtd); if (IS_ERR(sb)) goto out_error; @@ -74,8 +74,6 @@ static struct dentry *mount_mtd_aux(struct file_system_type *fs_type, int flags, pr_debug("MTDSB: New superblock for device %d (\"%s\")\n", mtd->index, mtd->name); - sb->s_flags = flags; - ret = fill_super(sb, data, flags & MS_SILENT ? 1 : 0); if (ret < 0) { deactivate_locked_super(sb); diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index 8c92a9ba8330..137d50396898 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -89,7 +89,7 @@ v9fs_fill_super(struct super_block *sb, struct v9fs_session_info *v9ses, if (v9ses->cache) sb->s_bdi->ra_pages = (VM_MAX_READAHEAD * 1024)/PAGE_CACHE_SIZE; - sb->s_flags = flags | MS_ACTIVE | MS_DIRSYNC | MS_NOATIME; + sb->s_flags |= MS_ACTIVE | MS_DIRSYNC | MS_NOATIME; if (!v9ses->cache) sb->s_flags |= MS_SYNCHRONOUS; @@ -137,7 +137,7 @@ static struct dentry *v9fs_mount(struct file_system_type *fs_type, int flags, goto close_session; } - sb = sget(fs_type, NULL, v9fs_set_super, v9ses); + sb = sget(fs_type, NULL, v9fs_set_super, flags, v9ses); if (IS_ERR(sb)) { retval = PTR_ERR(sb); goto clunk_fid; diff --git a/fs/afs/super.c b/fs/afs/super.c index f02b31e7e648..df8c6047c2a1 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -395,7 +395,7 @@ static struct dentry *afs_mount(struct file_system_type *fs_type, as->volume = vol; /* allocate a deviceless superblock */ - sb = sget(fs_type, afs_test_super, afs_set_super, as); + sb = sget(fs_type, afs_test_super, afs_set_super, flags, as); if (IS_ERR(sb)) { ret = PTR_ERR(sb); afs_put_volume(vol); @@ -406,7 +406,6 @@ static struct dentry *afs_mount(struct file_system_type *fs_type, if (!sb->s_root) { /* initial superblock/root creation */ _debug("create"); - sb->s_flags = flags; ret = afs_fill_super(sb, ¶ms); if (ret < 0) { deactivate_locked_super(sb); diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index e23991574fdf..b19d75567728 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -1068,7 +1068,8 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, } bdev = fs_devices->latest_bdev; - s = sget(fs_type, btrfs_test_super, btrfs_set_super, fs_info); + s = sget(fs_type, btrfs_test_super, btrfs_set_super, flags | MS_NOSEC, + fs_info); if (IS_ERR(s)) { error = PTR_ERR(s); goto error_close_devices; @@ -1082,7 +1083,6 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, } else { char b[BDEVNAME_SIZE]; - s->s_flags = flags | MS_NOSEC; strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); btrfs_sb(s)->bdev_holder = fs_type; error = btrfs_fill_super(s, fs_devices, data, diff --git a/fs/ceph/super.c b/fs/ceph/super.c index 1e67dd7305a4..7076109f014d 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -871,7 +871,7 @@ static struct dentry *ceph_mount(struct file_system_type *fs_type, if (ceph_test_opt(fsc->client, NOSHARE)) compare_super = NULL; - sb = sget(fs_type, compare_super, ceph_set_super, fsc); + sb = sget(fs_type, compare_super, ceph_set_super, flags, fsc); if (IS_ERR(sb)) { res = ERR_CAST(sb); goto out; diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index c0c2751a7573..a7610cfedf0a 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -637,7 +637,10 @@ cifs_do_mount(struct file_system_type *fs_type, mnt_data.cifs_sb = cifs_sb; mnt_data.flags = flags; - sb = sget(fs_type, cifs_match_super, cifs_set_super, &mnt_data); + /* BB should we make this contingent on mount parm? */ + flags |= MS_NODIRATIME | MS_NOATIME; + + sb = sget(fs_type, cifs_match_super, cifs_set_super, flags, &mnt_data); if (IS_ERR(sb)) { root = ERR_CAST(sb); cifs_umount(cifs_sb); @@ -648,10 +651,6 @@ cifs_do_mount(struct file_system_type *fs_type, cFYI(1, "Use existing superblock"); cifs_umount(cifs_sb); } else { - sb->s_flags = flags; - /* BB should we make this contingent on mount parm? */ - sb->s_flags |= MS_NODIRATIME | MS_NOATIME; - rc = cifs_read_super(sb); if (rc) { root = ERR_PTR(rc); diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index 979c1e309c73..14afbabe6546 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -439,15 +439,15 @@ static struct dentry *devpts_mount(struct file_system_type *fs_type, return ERR_PTR(error); if (opts.newinstance) - s = sget(fs_type, NULL, set_anon_super, NULL); + s = sget(fs_type, NULL, set_anon_super, flags, NULL); else - s = sget(fs_type, compare_init_pts_sb, set_anon_super, NULL); + s = sget(fs_type, compare_init_pts_sb, set_anon_super, flags, + NULL); if (IS_ERR(s)) return ERR_CAST(s); if (!s->s_root) { - s->s_flags = flags; error = devpts_fill_super(s, data, flags & MS_SILENT ? 1 : 0); if (error) goto out_undo_sget; diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 68954937a071..7edeb3d893c1 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -499,13 +499,12 @@ static struct dentry *ecryptfs_mount(struct file_system_type *fs_type, int flags goto out; } - s = sget(fs_type, NULL, set_anon_super, NULL); + s = sget(fs_type, NULL, set_anon_super, flags, NULL); if (IS_ERR(s)) { rc = PTR_ERR(s); goto out; } - s->s_flags = flags; rc = bdi_setup_and_register(&sbi->bdi, "ecryptfs", BDI_CAP_MAP_COPY); if (rc) goto out1; diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index b8c250fc4922..6c906078f657 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -1286,7 +1286,7 @@ static struct dentry *gfs2_mount(struct file_system_type *fs_type, int flags, error = -EBUSY; goto error_bdev; } - s = sget(fs_type, test_gfs2_super, set_gfs2_super, bdev); + s = sget(fs_type, test_gfs2_super, set_gfs2_super, flags, bdev); mutex_unlock(&bdev->bd_fsfreeze_mutex); error = PTR_ERR(s); if (IS_ERR(s)) @@ -1316,7 +1316,6 @@ static struct dentry *gfs2_mount(struct file_system_type *fs_type, int flags, } else { char b[BDEVNAME_SIZE]; - s->s_flags = flags; s->s_mode = mode; strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); sb_set_blocksize(s, block_size(bdev)); @@ -1360,7 +1359,7 @@ static struct dentry *gfs2_mount_meta(struct file_system_type *fs_type, dev_name, error); return ERR_PTR(error); } - s = sget(&gfs2_fs_type, test_gfs2_super, set_meta_super, + s = sget(&gfs2_fs_type, test_gfs2_super, set_meta_super, flags, path.dentry->d_inode->i_sb->s_bdev); path_put(&path); if (IS_ERR(s)) { diff --git a/fs/libfs.c b/fs/libfs.c index ebd03f6910d5..a74cb1725ac6 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -222,15 +222,15 @@ struct dentry *mount_pseudo(struct file_system_type *fs_type, char *name, const struct super_operations *ops, const struct dentry_operations *dops, unsigned long magic) { - struct super_block *s = sget(fs_type, NULL, set_anon_super, NULL); + struct super_block *s; struct dentry *dentry; struct inode *root; struct qstr d_name = QSTR_INIT(name, strlen(name)); + s = sget(fs_type, NULL, set_anon_super, MS_NOUSER, NULL); if (IS_ERR(s)) return ERR_CAST(s); - s->s_flags = MS_NOUSER; s->s_maxbytes = MAX_LFS_FILESIZE; s->s_blocksize = PAGE_SIZE; s->s_blocksize_bits = PAGE_SHIFT; diff --git a/fs/logfs/super.c b/fs/logfs/super.c index 97bca623d893..345c24b8a6f8 100644 --- a/fs/logfs/super.c +++ b/fs/logfs/super.c @@ -519,7 +519,7 @@ static struct dentry *logfs_get_sb_device(struct logfs_super *super, log_super("LogFS: Start mount %x\n", mount_count++); err = -EINVAL; - sb = sget(type, logfs_sb_test, logfs_sb_set, super); + sb = sget(type, logfs_sb_test, logfs_sb_set, flags | MS_NOATIME, super); if (IS_ERR(sb)) { super->s_devops->put_device(super); kfree(super); @@ -542,7 +542,6 @@ static struct dentry *logfs_get_sb_device(struct logfs_super *super, sb->s_maxbytes = (1ull << 43) - 1; sb->s_max_links = LOGFS_LINK_MAX; sb->s_op = &logfs_super_operations; - sb->s_flags = flags | MS_NOATIME; err = logfs_read_sb(sb, sb->s_flags & MS_RDONLY); if (err) diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 06228192f64e..8b2a2977b720 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -2419,7 +2419,7 @@ static struct dentry *nfs_fs_mount_common(struct file_system_type *fs_type, sb_mntdata.mntflags |= MS_SYNCHRONOUS; /* Get a superblock - note that we may end up sharing one that already exists */ - s = sget(fs_type, compare_super, nfs_set_super, &sb_mntdata); + s = sget(fs_type, compare_super, nfs_set_super, flags, &sb_mntdata); if (IS_ERR(s)) { mntroot = ERR_CAST(s); goto out_err_nosb; diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index 1099a76cee59..d57c42f974ea 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -1288,7 +1288,8 @@ nilfs_mount(struct file_system_type *fs_type, int flags, err = -EBUSY; goto failed; } - s = sget(fs_type, nilfs_test_bdev_super, nilfs_set_bdev_super, sd.bdev); + s = sget(fs_type, nilfs_test_bdev_super, nilfs_set_bdev_super, flags, + sd.bdev); mutex_unlock(&sd.bdev->bd_fsfreeze_mutex); if (IS_ERR(s)) { err = PTR_ERR(s); @@ -1301,7 +1302,6 @@ nilfs_mount(struct file_system_type *fs_type, int flags, s_new = true; /* New superblock instance created */ - s->s_flags = flags; s->s_mode = mode; strlcpy(s->s_id, bdevname(sd.bdev, b), sizeof(s->s_id)); sb_set_blocksize(s, block_size(sd.bdev)); diff --git a/fs/proc/root.c b/fs/proc/root.c index 568b20290c75..9a2d9fd7cadd 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -111,7 +111,7 @@ static struct dentry *proc_mount(struct file_system_type *fs_type, options = data; } - sb = sget(fs_type, proc_test_super, proc_set_super, ns); + sb = sget(fs_type, proc_test_super, proc_set_super, flags, ns); if (IS_ERR(sb)) return ERR_CAST(sb); @@ -121,7 +121,6 @@ static struct dentry *proc_mount(struct file_system_type *fs_type, } if (!sb->s_root) { - sb->s_flags = flags; err = proc_fill_super(sb); if (err) { deactivate_locked_super(sb); diff --git a/fs/reiserfs/procfs.c b/fs/reiserfs/procfs.c index 2c1ade692cc8..e60e87035bb3 100644 --- a/fs/reiserfs/procfs.c +++ b/fs/reiserfs/procfs.c @@ -403,7 +403,7 @@ static void *r_start(struct seq_file *m, loff_t * pos) if (l) return NULL; - if (IS_ERR(sget(&reiserfs_fs_type, test_sb, set_sb, s))) + if (IS_ERR(sget(&reiserfs_fs_type, test_sb, set_sb, 0, s))) return NULL; up_write(&s->s_umount); diff --git a/fs/super.c b/fs/super.c index cf001775617f..c743fb3be4b8 100644 --- a/fs/super.c +++ b/fs/super.c @@ -105,11 +105,12 @@ static int prune_super(struct shrinker *shrink, struct shrink_control *sc) /** * alloc_super - create new superblock * @type: filesystem type superblock should belong to + * @flags: the mount flags * * Allocates and initializes a new &struct super_block. alloc_super() * returns a pointer new superblock or %NULL if allocation had failed. */ -static struct super_block *alloc_super(struct file_system_type *type) +static struct super_block *alloc_super(struct file_system_type *type, int flags) { struct super_block *s = kzalloc(sizeof(struct super_block), GFP_USER); static const struct super_operations default_op; @@ -136,6 +137,7 @@ static struct super_block *alloc_super(struct file_system_type *type) #else INIT_LIST_HEAD(&s->s_files); #endif + s->s_flags = flags; s->s_bdi = &default_backing_dev_info; INIT_HLIST_NODE(&s->s_instances); INIT_HLIST_BL_HEAD(&s->s_anon); @@ -415,11 +417,13 @@ EXPORT_SYMBOL(generic_shutdown_super); * @type: filesystem type superblock should belong to * @test: comparison callback * @set: setup callback + * @flags: mount flags * @data: argument to each of them */ struct super_block *sget(struct file_system_type *type, int (*test)(struct super_block *,void *), int (*set)(struct super_block *,void *), + int flags, void *data) { struct super_block *s = NULL; @@ -450,7 +454,7 @@ retry: } if (!s) { spin_unlock(&sb_lock); - s = alloc_super(type); + s = alloc_super(type, flags); if (!s) return ERR_PTR(-ENOMEM); goto retry; @@ -925,13 +929,12 @@ struct dentry *mount_ns(struct file_system_type *fs_type, int flags, { struct super_block *sb; - sb = sget(fs_type, ns_test_super, ns_set_super, data); + sb = sget(fs_type, ns_test_super, ns_set_super, flags, data); if (IS_ERR(sb)) return ERR_CAST(sb); if (!sb->s_root) { int err; - sb->s_flags = flags; err = fill_super(sb, data, flags & MS_SILENT ? 1 : 0); if (err) { deactivate_locked_super(sb); @@ -992,7 +995,8 @@ struct dentry *mount_bdev(struct file_system_type *fs_type, error = -EBUSY; goto error_bdev; } - s = sget(fs_type, test_bdev_super, set_bdev_super, bdev); + s = sget(fs_type, test_bdev_super, set_bdev_super, flags | MS_NOSEC, + bdev); mutex_unlock(&bdev->bd_fsfreeze_mutex); if (IS_ERR(s)) goto error_s; @@ -1017,7 +1021,6 @@ struct dentry *mount_bdev(struct file_system_type *fs_type, } else { char b[BDEVNAME_SIZE]; - s->s_flags = flags | MS_NOSEC; s->s_mode = mode; strlcpy(s->s_id, bdevname(bdev, b), sizeof(s->s_id)); sb_set_blocksize(s, block_size(bdev)); @@ -1062,13 +1065,11 @@ struct dentry *mount_nodev(struct file_system_type *fs_type, int (*fill_super)(struct super_block *, void *, int)) { int error; - struct super_block *s = sget(fs_type, NULL, set_anon_super, NULL); + struct super_block *s = sget(fs_type, NULL, set_anon_super, flags, NULL); if (IS_ERR(s)) return ERR_CAST(s); - s->s_flags = flags; - error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); if (error) { deactivate_locked_super(s); @@ -1091,11 +1092,10 @@ struct dentry *mount_single(struct file_system_type *fs_type, struct super_block *s; int error; - s = sget(fs_type, compare_single, set_anon_super, NULL); + s = sget(fs_type, compare_single, set_anon_super, flags, NULL); if (IS_ERR(s)) return ERR_CAST(s); if (!s->s_root) { - s->s_flags = flags; error = fill_super(s, data, flags & MS_SILENT ? 1 : 0); if (error) { deactivate_locked_super(s); diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index c15a7a3572e9..71eb7e253927 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -118,13 +118,12 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, for (type = KOBJ_NS_TYPE_NONE; type < KOBJ_NS_TYPES; type++) info->ns[type] = kobj_ns_grab_current(type); - sb = sget(fs_type, sysfs_test_super, sysfs_set_super, info); + sb = sget(fs_type, sysfs_test_super, sysfs_set_super, flags, info); if (IS_ERR(sb) || sb->s_fs_info != info) free_sysfs_super_info(info); if (IS_ERR(sb)) return ERR_CAST(sb); if (!sb->s_root) { - sb->s_flags = flags; error = sysfs_fill_super(sb, data, flags & MS_SILENT ? 1 : 0); if (error) { deactivate_locked_super(sb); diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 5862dd9d2784..1c766c39c038 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -2136,7 +2136,7 @@ static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags, dbg_gen("opened ubi%d_%d", c->vi.ubi_num, c->vi.vol_id); - sb = sget(fs_type, sb_test, sb_set, c); + sb = sget(fs_type, sb_test, sb_set, flags, c); if (IS_ERR(sb)) { err = PTR_ERR(sb); kfree(c); @@ -2153,7 +2153,6 @@ static struct dentry *ubifs_mount(struct file_system_type *fs_type, int flags, goto out_deact; } } else { - sb->s_flags = flags; err = ubifs_fill_super(sb, data, flags & MS_SILENT ? 1 : 0); if (err) goto out_deact; diff --git a/include/linux/fs.h b/include/linux/fs.h index 2f857e9eeb3a..48548bdd7722 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1914,7 +1914,7 @@ void free_anon_bdev(dev_t); struct super_block *sget(struct file_system_type *type, int (*test)(struct super_block *,void *), int (*set)(struct super_block *,void *), - void *data); + int flags, void *data); extern struct dentry *mount_pseudo(struct file_system_type *, char *, const struct super_operations *ops, const struct dentry_operations *dops, diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 0cd1314acdaf..af2b5641fc8b 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -1587,7 +1587,7 @@ static struct dentry *cgroup_mount(struct file_system_type *fs_type, opts.new_root = new_root; /* Locate an existing or new sb for this hierarchy */ - sb = sget(fs_type, cgroup_test_super, cgroup_set_super, &opts); + sb = sget(fs_type, cgroup_test_super, cgroup_set_super, 0, &opts); if (IS_ERR(sb)) { ret = PTR_ERR(sb); cgroup_drop_root(opts.new_root); -- cgit v1.2.3 From 775f4b297b780601e61787b766f306ed3e1d23eb Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Mon, 2 Jul 2012 07:52:16 -0400 Subject: random: make 'add_interrupt_randomness()' do something sane We've been moving away from add_interrupt_randomness() for various reasons: it's too expensive to do on every interrupt, and flooding the CPU with interrupts could theoretically cause bogus floods of entropy from a somewhat externally controllable source. This solves both problems by limiting the actual randomness addition to just once a second or after 64 interrupts, whicever comes first. During that time, the interrupt cycle data is buffered up in a per-cpu pool. Also, we make sure the the nonblocking pool used by urandom is initialized before we start feeding the normal input pool. This assures that /dev/urandom is returning unpredictable data as soon as possible. (Based on an original patch by Linus, but significantly modified by tytso.) Tested-by: Eric Wustrow Reported-by: Eric Wustrow Reported-by: Nadia Heninger Reported-by: Zakir Durumeric Reported-by: J. Alex Halderman . Signed-off-by: Linus Torvalds Signed-off-by: "Theodore Ts'o" Cc: stable@vger.kernel.org --- drivers/char/random.c | 103 ++++++++++++++++++++++++++++++++++++++-------- drivers/mfd/ab3100-core.c | 2 - include/linux/random.h | 2 +- kernel/irq/handle.c | 7 ++-- 4 files changed, 90 insertions(+), 24 deletions(-) (limited to 'kernel') diff --git a/drivers/char/random.c b/drivers/char/random.c index cb541b9a5231..9fcceace239c 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -127,19 +127,15 @@ * * void add_input_randomness(unsigned int type, unsigned int code, * unsigned int value); - * void add_interrupt_randomness(int irq); + * void add_interrupt_randomness(int irq, int irq_flags); * void add_disk_randomness(struct gendisk *disk); * * add_input_randomness() uses the input layer interrupt timing, as well as * the event type information from the hardware. * - * add_interrupt_randomness() uses the inter-interrupt timing as random - * inputs to the entropy pool. Note that not all interrupts are good - * sources of randomness! For example, the timer interrupts is not a - * good choice, because the periodicity of the interrupts is too - * regular, and hence predictable to an attacker. Network Interface - * Controller interrupts are a better measure, since the timing of the - * NIC interrupts are more unpredictable. + * add_interrupt_randomness() uses the interrupt timing as random + * inputs to the entropy pool. Using the cycle counters and the irq source + * as inputs, it feeds the randomness roughly once a second. * * add_disk_randomness() uses what amounts to the seek time of block * layer request events, on a per-disk_devt basis, as input to the @@ -248,6 +244,7 @@ #include #include #include +#include #ifdef CONFIG_GENERIC_HARDIRQS # include @@ -256,6 +253,7 @@ #include #include #include +#include #include /* @@ -421,7 +419,9 @@ struct entropy_store { spinlock_t lock; unsigned add_ptr; int entropy_count; + int entropy_total; int input_rotate; + unsigned int initialized:1; __u8 last_data[EXTRACT_SIZE]; }; @@ -454,6 +454,10 @@ static struct entropy_store nonblocking_pool = { .pool = nonblocking_pool_data }; +static __u32 const twist_table[8] = { + 0x00000000, 0x3b6e20c8, 0x76dc4190, 0x4db26158, + 0xedb88320, 0xd6d6a3e8, 0x9b64c2b0, 0xa00ae278 }; + /* * This function adds bytes into the entropy "pool". It does not * update the entropy estimate. The caller should call @@ -467,9 +471,6 @@ static struct entropy_store nonblocking_pool = { static void mix_pool_bytes_extract(struct entropy_store *r, const void *in, int nbytes, __u8 out[64]) { - static __u32 const twist_table[8] = { - 0x00000000, 0x3b6e20c8, 0x76dc4190, 0x4db26158, - 0xedb88320, 0xd6d6a3e8, 0x9b64c2b0, 0xa00ae278 }; unsigned long i, j, tap1, tap2, tap3, tap4, tap5; int input_rotate; int wordmask = r->poolinfo->poolwords - 1; @@ -528,6 +529,36 @@ static void mix_pool_bytes(struct entropy_store *r, const void *in, int bytes) mix_pool_bytes_extract(r, in, bytes, NULL); } +struct fast_pool { + __u32 pool[4]; + unsigned long last; + unsigned short count; + unsigned char rotate; + unsigned char last_timer_intr; +}; + +/* + * This is a fast mixing routine used by the interrupt randomness + * collector. It's hardcoded for an 128 bit pool and assumes that any + * locks that might be needed are taken by the caller. + */ +static void fast_mix(struct fast_pool *f, const void *in, int nbytes) +{ + const char *bytes = in; + __u32 w; + unsigned i = f->count; + unsigned input_rotate = f->rotate; + + while (nbytes--) { + w = rol32(*bytes++, input_rotate & 31) ^ f->pool[i & 3] ^ + f->pool[(i + 1) & 3]; + f->pool[i & 3] = (w >> 3) ^ twist_table[w & 7]; + input_rotate += (i++ & 3) ? 7 : 14; + } + f->count = i; + f->rotate = input_rotate; +} + /* * Credit (or debit) the entropy store with n bits of entropy */ @@ -551,6 +582,12 @@ static void credit_entropy_bits(struct entropy_store *r, int nbits) entropy_count = r->poolinfo->POOLBITS; r->entropy_count = entropy_count; + if (!r->initialized && nbits > 0) { + r->entropy_total += nbits; + if (r->entropy_total > 128) + r->initialized = 1; + } + /* should we wake readers? */ if (r == &input_pool && entropy_count >= random_read_wakeup_thresh) { wake_up_interruptible(&random_read_wait); @@ -700,17 +737,48 @@ void add_input_randomness(unsigned int type, unsigned int code, } EXPORT_SYMBOL_GPL(add_input_randomness); -void add_interrupt_randomness(int irq) +static DEFINE_PER_CPU(struct fast_pool, irq_randomness); + +void add_interrupt_randomness(int irq, int irq_flags) { - struct timer_rand_state *state; + struct entropy_store *r; + struct fast_pool *fast_pool = &__get_cpu_var(irq_randomness); + struct pt_regs *regs = get_irq_regs(); + unsigned long now = jiffies; + __u32 input[4], cycles = get_cycles(); + + input[0] = cycles ^ jiffies; + input[1] = irq; + if (regs) { + __u64 ip = instruction_pointer(regs); + input[2] = ip; + input[3] = ip >> 32; + } - state = get_timer_rand_state(irq); + fast_mix(fast_pool, input, sizeof(input)); - if (state == NULL) + if ((fast_pool->count & 1023) && + !time_after(now, fast_pool->last + HZ)) return; - DEBUG_ENT("irq event %d\n", irq); - add_timer_randomness(state, 0x100 + irq); + fast_pool->last = now; + + r = nonblocking_pool.initialized ? &input_pool : &nonblocking_pool; + mix_pool_bytes(r, &fast_pool->pool, sizeof(fast_pool->pool)); + /* + * If we don't have a valid cycle counter, and we see + * back-to-back timer interrupts, then skip giving credit for + * any entropy. + */ + if (cycles == 0) { + if (irq_flags & __IRQF_TIMER) { + if (fast_pool->last_timer_intr) + return; + fast_pool->last_timer_intr = 1; + } else + fast_pool->last_timer_intr = 0; + } + credit_entropy_bits(r, 1); } #ifdef CONFIG_BLOCK @@ -971,6 +1039,7 @@ static void init_std_data(struct entropy_store *r) spin_lock_irqsave(&r->lock, flags); r->entropy_count = 0; + r->entropy_total = 0; spin_unlock_irqrestore(&r->lock, flags); now = ktime_get_real(); diff --git a/drivers/mfd/ab3100-core.c b/drivers/mfd/ab3100-core.c index 1efad20fb175..9522d6bda4f7 100644 --- a/drivers/mfd/ab3100-core.c +++ b/drivers/mfd/ab3100-core.c @@ -409,8 +409,6 @@ static irqreturn_t ab3100_irq_handler(int irq, void *data) u32 fatevent; int err; - add_interrupt_randomness(irq); - err = ab3100_get_register_page_interruptible(ab3100, AB3100_EVENTA1, event_regs, 3); if (err) diff --git a/include/linux/random.h b/include/linux/random.h index 8f74538c96db..6ef39d7f2db1 100644 --- a/include/linux/random.h +++ b/include/linux/random.h @@ -52,7 +52,7 @@ extern void rand_initialize_irq(int irq); extern void add_input_randomness(unsigned int type, unsigned int code, unsigned int value); -extern void add_interrupt_randomness(int irq); +extern void add_interrupt_randomness(int irq, int irq_flags); extern void get_random_bytes(void *buf, int nbytes); void generate_random_uuid(unsigned char uuid_out[16]); diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index bdb180325551..131ca176b497 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -133,7 +133,7 @@ irqreturn_t handle_irq_event_percpu(struct irq_desc *desc, struct irqaction *action) { irqreturn_t retval = IRQ_NONE; - unsigned int random = 0, irq = desc->irq_data.irq; + unsigned int flags = 0, irq = desc->irq_data.irq; do { irqreturn_t res; @@ -161,7 +161,7 @@ handle_irq_event_percpu(struct irq_desc *desc, struct irqaction *action) /* Fall through to add to randomness */ case IRQ_HANDLED: - random |= action->flags; + flags |= action->flags; break; default: @@ -172,8 +172,7 @@ handle_irq_event_percpu(struct irq_desc *desc, struct irqaction *action) action = action->next; } while (action); - if (random & IRQF_SAMPLE_RANDOM) - add_interrupt_randomness(irq); + add_interrupt_randomness(irq, flags); if (!noirqdebug) note_interrupt(irq, desc, retval); -- cgit v1.2.3 From 70498253186586e5dca7bc3ebd3415203b059fbc Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Mon, 16 Jul 2012 18:35:29 -0700 Subject: kmsg - properly print over-long continuation lines Reserve PREFIX_MAX bytes in the LOG_LINE_MAX line when buffering a continuation line, to be able to properly prefix the LOG_LINE_MAX line with the syslog prefix and timestamp when printing it. Reported-By: Dave Jones Signed-off-by: Kay Sievers Cc: stable Signed-off-by: Greg Kroah-Hartman --- kernel/printk.c | 33 +++++++++++++++++++-------------- 1 file changed, 19 insertions(+), 14 deletions(-) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index 177fa49357a5..d87ca5c6a989 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -235,7 +235,8 @@ static u32 log_next_idx; static u64 clear_seq; static u32 clear_idx; -#define LOG_LINE_MAX 1024 +#define PREFIX_MAX 32 +#define LOG_LINE_MAX 1024 - PREFIX_MAX /* record buffer */ #if defined(CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS) @@ -876,7 +877,7 @@ static size_t msg_print_text(const struct log *msg, enum log_flags prev, if (buf) { if (print_prefix(msg, syslog, NULL) + - text_len + 1>= size - len) + text_len + 1 >= size - len) break; if (prefix) @@ -907,7 +908,7 @@ static int syslog_print(char __user *buf, int size) struct log *msg; int len = 0; - text = kmalloc(LOG_LINE_MAX, GFP_KERNEL); + text = kmalloc(LOG_LINE_MAX + PREFIX_MAX, GFP_KERNEL); if (!text) return -ENOMEM; @@ -930,7 +931,8 @@ static int syslog_print(char __user *buf, int size) skip = syslog_partial; msg = log_from_idx(syslog_idx); - n = msg_print_text(msg, syslog_prev, true, text, LOG_LINE_MAX); + n = msg_print_text(msg, syslog_prev, true, text, + LOG_LINE_MAX + PREFIX_MAX); if (n - syslog_partial <= size) { /* message fits into buffer, move forward */ syslog_idx = log_next(syslog_idx); @@ -969,7 +971,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) char *text; int len = 0; - text = kmalloc(LOG_LINE_MAX, GFP_KERNEL); + text = kmalloc(LOG_LINE_MAX + PREFIX_MAX, GFP_KERNEL); if (!text) return -ENOMEM; @@ -1022,7 +1024,8 @@ static int syslog_print_all(char __user *buf, int size, bool clear) struct log *msg = log_from_idx(idx); int textlen; - textlen = msg_print_text(msg, prev, true, text, LOG_LINE_MAX); + textlen = msg_print_text(msg, prev, true, text, + LOG_LINE_MAX + PREFIX_MAX); if (textlen < 0) { len = textlen; break; @@ -1367,15 +1370,15 @@ static struct cont { bool flushed:1; /* buffer sealed and committed */ } cont; -static void cont_flush(void) +static void cont_flush(enum log_flags flags) { if (cont.flushed) return; if (cont.len == 0) return; - log_store(cont.facility, cont.level, LOG_NOCONS, cont.ts_nsec, - NULL, 0, cont.buf, cont.len); + log_store(cont.facility, cont.level, LOG_NOCONS | flags, + cont.ts_nsec, NULL, 0, cont.buf, cont.len); cont.flushed = true; } @@ -1386,7 +1389,8 @@ static bool cont_add(int facility, int level, const char *text, size_t len) return false; if (cont.len + len > sizeof(cont.buf)) { - cont_flush(); + /* the line gets too long, split it up in separate records */ + cont_flush(LOG_CONT); return false; } @@ -1522,7 +1526,7 @@ asmlinkage int vprintk_emit(int facility, int level, * or another task also prints continuation lines. */ if (cont.len && (lflags & LOG_PREFIX || cont.owner != current)) - cont_flush(); + cont_flush(0); /* buffer line if possible, otherwise store it right away */ if (!cont_add(facility, level, text, text_len)) @@ -1540,7 +1544,7 @@ asmlinkage int vprintk_emit(int facility, int level, if (cont.len && cont.owner == current) { if (!(lflags & LOG_PREFIX)) stored = cont_add(facility, level, text, text_len); - cont_flush(); + cont_flush(0); } if (!stored) @@ -1633,7 +1637,8 @@ EXPORT_SYMBOL(printk); #else -#define LOG_LINE_MAX 0 +#define LOG_LINE_MAX 0 +#define PREFIX_MAX 0 static struct cont { size_t len; size_t cons; @@ -1938,7 +1943,7 @@ static enum log_flags console_prev; */ void console_unlock(void) { - static char text[LOG_LINE_MAX]; + static char text[LOG_LINE_MAX + PREFIX_MAX]; static u64 seen_seq; unsigned long flags; bool wake_klogd = false; -- cgit v1.2.3 From 96efedf1491cdf0616e5e4fff0711cebf20f69c7 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Mon, 16 Jul 2012 18:35:29 -0700 Subject: kmsg - avoid warning for CONFIG_PRINTK=n compilations Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- kernel/printk.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index d87ca5c6a989..6c3d5bf14da2 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -216,6 +216,7 @@ struct log { */ static DEFINE_RAW_SPINLOCK(logbuf_lock); +#ifdef CONFIG_PRINTK /* the next printk record to read by syslog(READ) or /proc/kmsg */ static u64 syslog_seq; static u32 syslog_idx; @@ -228,7 +229,6 @@ static u32 log_first_idx; /* index and sequence number of the next record to store in the buffer */ static u64 log_next_seq; -#ifdef CONFIG_PRINTK static u32 log_next_idx; /* the next printk record to read after the last 'clear' command */ @@ -1635,10 +1635,17 @@ asmlinkage int printk(const char *fmt, ...) } EXPORT_SYMBOL(printk); -#else +#else /* CONFIG_PRINTK */ #define LOG_LINE_MAX 0 #define PREFIX_MAX 0 +#define LOG_LINE_MAX 0 +static u64 syslog_seq; +static u32 syslog_idx; +static enum log_flags syslog_prev; +static u64 log_first_seq; +static u32 log_first_idx; +static u64 log_next_seq; static struct cont { size_t len; size_t cons; -- cgit v1.2.3 From d39f3d77c9b1fe7cc33a14beb4a4849af0a4ac22 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Mon, 16 Jul 2012 18:35:30 -0700 Subject: kmsg - export "continuation record" flag to /dev/kmsg In some cases we are forced to store individual records for a continuation line print. Export a flag to allow the external re-construction of the line. The flag allows us to apply a similar logic externally which is used internally when the console, /proc/kmsg or the syslog() output is printed. $ cat /dev/kmsg 4,165,0,-;Free swap = 0kB 4,166,0,-;Total swap = 0kB 6,167,0,c;[ 4,168,0,+;0 4,169,0,+;1 4,170,0,+;2 4,171,0,+;3 4,172,0,+;] 6,173,0,-;[0 1 2 3 ] 6,174,0,-;Console: colour VGA+ 80x25 6,175,0,-;console [tty0] enabled Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/dev-kmsg | 29 ++++++++++++++++++++--------- kernel/printk.c | 23 +++++++++++++++++++++-- 2 files changed, 41 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/Documentation/ABI/testing/dev-kmsg b/Documentation/ABI/testing/dev-kmsg index 281ecc5f9709..7e7e07a82e0e 100644 --- a/Documentation/ABI/testing/dev-kmsg +++ b/Documentation/ABI/testing/dev-kmsg @@ -58,16 +58,18 @@ Description: The /dev/kmsg character device node provides userspace access The output format consists of a prefix carrying the syslog prefix including priority and facility, the 64 bit message - sequence number and the monotonic timestamp in microseconds. - The values are separated by a ','. Future extensions might - add more comma separated values before the terminating ';'. - Unknown values should be gracefully ignored. + sequence number and the monotonic timestamp in microseconds, + and a flag field. All fields are separated by a ','. + + Future extensions might add more comma separated values before + the terminating ';'. Unknown fields and values should be + gracefully ignored. The human readable text string starts directly after the ';' and is terminated by a '\n'. Untrusted values derived from hardware or other facilities are printed, therefore - all non-printable characters in the log message are escaped - by "\x00" C-style hex encoding. + all non-printable characters and '\' itself in the log message + are escaped by "\x00" C-style hex encoding. A line starting with ' ', is a continuation line, adding key/value pairs to the log message, which provide the machine @@ -75,11 +77,11 @@ Description: The /dev/kmsg character device node provides userspace access userspace. Example: - 7,160,424069;pci_root PNP0A03:00: host bridge window [io 0x0000-0x0cf7] (ignored) + 7,160,424069,-;pci_root PNP0A03:00: host bridge window [io 0x0000-0x0cf7] (ignored) SUBSYSTEM=acpi DEVICE=+acpi:PNP0A03:00 - 6,339,5140900;NET: Registered protocol family 10 - 30,340,5690716;udevd[80]: starting version 181 + 6,339,5140900,-;NET: Registered protocol family 10 + 30,340,5690716,-;udevd[80]: starting version 181 The DEVICE= key uniquely identifies devices the following way: b12:8 - block dev_t @@ -87,4 +89,13 @@ Description: The /dev/kmsg character device node provides userspace access n8 - netdev ifindex +sound:card0 - subsystem:devname + The flags field carries '-' by default. A 'c' indicates a + fragment of a line. All following fragments are flagged with + '+'. Note, that these hints about continuation lines are not + neccessarily correct, and the stream could be interleaved with + unrelated messages, but merging the lines in the output + usually produces better human readable results. A similar + logic is used internally when messages are printed to the + console, /proc/kmsg or the syslog() syscall. + Users: dmesg(1), userspace kernel log consumers diff --git a/kernel/printk.c b/kernel/printk.c index 6c3d5bf14da2..a41106e19077 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -361,6 +361,7 @@ static void log_store(int facility, int level, struct devkmsg_user { u64 seq; u32 idx; + enum log_flags prev; struct mutex lock; char buf[8192]; }; @@ -426,6 +427,7 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, struct log *msg; u64 ts_usec; size_t i; + char cont = '-'; size_t len; ssize_t ret; @@ -463,8 +465,25 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, msg = log_from_idx(user->idx); ts_usec = msg->ts_nsec; do_div(ts_usec, 1000); - len = sprintf(user->buf, "%u,%llu,%llu;", - (msg->facility << 3) | msg->level, user->seq, ts_usec); + + /* + * If we couldn't merge continuation line fragments during the print, + * export the stored flags to allow an optional external merge of the + * records. Merging the records isn't always neccessarily correct, like + * when we hit a race during printing. In most cases though, it produces + * better readable output. 'c' in the record flags mark the first + * fragment of a line, '+' the following. + */ + if (msg->flags & LOG_CONT && !(user->prev & LOG_CONT)) + cont = 'c'; + else if ((msg->flags & LOG_CONT) || + ((user->prev & LOG_CONT) && !(msg->flags & LOG_PREFIX))) + cont = '+'; + + len = sprintf(user->buf, "%u,%llu,%llu,%c;", + (msg->facility << 3) | msg->level, + user->seq, ts_usec, cont); + user->prev = msg->flags; /* escape non-printable characters */ for (i = 0; i < msg->text_len; i++) { -- cgit v1.2.3 From eab072609e11a357181806ab5a5c309ef6eb76f5 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Mon, 16 Jul 2012 18:35:30 -0700 Subject: kmsg - do not flush partial lines when the console is busy Fragments of continuation lines are flushed to the console immediately. In case the console is locked, the fragment must be queued up in the cont buffer. If the the console is busy and the continuation line is complete, but no part of it was written to the console up to this point, we can just store the entire line as a regular record and free the buffer earlier. If the console is busy and earlier messages are already queued up, we should not flush the fragments of continuation lines, but store them after the queued up messages, to ensure the proper ordering. This keeps the console output better readable in case printk()s race against each other, or we receive over-long continuation lines we need to flush. Signed-off-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- kernel/printk.c | 93 +++++++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 68 insertions(+), 25 deletions(-) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index a41106e19077..4da2377131b0 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -231,6 +231,11 @@ static u32 log_first_idx; static u64 log_next_seq; static u32 log_next_idx; +/* the next printk record to write to the console */ +static u64 console_seq; +static u32 console_idx; +static enum log_flags console_prev; + /* the next printk record to read after the last 'clear' command */ static u64 clear_seq; static u32 clear_idx; @@ -1386,6 +1391,7 @@ static struct cont { u64 ts_nsec; /* time of first print */ u8 level; /* log level of first message */ u8 facility; /* log level of first message */ + enum log_flags flags; /* prefix, newline flags */ bool flushed:1; /* buffer sealed and committed */ } cont; @@ -1396,10 +1402,25 @@ static void cont_flush(enum log_flags flags) if (cont.len == 0) return; - log_store(cont.facility, cont.level, LOG_NOCONS | flags, - cont.ts_nsec, NULL, 0, cont.buf, cont.len); - - cont.flushed = true; + if (cont.cons) { + /* + * If a fragment of this line was directly flushed to the + * console; wait for the console to pick up the rest of the + * line. LOG_NOCONS suppresses a duplicated output. + */ + log_store(cont.facility, cont.level, flags | LOG_NOCONS, + cont.ts_nsec, NULL, 0, cont.buf, cont.len); + cont.flags = flags; + cont.flushed = true; + } else { + /* + * If no fragment of this line ever reached the console, + * just submit it to the store and free the buffer. + */ + log_store(cont.facility, cont.level, flags, 0, + NULL, 0, cont.buf, cont.len); + cont.len = 0; + } } static bool cont_add(int facility, int level, const char *text, size_t len) @@ -1418,12 +1439,17 @@ static bool cont_add(int facility, int level, const char *text, size_t len) cont.level = level; cont.owner = current; cont.ts_nsec = local_clock(); + cont.flags = 0; cont.cons = 0; cont.flushed = false; } memcpy(cont.buf + cont.len, text, len); cont.len += len; + + if (cont.len > (sizeof(cont.buf) * 80) / 100) + cont_flush(LOG_CONT); + return true; } @@ -1432,7 +1458,7 @@ static size_t cont_print_text(char *text, size_t size) size_t textlen = 0; size_t len; - if (cont.cons == 0) { + if (cont.cons == 0 && (console_prev & LOG_NEWLINE)) { textlen += print_time(cont.ts_nsec, text); size -= textlen; } @@ -1447,7 +1473,8 @@ static size_t cont_print_text(char *text, size_t size) } if (cont.flushed) { - text[textlen++] = '\n'; + if (cont.flags & LOG_NEWLINE) + text[textlen++] = '\n'; /* got everything, release buffer */ cont.len = 0; } @@ -1545,7 +1572,7 @@ asmlinkage int vprintk_emit(int facility, int level, * or another task also prints continuation lines. */ if (cont.len && (lflags & LOG_PREFIX || cont.owner != current)) - cont_flush(0); + cont_flush(LOG_NEWLINE); /* buffer line if possible, otherwise store it right away */ if (!cont_add(facility, level, text, text_len)) @@ -1563,7 +1590,7 @@ asmlinkage int vprintk_emit(int facility, int level, if (cont.len && cont.owner == current) { if (!(lflags & LOG_PREFIX)) stored = cont_add(facility, level, text, text_len); - cont_flush(0); + cont_flush(LOG_NEWLINE); } if (!stored) @@ -1661,10 +1688,13 @@ EXPORT_SYMBOL(printk); #define LOG_LINE_MAX 0 static u64 syslog_seq; static u32 syslog_idx; +static u64 console_seq; +static u32 console_idx; static enum log_flags syslog_prev; static u64 log_first_seq; static u32 log_first_idx; static u64 log_next_seq; +static enum log_flags console_prev; static struct cont { size_t len; size_t cons; @@ -1948,10 +1978,34 @@ void wake_up_klogd(void) this_cpu_or(printk_pending, PRINTK_PENDING_WAKEUP); } -/* the next printk record to write to the console */ -static u64 console_seq; -static u32 console_idx; -static enum log_flags console_prev; +static void console_cont_flush(char *text, size_t size) +{ + unsigned long flags; + size_t len; + + raw_spin_lock_irqsave(&logbuf_lock, flags); + + if (!cont.len) + goto out; + + /* + * We still queue earlier records, likely because the console was + * busy. The earlier ones need to be printed before this one, we + * did not flush any fragment so far, so just let it queue up. + */ + if (console_seq < log_next_seq && !cont.cons) + goto out; + + len = cont_print_text(text, size); + raw_spin_unlock(&logbuf_lock); + stop_critical_timings(); + call_console_drivers(cont.level, text, len); + start_critical_timings(); + local_irq_restore(flags); + return; +out: + raw_spin_unlock_irqrestore(&logbuf_lock, flags); +} /** * console_unlock - unlock the console system @@ -1983,19 +2037,7 @@ void console_unlock(void) console_may_schedule = 0; /* flush buffered message fragment immediately to console */ - raw_spin_lock_irqsave(&logbuf_lock, flags); - if (cont.len && (cont.cons < cont.len || cont.flushed)) { - size_t len; - - len = cont_print_text(text, sizeof(text)); - raw_spin_unlock(&logbuf_lock); - stop_critical_timings(); - call_console_drivers(cont.level, text, len); - start_critical_timings(); - local_irq_restore(flags); - } else - raw_spin_unlock_irqrestore(&logbuf_lock, flags); - + console_cont_flush(text, sizeof(text)); again: for (;;) { struct log *msg; @@ -2032,6 +2074,7 @@ skip: * will properly dump everything later. */ msg->flags &= ~LOG_NOCONS; + console_prev = msg->flags; goto skip; } -- cgit v1.2.3 From b2ad368bebc0f772613668e893fa176396e9094c Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Mon, 9 Jul 2012 17:10:39 -0700 Subject: tracing: Fix initialization failure path in tracing_set_tracer() If tracer->init() fails, current code will leave current_tracer pointing to an unusable tracer, which at best makes 'current_tracer' report inaccurate value. Fix the issue by pointing current_tracer to nop tracer, and only update current_tracer with the new one after all the initialization succeeds. Signed-off-by: Anton Vorontsov Acked-by: Steven Rostedt Signed-off-by: Greg Kroah-Hartman --- kernel/trace/trace.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 49249c28690d..44ee11e31b82 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -3172,10 +3172,10 @@ static int tracing_set_tracer(const char *buf) } destroy_trace_option_files(topts); - current_trace = t; + current_trace = &nop_trace; - topts = create_trace_option_files(current_trace); - if (current_trace->use_max_tr) { + topts = create_trace_option_files(t); + if (t->use_max_tr) { int cpu; /* we need to make per cpu buffer sizes equivalent */ for_each_tracing_cpu(cpu) { @@ -3195,6 +3195,7 @@ static int tracing_set_tracer(const char *buf) goto out; } + current_trace = t; trace_branch_enable(tr); out: mutex_unlock(&trace_types_lock); -- cgit v1.2.3 From 21f679404a0c28bd5b1b3aff2a7218bbff4cb43d Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Mon, 9 Jul 2012 17:10:42 -0700 Subject: tracing/function: Introduce persistent trace option This patch introduces 'func_ptrace' option, now available in /sys/kernel/debug/tracing/options when function tracer is selected. The patch also adds some tiny code that calls back to pstore to record the trace. The callback is no-op when PSTORE=n. Signed-off-by: Anton Vorontsov Acked-by: Steven Rostedt Signed-off-by: Greg Kroah-Hartman --- kernel/trace/trace_functions.c | 25 ++++++++++++++++++++----- 1 file changed, 20 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index c7b0c6a7db09..13770abd7a12 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include "trace.h" @@ -74,6 +75,14 @@ function_trace_call_preempt_only(unsigned long ip, unsigned long parent_ip) preempt_enable_notrace(); } +/* Our two options */ +enum { + TRACE_FUNC_OPT_STACK = 0x1, + TRACE_FUNC_OPT_PSTORE = 0x2, +}; + +static struct tracer_flags func_flags; + static void function_trace_call(unsigned long ip, unsigned long parent_ip) { @@ -97,6 +106,12 @@ function_trace_call(unsigned long ip, unsigned long parent_ip) disabled = atomic_inc_return(&data->disabled); if (likely(disabled == 1)) { + /* + * So far tracing doesn't support multiple buffers, so + * we make an explicit call for now. + */ + if (unlikely(func_flags.val & TRACE_FUNC_OPT_PSTORE)) + pstore_ftrace_call(ip, parent_ip); pc = preempt_count(); trace_function(tr, ip, parent_ip, flags, pc); } @@ -158,14 +173,12 @@ static struct ftrace_ops trace_stack_ops __read_mostly = .flags = FTRACE_OPS_FL_GLOBAL, }; -/* Our two options */ -enum { - TRACE_FUNC_OPT_STACK = 0x1, -}; - static struct tracer_opt func_opts[] = { #ifdef CONFIG_STACKTRACE { TRACER_OPT(func_stack_trace, TRACE_FUNC_OPT_STACK) }, +#endif +#ifdef CONFIG_PSTORE_FTRACE + { TRACER_OPT(func_pstore, TRACE_FUNC_OPT_PSTORE) }, #endif { } /* Always set a last empty entry */ }; @@ -217,6 +230,8 @@ static int func_set_flag(u32 old_flags, u32 bit, int set) register_ftrace_function(&trace_ops); } + return 0; + } else if (bit == TRACE_FUNC_OPT_PSTORE) { return 0; } -- cgit v1.2.3 From f555f1231a69846d57099760f9c361982600ffa2 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Mon, 9 Jul 2012 17:10:46 -0700 Subject: tracing/function: Convert func_set_flag() to a switch statement Since the function accepts just one bit, we can use the switch construction instead of if/else if/... Just a cosmetic change, there should be no functional changes. Suggested-by: Steven Rostedt Signed-off-by: Anton Vorontsov Acked-by: Steven Rostedt Signed-off-by: Greg Kroah-Hartman --- kernel/trace/trace_functions.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 13770abd7a12..a426f410c060 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -217,10 +217,11 @@ static void tracing_stop_function_trace(void) static int func_set_flag(u32 old_flags, u32 bit, int set) { - if (bit == TRACE_FUNC_OPT_STACK) { + switch (bit) { + case TRACE_FUNC_OPT_STACK: /* do nothing if already set */ if (!!set == !!(func_flags.val & TRACE_FUNC_OPT_STACK)) - return 0; + break; if (set) { unregister_ftrace_function(&trace_ops); @@ -230,12 +231,14 @@ static int func_set_flag(u32 old_flags, u32 bit, int set) register_ftrace_function(&trace_ops); } - return 0; - } else if (bit == TRACE_FUNC_OPT_PSTORE) { - return 0; + break; + case TRACE_FUNC_OPT_PSTORE: + break; + default: + return -EINVAL; } - return -EINVAL; + return 0; } static struct tracer function_trace __read_mostly = -- cgit v1.2.3 From 6575820221f7a4dd6eadecf7bf83cdd154335eda Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 17 Jul 2012 12:39:26 -0700 Subject: workqueue: perform cpu down operations from low priority cpu_notifier() Currently, all workqueue cpu hotplug operations run off CPU_PRI_WORKQUEUE which is higher than normal notifiers. This is to ensure that workqueue is up and running while bringing up a CPU before other notifiers try to use workqueue on the CPU. Per-cpu workqueues are supposed to remain working and bound to the CPU for normal CPU_DOWN_PREPARE notifiers. This holds mostly true even with workqueue offlining running with higher priority because workqueue CPU_DOWN_PREPARE only creates a bound trustee thread which runs the per-cpu workqueue without concurrency management without explicitly detaching the existing workers. However, if the trustee needs to create new workers, it creates unbound workers which may wander off to other CPUs while CPU_DOWN_PREPARE notifiers are in progress. Furthermore, if the CPU down is cancelled, the per-CPU workqueue may end up with workers which aren't bound to the CPU. While reliably reproducible with a convoluted artificial test-case involving scheduling and flushing CPU burning work items from CPU down notifiers, this isn't very likely to happen in the wild, and, even when it happens, the effects are likely to be hidden by the following successful CPU down. Fix it by using different priorities for up and down notifiers - high priority for up operations and low priority for down operations. Workqueue cpu hotplug operations will soon go through further cleanup. Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org Acked-by: "Rafael J. Wysocki" --- include/linux/cpu.h | 5 +++-- kernel/workqueue.c | 38 +++++++++++++++++++++++++++++++++++++- 2 files changed, 40 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/include/linux/cpu.h b/include/linux/cpu.h index 2e9b9ebbeb78..ce7a074f2519 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -73,8 +73,9 @@ enum { /* migration should happen before other stuff but after perf */ CPU_PRI_PERF = 20, CPU_PRI_MIGRATION = 10, - /* prepare workqueues for other notifiers */ - CPU_PRI_WORKQUEUE = 5, + /* bring up workqueues before normal notifiers and down after */ + CPU_PRI_WORKQUEUE_UP = 5, + CPU_PRI_WORKQUEUE_DOWN = -5, }; #define CPU_ONLINE 0x0002 /* CPU (unsigned)v is up */ diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 4fa9e3552f1e..f59b7fd26e26 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -3644,6 +3644,41 @@ err_destroy: return NOTIFY_BAD; } +/* + * Workqueues should be brought up before normal priority CPU notifiers. + * This will be registered high priority CPU notifier. + */ +static int __devinit workqueue_cpu_up_callback(struct notifier_block *nfb, + unsigned long action, + void *hcpu) +{ + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_UP_PREPARE: + case CPU_UP_CANCELED: + case CPU_DOWN_FAILED: + case CPU_ONLINE: + return workqueue_cpu_callback(nfb, action, hcpu); + } + return NOTIFY_OK; +} + +/* + * Workqueues should be brought down after normal priority CPU notifiers. + * This will be registered as low priority CPU notifier. + */ +static int __devinit workqueue_cpu_down_callback(struct notifier_block *nfb, + unsigned long action, + void *hcpu) +{ + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_DOWN_PREPARE: + case CPU_DYING: + case CPU_POST_DEAD: + return workqueue_cpu_callback(nfb, action, hcpu); + } + return NOTIFY_OK; +} + #ifdef CONFIG_SMP struct work_for_cpu { @@ -3839,7 +3874,8 @@ static int __init init_workqueues(void) unsigned int cpu; int i; - cpu_notifier(workqueue_cpu_callback, CPU_PRI_WORKQUEUE); + cpu_notifier(workqueue_cpu_up_callback, CPU_PRI_WORKQUEUE_UP); + cpu_notifier(workqueue_cpu_down_callback, CPU_PRI_WORKQUEUE_DOWN); /* initialize gcwqs */ for_each_gcwq_cpu(cpu) { -- cgit v1.2.3 From f2d5a0ee06c1813f985bb9386f3ccc0d0315720f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 17 Jul 2012 12:39:26 -0700 Subject: workqueue: drop CPU_DYING notifier operation Workqueue used CPU_DYING notification to mark GCWQ_DISASSOCIATED. This was necessary because workqueue's CPU_DOWN_PREPARE happened before other DOWN_PREPARE notifiers and workqueue needed to stay associated across the rest of DOWN_PREPARE. After the previous patch, workqueue's DOWN_PREPARE happens after others and can set GCWQ_DISASSOCIATED directly. Drop CPU_DYING and let the trustee set GCWQ_DISASSOCIATED after disabling concurrency management. Signed-off-by: Tejun Heo Acked-by: "Rafael J. Wysocki" --- kernel/workqueue.c | 29 +++++++++++++---------------- 1 file changed, 13 insertions(+), 16 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index f59b7fd26e26..1405fb98c0b1 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1250,11 +1250,11 @@ static void worker_leave_idle(struct worker *worker) * verbatim as it's best effort and blocking and gcwq may be * [dis]associated in the meantime. * - * This function tries set_cpus_allowed() and locks gcwq and verifies - * the binding against GCWQ_DISASSOCIATED which is set during - * CPU_DYING and cleared during CPU_ONLINE, so if the worker enters - * idle state or fetches works without dropping lock, it can guarantee - * the scheduling requirement described in the first paragraph. + * This function tries set_cpus_allowed() and locks gcwq and verifies the + * binding against %GCWQ_DISASSOCIATED which is set during + * %CPU_DOWN_PREPARE and cleared during %CPU_ONLINE, so if the worker + * enters idle state or fetches works without dropping lock, it can + * guarantee the scheduling requirement described in the first paragraph. * * CONTEXT: * Might sleep. Called without any lock but returns with gcwq->lock @@ -3349,6 +3349,12 @@ static int __cpuinit trustee_thread(void *__gcwq) rc = trustee_wait_event(!gcwq_is_managing_workers(gcwq)); BUG_ON(rc < 0); + /* + * We've claimed all manager positions. Make all workers unbound + * and set DISASSOCIATED. Before this, all workers except for the + * ones which are still executing works from before the last CPU + * down must be on the cpu. After this, they may become diasporas. + */ for_each_worker_pool(pool, gcwq) { pool->flags |= POOL_MANAGING_WORKERS; @@ -3359,6 +3365,8 @@ static int __cpuinit trustee_thread(void *__gcwq) for_each_busy_worker(worker, i, pos, gcwq) worker->flags |= WORKER_ROGUE; + gcwq->flags |= GCWQ_DISASSOCIATED; + /* * Call schedule() so that we cross rq->lock and thus can * guarantee sched callbacks see the rogue flag. This is @@ -3582,16 +3590,6 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, } break; - case CPU_DYING: - /* - * Before this, the trustee and all workers except for - * the ones which are still executing works from - * before the last CPU down must be on the cpu. After - * this, they'll all be diasporas. - */ - gcwq->flags |= GCWQ_DISASSOCIATED; - break; - case CPU_POST_DEAD: gcwq->trustee_state = TRUSTEE_BUTCHER; /* fall through */ @@ -3672,7 +3670,6 @@ static int __devinit workqueue_cpu_down_callback(struct notifier_block *nfb, { switch (action & ~CPU_TASKS_FROZEN) { case CPU_DOWN_PREPARE: - case CPU_DYING: case CPU_POST_DEAD: return workqueue_cpu_callback(nfb, action, hcpu); } -- cgit v1.2.3 From 403c821d452c03be4ced571ac91339a9d3631b17 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 17 Jul 2012 12:39:27 -0700 Subject: workqueue: ROGUE workers are UNBOUND workers Currently, WORKER_UNBOUND is used to mark workers for the unbound global_cwq and WORKER_ROGUE is used to mark workers for disassociated per-cpu global_cwqs. Both are used to make the marked worker skip concurrency management and the only place they make any difference is in worker_enter_idle() where WORKER_ROGUE is used to skip scheduling idle timer, which can easily be replaced with trustee state testing. This patch replaces WORKER_ROGUE with WORKER_UNBOUND and drops WORKER_ROGUE. This is to prepare for removing trustee and handling disassociated global_cwqs as unbound. Signed-off-by: Tejun Heo Acked-by: "Rafael J. Wysocki" --- kernel/workqueue.c | 46 +++++++++++++++++++++------------------------- 1 file changed, 21 insertions(+), 25 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 1405fb98c0b1..af512927c607 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -58,13 +58,12 @@ enum { WORKER_DIE = 1 << 1, /* die die die */ WORKER_IDLE = 1 << 2, /* is idle */ WORKER_PREP = 1 << 3, /* preparing to run works */ - WORKER_ROGUE = 1 << 4, /* not bound to any cpu */ WORKER_REBIND = 1 << 5, /* mom is home, come back */ WORKER_CPU_INTENSIVE = 1 << 6, /* cpu intensive */ WORKER_UNBOUND = 1 << 7, /* worker is unbound */ - WORKER_NOT_RUNNING = WORKER_PREP | WORKER_ROGUE | WORKER_REBIND | - WORKER_CPU_INTENSIVE | WORKER_UNBOUND, + WORKER_NOT_RUNNING = WORKER_PREP | WORKER_REBIND | WORKER_UNBOUND | + WORKER_CPU_INTENSIVE, /* gcwq->trustee_state */ TRUSTEE_START = 0, /* start */ @@ -1198,7 +1197,7 @@ static void worker_enter_idle(struct worker *worker) /* idle_list is LIFO */ list_add(&worker->entry, &pool->idle_list); - if (likely(!(worker->flags & WORKER_ROGUE))) { + if (likely(gcwq->trustee_state != TRUSTEE_DONE)) { if (too_many_workers(pool) && !timer_pending(&pool->idle_timer)) mod_timer(&pool->idle_timer, jiffies + IDLE_WORKER_TIMEOUT); @@ -1207,7 +1206,7 @@ static void worker_enter_idle(struct worker *worker) /* * Sanity check nr_running. Because trustee releases gcwq->lock - * between setting %WORKER_ROGUE and zapping nr_running, the + * between setting %WORKER_UNBOUND and zapping nr_running, the * warning may trigger spuriously. Check iff trustee is idle. */ WARN_ON_ONCE(gcwq->trustee_state == TRUSTEE_DONE && @@ -1301,10 +1300,10 @@ __acquires(&gcwq->lock) } /* - * Function for worker->rebind_work used to rebind rogue busy workers - * to the associated cpu which is coming back online. This is - * scheduled by cpu up but can race with other cpu hotplug operations - * and may be executed twice without intervening cpu down. + * Function for worker->rebind_work used to rebind unbound busy workers to + * the associated cpu which is coming back online. This is scheduled by + * cpu up but can race with other cpu hotplug operations and may be + * executed twice without intervening cpu down. */ static void worker_rebind_fn(struct work_struct *work) { @@ -1385,9 +1384,8 @@ static struct worker *create_worker(struct worker_pool *pool, bool bind) set_user_nice(worker->task, HIGHPRI_NICE_LEVEL); /* - * A rogue worker will become a regular one if CPU comes - * online later on. Make sure every worker has - * PF_THREAD_BOUND set. + * An unbound worker will become a regular one if CPU comes online + * later on. Make sure every worker has PF_THREAD_BOUND set. */ if (bind && !on_unbound_cpu) kthread_bind(worker->task, gcwq->cpu); @@ -3215,11 +3213,10 @@ EXPORT_SYMBOL_GPL(work_busy); * gcwqs serve mix of short, long and very long running works making * blocked draining impractical. * - * This is solved by allowing a gcwq to be detached from CPU, running - * it with unbound (rogue) workers and allowing it to be reattached - * later if the cpu comes back online. A separate thread is created - * to govern a gcwq in such state and is called the trustee of the - * gcwq. + * This is solved by allowing a gcwq to be detached from CPU, running it + * with unbound workers and allowing it to be reattached later if the cpu + * comes back online. A separate thread is created to govern a gcwq in + * such state and is called the trustee of the gcwq. * * Trustee states and their descriptions. * @@ -3359,19 +3356,18 @@ static int __cpuinit trustee_thread(void *__gcwq) pool->flags |= POOL_MANAGING_WORKERS; list_for_each_entry(worker, &pool->idle_list, entry) - worker->flags |= WORKER_ROGUE; + worker->flags |= WORKER_UNBOUND; } for_each_busy_worker(worker, i, pos, gcwq) - worker->flags |= WORKER_ROGUE; + worker->flags |= WORKER_UNBOUND; gcwq->flags |= GCWQ_DISASSOCIATED; /* - * Call schedule() so that we cross rq->lock and thus can - * guarantee sched callbacks see the rogue flag. This is - * necessary as scheduler callbacks may be invoked from other - * cpus. + * Call schedule() so that we cross rq->lock and thus can guarantee + * sched callbacks see the unbound flag. This is necessary as + * scheduler callbacks may be invoked from other cpus. */ spin_unlock_irq(&gcwq->lock); schedule(); @@ -3439,7 +3435,7 @@ static int __cpuinit trustee_thread(void *__gcwq) worker = create_worker(pool, false); spin_lock_irq(&gcwq->lock); if (worker) { - worker->flags |= WORKER_ROGUE; + worker->flags |= WORKER_UNBOUND; start_worker(worker); } } @@ -3488,7 +3484,7 @@ static int __cpuinit trustee_thread(void *__gcwq) * rebinding is scheduled. */ worker->flags |= WORKER_REBIND; - worker->flags &= ~WORKER_ROGUE; + worker->flags &= ~WORKER_UNBOUND; /* queue rebind_work, wq doesn't matter, use the default one */ if (test_and_set_bit(WORK_STRUCT_PENDING_BIT, -- cgit v1.2.3 From 6037315269d62bf967286ae2670fdd6b6acedab9 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 17 Jul 2012 12:39:27 -0700 Subject: workqueue: use mutex for global_cwq manager exclusion POOL_MANAGING_WORKERS is used to ensure that at most one worker takes the manager role at any given time on a given global_cwq. Trustee later hitched on it to assume manager adding blocking wait for the bit. As trustee already needed a custom wait mechanism, waiting for MANAGING_WORKERS was rolled into the same mechanism. Trustee is scheduled to be removed. This patch separates out MANAGING_WORKERS wait into per-pool mutex. Workers use mutex_trylock() to test for manager role and trustee uses mutex_lock() to claim manager roles. gcwq_claim/release_management() helpers are added to grab and release manager roles of all pools on a global_cwq. gcwq_claim_management() always grabs pool manager mutexes in ascending pool index order and uses pool index as lockdep subclass. Signed-off-by: Tejun Heo Acked-by: "Rafael J. Wysocki" --- kernel/workqueue.c | 65 ++++++++++++++++++++++-------------------------------- 1 file changed, 26 insertions(+), 39 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index af512927c607..f7a00697d150 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -51,7 +51,6 @@ enum { /* pool flags */ POOL_MANAGE_WORKERS = 1 << 0, /* need to manage workers */ - POOL_MANAGING_WORKERS = 1 << 1, /* managing workers */ /* worker flags */ WORKER_STARTED = 1 << 0, /* started */ @@ -155,6 +154,7 @@ struct worker_pool { struct timer_list idle_timer; /* L: worker idle timeout */ struct timer_list mayday_timer; /* L: SOS timer for workers */ + struct mutex manager_mutex; /* mutex manager should hold */ struct ida worker_ida; /* L: for worker IDs */ struct worker *first_idle; /* L: first idle worker */ }; @@ -644,7 +644,7 @@ static bool need_to_manage_workers(struct worker_pool *pool) /* Do we have too many workers and should some go away? */ static bool too_many_workers(struct worker_pool *pool) { - bool managing = pool->flags & POOL_MANAGING_WORKERS; + bool managing = mutex_is_locked(&pool->manager_mutex); int nr_idle = pool->nr_idle + managing; /* manager is considered idle */ int nr_busy = pool->nr_workers - nr_idle; @@ -1655,14 +1655,12 @@ static bool maybe_destroy_workers(struct worker_pool *pool) static bool manage_workers(struct worker *worker) { struct worker_pool *pool = worker->pool; - struct global_cwq *gcwq = pool->gcwq; bool ret = false; - if (pool->flags & POOL_MANAGING_WORKERS) + if (!mutex_trylock(&pool->manager_mutex)) return ret; pool->flags &= ~POOL_MANAGE_WORKERS; - pool->flags |= POOL_MANAGING_WORKERS; /* * Destroy and then create so that may_start_working() is true @@ -1671,15 +1669,7 @@ static bool manage_workers(struct worker *worker) ret |= maybe_destroy_workers(pool); ret |= maybe_create_worker(pool); - pool->flags &= ~POOL_MANAGING_WORKERS; - - /* - * The trustee might be waiting to take over the manager - * position, tell it we're done. - */ - if (unlikely(gcwq->trustee)) - wake_up_all(&gcwq->trustee_wait); - + mutex_unlock(&pool->manager_mutex); return ret; } @@ -3255,6 +3245,24 @@ EXPORT_SYMBOL_GPL(work_busy); * ----------------> RELEASE -------------- */ +/* claim manager positions of all pools */ +static void gcwq_claim_management(struct global_cwq *gcwq) +{ + struct worker_pool *pool; + + for_each_worker_pool(pool, gcwq) + mutex_lock_nested(&pool->manager_mutex, pool - gcwq->pools); +} + +/* release manager positions */ +static void gcwq_release_management(struct global_cwq *gcwq) +{ + struct worker_pool *pool; + + for_each_worker_pool(pool, gcwq) + mutex_unlock(&pool->manager_mutex); +} + /** * trustee_wait_event_timeout - timed event wait for trustee * @cond: condition to wait for @@ -3304,16 +3312,6 @@ EXPORT_SYMBOL_GPL(work_busy); __ret1 < 0 ? -1 : 0; \ }) -static bool gcwq_is_managing_workers(struct global_cwq *gcwq) -{ - struct worker_pool *pool; - - for_each_worker_pool(pool, gcwq) - if (pool->flags & POOL_MANAGING_WORKERS) - return true; - return false; -} - static bool gcwq_has_idle_workers(struct global_cwq *gcwq) { struct worker_pool *pool; @@ -3336,15 +3334,8 @@ static int __cpuinit trustee_thread(void *__gcwq) BUG_ON(gcwq->cpu != smp_processor_id()); + gcwq_claim_management(gcwq); spin_lock_irq(&gcwq->lock); - /* - * Claim the manager position and make all workers rogue. - * Trustee must be bound to the target cpu and can't be - * cancelled. - */ - BUG_ON(gcwq->cpu != smp_processor_id()); - rc = trustee_wait_event(!gcwq_is_managing_workers(gcwq)); - BUG_ON(rc < 0); /* * We've claimed all manager positions. Make all workers unbound @@ -3352,12 +3343,9 @@ static int __cpuinit trustee_thread(void *__gcwq) * ones which are still executing works from before the last CPU * down must be on the cpu. After this, they may become diasporas. */ - for_each_worker_pool(pool, gcwq) { - pool->flags |= POOL_MANAGING_WORKERS; - + for_each_worker_pool(pool, gcwq) list_for_each_entry(worker, &pool->idle_list, entry) worker->flags |= WORKER_UNBOUND; - } for_each_busy_worker(worker, i, pos, gcwq) worker->flags |= WORKER_UNBOUND; @@ -3497,9 +3485,7 @@ static int __cpuinit trustee_thread(void *__gcwq) work_color_to_flags(WORK_NO_COLOR)); } - /* relinquish manager role */ - for_each_worker_pool(pool, gcwq) - pool->flags &= ~POOL_MANAGING_WORKERS; + gcwq_release_management(gcwq); /* notify completion */ gcwq->trustee = NULL; @@ -3894,6 +3880,7 @@ static int __init init_workqueues(void) setup_timer(&pool->mayday_timer, gcwq_mayday_timeout, (unsigned long)pool); + mutex_init(&pool->manager_mutex); ida_init(&pool->worker_ida); } -- cgit v1.2.3 From bc2ae0f5bb2f39e6db06a62f9d353e4601a332a1 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 17 Jul 2012 12:39:27 -0700 Subject: workqueue: drop @bind from create_worker() Currently, create_worker()'s callers are responsible for deciding whether the newly created worker should be bound to the associated CPU and create_worker() sets WORKER_UNBOUND only for the workers for the unbound global_cwq. Creation during normal operation is always via maybe_create_worker() and @bind is true. For workers created during hotplug, @bind is false. Normal operation path is planned to be used even while the CPU is going through hotplug operations or offline and this static decision won't work. Drop @bind from create_worker() and decide whether to bind by looking at GCWQ_DISASSOCIATED. create_worker() will also set WORKER_UNBOUND autmatically if disassociated. To avoid flipping GCWQ_DISASSOCIATED while create_worker() is in progress, the flag is now allowed to be changed only while holding all manager_mutexes on the global_cwq. This requires that GCWQ_DISASSOCIATED is not cleared behind trustee's back. CPU_ONLINE no longer clears DISASSOCIATED before flushing trustee, which clears DISASSOCIATED before rebinding remaining workers if asked to release. For cases where trustee isn't around, CPU_ONLINE clears DISASSOCIATED after flushing trustee. Also, now, first_idle has UNBOUND set on creation which is explicitly cleared by CPU_ONLINE while binding it. These convolutions will soon be removed by further simplification of CPU hotplug path. Signed-off-by: Tejun Heo Acked-by: "Rafael J. Wysocki" --- kernel/workqueue.c | 64 ++++++++++++++++++++++++++++++++++++++---------------- 1 file changed, 45 insertions(+), 19 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index f7a00697d150..e1d05e51a80a 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -45,7 +45,22 @@ #include "workqueue_sched.h" enum { - /* global_cwq flags */ + /* + * global_cwq flags + * + * A bound gcwq is either associated or disassociated with its CPU. + * While associated (!DISASSOCIATED), all workers are bound to the + * CPU and none has %WORKER_UNBOUND set and concurrency management + * is in effect. + * + * While DISASSOCIATED, the cpu may be offline and all workers have + * %WORKER_UNBOUND set and concurrency management disabled, and may + * be executing on any CPU. The gcwq behaves as an unbound one. + * + * Note that DISASSOCIATED can be flipped only while holding + * managership of all pools on the gcwq to avoid changing binding + * state while create_worker() is in progress. + */ GCWQ_DISASSOCIATED = 1 << 0, /* cpu can't serve workers */ GCWQ_FREEZING = 1 << 1, /* freeze in progress */ @@ -1334,7 +1349,6 @@ static struct worker *alloc_worker(void) /** * create_worker - create a new workqueue worker * @pool: pool the new worker will belong to - * @bind: whether to set affinity to @cpu or not * * Create a new worker which is bound to @pool. The returned worker * can be started by calling start_worker() or destroyed using @@ -1346,10 +1360,9 @@ static struct worker *alloc_worker(void) * RETURNS: * Pointer to the newly created worker. */ -static struct worker *create_worker(struct worker_pool *pool, bool bind) +static struct worker *create_worker(struct worker_pool *pool) { struct global_cwq *gcwq = pool->gcwq; - bool on_unbound_cpu = gcwq->cpu == WORK_CPU_UNBOUND; const char *pri = worker_pool_pri(pool) ? "H" : ""; struct worker *worker = NULL; int id = -1; @@ -1370,7 +1383,7 @@ static struct worker *create_worker(struct worker_pool *pool, bool bind) worker->pool = pool; worker->id = id; - if (!on_unbound_cpu) + if (gcwq->cpu != WORK_CPU_UNBOUND) worker->task = kthread_create_on_node(worker_thread, worker, cpu_to_node(gcwq->cpu), "kworker/%u:%d%s", gcwq->cpu, id, pri); @@ -1384,15 +1397,19 @@ static struct worker *create_worker(struct worker_pool *pool, bool bind) set_user_nice(worker->task, HIGHPRI_NICE_LEVEL); /* - * An unbound worker will become a regular one if CPU comes online - * later on. Make sure every worker has PF_THREAD_BOUND set. + * Determine CPU binding of the new worker depending on + * %GCWQ_DISASSOCIATED. The caller is responsible for ensuring the + * flag remains stable across this function. See the comments + * above the flag definition for details. + * + * As an unbound worker may later become a regular one if CPU comes + * online, make sure every worker has %PF_THREAD_BOUND set. */ - if (bind && !on_unbound_cpu) + if (!(gcwq->flags & GCWQ_DISASSOCIATED)) { kthread_bind(worker->task, gcwq->cpu); - else { + } else { worker->task->flags |= PF_THREAD_BOUND; - if (on_unbound_cpu) - worker->flags |= WORKER_UNBOUND; + worker->flags |= WORKER_UNBOUND; } return worker; @@ -1568,7 +1585,7 @@ restart: while (true) { struct worker *worker; - worker = create_worker(pool, true); + worker = create_worker(pool); if (worker) { del_timer_sync(&pool->mayday_timer); spin_lock_irq(&gcwq->lock); @@ -3420,12 +3437,10 @@ static int __cpuinit trustee_thread(void *__gcwq) if (need_to_create_worker(pool)) { spin_unlock_irq(&gcwq->lock); - worker = create_worker(pool, false); + worker = create_worker(pool); spin_lock_irq(&gcwq->lock); - if (worker) { - worker->flags |= WORKER_UNBOUND; + if (worker) start_worker(worker); - } } } @@ -3463,6 +3478,10 @@ static int __cpuinit trustee_thread(void *__gcwq) for_each_worker_pool(pool, gcwq) WARN_ON(!list_empty(&pool->idle_list)); + /* if we're reassociating, clear DISASSOCIATED */ + if (gcwq->trustee_state == TRUSTEE_RELEASE) + gcwq->flags &= ~GCWQ_DISASSOCIATED; + for_each_busy_worker(worker, i, pos, gcwq) { struct work_struct *rebind_work = &worker->rebind_work; @@ -3546,7 +3565,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, i = 0; for_each_worker_pool(pool, gcwq) { BUG_ON(pool->first_idle); - new_workers[i] = create_worker(pool, false); + new_workers[i] = create_worker(pool); if (!new_workers[i++]) goto err_destroy; } @@ -3584,13 +3603,19 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, case CPU_DOWN_FAILED: case CPU_ONLINE: - gcwq->flags &= ~GCWQ_DISASSOCIATED; if (gcwq->trustee_state != TRUSTEE_DONE) { gcwq->trustee_state = TRUSTEE_RELEASE; wake_up_process(gcwq->trustee); wait_trustee_state(gcwq, TRUSTEE_DONE); } + /* + * Either DISASSOCIATED is already cleared or no worker is + * left on the gcwq. Safe to clear DISASSOCIATED without + * claiming managers. + */ + gcwq->flags &= ~GCWQ_DISASSOCIATED; + /* * Trustee is done and there might be no worker left. * Put the first_idle in and request a real manager to @@ -3601,6 +3626,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, kthread_bind(pool->first_idle->task, cpu); spin_lock_irq(&gcwq->lock); pool->flags |= POOL_MANAGE_WORKERS; + pool->first_idle->flags &= ~WORKER_UNBOUND; start_worker(pool->first_idle); pool->first_idle = NULL; } @@ -3899,7 +3925,7 @@ static int __init init_workqueues(void) for_each_worker_pool(pool, gcwq) { struct worker *worker; - worker = create_worker(pool, true); + worker = create_worker(pool); BUG_ON(!worker); spin_lock_irq(&gcwq->lock); start_worker(worker); -- cgit v1.2.3 From 25511a477657884d2164f338341fa89652610507 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 17 Jul 2012 12:39:27 -0700 Subject: workqueue: reimplement CPU online rebinding to handle idle workers Currently, if there are left workers when a CPU is being brough back online, the trustee kills all idle workers and scheduled rebind_work so that they re-bind to the CPU after the currently executing work is finished. This works for busy workers because concurrency management doesn't try to wake up them from scheduler callbacks, which require the target task to be on the local run queue. The busy worker bumps concurrency counter appropriately as it clears WORKER_UNBOUND from the rebind work item and it's bound to the CPU before returning to the idle state. To reduce CPU on/offlining overhead (as many embedded systems use it for powersaving) and simplify the code path, workqueue is planned to be modified to retain idle workers across CPU on/offlining. This patch reimplements CPU online rebinding such that it can also handle idle workers. As noted earlier, due to the local wakeup requirement, rebinding idle workers is tricky. All idle workers must be re-bound before scheduler callbacks are enabled. This is achieved by interlocking idle re-binding. Idle workers are requested to re-bind and then hold until all idle re-binding is complete so that no bound worker starts executing work item. Only after all idle workers are re-bound and parked, CPU_ONLINE proceeds to release them and queue rebind work item to busy workers thus guaranteeing scheduler callbacks aren't invoked until all idle workers are ready. worker_rebind_fn() is renamed to busy_worker_rebind_fn() and idle_worker_rebind() for idle workers is added. Rebinding logic is moved to rebind_workers() and now called from CPU_ONLINE after flushing trustee. While at it, add CPU sanity check in worker_thread(). Note that now a worker may become idle or the manager between trustee release and rebinding during CPU_ONLINE. As the previous patch updated create_worker() so that it can be used by regular manager while unbound and this patch implements idle re-binding, this is safe. This prepares for removal of trustee and keeping idle workers across CPU hotplugs. Signed-off-by: Tejun Heo Acked-by: "Rafael J. Wysocki" --- kernel/workqueue.c | 215 +++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 166 insertions(+), 49 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index e1d05e51a80a..6927fecae412 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -133,6 +133,7 @@ enum { struct global_cwq; struct worker_pool; +struct idle_rebind; /* * The poor guys doing the actual heavy lifting. All on-duty workers @@ -154,7 +155,10 @@ struct worker { unsigned long last_active; /* L: last active timestamp */ unsigned int flags; /* X: flags */ int id; /* I: worker id */ - struct work_struct rebind_work; /* L: rebind worker to cpu */ + + /* for rebinding worker to CPU */ + struct idle_rebind *idle_rebind; /* L: for idle worker */ + struct work_struct rebind_work; /* L: for busy worker */ }; struct worker_pool { @@ -190,6 +194,8 @@ struct global_cwq { struct worker_pool pools[2]; /* normal and highpri pools */ + wait_queue_head_t rebind_hold; /* rebind hold wait */ + struct task_struct *trustee; /* L: for gcwq shutdown */ unsigned int trustee_state; /* L: trustee state */ wait_queue_head_t trustee_wait; /* trustee wait */ @@ -1314,13 +1320,37 @@ __acquires(&gcwq->lock) } } +struct idle_rebind { + int cnt; /* # workers to be rebound */ + struct completion done; /* all workers rebound */ +}; + +/* + * Rebind an idle @worker to its CPU. During CPU onlining, this has to + * happen synchronously for idle workers. worker_thread() will test + * %WORKER_REBIND before leaving idle and call this function. + */ +static void idle_worker_rebind(struct worker *worker) +{ + struct global_cwq *gcwq = worker->pool->gcwq; + + /* CPU must be online at this point */ + WARN_ON(!worker_maybe_bind_and_lock(worker)); + if (!--worker->idle_rebind->cnt) + complete(&worker->idle_rebind->done); + spin_unlock_irq(&worker->pool->gcwq->lock); + + /* we did our part, wait for rebind_workers() to finish up */ + wait_event(gcwq->rebind_hold, !(worker->flags & WORKER_REBIND)); +} + /* - * Function for worker->rebind_work used to rebind unbound busy workers to + * Function for @worker->rebind.work used to rebind unbound busy workers to * the associated cpu which is coming back online. This is scheduled by * cpu up but can race with other cpu hotplug operations and may be * executed twice without intervening cpu down. */ -static void worker_rebind_fn(struct work_struct *work) +static void busy_worker_rebind_fn(struct work_struct *work) { struct worker *worker = container_of(work, struct worker, rebind_work); struct global_cwq *gcwq = worker->pool->gcwq; @@ -1331,6 +1361,112 @@ static void worker_rebind_fn(struct work_struct *work) spin_unlock_irq(&gcwq->lock); } +/** + * rebind_workers - rebind all workers of a gcwq to the associated CPU + * @gcwq: gcwq of interest + * + * @gcwq->cpu is coming online. Rebind all workers to the CPU. Rebinding + * is different for idle and busy ones. + * + * The idle ones should be rebound synchronously and idle rebinding should + * be complete before any worker starts executing work items with + * concurrency management enabled; otherwise, scheduler may oops trying to + * wake up non-local idle worker from wq_worker_sleeping(). + * + * This is achieved by repeatedly requesting rebinding until all idle + * workers are known to have been rebound under @gcwq->lock and holding all + * idle workers from becoming busy until idle rebinding is complete. + * + * Once idle workers are rebound, busy workers can be rebound as they + * finish executing their current work items. Queueing the rebind work at + * the head of their scheduled lists is enough. Note that nr_running will + * be properbly bumped as busy workers rebind. + * + * On return, all workers are guaranteed to either be bound or have rebind + * work item scheduled. + */ +static void rebind_workers(struct global_cwq *gcwq) + __releases(&gcwq->lock) __acquires(&gcwq->lock) +{ + struct idle_rebind idle_rebind; + struct worker_pool *pool; + struct worker *worker; + struct hlist_node *pos; + int i; + + lockdep_assert_held(&gcwq->lock); + + for_each_worker_pool(pool, gcwq) + lockdep_assert_held(&pool->manager_mutex); + + /* + * Rebind idle workers. Interlocked both ways. We wait for + * workers to rebind via @idle_rebind.done. Workers will wait for + * us to finish up by watching %WORKER_REBIND. + */ + init_completion(&idle_rebind.done); +retry: + idle_rebind.cnt = 1; + INIT_COMPLETION(idle_rebind.done); + + /* set REBIND and kick idle ones, we'll wait for these later */ + for_each_worker_pool(pool, gcwq) { + list_for_each_entry(worker, &pool->idle_list, entry) { + if (worker->flags & WORKER_REBIND) + continue; + + /* morph UNBOUND to REBIND */ + worker->flags &= ~WORKER_UNBOUND; + worker->flags |= WORKER_REBIND; + + idle_rebind.cnt++; + worker->idle_rebind = &idle_rebind; + + /* worker_thread() will call idle_worker_rebind() */ + wake_up_process(worker->task); + } + } + + if (--idle_rebind.cnt) { + spin_unlock_irq(&gcwq->lock); + wait_for_completion(&idle_rebind.done); + spin_lock_irq(&gcwq->lock); + /* busy ones might have become idle while waiting, retry */ + goto retry; + } + + /* + * All idle workers are rebound and waiting for %WORKER_REBIND to + * be cleared inside idle_worker_rebind(). Clear and release. + * Clearing %WORKER_REBIND from this foreign context is safe + * because these workers are still guaranteed to be idle. + */ + for_each_worker_pool(pool, gcwq) + list_for_each_entry(worker, &pool->idle_list, entry) + worker->flags &= ~WORKER_REBIND; + + wake_up_all(&gcwq->rebind_hold); + + /* rebind busy workers */ + for_each_busy_worker(worker, i, pos, gcwq) { + struct work_struct *rebind_work = &worker->rebind_work; + + /* morph UNBOUND to REBIND */ + worker->flags &= ~WORKER_UNBOUND; + worker->flags |= WORKER_REBIND; + + if (test_and_set_bit(WORK_STRUCT_PENDING_BIT, + work_data_bits(rebind_work))) + continue; + + /* wq doesn't matter, use the default one */ + debug_work_activate(rebind_work); + insert_work(get_cwq(gcwq->cpu, system_wq), rebind_work, + worker->scheduled.next, + work_color_to_flags(WORK_NO_COLOR)); + } +} + static struct worker *alloc_worker(void) { struct worker *worker; @@ -1339,7 +1475,7 @@ static struct worker *alloc_worker(void) if (worker) { INIT_LIST_HEAD(&worker->entry); INIT_LIST_HEAD(&worker->scheduled); - INIT_WORK(&worker->rebind_work, worker_rebind_fn); + INIT_WORK(&worker->rebind_work, busy_worker_rebind_fn); /* on creation a worker is in !idle && prep state */ worker->flags = WORKER_PREP; } @@ -1829,6 +1965,9 @@ __acquires(&gcwq->lock) lockdep_copy_map(&lockdep_map, &work->lockdep_map); #endif + WARN_ON_ONCE(!(worker->flags & (WORKER_UNBOUND | WORKER_REBIND)) && + raw_smp_processor_id() != gcwq->cpu); + /* * A single work shouldn't be executed concurrently by * multiple workers on a single cpu. Check whether anyone is @@ -1946,11 +2085,20 @@ static int worker_thread(void *__worker) woke_up: spin_lock_irq(&gcwq->lock); - /* DIE can be set only while we're idle, checking here is enough */ - if (worker->flags & WORKER_DIE) { + /* + * DIE can be set only while idle and REBIND set while busy has + * @worker->rebind_work scheduled. Checking here is enough. + */ + if (unlikely(worker->flags & (WORKER_REBIND | WORKER_DIE))) { spin_unlock_irq(&gcwq->lock); - worker->task->flags &= ~PF_WQ_WORKER; - return 0; + + if (worker->flags & WORKER_DIE) { + worker->task->flags &= ~PF_WQ_WORKER; + return 0; + } + + idle_worker_rebind(worker); + goto woke_up; } worker_leave_idle(worker); @@ -3468,42 +3616,6 @@ static int __cpuinit trustee_thread(void *__gcwq) } } while (i && rc >= 0); - /* - * At this point, either draining has completed and no worker - * is left, or cpu down has been canceled or the cpu is being - * brought back up. There shouldn't be any idle one left. - * Tell the remaining busy ones to rebind once it finishes the - * currently scheduled works by scheduling the rebind_work. - */ - for_each_worker_pool(pool, gcwq) - WARN_ON(!list_empty(&pool->idle_list)); - - /* if we're reassociating, clear DISASSOCIATED */ - if (gcwq->trustee_state == TRUSTEE_RELEASE) - gcwq->flags &= ~GCWQ_DISASSOCIATED; - - for_each_busy_worker(worker, i, pos, gcwq) { - struct work_struct *rebind_work = &worker->rebind_work; - - /* - * Rebind_work may race with future cpu hotplug - * operations. Use a separate flag to mark that - * rebinding is scheduled. - */ - worker->flags |= WORKER_REBIND; - worker->flags &= ~WORKER_UNBOUND; - - /* queue rebind_work, wq doesn't matter, use the default one */ - if (test_and_set_bit(WORK_STRUCT_PENDING_BIT, - work_data_bits(rebind_work))) - continue; - - debug_work_activate(rebind_work); - insert_work(get_cwq(gcwq->cpu, system_wq), rebind_work, - worker->scheduled.next, - work_color_to_flags(WORK_NO_COLOR)); - } - gcwq_release_management(gcwq); /* notify completion */ @@ -3609,13 +3721,16 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, wait_trustee_state(gcwq, TRUSTEE_DONE); } - /* - * Either DISASSOCIATED is already cleared or no worker is - * left on the gcwq. Safe to clear DISASSOCIATED without - * claiming managers. - */ + spin_unlock_irq(&gcwq->lock); + gcwq_claim_management(gcwq); + spin_lock_irq(&gcwq->lock); + gcwq->flags &= ~GCWQ_DISASSOCIATED; + rebind_workers(gcwq); + + gcwq_release_management(gcwq); + /* * Trustee is done and there might be no worker left. * Put the first_idle in and request a real manager to @@ -3910,6 +4025,8 @@ static int __init init_workqueues(void) ida_init(&pool->worker_ida); } + init_waitqueue_head(&gcwq->rebind_hold); + gcwq->trustee_state = TRUSTEE_DONE; init_waitqueue_head(&gcwq->trustee_wait); } -- cgit v1.2.3 From 3ce63377305b694f53e7dd0c72907591c5344224 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 17 Jul 2012 12:39:27 -0700 Subject: workqueue: don't butcher idle workers on an offline CPU Currently, during CPU offlining, after all pending work items are drained, the trustee butchers all workers. Also, on CPU onlining failure, workqueue_cpu_callback() ensures that the first idle worker is destroyed. Combined, these guarantee that an offline CPU doesn't have any worker for it once all the lingering work items are finished. This guarantee isn't really necessary and makes CPU on/offlining more expensive than needs to be, especially for platforms which use CPU hotplug for powersaving. This patch lets offline CPUs removes idle worker butchering from the trustee and let a CPU which failed onlining keep the created first worker. The first worker is created if the CPU doesn't have any during CPU_DOWN_PREPARE and started right away. If onlining succeeds, the rebind_workers() call in CPU_ONLINE will rebind it like any other workers. If onlining fails, the worker is left alone till the next try. This makes CPU hotplugs cheaper by allowing global_cwqs to keep workers across them and simplifies code. Note that trustee doesn't re-arm idle timer when it's done and thus the disassociated global_cwq will keep all workers until it comes back online. This will be improved by further patches. Signed-off-by: Tejun Heo Acked-by: "Rafael J. Wysocki" --- kernel/workqueue.c | 94 ++++++++---------------------------------------------- 1 file changed, 14 insertions(+), 80 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 6927fecae412..acfabb22e2c4 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -175,7 +175,6 @@ struct worker_pool { struct mutex manager_mutex; /* mutex manager should hold */ struct ida worker_ida; /* L: for worker IDs */ - struct worker *first_idle; /* L: first idle worker */ }; /* @@ -3477,16 +3476,6 @@ static void gcwq_release_management(struct global_cwq *gcwq) __ret1 < 0 ? -1 : 0; \ }) -static bool gcwq_has_idle_workers(struct global_cwq *gcwq) -{ - struct worker_pool *pool; - - for_each_worker_pool(pool, gcwq) - if (!list_empty(&pool->idle_list)) - return true; - return false; -} - static int __cpuinit trustee_thread(void *__gcwq) { struct global_cwq *gcwq = __gcwq; @@ -3494,7 +3483,6 @@ static int __cpuinit trustee_thread(void *__gcwq) struct worker *worker; struct work_struct *work; struct hlist_node *pos; - long rc; int i; BUG_ON(gcwq->cpu != smp_processor_id()); @@ -3597,25 +3585,6 @@ static int __cpuinit trustee_thread(void *__gcwq) break; } - /* - * Either all works have been scheduled and cpu is down, or - * cpu down has already been canceled. Wait for and butcher - * all workers till we're canceled. - */ - do { - rc = trustee_wait_event(gcwq_has_idle_workers(gcwq)); - - i = 0; - for_each_worker_pool(pool, gcwq) { - while (!list_empty(&pool->idle_list)) { - worker = list_first_entry(&pool->idle_list, - struct worker, entry); - destroy_worker(worker); - } - i |= pool->nr_workers; - } - } while (i && rc >= 0); - gcwq_release_management(gcwq); /* notify completion */ @@ -3658,10 +3627,8 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, unsigned int cpu = (unsigned long)hcpu; struct global_cwq *gcwq = get_gcwq(cpu); struct task_struct *new_trustee = NULL; - struct worker *new_workers[NR_WORKER_POOLS] = { }; struct worker_pool *pool; unsigned long flags; - int i; action &= ~CPU_TASKS_FROZEN; @@ -3672,14 +3639,22 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, if (IS_ERR(new_trustee)) return notifier_from_errno(PTR_ERR(new_trustee)); kthread_bind(new_trustee, cpu); - /* fall through */ + break; + case CPU_UP_PREPARE: - i = 0; for_each_worker_pool(pool, gcwq) { - BUG_ON(pool->first_idle); - new_workers[i] = create_worker(pool); - if (!new_workers[i++]) - goto err_destroy; + struct worker *worker; + + if (pool->nr_workers) + continue; + + worker = create_worker(pool); + if (!worker) + return NOTIFY_BAD; + + spin_lock_irq(&gcwq->lock); + start_worker(worker); + spin_unlock_irq(&gcwq->lock); } } @@ -3694,23 +3669,10 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, gcwq->trustee_state = TRUSTEE_START; wake_up_process(gcwq->trustee); wait_trustee_state(gcwq, TRUSTEE_IN_CHARGE); - /* fall through */ - case CPU_UP_PREPARE: - i = 0; - for_each_worker_pool(pool, gcwq) { - BUG_ON(pool->first_idle); - pool->first_idle = new_workers[i++]; - } break; case CPU_POST_DEAD: gcwq->trustee_state = TRUSTEE_BUTCHER; - /* fall through */ - case CPU_UP_CANCELED: - for_each_worker_pool(pool, gcwq) { - destroy_worker(pool->first_idle); - pool->first_idle = NULL; - } break; case CPU_DOWN_FAILED: @@ -3730,39 +3692,12 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, rebind_workers(gcwq); gcwq_release_management(gcwq); - - /* - * Trustee is done and there might be no worker left. - * Put the first_idle in and request a real manager to - * take a look. - */ - for_each_worker_pool(pool, gcwq) { - spin_unlock_irq(&gcwq->lock); - kthread_bind(pool->first_idle->task, cpu); - spin_lock_irq(&gcwq->lock); - pool->flags |= POOL_MANAGE_WORKERS; - pool->first_idle->flags &= ~WORKER_UNBOUND; - start_worker(pool->first_idle); - pool->first_idle = NULL; - } break; } spin_unlock_irqrestore(&gcwq->lock, flags); return notifier_from_errno(0); - -err_destroy: - if (new_trustee) - kthread_stop(new_trustee); - - spin_lock_irqsave(&gcwq->lock, flags); - for (i = 0; i < NR_WORKER_POOLS; i++) - if (new_workers[i]) - destroy_worker(new_workers[i]); - spin_unlock_irqrestore(&gcwq->lock, flags); - - return NOTIFY_BAD; } /* @@ -3775,7 +3710,6 @@ static int __devinit workqueue_cpu_up_callback(struct notifier_block *nfb, { switch (action & ~CPU_TASKS_FROZEN) { case CPU_UP_PREPARE: - case CPU_UP_CANCELED: case CPU_DOWN_FAILED: case CPU_ONLINE: return workqueue_cpu_callback(nfb, action, hcpu); -- cgit v1.2.3 From 628c78e7ea19d5b70d2b6a59030362168cdbe1ad Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 17 Jul 2012 12:39:27 -0700 Subject: workqueue: remove CPU offline trustee With the previous changes, a disassociated global_cwq now can run as an unbound one on its own - it can create workers as necessary to drain remaining works after the CPU has been brought down and manage the number of workers using the usual idle timer mechanism making trustee completely redundant except for the actual unbinding operation. This patch removes the trustee and let a disassociated global_cwq manage itself. Unbinding is moved to a work item (for CPU affinity) which is scheduled and flushed from CPU_DONW_PREPARE. This patch moves nr_running clearing outside gcwq and manager locks to simplify the code. As nr_running is unused at the point, this is safe. Signed-off-by: Tejun Heo Acked-by: "Rafael J. Wysocki" --- kernel/workqueue.c | 288 +++++++---------------------------------------------- 1 file changed, 36 insertions(+), 252 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index acfabb22e2c4..d1545daa74ad 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -79,13 +79,6 @@ enum { WORKER_NOT_RUNNING = WORKER_PREP | WORKER_REBIND | WORKER_UNBOUND | WORKER_CPU_INTENSIVE, - /* gcwq->trustee_state */ - TRUSTEE_START = 0, /* start */ - TRUSTEE_IN_CHARGE = 1, /* trustee in charge of gcwq */ - TRUSTEE_BUTCHER = 2, /* butcher workers */ - TRUSTEE_RELEASE = 3, /* release workers */ - TRUSTEE_DONE = 4, /* trustee is done */ - NR_WORKER_POOLS = 2, /* # worker pools per gcwq */ BUSY_WORKER_HASH_ORDER = 6, /* 64 pointers */ @@ -100,7 +93,6 @@ enum { (min two ticks) */ MAYDAY_INTERVAL = HZ / 10, /* and then every 100ms */ CREATE_COOLDOWN = HZ, /* time to breath after fail */ - TRUSTEE_COOLDOWN = HZ / 10, /* for trustee draining */ /* * Rescue workers are used only on emergencies and shared by @@ -194,10 +186,6 @@ struct global_cwq { struct worker_pool pools[2]; /* normal and highpri pools */ wait_queue_head_t rebind_hold; /* rebind hold wait */ - - struct task_struct *trustee; /* L: for gcwq shutdown */ - unsigned int trustee_state; /* L: trustee state */ - wait_queue_head_t trustee_wait; /* trustee wait */ } ____cacheline_aligned_in_smp; /* @@ -753,11 +741,11 @@ struct task_struct *wq_worker_sleeping(struct task_struct *task, * worklist not empty test sequence is in insert_work(). * Please read comment there. * - * NOT_RUNNING is clear. This means that trustee is not in - * charge and we're running on the local cpu w/ rq lock held - * and preemption disabled, which in turn means that none else - * could be manipulating idle_list, so dereferencing idle_list - * without gcwq lock is safe. + * NOT_RUNNING is clear. This means that we're bound to and + * running on the local cpu w/ rq lock held and preemption + * disabled, which in turn means that none else could be + * manipulating idle_list, so dereferencing idle_list without gcwq + * lock is safe. */ if (atomic_dec_and_test(nr_running) && !list_empty(&pool->worklist)) to_wakeup = first_worker(pool); @@ -1217,19 +1205,16 @@ static void worker_enter_idle(struct worker *worker) /* idle_list is LIFO */ list_add(&worker->entry, &pool->idle_list); - if (likely(gcwq->trustee_state != TRUSTEE_DONE)) { - if (too_many_workers(pool) && !timer_pending(&pool->idle_timer)) - mod_timer(&pool->idle_timer, - jiffies + IDLE_WORKER_TIMEOUT); - } else - wake_up_all(&gcwq->trustee_wait); + if (too_many_workers(pool) && !timer_pending(&pool->idle_timer)) + mod_timer(&pool->idle_timer, jiffies + IDLE_WORKER_TIMEOUT); /* - * Sanity check nr_running. Because trustee releases gcwq->lock - * between setting %WORKER_UNBOUND and zapping nr_running, the - * warning may trigger spuriously. Check iff trustee is idle. + * Sanity check nr_running. Because gcwq_unbind_fn() releases + * gcwq->lock between setting %WORKER_UNBOUND and zapping + * nr_running, the warning may trigger spuriously. Check iff + * unbind is not in progress. */ - WARN_ON_ONCE(gcwq->trustee_state == TRUSTEE_DONE && + WARN_ON_ONCE(!(gcwq->flags & GCWQ_DISASSOCIATED) && pool->nr_workers == pool->nr_idle && atomic_read(get_pool_nr_running(pool))); } @@ -3367,46 +3352,9 @@ EXPORT_SYMBOL_GPL(work_busy); * gcwqs serve mix of short, long and very long running works making * blocked draining impractical. * - * This is solved by allowing a gcwq to be detached from CPU, running it - * with unbound workers and allowing it to be reattached later if the cpu - * comes back online. A separate thread is created to govern a gcwq in - * such state and is called the trustee of the gcwq. - * - * Trustee states and their descriptions. - * - * START Command state used on startup. On CPU_DOWN_PREPARE, a - * new trustee is started with this state. - * - * IN_CHARGE Once started, trustee will enter this state after - * assuming the manager role and making all existing - * workers rogue. DOWN_PREPARE waits for trustee to - * enter this state. After reaching IN_CHARGE, trustee - * tries to execute the pending worklist until it's empty - * and the state is set to BUTCHER, or the state is set - * to RELEASE. - * - * BUTCHER Command state which is set by the cpu callback after - * the cpu has went down. Once this state is set trustee - * knows that there will be no new works on the worklist - * and once the worklist is empty it can proceed to - * killing idle workers. - * - * RELEASE Command state which is set by the cpu callback if the - * cpu down has been canceled or it has come online - * again. After recognizing this state, trustee stops - * trying to drain or butcher and clears ROGUE, rebinds - * all remaining workers back to the cpu and releases - * manager role. - * - * DONE Trustee will enter this state after BUTCHER or RELEASE - * is complete. - * - * trustee CPU draining - * took over down complete - * START -----------> IN_CHARGE -----------> BUTCHER -----------> DONE - * | | ^ - * | CPU is back online v return workers | - * ----------------> RELEASE -------------- + * This is solved by allowing a gcwq to be disassociated from the CPU + * running as an unbound one and allowing it to be reattached later if the + * cpu comes back online. */ /* claim manager positions of all pools */ @@ -3427,61 +3375,11 @@ static void gcwq_release_management(struct global_cwq *gcwq) mutex_unlock(&pool->manager_mutex); } -/** - * trustee_wait_event_timeout - timed event wait for trustee - * @cond: condition to wait for - * @timeout: timeout in jiffies - * - * wait_event_timeout() for trustee to use. Handles locking and - * checks for RELEASE request. - * - * CONTEXT: - * spin_lock_irq(gcwq->lock) which may be released and regrabbed - * multiple times. To be used by trustee. - * - * RETURNS: - * Positive indicating left time if @cond is satisfied, 0 if timed - * out, -1 if canceled. - */ -#define trustee_wait_event_timeout(cond, timeout) ({ \ - long __ret = (timeout); \ - while (!((cond) || (gcwq->trustee_state == TRUSTEE_RELEASE)) && \ - __ret) { \ - spin_unlock_irq(&gcwq->lock); \ - __wait_event_timeout(gcwq->trustee_wait, (cond) || \ - (gcwq->trustee_state == TRUSTEE_RELEASE), \ - __ret); \ - spin_lock_irq(&gcwq->lock); \ - } \ - gcwq->trustee_state == TRUSTEE_RELEASE ? -1 : (__ret); \ -}) - -/** - * trustee_wait_event - event wait for trustee - * @cond: condition to wait for - * - * wait_event() for trustee to use. Automatically handles locking and - * checks for CANCEL request. - * - * CONTEXT: - * spin_lock_irq(gcwq->lock) which may be released and regrabbed - * multiple times. To be used by trustee. - * - * RETURNS: - * 0 if @cond is satisfied, -1 if canceled. - */ -#define trustee_wait_event(cond) ({ \ - long __ret1; \ - __ret1 = trustee_wait_event_timeout(cond, MAX_SCHEDULE_TIMEOUT);\ - __ret1 < 0 ? -1 : 0; \ -}) - -static int __cpuinit trustee_thread(void *__gcwq) +static void gcwq_unbind_fn(struct work_struct *work) { - struct global_cwq *gcwq = __gcwq; + struct global_cwq *gcwq = get_gcwq(smp_processor_id()); struct worker_pool *pool; struct worker *worker; - struct work_struct *work; struct hlist_node *pos; int i; @@ -3505,119 +3403,29 @@ static int __cpuinit trustee_thread(void *__gcwq) gcwq->flags |= GCWQ_DISASSOCIATED; + spin_unlock_irq(&gcwq->lock); + gcwq_release_management(gcwq); + /* * Call schedule() so that we cross rq->lock and thus can guarantee - * sched callbacks see the unbound flag. This is necessary as - * scheduler callbacks may be invoked from other cpus. + * sched callbacks see the %WORKER_UNBOUND flag. This is necessary + * as scheduler callbacks may be invoked from other cpus. */ - spin_unlock_irq(&gcwq->lock); schedule(); - spin_lock_irq(&gcwq->lock); /* - * Sched callbacks are disabled now. Zap nr_running. After - * this, nr_running stays zero and need_more_worker() and - * keep_working() are always true as long as the worklist is - * not empty. + * Sched callbacks are disabled now. Zap nr_running. After this, + * nr_running stays zero and need_more_worker() and keep_working() + * are always true as long as the worklist is not empty. @gcwq now + * behaves as unbound (in terms of concurrency management) gcwq + * which is served by workers tied to the CPU. + * + * On return from this function, the current worker would trigger + * unbound chain execution of pending work items if other workers + * didn't already. */ for_each_worker_pool(pool, gcwq) atomic_set(get_pool_nr_running(pool), 0); - - spin_unlock_irq(&gcwq->lock); - for_each_worker_pool(pool, gcwq) - del_timer_sync(&pool->idle_timer); - spin_lock_irq(&gcwq->lock); - - /* - * We're now in charge. Notify and proceed to drain. We need - * to keep the gcwq running during the whole CPU down - * procedure as other cpu hotunplug callbacks may need to - * flush currently running tasks. - */ - gcwq->trustee_state = TRUSTEE_IN_CHARGE; - wake_up_all(&gcwq->trustee_wait); - - /* - * The original cpu is in the process of dying and may go away - * anytime now. When that happens, we and all workers would - * be migrated to other cpus. Try draining any left work. We - * want to get it over with ASAP - spam rescuers, wake up as - * many idlers as necessary and create new ones till the - * worklist is empty. Note that if the gcwq is frozen, there - * may be frozen works in freezable cwqs. Don't declare - * completion while frozen. - */ - while (true) { - bool busy = false; - - for_each_worker_pool(pool, gcwq) - busy |= pool->nr_workers != pool->nr_idle; - - if (!busy && !(gcwq->flags & GCWQ_FREEZING) && - gcwq->trustee_state != TRUSTEE_IN_CHARGE) - break; - - for_each_worker_pool(pool, gcwq) { - int nr_works = 0; - - list_for_each_entry(work, &pool->worklist, entry) { - send_mayday(work); - nr_works++; - } - - list_for_each_entry(worker, &pool->idle_list, entry) { - if (!nr_works--) - break; - wake_up_process(worker->task); - } - - if (need_to_create_worker(pool)) { - spin_unlock_irq(&gcwq->lock); - worker = create_worker(pool); - spin_lock_irq(&gcwq->lock); - if (worker) - start_worker(worker); - } - } - - /* give a breather */ - if (trustee_wait_event_timeout(false, TRUSTEE_COOLDOWN) < 0) - break; - } - - gcwq_release_management(gcwq); - - /* notify completion */ - gcwq->trustee = NULL; - gcwq->trustee_state = TRUSTEE_DONE; - wake_up_all(&gcwq->trustee_wait); - spin_unlock_irq(&gcwq->lock); - return 0; -} - -/** - * wait_trustee_state - wait for trustee to enter the specified state - * @gcwq: gcwq the trustee of interest belongs to - * @state: target state to wait for - * - * Wait for the trustee to reach @state. DONE is already matched. - * - * CONTEXT: - * spin_lock_irq(gcwq->lock) which may be released and regrabbed - * multiple times. To be used by cpu_callback. - */ -static void __cpuinit wait_trustee_state(struct global_cwq *gcwq, int state) -__releases(&gcwq->lock) -__acquires(&gcwq->lock) -{ - if (!(gcwq->trustee_state == state || - gcwq->trustee_state == TRUSTEE_DONE)) { - spin_unlock_irq(&gcwq->lock); - __wait_event(gcwq->trustee_wait, - gcwq->trustee_state == state || - gcwq->trustee_state == TRUSTEE_DONE); - spin_lock_irq(&gcwq->lock); - } } static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, @@ -3626,19 +3434,18 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, { unsigned int cpu = (unsigned long)hcpu; struct global_cwq *gcwq = get_gcwq(cpu); - struct task_struct *new_trustee = NULL; struct worker_pool *pool; + struct work_struct unbind_work; unsigned long flags; action &= ~CPU_TASKS_FROZEN; switch (action) { case CPU_DOWN_PREPARE: - new_trustee = kthread_create(trustee_thread, gcwq, - "workqueue_trustee/%d\n", cpu); - if (IS_ERR(new_trustee)) - return notifier_from_errno(PTR_ERR(new_trustee)); - kthread_bind(new_trustee, cpu); + /* unbinding should happen on the local CPU */ + INIT_WORK_ONSTACK(&unbind_work, gcwq_unbind_fn); + schedule_work_on(cpu, &unbind_work); + flush_work(&unbind_work); break; case CPU_UP_PREPARE: @@ -3662,27 +3469,8 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, spin_lock_irqsave(&gcwq->lock, flags); switch (action) { - case CPU_DOWN_PREPARE: - /* initialize trustee and tell it to acquire the gcwq */ - BUG_ON(gcwq->trustee || gcwq->trustee_state != TRUSTEE_DONE); - gcwq->trustee = new_trustee; - gcwq->trustee_state = TRUSTEE_START; - wake_up_process(gcwq->trustee); - wait_trustee_state(gcwq, TRUSTEE_IN_CHARGE); - break; - - case CPU_POST_DEAD: - gcwq->trustee_state = TRUSTEE_BUTCHER; - break; - case CPU_DOWN_FAILED: case CPU_ONLINE: - if (gcwq->trustee_state != TRUSTEE_DONE) { - gcwq->trustee_state = TRUSTEE_RELEASE; - wake_up_process(gcwq->trustee); - wait_trustee_state(gcwq, TRUSTEE_DONE); - } - spin_unlock_irq(&gcwq->lock); gcwq_claim_management(gcwq); spin_lock_irq(&gcwq->lock); @@ -3727,7 +3515,6 @@ static int __devinit workqueue_cpu_down_callback(struct notifier_block *nfb, { switch (action & ~CPU_TASKS_FROZEN) { case CPU_DOWN_PREPARE: - case CPU_POST_DEAD: return workqueue_cpu_callback(nfb, action, hcpu); } return NOTIFY_OK; @@ -3960,9 +3747,6 @@ static int __init init_workqueues(void) } init_waitqueue_head(&gcwq->rebind_hold); - - gcwq->trustee_state = TRUSTEE_DONE; - init_waitqueue_head(&gcwq->trustee_wait); } /* create the initial worker */ -- cgit v1.2.3 From 8db25e7891a47e03db6f04344a9c92be16e391bb Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 17 Jul 2012 12:39:28 -0700 Subject: workqueue: simplify CPU hotplug code With trustee gone, CPU hotplug code can be simplified. * gcwq_claim/release_management() now grab and release gcwq lock too respectively and gained _and_lock and _and_unlock postfixes. * All CPU hotplug logic was implemented in workqueue_cpu_callback() which was called by workqueue_cpu_up/down_callback() for the correct priority. This was because up and down paths shared a lot of logic, which is no longer true. Remove workqueue_cpu_callback() and move all hotplug logic into the two actual callbacks. This patch doesn't make any functional changes. Signed-off-by: Tejun Heo Acked-by: "Rafael J. Wysocki" --- kernel/workqueue.c | 79 +++++++++++++++++------------------------------------- 1 file changed, 25 insertions(+), 54 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index d1545daa74ad..471996a81633 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -3358,19 +3358,21 @@ EXPORT_SYMBOL_GPL(work_busy); */ /* claim manager positions of all pools */ -static void gcwq_claim_management(struct global_cwq *gcwq) +static void gcwq_claim_management_and_lock(struct global_cwq *gcwq) { struct worker_pool *pool; for_each_worker_pool(pool, gcwq) mutex_lock_nested(&pool->manager_mutex, pool - gcwq->pools); + spin_lock_irq(&gcwq->lock); } /* release manager positions */ -static void gcwq_release_management(struct global_cwq *gcwq) +static void gcwq_release_management_and_unlock(struct global_cwq *gcwq) { struct worker_pool *pool; + spin_unlock_irq(&gcwq->lock); for_each_worker_pool(pool, gcwq) mutex_unlock(&pool->manager_mutex); } @@ -3385,8 +3387,7 @@ static void gcwq_unbind_fn(struct work_struct *work) BUG_ON(gcwq->cpu != smp_processor_id()); - gcwq_claim_management(gcwq); - spin_lock_irq(&gcwq->lock); + gcwq_claim_management_and_lock(gcwq); /* * We've claimed all manager positions. Make all workers unbound @@ -3403,8 +3404,7 @@ static void gcwq_unbind_fn(struct work_struct *work) gcwq->flags |= GCWQ_DISASSOCIATED; - spin_unlock_irq(&gcwq->lock); - gcwq_release_management(gcwq); + gcwq_release_management_and_unlock(gcwq); /* * Call schedule() so that we cross rq->lock and thus can guarantee @@ -3428,26 +3428,19 @@ static void gcwq_unbind_fn(struct work_struct *work) atomic_set(get_pool_nr_running(pool), 0); } -static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, - unsigned long action, - void *hcpu) +/* + * Workqueues should be brought up before normal priority CPU notifiers. + * This will be registered high priority CPU notifier. + */ +static int __devinit workqueue_cpu_up_callback(struct notifier_block *nfb, + unsigned long action, + void *hcpu) { unsigned int cpu = (unsigned long)hcpu; struct global_cwq *gcwq = get_gcwq(cpu); struct worker_pool *pool; - struct work_struct unbind_work; - unsigned long flags; - - action &= ~CPU_TASKS_FROZEN; - - switch (action) { - case CPU_DOWN_PREPARE: - /* unbinding should happen on the local CPU */ - INIT_WORK_ONSTACK(&unbind_work, gcwq_unbind_fn); - schedule_work_on(cpu, &unbind_work); - flush_work(&unbind_work); - break; + switch (action & ~CPU_TASKS_FROZEN) { case CPU_UP_PREPARE: for_each_worker_pool(pool, gcwq) { struct worker *worker; @@ -3463,45 +3456,16 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, start_worker(worker); spin_unlock_irq(&gcwq->lock); } - } - - /* some are called w/ irq disabled, don't disturb irq status */ - spin_lock_irqsave(&gcwq->lock, flags); + break; - switch (action) { case CPU_DOWN_FAILED: case CPU_ONLINE: - spin_unlock_irq(&gcwq->lock); - gcwq_claim_management(gcwq); - spin_lock_irq(&gcwq->lock); - + gcwq_claim_management_and_lock(gcwq); gcwq->flags &= ~GCWQ_DISASSOCIATED; - rebind_workers(gcwq); - - gcwq_release_management(gcwq); + gcwq_release_management_and_unlock(gcwq); break; } - - spin_unlock_irqrestore(&gcwq->lock, flags); - - return notifier_from_errno(0); -} - -/* - * Workqueues should be brought up before normal priority CPU notifiers. - * This will be registered high priority CPU notifier. - */ -static int __devinit workqueue_cpu_up_callback(struct notifier_block *nfb, - unsigned long action, - void *hcpu) -{ - switch (action & ~CPU_TASKS_FROZEN) { - case CPU_UP_PREPARE: - case CPU_DOWN_FAILED: - case CPU_ONLINE: - return workqueue_cpu_callback(nfb, action, hcpu); - } return NOTIFY_OK; } @@ -3513,9 +3477,16 @@ static int __devinit workqueue_cpu_down_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { + unsigned int cpu = (unsigned long)hcpu; + struct work_struct unbind_work; + switch (action & ~CPU_TASKS_FROZEN) { case CPU_DOWN_PREPARE: - return workqueue_cpu_callback(nfb, action, hcpu); + /* unbinding should happen on the local CPU */ + INIT_WORK_ONSTACK(&unbind_work, gcwq_unbind_fn); + schedule_work_on(cpu, &unbind_work); + flush_work(&unbind_work); + break; } return NOTIFY_OK; } -- cgit v1.2.3 From 11388c87d2abca1f01975ced28ce9eacea239104 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 19 Jul 2012 00:00:58 +0200 Subject: PM / Sleep: Require CAP_BLOCK_SUSPEND to use wake_lock/wake_unlock Require processes wanting to use the wake_lock/wake_unlock sysfs files to have the CAP_BLOCK_SUSPEND capability, which also is required for the eventpoll EPOLLWAKEUP flag to be effective, so that all interfaces related to blocking autosleep depend on the same capability. Signed-off-by: Rafael J. Wysocki Cc: stable@vger.kernel.org Acked-by: Michael Kerrisk --- kernel/power/wakelock.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'kernel') diff --git a/kernel/power/wakelock.c b/kernel/power/wakelock.c index c8fba3380076..8f50de394d22 100644 --- a/kernel/power/wakelock.c +++ b/kernel/power/wakelock.c @@ -9,6 +9,7 @@ * manipulate wakelocks on Android. */ +#include #include #include #include @@ -188,6 +189,9 @@ int pm_wake_lock(const char *buf) size_t len; int ret = 0; + if (!capable(CAP_BLOCK_SUSPEND)) + return -EPERM; + while (*str && !isspace(*str)) str++; @@ -231,6 +235,9 @@ int pm_wake_unlock(const char *buf) size_t len; int ret = 0; + if (!capable(CAP_BLOCK_SUSPEND)) + return -EPERM; + len = strlen(buf); if (!len) return -EINVAL; -- cgit v1.2.3 From eea03c20ae38a55405c0865ed9adfccc400e4c8e Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Wed, 18 Jul 2012 18:15:46 -0700 Subject: Make wait_for_device_probe() also do scsi_complete_async_scans() Commit a7a20d103994 ("sd: limit the scope of the async probe domain") make the SCSI device probing run device discovery in it's own async domain. However, as a result, the partition detection was no longer synchronized by async_synchronize_full() (which, despite the name, only synchronizes the global async space, not all of them). Which in turn meant that "wait_for_device_probe()" would not wait for the SCSI partitions to be parsed. And "wait_for_device_probe()" was what the boot time init code relied on for mounting the root filesystem. Now, most people never noticed this, because not only is it timing-dependent, but modern distributions all use initrd. So the root filesystem isn't actually on a disk at all. And then before they actually mount the final disk filesystem, they will have loaded the scsi-wait-scan module, which not only does the expected wait_for_device_probe(), but also does scsi_complete_async_scans(). [ Side note: scsi_complete_async_scans() had also been partially broken, but that was fixed in commit 43a8d39d0137 ("fix async probe regression"), so that same commit a7a20d103994 had actually broken setups even if you used scsi-wait-scan explicitly ] Solve this problem by just moving the scsi_complete_async_scans() call into wait_for_device_probe(). Everybody who wants to wait for device probing to finish really wants the SCSI probing to complete, so there's no reason not to do this. So now "wait_for_device_probe()" really does what the name implies, and properly waits for device probing to finish. This also removes the now unnecessary extra calls to scsi_complete_async_scans(). Reported-and-tested-by: Artem S. Tashkinov Cc: Dan Williams Cc: Alan Stern Cc: James Bottomley Cc: Borislav Petkov Cc: linux-scsi Signed-off-by: Linus Torvalds --- drivers/base/dd.c | 2 ++ drivers/scsi/scsi_wait_scan.c | 5 ----- include/linux/device.h | 2 -- kernel/power/hibernate.c | 8 -------- kernel/power/user.c | 2 -- 5 files changed, 2 insertions(+), 17 deletions(-) (limited to 'kernel') diff --git a/drivers/base/dd.c b/drivers/base/dd.c index dcb8a6e48692..4b01ab3d2c24 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -24,6 +24,7 @@ #include #include #include +#include #include "base.h" #include "power/power.h" @@ -332,6 +333,7 @@ void wait_for_device_probe(void) /* wait for the known devices to complete their probing */ wait_event(probe_waitqueue, atomic_read(&probe_count) == 0); async_synchronize_full(); + scsi_complete_async_scans(); } EXPORT_SYMBOL_GPL(wait_for_device_probe); diff --git a/drivers/scsi/scsi_wait_scan.c b/drivers/scsi/scsi_wait_scan.c index ae7814874618..072734538876 100644 --- a/drivers/scsi/scsi_wait_scan.c +++ b/drivers/scsi/scsi_wait_scan.c @@ -22,11 +22,6 @@ static int __init wait_scan_init(void) * and might not yet have reached the scsi async scanning */ wait_for_device_probe(); - /* - * and then we wait for the actual asynchronous scsi scan - * to finish. - */ - scsi_complete_async_scans(); return 0; } diff --git a/include/linux/device.h b/include/linux/device.h index 161d96241b1b..6de94151ff6f 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -865,8 +865,6 @@ extern int (*platform_notify_remove)(struct device *dev); extern struct device *get_device(struct device *dev); extern void put_device(struct device *dev); -extern void wait_for_device_probe(void); - #ifdef CONFIG_DEVTMPFS extern int devtmpfs_create_node(struct device *dev); extern int devtmpfs_delete_node(struct device *dev); diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 8b53db38a279..238025f5472e 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -27,7 +27,6 @@ #include #include #include -#include #include "power.h" @@ -748,13 +747,6 @@ static int software_resume(void) async_synchronize_full(); } - /* - * We can't depend on SCSI devices being available after loading - * one of their modules until scsi_complete_async_scans() is - * called and the resume device usually is a SCSI one. - */ - scsi_complete_async_scans(); - swsusp_resume_device = name_to_dev_t(resume_file); if (!swsusp_resume_device) { error = -ENODEV; diff --git a/kernel/power/user.c b/kernel/power/user.c index 91b0fd021a95..4ed81e74f86f 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -24,7 +24,6 @@ #include #include #include -#include #include @@ -84,7 +83,6 @@ static int snapshot_open(struct inode *inode, struct file *filp) * appear. */ wait_for_device_probe(); - scsi_complete_async_scans(); data->swap = -1; data->mode = O_WRONLY; -- cgit v1.2.3 From c5857ccf293968348e5eb4ebedc68074de3dcda6 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Sat, 14 Jul 2012 20:27:52 -0400 Subject: random: remove rand_initialize_irq() With the new interrupt sampling system, we are no longer using the timer_rand_state structure in the irq descriptor, so we can stop initializing it now. [ Merged in fixes from Sedat to find some last missing references to rand_initialize_irq() ] Signed-off-by: "Theodore Ts'o" Signed-off-by: Sedat Dilek --- arch/ia64/kernel/irq_ia64.c | 1 - drivers/char/random.c | 55 --------------------------------------------- drivers/mfd/ab3100-core.c | 3 --- include/linux/irqdesc.h | 1 - include/linux/random.h | 2 -- kernel/irq/manage.c | 17 -------------- 6 files changed, 79 deletions(-) (limited to 'kernel') diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c index 5c3e0888265a..1034884b77da 100644 --- a/arch/ia64/kernel/irq_ia64.c +++ b/arch/ia64/kernel/irq_ia64.c @@ -23,7 +23,6 @@ #include #include #include -#include /* for rand_initialize_irq() */ #include #include #include diff --git a/drivers/char/random.c b/drivers/char/random.c index e3180852ec85..9793b40f5754 100644 --- a/drivers/char/random.c +++ b/drivers/char/random.c @@ -634,43 +634,6 @@ struct timer_rand_state { unsigned dont_count_entropy:1; }; -#ifndef CONFIG_GENERIC_HARDIRQS - -static struct timer_rand_state *irq_timer_state[NR_IRQS]; - -static struct timer_rand_state *get_timer_rand_state(unsigned int irq) -{ - return irq_timer_state[irq]; -} - -static void set_timer_rand_state(unsigned int irq, - struct timer_rand_state *state) -{ - irq_timer_state[irq] = state; -} - -#else - -static struct timer_rand_state *get_timer_rand_state(unsigned int irq) -{ - struct irq_desc *desc; - - desc = irq_to_desc(irq); - - return desc->timer_rand_state; -} - -static void set_timer_rand_state(unsigned int irq, - struct timer_rand_state *state) -{ - struct irq_desc *desc; - - desc = irq_to_desc(irq); - - desc->timer_rand_state = state; -} -#endif - /* * Add device- or boot-specific data to the input and nonblocking * pools to help initialize them to unique values. @@ -1133,24 +1096,6 @@ static int rand_initialize(void) } module_init(rand_initialize); -void rand_initialize_irq(int irq) -{ - struct timer_rand_state *state; - - state = get_timer_rand_state(irq); - - if (state) - return; - - /* - * If kzalloc returns null, we just won't use that entropy - * source. - */ - state = kzalloc(sizeof(struct timer_rand_state), GFP_KERNEL); - if (state) - set_timer_rand_state(irq, state); -} - #ifdef CONFIG_BLOCK void rand_initialize_disk(struct gendisk *disk) { diff --git a/drivers/mfd/ab3100-core.c b/drivers/mfd/ab3100-core.c index 9522d6bda4f7..1287645b984d 100644 --- a/drivers/mfd/ab3100-core.c +++ b/drivers/mfd/ab3100-core.c @@ -931,9 +931,6 @@ static int __devinit ab3100_probe(struct i2c_client *client, err = request_threaded_irq(client->irq, NULL, ab3100_irq_handler, IRQF_ONESHOT, "ab3100-core", ab3100); - /* This real unpredictable IRQ is of course sampled for entropy */ - rand_initialize_irq(client->irq); - if (err) goto exit_no_irq; diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h index f1e2527006bd..9a323d12de1c 100644 --- a/include/linux/irqdesc.h +++ b/include/linux/irqdesc.h @@ -39,7 +39,6 @@ struct module; */ struct irq_desc { struct irq_data irq_data; - struct timer_rand_state *timer_rand_state; unsigned int __percpu *kstat_irqs; irq_flow_handler_t handle_irq; #ifdef CONFIG_IRQ_PREFLOW_FASTEOI diff --git a/include/linux/random.h b/include/linux/random.h index 29e217a7e6d0..ac621ce886ca 100644 --- a/include/linux/random.h +++ b/include/linux/random.h @@ -48,8 +48,6 @@ struct rnd_state { #ifdef __KERNEL__ -extern void rand_initialize_irq(int irq); - extern void add_device_randomness(const void *, unsigned int); extern void add_input_randomness(unsigned int type, unsigned int code, unsigned int value); diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 8c548232ba39..5e42eb119677 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -893,22 +893,6 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) return -ENOSYS; if (!try_module_get(desc->owner)) return -ENODEV; - /* - * Some drivers like serial.c use request_irq() heavily, - * so we have to be careful not to interfere with a - * running system. - */ - if (new->flags & IRQF_SAMPLE_RANDOM) { - /* - * This function might sleep, we want to call it first, - * outside of the atomic block. - * Yes, this might clear the entropy pool if the wrong - * driver is attempted to be loaded, without actually - * installing a new handler, but is this really a problem, - * only the sysadmin is able to do this. - */ - rand_initialize_irq(irq); - } /* * Check whether the interrupt nests into another interrupt @@ -1354,7 +1338,6 @@ EXPORT_SYMBOL(free_irq); * Flags: * * IRQF_SHARED Interrupt is shared - * IRQF_SAMPLE_RANDOM The interrupt can be used for entropy * IRQF_TRIGGER_* Specify active edge(s) or level * */ -- cgit v1.2.3 From 6791457a090d9a234a40b501c2536f0aefaeae4b Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Wed, 18 Jul 2012 13:18:12 -0400 Subject: printk: Export struct log size and member offsets through vmcoreinfo There are tools like makedumpfile and vmcore-dmesg which can extract kernel log buffer from vmcore. Since we introduced structured logging, that functionality is broken. Now user space tools need to know about "struct log" and offsets of various fields to be able to parse struct log data and extract text message or dictonary. This patch exports some of the fields. Currently I am not exporting log "level" info as that is a bitfield and offsetof() bitfields can't be calculated. But if people start asking for log level info in the output then we probably either need to seprate out "level" or use bit shift operations for flags and level. Signed-off-by: Vivek Goyal Acked-by: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- kernel/printk.c | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index 4da2377131b0..449364f07a1e 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -671,6 +671,15 @@ void log_buf_kexec_setup(void) VMCOREINFO_SYMBOL(log_buf_len); VMCOREINFO_SYMBOL(log_first_idx); VMCOREINFO_SYMBOL(log_next_idx); + /* + * Export struct log size and field offsets. User space tools can + * parse it and detect any changes to structure down the line. + */ + VMCOREINFO_STRUCT_SIZE(log); + VMCOREINFO_OFFSET(log, ts_nsec); + VMCOREINFO_OFFSET(log, len); + VMCOREINFO_OFFSET(log, text_len); + VMCOREINFO_OFFSET(log, dict_len); } #endif -- cgit v1.2.3 From 2955b47d2c1983998a8c5915cb96884e67f7cb53 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Mon, 9 Jul 2012 19:33:25 -0700 Subject: [SCSI] async: introduce 'async_domain' type This is in preparation for teaching async_synchronize_full() to sync all pending async work, and not just on the async_running domain. This conversion is functionally equivalent, just embedding the existing list in a new async_domain type. The .registered attribute is used in a later patch to distinguish between domains that want to be flushed by async_synchronize_full() versus those that only expect async_synchronize_{full|cookie}_domain to be used for flushing. [jejb: add async.h to scsi_priv.h for struct async_domain] Signed-off-by: Dan Williams Acked-by: Arjan van de Ven Acked-by: Mark Brown Tested-by: Eldad Zack Signed-off-by: James Bottomley --- drivers/regulator/core.c | 2 +- drivers/scsi/libsas/sas_ata.c | 2 +- drivers/scsi/scsi.c | 3 ++- drivers/scsi/scsi_priv.h | 3 ++- include/linux/async.h | 35 +++++++++++++++++++++++++++++++---- kernel/async.c | 35 +++++++++++++++++------------------ sound/soc/soc-dapm.c | 2 +- 7 files changed, 55 insertions(+), 27 deletions(-) (limited to 'kernel') diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index 8b4b3829d9e7..6c74546fc3cd 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -2744,7 +2744,7 @@ static void regulator_bulk_enable_async(void *data, async_cookie_t cookie) int regulator_bulk_enable(int num_consumers, struct regulator_bulk_data *consumers) { - LIST_HEAD(async_domain); + ASYNC_DOMAIN_EXCLUSIVE(async_domain); int i; int ret = 0; diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c index bec3bc8aab0c..a59fcdc8fd63 100644 --- a/drivers/scsi/libsas/sas_ata.c +++ b/drivers/scsi/libsas/sas_ata.c @@ -742,7 +742,7 @@ static void async_sas_ata_eh(void *data, async_cookie_t cookie) void sas_ata_strategy_handler(struct Scsi_Host *shost) { struct sas_ha_struct *sas_ha = SHOST_TO_SAS_HA(shost); - LIST_HEAD(async); + ASYNC_DOMAIN_EXCLUSIVE(async); int i; /* it's ok to defer revalidation events during ata eh, these diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index bbbc9c918d4c..4cade886a50a 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -54,6 +54,7 @@ #include #include #include +#include #include #include @@ -91,7 +92,7 @@ EXPORT_SYMBOL(scsi_logging_level); #endif /* sd, scsi core and power management need to coordinate flushing async actions */ -LIST_HEAD(scsi_sd_probe_domain); +ASYNC_DOMAIN(scsi_sd_probe_domain); EXPORT_SYMBOL(scsi_sd_probe_domain); /* NB: These are exposed through /proc/scsi/scsi and form part of the ABI. diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h index 13d74da5dfab..8f9a0cadc296 100644 --- a/drivers/scsi/scsi_priv.h +++ b/drivers/scsi/scsi_priv.h @@ -2,6 +2,7 @@ #define _SCSI_PRIV_H #include +#include #include struct request_queue; @@ -163,7 +164,7 @@ static inline int scsi_autopm_get_host(struct Scsi_Host *h) { return 0; } static inline void scsi_autopm_put_host(struct Scsi_Host *h) {} #endif /* CONFIG_PM_RUNTIME */ -extern struct list_head scsi_sd_probe_domain; +extern struct async_domain scsi_sd_probe_domain; /* * internal scsi timeout functions: for use by mid-layer and transport diff --git a/include/linux/async.h b/include/linux/async.h index 68a9530196f2..364e7ff16c08 100644 --- a/include/linux/async.h +++ b/include/linux/async.h @@ -9,19 +9,46 @@ * as published by the Free Software Foundation; version 2 * of the License. */ +#ifndef __ASYNC_H__ +#define __ASYNC_H__ #include #include typedef u64 async_cookie_t; typedef void (async_func_ptr) (void *data, async_cookie_t cookie); +struct async_domain { + struct list_head node; + struct list_head domain; + int count; + unsigned registered:1; +}; + +/* + * domain participates in global async_synchronize_full + */ +#define ASYNC_DOMAIN(_name) \ + struct async_domain _name = { .node = LIST_HEAD_INIT(_name.node), \ + .domain = LIST_HEAD_INIT(_name.domain), \ + .count = 0, \ + .registered = 1 } + +/* + * domain is free to go out of scope as soon as all pending work is + * complete, this domain does not participate in async_synchronize_full + */ +#define ASYNC_DOMAIN_EXCLUSIVE(_name) \ + struct async_domain _name = { .node = LIST_HEAD_INIT(_name.node), \ + .domain = LIST_HEAD_INIT(_name.domain), \ + .count = 0, \ + .registered = 0 } extern async_cookie_t async_schedule(async_func_ptr *ptr, void *data); extern async_cookie_t async_schedule_domain(async_func_ptr *ptr, void *data, - struct list_head *list); + struct async_domain *domain); extern void async_synchronize_full(void); -extern void async_synchronize_full_domain(struct list_head *list); +extern void async_synchronize_full_domain(struct async_domain *domain); extern void async_synchronize_cookie(async_cookie_t cookie); extern void async_synchronize_cookie_domain(async_cookie_t cookie, - struct list_head *list); - + struct async_domain *domain); +#endif diff --git a/kernel/async.c b/kernel/async.c index bd0c168a3bbe..ba5491dfa991 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -62,7 +62,7 @@ static async_cookie_t next_cookie = 1; #define MAX_WORK 32768 static LIST_HEAD(async_pending); -static LIST_HEAD(async_running); +static ASYNC_DOMAIN(async_running); static DEFINE_SPINLOCK(async_lock); struct async_entry { @@ -71,7 +71,7 @@ struct async_entry { async_cookie_t cookie; async_func_ptr *func; void *data; - struct list_head *running; + struct async_domain *running; }; static DECLARE_WAIT_QUEUE_HEAD(async_done); @@ -82,13 +82,12 @@ static atomic_t entry_count; /* * MUST be called with the lock held! */ -static async_cookie_t __lowest_in_progress(struct list_head *running) +static async_cookie_t __lowest_in_progress(struct async_domain *running) { struct async_entry *entry; - if (!list_empty(running)) { - entry = list_first_entry(running, - struct async_entry, list); + if (!list_empty(&running->domain)) { + entry = list_first_entry(&running->domain, typeof(*entry), list); return entry->cookie; } @@ -99,7 +98,7 @@ static async_cookie_t __lowest_in_progress(struct list_head *running) return next_cookie; /* "infinity" value */ } -static async_cookie_t lowest_in_progress(struct list_head *running) +static async_cookie_t lowest_in_progress(struct async_domain *running) { unsigned long flags; async_cookie_t ret; @@ -119,10 +118,11 @@ static void async_run_entry_fn(struct work_struct *work) container_of(work, struct async_entry, work); unsigned long flags; ktime_t uninitialized_var(calltime), delta, rettime; + struct async_domain *running = entry->running; /* 1) move self to the running queue */ spin_lock_irqsave(&async_lock, flags); - list_move_tail(&entry->list, entry->running); + list_move_tail(&entry->list, &running->domain); spin_unlock_irqrestore(&async_lock, flags); /* 2) run (and print duration) */ @@ -156,7 +156,7 @@ static void async_run_entry_fn(struct work_struct *work) wake_up(&async_done); } -static async_cookie_t __async_schedule(async_func_ptr *ptr, void *data, struct list_head *running) +static async_cookie_t __async_schedule(async_func_ptr *ptr, void *data, struct async_domain *running) { struct async_entry *entry; unsigned long flags; @@ -223,7 +223,7 @@ EXPORT_SYMBOL_GPL(async_schedule); * Note: This function may be called from atomic or non-atomic contexts. */ async_cookie_t async_schedule_domain(async_func_ptr *ptr, void *data, - struct list_head *running) + struct async_domain *running) { return __async_schedule(ptr, data, running); } @@ -238,20 +238,20 @@ void async_synchronize_full(void) { do { async_synchronize_cookie(next_cookie); - } while (!list_empty(&async_running) || !list_empty(&async_pending)); + } while (!list_empty(&async_running.domain) || !list_empty(&async_pending)); } EXPORT_SYMBOL_GPL(async_synchronize_full); /** * async_synchronize_full_domain - synchronize all asynchronous function within a certain domain - * @list: running list to synchronize on + * @domain: running list to synchronize on * * This function waits until all asynchronous function calls for the - * synchronization domain specified by the running list @list have been done. + * synchronization domain specified by the running list @domain have been done. */ -void async_synchronize_full_domain(struct list_head *list) +void async_synchronize_full_domain(struct async_domain *domain) { - async_synchronize_cookie_domain(next_cookie, list); + async_synchronize_cookie_domain(next_cookie, domain); } EXPORT_SYMBOL_GPL(async_synchronize_full_domain); @@ -261,11 +261,10 @@ EXPORT_SYMBOL_GPL(async_synchronize_full_domain); * @running: running list to synchronize on * * This function waits until all asynchronous function calls for the - * synchronization domain specified by the running list @list submitted + * synchronization domain specified by running list @running submitted * prior to @cookie have been done. */ -void async_synchronize_cookie_domain(async_cookie_t cookie, - struct list_head *running) +void async_synchronize_cookie_domain(async_cookie_t cookie, struct async_domain *running) { ktime_t uninitialized_var(starttime), delta, endtime; diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index 89eae93445cf..fa1e31206892 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -1545,7 +1545,7 @@ static int dapm_power_widgets(struct snd_soc_dapm_context *dapm, int event) struct snd_soc_dapm_context *d; LIST_HEAD(up_list); LIST_HEAD(down_list); - LIST_HEAD(async_domain); + ASYNC_DOMAIN_EXCLUSIVE(async_domain); enum snd_soc_bias_level bias; trace_snd_soc_dapm_start(card); -- cgit v1.2.3 From a4683487f90bfe3049686fc5c566bdc1ad03ace6 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Mon, 9 Jul 2012 19:33:30 -0700 Subject: [SCSI] async: make async_synchronize_full() flush all work regardless of domain In response to an async related regression James noted: "My theory is that this is an init problem: The assumption in a lot of our code is that async_synchronize_full() waits for everything ... even the domain specific async schedules, which isn't true." ...so make this assumption true. Each domain, including the default one, registers itself on a global domain list when work is scheduled. Once all entries complete it exits that list. Waiting for the list to be empty syncs all in-flight work across all domains. Domains can opt-out of global syncing if they are declared as exclusive ASYNC_DOMAIN_EXCLUSIVE(). All stack-based domains have been declared exclusive since the domain may go out of scope as soon as the last work item completes. Statically declared domains are mostly ok, but async_unregister_domain() is there to close any theoretical races with pending async_synchronize_full waiters at module removal time. Signed-off-by: Dan Williams Acked-by: Arjan van de Ven Reported-by: Meelis Roos Reported-by: Eldad Zack Tested-by: Eldad Zack Signed-off-by: James Bottomley --- drivers/scsi/scsi.c | 1 + include/linux/async.h | 1 + kernel/async.c | 43 +++++++++++++++++++++++++++++++++++++++++-- 3 files changed, 43 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index 4cade886a50a..2936b447cae9 100644 --- a/drivers/scsi/scsi.c +++ b/drivers/scsi/scsi.c @@ -1355,6 +1355,7 @@ static void __exit exit_scsi(void) scsi_exit_devinfo(); scsi_exit_procfs(); scsi_exit_queue(); + async_unregister_domain(&scsi_sd_probe_domain); } subsys_initcall(init_scsi); diff --git a/include/linux/async.h b/include/linux/async.h index 364e7ff16c08..7a24fe9b44b4 100644 --- a/include/linux/async.h +++ b/include/linux/async.h @@ -46,6 +46,7 @@ struct async_domain { extern async_cookie_t async_schedule(async_func_ptr *ptr, void *data); extern async_cookie_t async_schedule_domain(async_func_ptr *ptr, void *data, struct async_domain *domain); +void async_unregister_domain(struct async_domain *domain); extern void async_synchronize_full(void); extern void async_synchronize_full_domain(struct async_domain *domain); extern void async_synchronize_cookie(async_cookie_t cookie); diff --git a/kernel/async.c b/kernel/async.c index ba5491dfa991..9d3118384858 100644 --- a/kernel/async.c +++ b/kernel/async.c @@ -63,7 +63,9 @@ static async_cookie_t next_cookie = 1; static LIST_HEAD(async_pending); static ASYNC_DOMAIN(async_running); +static LIST_HEAD(async_domains); static DEFINE_SPINLOCK(async_lock); +static DEFINE_MUTEX(async_register_mutex); struct async_entry { struct list_head list; @@ -145,6 +147,8 @@ static void async_run_entry_fn(struct work_struct *work) /* 3) remove self from the running queue */ spin_lock_irqsave(&async_lock, flags); list_del(&entry->list); + if (running->registered && --running->count == 0) + list_del_init(&running->node); /* 4) free the entry */ kfree(entry); @@ -187,6 +191,8 @@ static async_cookie_t __async_schedule(async_func_ptr *ptr, void *data, struct a spin_lock_irqsave(&async_lock, flags); newcookie = entry->cookie = next_cookie++; list_add_tail(&entry->list, &async_pending); + if (running->registered && running->count++ == 0) + list_add_tail(&running->node, &async_domains); atomic_inc(&entry_count); spin_unlock_irqrestore(&async_lock, flags); @@ -236,12 +242,42 @@ EXPORT_SYMBOL_GPL(async_schedule_domain); */ void async_synchronize_full(void) { + mutex_lock(&async_register_mutex); do { - async_synchronize_cookie(next_cookie); - } while (!list_empty(&async_running.domain) || !list_empty(&async_pending)); + struct async_domain *domain = NULL; + + spin_lock_irq(&async_lock); + if (!list_empty(&async_domains)) + domain = list_first_entry(&async_domains, typeof(*domain), node); + spin_unlock_irq(&async_lock); + + async_synchronize_cookie_domain(next_cookie, domain); + } while (!list_empty(&async_domains)); + mutex_unlock(&async_register_mutex); } EXPORT_SYMBOL_GPL(async_synchronize_full); +/** + * async_unregister_domain - ensure no more anonymous waiters on this domain + * @domain: idle domain to flush out of any async_synchronize_full instances + * + * async_synchronize_{cookie|full}_domain() are not flushed since callers + * of these routines should know the lifetime of @domain + * + * Prefer ASYNC_DOMAIN_EXCLUSIVE() declarations over flushing + */ +void async_unregister_domain(struct async_domain *domain) +{ + mutex_lock(&async_register_mutex); + spin_lock_irq(&async_lock); + WARN_ON(!domain->registered || !list_empty(&domain->node) || + !list_empty(&domain->domain)); + domain->registered = 0; + spin_unlock_irq(&async_lock); + mutex_unlock(&async_register_mutex); +} +EXPORT_SYMBOL_GPL(async_unregister_domain); + /** * async_synchronize_full_domain - synchronize all asynchronous function within a certain domain * @domain: running list to synchronize on @@ -268,6 +304,9 @@ void async_synchronize_cookie_domain(async_cookie_t cookie, struct async_domain { ktime_t uninitialized_var(starttime), delta, endtime; + if (!running) + return; + if (initcall_debug && system_state == SYSTEM_BOOTING) { printk(KERN_DEBUG "async_waiting @ %i\n", task_pid_nr(current)); starttime = ktime_get(); -- cgit v1.2.3 From bc792e612e78a24ae0b30cc5b85f2368379ba4d4 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Fri, 20 Jul 2012 17:27:37 -0700 Subject: kdb: Revive dmesg command The kgdb dmesg command is broken after the printk rework. The old logic in kdb code makes no sense in terms of current printk/logging storage format, and KDB simply hangs forever. This patch revives the command by switching to kmsg_dumper iterator. The code is now much more simpler and shorter. Signed-off-by: Anton Vorontsov Signed-off-by: Linus Torvalds --- kernel/debug/kdb/kdb_main.c | 91 ++++++++++++++++----------------------------- 1 file changed, 33 insertions(+), 58 deletions(-) (limited to 'kernel') diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c index 67b847dfa2bb..df17c935d3c6 100644 --- a/kernel/debug/kdb/kdb_main.c +++ b/kernel/debug/kdb/kdb_main.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -2040,8 +2041,15 @@ static int kdb_env(int argc, const char **argv) */ static int kdb_dmesg(int argc, const char **argv) { - char *syslog_data[4], *start, *end, c = '\0', *p; - int diag, logging, logsize, lines = 0, adjust = 0, n; + int diag; + int logging; + int lines = 0; + int adjust = 0; + int n = 0; + int skip = 0; + struct kmsg_dumper dumper = { .active = 1 }; + size_t len; + char buf[201]; if (argc > 2) return KDB_ARGCOUNT; @@ -2064,22 +2072,10 @@ static int kdb_dmesg(int argc, const char **argv) kdb_set(2, setargs); } - /* syslog_data[0,1] physical start, end+1. syslog_data[2,3] - * logical start, end+1. */ - kdb_syslog_data(syslog_data); - if (syslog_data[2] == syslog_data[3]) - return 0; - logsize = syslog_data[1] - syslog_data[0]; - start = syslog_data[2]; - end = syslog_data[3]; -#define KDB_WRAP(p) (((p - syslog_data[0]) % logsize) + syslog_data[0]) - for (n = 0, p = start; p < end; ++p) { - c = *KDB_WRAP(p); - if (c == '\n') - ++n; - } - if (c != '\n') - ++n; + kmsg_dump_rewind(&dumper); + while (kmsg_dump_get_line(&dumper, 1, NULL, 0, NULL)) + n++; + if (lines < 0) { if (adjust >= n) kdb_printf("buffer only contains %d lines, nothing " @@ -2087,21 +2083,11 @@ static int kdb_dmesg(int argc, const char **argv) else if (adjust - lines >= n) kdb_printf("buffer only contains %d lines, last %d " "lines printed\n", n, n - adjust); - if (adjust) { - for (; start < end && adjust; ++start) { - if (*KDB_WRAP(start) == '\n') - --adjust; - } - if (start < end) - ++start; - } - for (p = start; p < end && lines; ++p) { - if (*KDB_WRAP(p) == '\n') - ++lines; - } - end = p; + skip = adjust; + lines = abs(lines); } else if (lines > 0) { - int skip = n - (adjust + lines); + skip = n - lines - adjust; + lines = abs(lines); if (adjust >= n) { kdb_printf("buffer only contains %d lines, " "nothing printed\n", n); @@ -2112,35 +2098,24 @@ static int kdb_dmesg(int argc, const char **argv) kdb_printf("buffer only contains %d lines, first " "%d lines printed\n", n, lines); } - for (; start < end && skip; ++start) { - if (*KDB_WRAP(start) == '\n') - --skip; - } - for (p = start; p < end && lines; ++p) { - if (*KDB_WRAP(p) == '\n') - --lines; - } - end = p; + } else { + lines = n; } - /* Do a line at a time (max 200 chars) to reduce protocol overhead */ - c = '\n'; - while (start != end) { - char buf[201]; - p = buf; - if (KDB_FLAG(CMD_INTERRUPT)) - return 0; - while (start < end && (c = *KDB_WRAP(start)) && - (p - buf) < sizeof(buf)-1) { - ++start; - *p++ = c; - if (c == '\n') - break; + + if (skip >= n || skip < 0) + return 0; + + kmsg_dump_rewind(&dumper); + while (kmsg_dump_get_line(&dumper, 1, buf, sizeof(buf), &len)) { + if (skip) { + skip--; + continue; } - *p = '\0'; - kdb_printf("%s", buf); + if (!lines--) + break; + + kdb_printf("%.*s\n", (int)len - 1, buf); } - if (c != '\n') - kdb_printf("\n"); return 0; } -- cgit v1.2.3 From 1b499d05eecbe04969516717a8e15afb6ad80689 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Fri, 20 Jul 2012 17:27:54 -0700 Subject: printk: Remove kdb_syslog_data The function is no longer needed, so remove it. Signed-off-by: Anton Vorontsov Signed-off-by: Linus Torvalds --- kernel/debug/kdb/kdb_private.h | 1 - kernel/printk.c | 15 --------------- 2 files changed, 16 deletions(-) (limited to 'kernel') diff --git a/kernel/debug/kdb/kdb_private.h b/kernel/debug/kdb/kdb_private.h index 47c4e56e513b..392ec6a25844 100644 --- a/kernel/debug/kdb/kdb_private.h +++ b/kernel/debug/kdb/kdb_private.h @@ -205,7 +205,6 @@ extern char kdb_grep_string[]; extern int kdb_grep_leading; extern int kdb_grep_trailing; extern char *kdb_cmds[]; -extern void kdb_syslog_data(char *syslog_data[]); extern unsigned long kdb_task_state_string(const char *); extern char kdb_task_state_char (const struct task_struct *); extern unsigned long kdb_task_state(const struct task_struct *p, diff --git a/kernel/printk.c b/kernel/printk.c index 177fa49357a5..c8129678dfbf 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1192,21 +1192,6 @@ SYSCALL_DEFINE3(syslog, int, type, char __user *, buf, int, len) return do_syslog(type, buf, len, SYSLOG_FROM_CALL); } -#ifdef CONFIG_KGDB_KDB -/* kdb dmesg command needs access to the syslog buffer. do_syslog() - * uses locks so it cannot be used during debugging. Just tell kdb - * where the start and end of the physical and logical logs are. This - * is equivalent to do_syslog(3). - */ -void kdb_syslog_data(char *syslog_data[4]) -{ - syslog_data[0] = log_buf; - syslog_data[1] = log_buf + log_buf_len; - syslog_data[2] = log_buf + log_first_idx; - syslog_data[3] = log_buf + log_next_idx; -} -#endif /* CONFIG_KGDB_KDB */ - static bool __read_mostly ignore_loglevel; static int __init ignore_loglevel_setup(char *str) -- cgit v1.2.3 From 533827c921c34310f63e859e1d6d0feec439657d Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Fri, 20 Jul 2012 17:28:07 -0700 Subject: printk: Implement some unlocked kmsg_dump functions If used from KDB, the locked variants are prone to deadlocks (suppose we got to the debugger w/ the logbuf lock held). So, we have to implement a few routines that grab no logbuf lock. Yet we don't need these functions in modules, so we don't export them. Signed-off-by: Anton Vorontsov Signed-off-by: Linus Torvalds --- include/linux/kmsg_dump.h | 16 +++++++++++ kernel/printk.c | 68 ++++++++++++++++++++++++++++++++++++++--------- 2 files changed, 71 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/include/linux/kmsg_dump.h b/include/linux/kmsg_dump.h index d6bd50110ec2..2e7a1e032c71 100644 --- a/include/linux/kmsg_dump.h +++ b/include/linux/kmsg_dump.h @@ -55,12 +55,17 @@ struct kmsg_dumper { #ifdef CONFIG_PRINTK void kmsg_dump(enum kmsg_dump_reason reason); +bool kmsg_dump_get_line_nolock(struct kmsg_dumper *dumper, bool syslog, + char *line, size_t size, size_t *len); + bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog, char *line, size_t size, size_t *len); bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, char *buf, size_t size, size_t *len); +void kmsg_dump_rewind_nolock(struct kmsg_dumper *dumper); + void kmsg_dump_rewind(struct kmsg_dumper *dumper); int kmsg_dump_register(struct kmsg_dumper *dumper); @@ -71,6 +76,13 @@ static inline void kmsg_dump(enum kmsg_dump_reason reason) { } +static inline bool kmsg_dump_get_line_nolock(struct kmsg_dumper *dumper, + bool syslog, const char *line, + size_t size, size_t *len) +{ + return false; +} + static inline bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog, const char *line, size_t size, size_t *len) { @@ -83,6 +95,10 @@ static inline bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, return false; } +static inline void kmsg_dump_rewind_nolock(struct kmsg_dumper *dumper) +{ +} + static inline void kmsg_dump_rewind(struct kmsg_dumper *dumper) { } diff --git a/kernel/printk.c b/kernel/printk.c index c8129678dfbf..ac4bc9e79465 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -2510,7 +2510,7 @@ void kmsg_dump(enum kmsg_dump_reason reason) } /** - * kmsg_dump_get_line - retrieve one kmsg log line + * kmsg_dump_get_line_nolock - retrieve one kmsg log line (unlocked version) * @dumper: registered kmsg dumper * @syslog: include the "<4>" prefixes * @line: buffer to copy the line to @@ -2525,11 +2525,12 @@ void kmsg_dump(enum kmsg_dump_reason reason) * * A return value of FALSE indicates that there are no more records to * read. + * + * The function is similar to kmsg_dump_get_line(), but grabs no locks. */ -bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog, - char *line, size_t size, size_t *len) +bool kmsg_dump_get_line_nolock(struct kmsg_dumper *dumper, bool syslog, + char *line, size_t size, size_t *len) { - unsigned long flags; struct log *msg; size_t l = 0; bool ret = false; @@ -2537,7 +2538,6 @@ bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog, if (!dumper->active) goto out; - raw_spin_lock_irqsave(&logbuf_lock, flags); if (dumper->cur_seq < log_first_seq) { /* messages are gone, move to first available one */ dumper->cur_seq = log_first_seq; @@ -2545,10 +2545,8 @@ bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog, } /* last entry */ - if (dumper->cur_seq >= log_next_seq) { - raw_spin_unlock_irqrestore(&logbuf_lock, flags); + if (dumper->cur_seq >= log_next_seq) goto out; - } msg = log_from_idx(dumper->cur_idx); l = msg_print_text(msg, 0, syslog, line, size); @@ -2556,12 +2554,41 @@ bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog, dumper->cur_idx = log_next(dumper->cur_idx); dumper->cur_seq++; ret = true; - raw_spin_unlock_irqrestore(&logbuf_lock, flags); out: if (len) *len = l; return ret; } + +/** + * kmsg_dump_get_line - retrieve one kmsg log line + * @dumper: registered kmsg dumper + * @syslog: include the "<4>" prefixes + * @line: buffer to copy the line to + * @size: maximum size of the buffer + * @len: length of line placed into buffer + * + * Start at the beginning of the kmsg buffer, with the oldest kmsg + * record, and copy one record into the provided buffer. + * + * Consecutive calls will return the next available record moving + * towards the end of the buffer with the youngest messages. + * + * A return value of FALSE indicates that there are no more records to + * read. + */ +bool kmsg_dump_get_line(struct kmsg_dumper *dumper, bool syslog, + char *line, size_t size, size_t *len) +{ + unsigned long flags; + bool ret; + + raw_spin_lock_irqsave(&logbuf_lock, flags); + ret = kmsg_dump_get_line_nolock(dumper, syslog, line, size, len); + raw_spin_unlock_irqrestore(&logbuf_lock, flags); + + return ret; +} EXPORT_SYMBOL_GPL(kmsg_dump_get_line); /** @@ -2663,6 +2690,24 @@ out: } EXPORT_SYMBOL_GPL(kmsg_dump_get_buffer); +/** + * kmsg_dump_rewind_nolock - reset the interator (unlocked version) + * @dumper: registered kmsg dumper + * + * Reset the dumper's iterator so that kmsg_dump_get_line() and + * kmsg_dump_get_buffer() can be called again and used multiple + * times within the same dumper.dump() callback. + * + * The function is similar to kmsg_dump_rewind(), but grabs no locks. + */ +void kmsg_dump_rewind_nolock(struct kmsg_dumper *dumper) +{ + dumper->cur_seq = clear_seq; + dumper->cur_idx = clear_idx; + dumper->next_seq = log_next_seq; + dumper->next_idx = log_next_idx; +} + /** * kmsg_dump_rewind - reset the interator * @dumper: registered kmsg dumper @@ -2676,10 +2721,7 @@ void kmsg_dump_rewind(struct kmsg_dumper *dumper) unsigned long flags; raw_spin_lock_irqsave(&logbuf_lock, flags); - dumper->cur_seq = clear_seq; - dumper->cur_idx = clear_idx; - dumper->next_seq = log_next_seq; - dumper->next_idx = log_next_idx; + kmsg_dump_rewind_nolock(dumper); raw_spin_unlock_irqrestore(&logbuf_lock, flags); } EXPORT_SYMBOL_GPL(kmsg_dump_rewind); -- cgit v1.2.3 From c064da47144b11be4697a4611f640086a663016a Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Fri, 20 Jul 2012 17:28:25 -0700 Subject: kdb: Switch to nolock variants of kmsg_dump functions The locked variants are prone to deadlocks (suppose we got to the debugger w/ the logbuf lock held), so let's switch to nolock variants. Signed-off-by: Anton Vorontsov Signed-off-by: Linus Torvalds --- kernel/debug/kdb/kdb_main.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c index df17c935d3c6..1f91413edb87 100644 --- a/kernel/debug/kdb/kdb_main.c +++ b/kernel/debug/kdb/kdb_main.c @@ -2072,8 +2072,8 @@ static int kdb_dmesg(int argc, const char **argv) kdb_set(2, setargs); } - kmsg_dump_rewind(&dumper); - while (kmsg_dump_get_line(&dumper, 1, NULL, 0, NULL)) + kmsg_dump_rewind_nolock(&dumper); + while (kmsg_dump_get_line_nolock(&dumper, 1, NULL, 0, NULL)) n++; if (lines < 0) { @@ -2105,8 +2105,8 @@ static int kdb_dmesg(int argc, const char **argv) if (skip >= n || skip < 0) return 0; - kmsg_dump_rewind(&dumper); - while (kmsg_dump_get_line(&dumper, 1, buf, sizeof(buf), &len)) { + kmsg_dump_rewind_nolock(&dumper); + while (kmsg_dump_get_line_nolock(&dumper, 1, buf, sizeof(buf), &len)) { if (skip) { skip--; continue; -- cgit v1.2.3 From 9a2e03d8ed518a61154f18d83d6466628e519f94 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 19 Jul 2012 13:52:53 -0700 Subject: kthread_worker: reorganize to prepare for flush_kthread_work() reimplementation Make the following two non-functional changes. * Separate out insert_kthread_work() from queue_kthread_work(). * Relocate struct kthread_flush_work and kthread_flush_work_fn() definitions above flush_kthread_work(). v2: Added lockdep_assert_held() in insert_kthread_work() as suggested by Andy Walls. Signed-off-by: Tejun Heo Acked-by: Andy Walls --- kernel/kthread.c | 42 ++++++++++++++++++++++++++---------------- 1 file changed, 26 insertions(+), 16 deletions(-) (limited to 'kernel') diff --git a/kernel/kthread.c b/kernel/kthread.c index 3d3de633702e..4bfbff36d447 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -378,6 +378,19 @@ repeat: } EXPORT_SYMBOL_GPL(kthread_worker_fn); +/* insert @work before @pos in @worker */ +static void insert_kthread_work(struct kthread_worker *worker, + struct kthread_work *work, + struct list_head *pos) +{ + lockdep_assert_held(&worker->lock); + + list_add_tail(&work->node, pos); + work->queue_seq++; + if (likely(worker->task)) + wake_up_process(worker->task); +} + /** * queue_kthread_work - queue a kthread_work * @worker: target kthread_worker @@ -395,10 +408,7 @@ bool queue_kthread_work(struct kthread_worker *worker, spin_lock_irqsave(&worker->lock, flags); if (list_empty(&work->node)) { - list_add_tail(&work->node, &worker->work_list); - work->queue_seq++; - if (likely(worker->task)) - wake_up_process(worker->task); + insert_kthread_work(worker, work, &worker->work_list); ret = true; } spin_unlock_irqrestore(&worker->lock, flags); @@ -406,6 +416,18 @@ bool queue_kthread_work(struct kthread_worker *worker, } EXPORT_SYMBOL_GPL(queue_kthread_work); +struct kthread_flush_work { + struct kthread_work work; + struct completion done; +}; + +static void kthread_flush_work_fn(struct kthread_work *work) +{ + struct kthread_flush_work *fwork = + container_of(work, struct kthread_flush_work, work); + complete(&fwork->done); +} + /** * flush_kthread_work - flush a kthread_work * @work: work to flush @@ -436,18 +458,6 @@ void flush_kthread_work(struct kthread_work *work) } EXPORT_SYMBOL_GPL(flush_kthread_work); -struct kthread_flush_work { - struct kthread_work work; - struct completion done; -}; - -static void kthread_flush_work_fn(struct kthread_work *work) -{ - struct kthread_flush_work *fwork = - container_of(work, struct kthread_flush_work, work); - complete(&fwork->done); -} - /** * flush_kthread_worker - flush all current works on a kthread_worker * @worker: worker to flush -- cgit v1.2.3 From 46f3d976213452350f9d10b0c2780c2681f7075b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 19 Jul 2012 13:52:53 -0700 Subject: kthread_worker: reimplement flush_kthread_work() to allow freeing the work item being executed kthread_worker provides minimalistic workqueue-like interface for users which need a dedicated worker thread (e.g. for realtime priority). It has basic queue, flush_work, flush_worker operations which mostly match the workqueue counterparts; however, due to the way flush_work() is implemented, it has a noticeable difference of not allowing work items to be freed while being executed. While the current users of kthread_worker are okay with the current behavior, the restriction does impede some valid use cases. Also, removing this difference isn't difficult and actually makes the code easier to understand. This patch reimplements flush_kthread_work() such that it uses a flush_work item instead of queue/done sequence numbers. Signed-off-by: Tejun Heo --- include/linux/kthread.h | 8 ++------ kernel/kthread.c | 48 +++++++++++++++++++++++++++--------------------- 2 files changed, 29 insertions(+), 27 deletions(-) (limited to 'kernel') diff --git a/include/linux/kthread.h b/include/linux/kthread.h index 0714b24c0e45..22ccf9dee177 100644 --- a/include/linux/kthread.h +++ b/include/linux/kthread.h @@ -49,8 +49,6 @@ extern int tsk_fork_get_node(struct task_struct *tsk); * can be queued and flushed using queue/flush_kthread_work() * respectively. Queued kthread_works are processed by a kthread * running kthread_worker_fn(). - * - * A kthread_work can't be freed while it is executing. */ struct kthread_work; typedef void (*kthread_work_func_t)(struct kthread_work *work); @@ -59,15 +57,14 @@ struct kthread_worker { spinlock_t lock; struct list_head work_list; struct task_struct *task; + struct kthread_work *current_work; }; struct kthread_work { struct list_head node; kthread_work_func_t func; wait_queue_head_t done; - atomic_t flushing; - int queue_seq; - int done_seq; + struct kthread_worker *worker; }; #define KTHREAD_WORKER_INIT(worker) { \ @@ -79,7 +76,6 @@ struct kthread_work { .node = LIST_HEAD_INIT((work).node), \ .func = (fn), \ .done = __WAIT_QUEUE_HEAD_INITIALIZER((work).done), \ - .flushing = ATOMIC_INIT(0), \ } #define DEFINE_KTHREAD_WORKER(worker) \ diff --git a/kernel/kthread.c b/kernel/kthread.c index 4bfbff36d447..b579af57ea10 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -360,16 +360,12 @@ repeat: struct kthread_work, node); list_del_init(&work->node); } + worker->current_work = work; spin_unlock_irq(&worker->lock); if (work) { __set_current_state(TASK_RUNNING); work->func(work); - smp_wmb(); /* wmb worker-b0 paired with flush-b1 */ - work->done_seq = work->queue_seq; - smp_mb(); /* mb worker-b1 paired with flush-b0 */ - if (atomic_read(&work->flushing)) - wake_up_all(&work->done); } else if (!freezing(current)) schedule(); @@ -386,7 +382,7 @@ static void insert_kthread_work(struct kthread_worker *worker, lockdep_assert_held(&worker->lock); list_add_tail(&work->node, pos); - work->queue_seq++; + work->worker = worker; if (likely(worker->task)) wake_up_process(worker->task); } @@ -436,25 +432,35 @@ static void kthread_flush_work_fn(struct kthread_work *work) */ void flush_kthread_work(struct kthread_work *work) { - int seq = work->queue_seq; + struct kthread_flush_work fwork = { + KTHREAD_WORK_INIT(fwork.work, kthread_flush_work_fn), + COMPLETION_INITIALIZER_ONSTACK(fwork.done), + }; + struct kthread_worker *worker; + bool noop = false; + +retry: + worker = work->worker; + if (!worker) + return; - atomic_inc(&work->flushing); + spin_lock_irq(&worker->lock); + if (work->worker != worker) { + spin_unlock_irq(&worker->lock); + goto retry; + } - /* - * mb flush-b0 paired with worker-b1, to make sure either - * worker sees the above increment or we see done_seq update. - */ - smp_mb__after_atomic_inc(); + if (!list_empty(&work->node)) + insert_kthread_work(worker, &fwork.work, work->node.next); + else if (worker->current_work == work) + insert_kthread_work(worker, &fwork.work, worker->work_list.next); + else + noop = true; - /* A - B <= 0 tests whether B is in front of A regardless of overflow */ - wait_event(work->done, seq - work->done_seq <= 0); - atomic_dec(&work->flushing); + spin_unlock_irq(&worker->lock); - /* - * rmb flush-b1 paired with worker-b0, to make sure our caller - * sees every change made by work->func(). - */ - smp_mb__after_atomic_dec(); + if (!noop) + wait_for_completion(&fwork.done); } EXPORT_SYMBOL_GPL(flush_kthread_work); -- cgit v1.2.3 From 6fec10a1a5866dda3cd6a825a521fc7c2f226ba5 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sun, 22 Jul 2012 10:16:34 -0700 Subject: workqueue: fix spurious CPU locality WARN from process_one_work() 25511a4776 "workqueue: reimplement CPU online rebinding to handle idle workers" added CPU locality sanity check in process_one_work(). It triggers if a worker is executing on a different CPU without UNBOUND or REBIND set. This works for all normal workers but rescuers can trigger this spuriously when they're serving the unbound or a disassociated global_cwq - rescuers don't have either flag set and thus its gcwq->cpu can be a different value including %WORK_CPU_UNBOUND. Fix it by additionally testing %GCWQ_DISASSOCIATED. Signed-off-by: Tejun Heo Reported-by: "Paul E. McKenney" LKML-Refence: <20120721213656.GA7783@linux.vnet.ibm.com> --- kernel/workqueue.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 471996a81633..692d97628a10 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1949,7 +1949,13 @@ __acquires(&gcwq->lock) lockdep_copy_map(&lockdep_map, &work->lockdep_map); #endif + /* + * Ensure we're on the correct CPU. DISASSOCIATED test is + * necessary to avoid spurious warnings from rescuers servicing the + * unbound or a disassociated gcwq. + */ WARN_ON_ONCE(!(worker->flags & (WORKER_UNBOUND | WORKER_REBIND)) && + !(gcwq->flags & GCWQ_DISASSOCIATED) && raw_smp_processor_id() != gcwq->cpu); /* -- cgit v1.2.3 From 7266702805f9d824f92ce5c4069eca65d0f21d28 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 15 Jul 2012 14:10:52 +0400 Subject: signal: make sure we don't get stopped with pending task_work Signed-off-by: Al Viro --- kernel/signal.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'kernel') diff --git a/kernel/signal.c b/kernel/signal.c index 677102789cf2..be4f856d52f8 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1971,6 +1971,13 @@ static void ptrace_do_notify(int signr, int exit_code, int why) void ptrace_notify(int exit_code) { BUG_ON((exit_code & (0x7f | ~0xffff)) != SIGTRAP); + if (unlikely(current->task_works)) { + if (test_and_clear_ti_thread_flag(current_thread_info(), + TIF_NOTIFY_RESUME)) { + smp_mb__after_clear_bit(); + task_work_run(); + } + } spin_lock_irq(¤t->sighand->siglock); ptrace_do_notify(SIGTRAP, exit_code, CLD_TRAPPED); @@ -2191,6 +2198,14 @@ int get_signal_to_deliver(siginfo_t *info, struct k_sigaction *return_ka, struct signal_struct *signal = current->signal; int signr; + if (unlikely(current->task_works)) { + if (test_and_clear_ti_thread_flag(current_thread_info(), + TIF_NOTIFY_RESUME)) { + smp_mb__after_clear_bit(); + task_work_run(); + } + } + if (unlikely(uprobe_deny_signal())) return 0; -- cgit v1.2.3 From 41f9d29f09ca0b22c3631e8a39676e74cda9bcc0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Tue, 26 Jun 2012 22:10:04 +0400 Subject: trimming task_work: kill ->data get rid of the only user of ->data; this is _not_ the final variant - in the end we'll have task_work and rcu_head identical and just use cred->rcu, at which point the separate allocation will be gone completely. Signed-off-by: Al Viro --- include/linux/task_work.h | 4 +--- kernel/irq/manage.c | 2 +- security/keys/internal.h | 4 ++++ security/keys/keyctl.c | 14 ++++++++------ security/keys/process_keys.c | 5 +++-- 5 files changed, 17 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/include/linux/task_work.h b/include/linux/task_work.h index 294d5d5e90b1..627421c0e108 100644 --- a/include/linux/task_work.h +++ b/include/linux/task_work.h @@ -10,14 +10,12 @@ typedef void (*task_work_func_t)(struct task_work *); struct task_work { struct hlist_node hlist; task_work_func_t func; - void *data; }; static inline void -init_task_work(struct task_work *twork, task_work_func_t func, void *data) +init_task_work(struct task_work *twork, task_work_func_t func) { twork->func = func; - twork->data = data; } int task_work_add(struct task_struct *task, struct task_work *twork, bool); diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 8c548232ba39..d1dd54734ce7 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -830,7 +830,7 @@ static int irq_thread(void *data) sched_setscheduler(current, SCHED_FIFO, ¶m); - init_task_work(&on_exit_work, irq_thread_dtor, NULL); + init_task_work(&on_exit_work, irq_thread_dtor); task_work_add(current, &on_exit_work, false); while (!irq_wait_for_interrupt(action)) { diff --git a/security/keys/internal.h b/security/keys/internal.h index 3dcbf86b0d31..b510a316874a 100644 --- a/security/keys/internal.h +++ b/security/keys/internal.h @@ -148,6 +148,10 @@ extern key_ref_t lookup_user_key(key_serial_t id, unsigned long flags, #define KEY_LOOKUP_PARTIAL 0x02 #define KEY_LOOKUP_FOR_UNLINK 0x04 +struct kludge { /* this will die off very soon */ + struct task_work twork; + struct cred *cred; +}; extern long join_session_keyring(const char *name); extern void key_change_session_keyring(struct task_work *twork); diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c index 0f5b3f027299..26723caaad05 100644 --- a/security/keys/keyctl.c +++ b/security/keys/keyctl.c @@ -1456,7 +1456,8 @@ long keyctl_session_to_parent(void) { struct task_struct *me, *parent; const struct cred *mycred, *pcred; - struct task_work *newwork, *oldwork; + struct kludge *newwork; + struct task_work *oldwork; key_ref_t keyring_r; struct cred *cred; int ret; @@ -1466,7 +1467,7 @@ long keyctl_session_to_parent(void) return PTR_ERR(keyring_r); ret = -ENOMEM; - newwork = kmalloc(sizeof(struct task_work), GFP_KERNEL); + newwork = kmalloc(sizeof(struct kludge), GFP_KERNEL); if (!newwork) goto error_keyring; @@ -1478,7 +1479,8 @@ long keyctl_session_to_parent(void) goto error_newwork; cred->tgcred->session_keyring = key_ref_to_ptr(keyring_r); - init_task_work(newwork, key_change_session_keyring, cred); + init_task_work(&newwork->twork, key_change_session_keyring); + newwork->cred = cred; me = current; rcu_read_lock(); @@ -1527,18 +1529,18 @@ long keyctl_session_to_parent(void) /* the replacement session keyring is applied just prior to userspace * restarting */ - ret = task_work_add(parent, newwork, true); + ret = task_work_add(parent, &newwork->twork, true); if (!ret) newwork = NULL; unlock: write_unlock_irq(&tasklist_lock); rcu_read_unlock(); if (oldwork) { - put_cred(oldwork->data); + put_cred(container_of(oldwork, struct kludge, twork)->cred); kfree(oldwork); } if (newwork) { - put_cred(newwork->data); + put_cred(newwork->cred); kfree(newwork); } return ret; diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c index 4ad54eea1ea4..c9b07c97d7f2 100644 --- a/security/keys/process_keys.c +++ b/security/keys/process_keys.c @@ -837,9 +837,10 @@ error: void key_change_session_keyring(struct task_work *twork) { const struct cred *old = current_cred(); - struct cred *new = twork->data; + struct kludge *p = container_of(twork, struct kludge, twork); + struct cred *new = p->cred; - kfree(twork); + kfree(p); if (unlikely(current->flags & PF_EXITING)) { put_cred(new); return; -- cgit v1.2.3 From 158e1645e07f3e9f7e4962d7a0997f5c3b98311b Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 27 Jun 2012 09:24:13 +0400 Subject: trim task_work: get rid of hlist layout based on Oleg's suggestion; single-linked list, task->task_works points to the last element, forward pointer from said last element points to head. I'd still prefer much more regular scheme with two pointers in task_work, but... Signed-off-by: Al Viro --- include/linux/sched.h | 2 +- include/linux/task_work.h | 4 +-- include/linux/tracehook.h | 2 +- kernel/fork.c | 2 +- kernel/task_work.c | 64 ++++++++++++++++++++++++----------------------- 5 files changed, 38 insertions(+), 36 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 4059c0f33f07..b9216ebc2789 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1405,7 +1405,7 @@ struct task_struct { int (*notifier)(void *priv); void *notifier_data; sigset_t *notifier_mask; - struct hlist_head task_works; + void *task_works; struct audit_context *audit_context; #ifdef CONFIG_AUDITSYSCALL diff --git a/include/linux/task_work.h b/include/linux/task_work.h index 627421c0e108..3b3e2c8d037b 100644 --- a/include/linux/task_work.h +++ b/include/linux/task_work.h @@ -8,7 +8,7 @@ struct task_work; typedef void (*task_work_func_t)(struct task_work *); struct task_work { - struct hlist_node hlist; + struct task_work *next; task_work_func_t func; }; @@ -24,7 +24,7 @@ void task_work_run(void); static inline void exit_task_work(struct task_struct *task) { - if (unlikely(!hlist_empty(&task->task_works))) + if (unlikely(task->task_works)) task_work_run(); } diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h index 6a4d82bedb03..1e98b5530425 100644 --- a/include/linux/tracehook.h +++ b/include/linux/tracehook.h @@ -192,7 +192,7 @@ static inline void tracehook_notify_resume(struct pt_regs *regs) * hlist_add_head(task->task_works); */ smp_mb__after_clear_bit(); - if (unlikely(!hlist_empty(¤t->task_works))) + if (unlikely(current->task_works)) task_work_run(); } diff --git a/kernel/fork.c b/kernel/fork.c index ab5211b9e622..bebabad59202 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1415,7 +1415,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, */ p->group_leader = p; INIT_LIST_HEAD(&p->thread_group); - INIT_HLIST_HEAD(&p->task_works); + p->task_works = NULL; /* Now that the task is set up, run cgroup callbacks if * necessary. We need to run them before the task is visible diff --git a/kernel/task_work.c b/kernel/task_work.c index 82d1c794066d..9b8948dbdc60 100644 --- a/kernel/task_work.c +++ b/kernel/task_work.c @@ -19,7 +19,12 @@ task_work_add(struct task_struct *task, struct task_work *twork, bool notify) */ raw_spin_lock_irqsave(&task->pi_lock, flags); if (likely(!(task->flags & PF_EXITING))) { - hlist_add_head(&twork->hlist, &task->task_works); + struct task_work *last = task->task_works; + struct task_work *first = last ? last->next : twork; + twork->next = first; + if (last) + last->next = twork; + task->task_works = twork; err = 0; } raw_spin_unlock_irqrestore(&task->pi_lock, flags); @@ -34,51 +39,48 @@ struct task_work * task_work_cancel(struct task_struct *task, task_work_func_t func) { unsigned long flags; - struct task_work *twork; - struct hlist_node *pos; + struct task_work *last, *res = NULL; raw_spin_lock_irqsave(&task->pi_lock, flags); - hlist_for_each_entry(twork, pos, &task->task_works, hlist) { - if (twork->func == func) { - hlist_del(&twork->hlist); - goto found; + last = task->task_works; + if (last) { + struct task_work *q = last, *p = q->next; + while (1) { + if (p->func == func) { + q->next = p->next; + if (p == last) + task->task_works = q == p ? NULL : q; + res = p; + break; + } + if (p == last) + break; + q = p; + p = q->next; } } - twork = NULL; - found: raw_spin_unlock_irqrestore(&task->pi_lock, flags); - - return twork; + return res; } void task_work_run(void) { struct task_struct *task = current; - struct hlist_head task_works; - struct hlist_node *pos; + struct task_work *p, *q; raw_spin_lock_irq(&task->pi_lock); - hlist_move_list(&task->task_works, &task_works); + p = task->task_works; + task->task_works = NULL; raw_spin_unlock_irq(&task->pi_lock); - if (unlikely(hlist_empty(&task_works))) + if (unlikely(!p)) return; - /* - * We use hlist to save the space in task_struct, but we want fifo. - * Find the last entry, the list should be short, then process them - * in reverse order. - */ - for (pos = task_works.first; pos->next; pos = pos->next) - ; - - for (;;) { - struct hlist_node **pprev = pos->pprev; - struct task_work *twork = container_of(pos, struct task_work, - hlist); - twork->func(twork); - if (pprev == &task_works.first) - break; - pos = container_of(pprev, struct hlist_node, next); + q = p->next; /* head */ + p->next = NULL; /* cut it */ + while (q) { + p = q->next; + q->func(q); + q = p; } } -- cgit v1.2.3 From 67d1214551e800f9fe7dc7c47a346d2df0fafed5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 27 Jun 2012 11:07:19 +0400 Subject: merge task_work and rcu_head, get rid of separate allocation for keyring case task_work and rcu_head are identical now; merge them (calling the result struct callback_head, rcu_head #define'd to it), kill separate allocation in security/keys since we can just use cred->rcu now. Signed-off-by: Al Viro --- include/linux/sched.h | 2 +- include/linux/task_work.h | 14 ++++---------- include/linux/types.h | 9 +++++---- kernel/irq/manage.c | 4 ++-- kernel/task_work.c | 14 +++++++------- security/keys/internal.h | 6 +----- security/keys/keyctl.c | 28 +++++++++------------------- security/keys/process_keys.c | 6 ++---- 8 files changed, 31 insertions(+), 52 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index b9216ebc2789..af3555cc760f 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1405,7 +1405,7 @@ struct task_struct { int (*notifier)(void *priv); void *notifier_data; sigset_t *notifier_mask; - void *task_works; + struct callback_head *task_works; struct audit_context *audit_context; #ifdef CONFIG_AUDITSYSCALL diff --git a/include/linux/task_work.h b/include/linux/task_work.h index 3b3e2c8d037b..fb46b03b1852 100644 --- a/include/linux/task_work.h +++ b/include/linux/task_work.h @@ -4,22 +4,16 @@ #include #include -struct task_work; -typedef void (*task_work_func_t)(struct task_work *); - -struct task_work { - struct task_work *next; - task_work_func_t func; -}; +typedef void (*task_work_func_t)(struct callback_head *); static inline void -init_task_work(struct task_work *twork, task_work_func_t func) +init_task_work(struct callback_head *twork, task_work_func_t func) { twork->func = func; } -int task_work_add(struct task_struct *task, struct task_work *twork, bool); -struct task_work *task_work_cancel(struct task_struct *, task_work_func_t); +int task_work_add(struct task_struct *task, struct callback_head *twork, bool); +struct callback_head *task_work_cancel(struct task_struct *, task_work_func_t); void task_work_run(void); static inline void exit_task_work(struct task_struct *task) diff --git a/include/linux/types.h b/include/linux/types.h index 9c1bd539ea70..bf0dd7524b2a 100644 --- a/include/linux/types.h +++ b/include/linux/types.h @@ -246,14 +246,15 @@ struct ustat { }; /** - * struct rcu_head - callback structure for use with RCU + * struct callback_head - callback structure for use with RCU and task_work * @next: next update requests in a list * @func: actual update function to call after the grace period. */ -struct rcu_head { - struct rcu_head *next; - void (*func)(struct rcu_head *head); +struct callback_head { + struct callback_head *next; + void (*func)(struct callback_head *head); }; +#define rcu_head callback_head #endif /* __KERNEL__ */ #endif /* __ASSEMBLY__ */ diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index d1dd54734ce7..814c9ef6bba1 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -781,7 +781,7 @@ static void wake_threads_waitq(struct irq_desc *desc) wake_up(&desc->wait_for_threads); } -static void irq_thread_dtor(struct task_work *unused) +static void irq_thread_dtor(struct callback_head *unused) { struct task_struct *tsk = current; struct irq_desc *desc; @@ -813,7 +813,7 @@ static void irq_thread_dtor(struct task_work *unused) */ static int irq_thread(void *data) { - struct task_work on_exit_work; + struct callback_head on_exit_work; static const struct sched_param param = { .sched_priority = MAX_USER_RT_PRIO/2, }; diff --git a/kernel/task_work.c b/kernel/task_work.c index 9b8948dbdc60..76266fb665dc 100644 --- a/kernel/task_work.c +++ b/kernel/task_work.c @@ -3,7 +3,7 @@ #include int -task_work_add(struct task_struct *task, struct task_work *twork, bool notify) +task_work_add(struct task_struct *task, struct callback_head *twork, bool notify) { unsigned long flags; int err = -ESRCH; @@ -19,8 +19,8 @@ task_work_add(struct task_struct *task, struct task_work *twork, bool notify) */ raw_spin_lock_irqsave(&task->pi_lock, flags); if (likely(!(task->flags & PF_EXITING))) { - struct task_work *last = task->task_works; - struct task_work *first = last ? last->next : twork; + struct callback_head *last = task->task_works; + struct callback_head *first = last ? last->next : twork; twork->next = first; if (last) last->next = twork; @@ -35,16 +35,16 @@ task_work_add(struct task_struct *task, struct task_work *twork, bool notify) return err; } -struct task_work * +struct callback_head * task_work_cancel(struct task_struct *task, task_work_func_t func) { unsigned long flags; - struct task_work *last, *res = NULL; + struct callback_head *last, *res = NULL; raw_spin_lock_irqsave(&task->pi_lock, flags); last = task->task_works; if (last) { - struct task_work *q = last, *p = q->next; + struct callback_head *q = last, *p = q->next; while (1) { if (p->func == func) { q->next = p->next; @@ -66,7 +66,7 @@ task_work_cancel(struct task_struct *task, task_work_func_t func) void task_work_run(void) { struct task_struct *task = current; - struct task_work *p, *q; + struct callback_head *p, *q; raw_spin_lock_irq(&task->pi_lock); p = task->task_works; diff --git a/security/keys/internal.h b/security/keys/internal.h index b510a316874a..c246ba5d43ab 100644 --- a/security/keys/internal.h +++ b/security/keys/internal.h @@ -148,12 +148,8 @@ extern key_ref_t lookup_user_key(key_serial_t id, unsigned long flags, #define KEY_LOOKUP_PARTIAL 0x02 #define KEY_LOOKUP_FOR_UNLINK 0x04 -struct kludge { /* this will die off very soon */ - struct task_work twork; - struct cred *cred; -}; extern long join_session_keyring(const char *name); -extern void key_change_session_keyring(struct task_work *twork); +extern void key_change_session_keyring(struct callback_head *twork); extern struct work_struct key_gc_work; extern unsigned key_gc_delay; diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c index 26723caaad05..0291b3f9397c 100644 --- a/security/keys/keyctl.c +++ b/security/keys/keyctl.c @@ -1456,8 +1456,7 @@ long keyctl_session_to_parent(void) { struct task_struct *me, *parent; const struct cred *mycred, *pcred; - struct kludge *newwork; - struct task_work *oldwork; + struct callback_head *newwork, *oldwork; key_ref_t keyring_r; struct cred *cred; int ret; @@ -1467,20 +1466,17 @@ long keyctl_session_to_parent(void) return PTR_ERR(keyring_r); ret = -ENOMEM; - newwork = kmalloc(sizeof(struct kludge), GFP_KERNEL); - if (!newwork) - goto error_keyring; /* our parent is going to need a new cred struct, a new tgcred struct * and new security data, so we allocate them here to prevent ENOMEM in * our parent */ cred = cred_alloc_blank(); if (!cred) - goto error_newwork; + goto error_keyring; + newwork = &cred->rcu; cred->tgcred->session_keyring = key_ref_to_ptr(keyring_r); - init_task_work(&newwork->twork, key_change_session_keyring); - newwork->cred = cred; + init_task_work(newwork, key_change_session_keyring); me = current; rcu_read_lock(); @@ -1529,24 +1525,18 @@ long keyctl_session_to_parent(void) /* the replacement session keyring is applied just prior to userspace * restarting */ - ret = task_work_add(parent, &newwork->twork, true); + ret = task_work_add(parent, newwork, true); if (!ret) newwork = NULL; unlock: write_unlock_irq(&tasklist_lock); rcu_read_unlock(); - if (oldwork) { - put_cred(container_of(oldwork, struct kludge, twork)->cred); - kfree(oldwork); - } - if (newwork) { - put_cred(newwork->cred); - kfree(newwork); - } + if (oldwork) + put_cred(container_of(oldwork, struct cred, rcu)); + if (newwork) + put_cred(cred); return ret; -error_newwork: - kfree(newwork); error_keyring: key_ref_put(keyring_r); return ret; diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c index c9b07c97d7f2..54339cfd6734 100644 --- a/security/keys/process_keys.c +++ b/security/keys/process_keys.c @@ -834,13 +834,11 @@ error: * Replace a process's session keyring on behalf of one of its children when * the target process is about to resume userspace execution. */ -void key_change_session_keyring(struct task_work *twork) +void key_change_session_keyring(struct callback_head *twork) { const struct cred *old = current_cred(); - struct kludge *p = container_of(twork, struct kludge, twork); - struct cred *new = p->cred; + struct cred *new = container_of(twork, struct cred, rcu); - kfree(p); if (unlikely(current->flags & PF_EXITING)) { put_cred(new); return; -- cgit v1.2.3 From ed3e694d78cc75fa79bf29698631b146fd27aa35 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 27 Jun 2012 11:31:24 +0400 Subject: move exit_task_work() past exit_files() et.al. ... and get rid of PF_EXITING check in task_work_add(). Signed-off-by: Al Viro --- kernel/exit.c | 6 ++---- kernel/task_work.c | 30 +++++++++++------------------- 2 files changed, 13 insertions(+), 23 deletions(-) (limited to 'kernel') diff --git a/kernel/exit.c b/kernel/exit.c index 2f59cc334516..d17f6c4ddfa9 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -953,14 +953,11 @@ void do_exit(long code) exit_signals(tsk); /* sets PF_EXITING */ /* * tsk->flags are checked in the futex code to protect against - * an exiting task cleaning up the robust pi futexes, and in - * task_work_add() to avoid the race with exit_task_work(). + * an exiting task cleaning up the robust pi futexes. */ smp_mb(); raw_spin_unlock_wait(&tsk->pi_lock); - exit_task_work(tsk); - if (unlikely(in_atomic())) printk(KERN_INFO "note: %s[%d] exited with preempt_count %d\n", current->comm, task_pid_nr(current), @@ -995,6 +992,7 @@ void do_exit(long code) exit_shm(tsk); exit_files(tsk); exit_fs(tsk); + exit_task_work(tsk); check_stack_usage(); exit_thread(); diff --git a/kernel/task_work.c b/kernel/task_work.c index 76266fb665dc..fb396089f66a 100644 --- a/kernel/task_work.c +++ b/kernel/task_work.c @@ -5,34 +5,26 @@ int task_work_add(struct task_struct *task, struct callback_head *twork, bool notify) { + struct callback_head *last, *first; unsigned long flags; - int err = -ESRCH; -#ifndef TIF_NOTIFY_RESUME - if (notify) - return -ENOTSUPP; -#endif /* - * We must not insert the new work if the task has already passed - * exit_task_work(). We rely on do_exit()->raw_spin_unlock_wait() - * and check PF_EXITING under pi_lock. + * Not inserting the new work if the task has already passed + * exit_task_work() is the responisbility of callers. */ raw_spin_lock_irqsave(&task->pi_lock, flags); - if (likely(!(task->flags & PF_EXITING))) { - struct callback_head *last = task->task_works; - struct callback_head *first = last ? last->next : twork; - twork->next = first; - if (last) - last->next = twork; - task->task_works = twork; - err = 0; - } + last = task->task_works; + first = last ? last->next : twork; + twork->next = first; + if (last) + last->next = twork; + task->task_works = twork; raw_spin_unlock_irqrestore(&task->pi_lock, flags); /* test_and_set_bit() implies mb(), see tracehook_notify_resume(). */ - if (likely(!err) && notify) + if (notify) set_notify_resume(task); - return err; + return 0; } struct callback_head * -- cgit v1.2.3 From a2d4c71d1559426155e5da8db3265bfa0d8d398d Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 27 Jun 2012 11:33:29 +0400 Subject: deal with task_work callbacks adding more work It doesn't matter on normal return to userland path (we'll recheck the NOTIFY_RESUME flag anyway), but in case of exit_task_work() we'll need that as soon as we get callbacks capable of triggering more task_work_add(). Signed-off-by: Al Viro --- kernel/task_work.c | 26 ++++++++++++++------------ 1 file changed, 14 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/kernel/task_work.c b/kernel/task_work.c index fb396089f66a..91d4e1742a0c 100644 --- a/kernel/task_work.c +++ b/kernel/task_work.c @@ -60,19 +60,21 @@ void task_work_run(void) struct task_struct *task = current; struct callback_head *p, *q; - raw_spin_lock_irq(&task->pi_lock); - p = task->task_works; - task->task_works = NULL; - raw_spin_unlock_irq(&task->pi_lock); + while (1) { + raw_spin_lock_irq(&task->pi_lock); + p = task->task_works; + task->task_works = NULL; + raw_spin_unlock_irq(&task->pi_lock); - if (unlikely(!p)) - return; + if (unlikely(!p)) + return; - q = p->next; /* head */ - p->next = NULL; /* cut it */ - while (q) { - p = q->next; - q->func(q); - q = p; + q = p->next; /* head */ + p->next = NULL; /* cut it */ + while (q) { + p = q->next; + q->func(q); + q = p; + } } } -- cgit v1.2.3 From d35be8bab9b0ce44bed4b9453f86ebf64062721e Mon Sep 17 00:00:00 2001 From: "Srivatsa S. Bhat" Date: Thu, 24 May 2012 19:46:26 +0530 Subject: CPU hotplug, cpusets, suspend: Don't modify cpusets during suspend/resume In the event of CPU hotplug, the kernel modifies the cpusets' cpus_allowed masks as and when necessary to ensure that the tasks belonging to the cpusets have some place (online CPUs) to run on. And regular CPU hotplug is destructive in the sense that the kernel doesn't remember the original cpuset configurations set by the user, across hotplug operations. However, suspend/resume (which uses CPU hotplug) is a special case in which the kernel has the responsibility to restore the system (during resume), to exactly the same state it was in before suspend. In order to achieve that, do the following: 1. Don't modify cpusets during suspend/resume. At all. In particular, don't move the tasks from one cpuset to another, and don't modify any cpuset's cpus_allowed mask. So, simply ignore cpusets during the CPU hotplug operations that are carried out in the suspend/resume path. 2. However, cpusets and sched domains are related. We just want to avoid altering cpusets alone. So, to keep the sched domains updated, build a single sched domain (containing all active cpus) during each of the CPU hotplug operations carried out in s/r path, effectively ignoring the cpusets' cpus_allowed masks. (Since userspace is frozen while doing all this, it will go unnoticed.) 3. During the last CPU online operation during resume, build the sched domains by looking up the (unaltered) cpusets' cpus_allowed masks. That will bring back the system to the same original state as it was in before suspend. Ultimately, this will not only solve the cpuset problem related to suspend resume (ie., restores the cpusets to exactly what it was before suspend, by not touching it at all) but also speeds up suspend/resume because we avoid running cpuset update code for every CPU being offlined/onlined. Signed-off-by: Srivatsa S. Bhat Signed-off-by: Peter Zijlstra Cc: Linus Torvalds Cc: Andrew Morton Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20120524141611.3692.20155.stgit@srivatsabhat.in.ibm.com Signed-off-by: Ingo Molnar --- kernel/cpuset.c | 3 +++ kernel/sched/core.c | 40 ++++++++++++++++++++++++++++++++++++---- 2 files changed, 39 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 8c8bd652dd12..746d1eeb5dbe 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -2054,6 +2054,9 @@ static void scan_for_empty_cpusets(struct cpuset *root) * (of no affect) on systems that are actively using CPU hotplug * but making no active use of cpusets. * + * The only exception to this is suspend/resume, where we don't + * modify cpusets at all. + * * This routine ensures that top_cpuset.cpus_allowed tracks * cpu_active_mask on each CPU hotplug (cpuhp) event. * diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 468bdd44c1ba..4c1d80c6b318 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7097,34 +7097,66 @@ match2: mutex_unlock(&sched_domains_mutex); } +static int num_cpus_frozen; /* used to mark begin/end of suspend/resume */ + /* * Update cpusets according to cpu_active mask. If cpusets are * disabled, cpuset_update_active_cpus() becomes a simple wrapper * around partition_sched_domains(). + * + * If we come here as part of a suspend/resume, don't touch cpusets because we + * want to restore it back to its original state upon resume anyway. */ static int cpuset_cpu_active(struct notifier_block *nfb, unsigned long action, void *hcpu) { - switch (action & ~CPU_TASKS_FROZEN) { + switch (action) { + case CPU_ONLINE_FROZEN: + case CPU_DOWN_FAILED_FROZEN: + + /* + * num_cpus_frozen tracks how many CPUs are involved in suspend + * resume sequence. As long as this is not the last online + * operation in the resume sequence, just build a single sched + * domain, ignoring cpusets. + */ + num_cpus_frozen--; + if (likely(num_cpus_frozen)) { + partition_sched_domains(1, NULL, NULL); + break; + } + + /* + * This is the last CPU online operation. So fall through and + * restore the original sched domains by considering the + * cpuset configurations. + */ + case CPU_ONLINE: case CPU_DOWN_FAILED: cpuset_update_active_cpus(); - return NOTIFY_OK; + break; default: return NOTIFY_DONE; } + return NOTIFY_OK; } static int cpuset_cpu_inactive(struct notifier_block *nfb, unsigned long action, void *hcpu) { - switch (action & ~CPU_TASKS_FROZEN) { + switch (action) { case CPU_DOWN_PREPARE: cpuset_update_active_cpus(); - return NOTIFY_OK; + break; + case CPU_DOWN_PREPARE_FROZEN: + num_cpus_frozen++; + partition_sched_domains(1, NULL, NULL); + break; default: return NOTIFY_DONE; } + return NOTIFY_OK; } void __init sched_init_smp(void) -- cgit v1.2.3 From 80d1fa6463d934969b7aebf04107fc133463f0f6 Mon Sep 17 00:00:00 2001 From: "Srivatsa S. Bhat" Date: Thu, 24 May 2012 19:46:41 +0530 Subject: cpusets, hotplug: Implement cpuset tree traversal in a helper function At present, the functions that deal with cpusets during CPU/Mem hotplug are quite messy, since a lot of the functionality is mixed up without clear separation. And this takes a toll on optimization as well. For example, the function cpuset_update_active_cpus() is called on both CPU offline and CPU online events; and it invokes scan_for_empty_cpusets(), which makes sense only for CPU offline events. And hence, the current code ends up unnecessarily traversing the cpuset tree during CPU online also. As a first step towards cleaning up those functions, encapsulate the cpuset tree traversal in a helper function, so as to facilitate upcoming changes. Signed-off-by: Srivatsa S. Bhat Signed-off-by: Peter Zijlstra Cc: Linus Torvalds Cc: Andrew Morton Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20120524141635.3692.893.stgit@srivatsabhat.in.ibm.com Signed-off-by: Ingo Molnar --- kernel/cpuset.c | 36 +++++++++++++++++++++++++++--------- 1 file changed, 27 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 746d1eeb5dbe..ba96349aa522 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -1989,6 +1989,32 @@ static void remove_tasks_in_empty_cpuset(struct cpuset *cs) move_member_tasks_to_cpuset(cs, parent); } +/* + * Helper function to traverse cpusets. + * It can be used to walk the cpuset tree from top to bottom, completing + * one layer before dropping down to the next (thus always processing a + * node before any of its children). + */ +static struct cpuset *cpuset_next(struct list_head *queue) +{ + struct cpuset *cp; + struct cpuset *child; /* scans child cpusets of cp */ + struct cgroup *cont; + + if (list_empty(queue)) + return NULL; + + cp = list_first_entry(queue, struct cpuset, stack_list); + list_del(queue->next); + list_for_each_entry(cont, &cp->css.cgroup->children, sibling) { + child = cgroup_cs(cont); + list_add_tail(&child->stack_list, queue); + } + + return cp; +} + + /* * Walk the specified cpuset subtree and look for empty cpusets. * The tasks of such cpuset must be moved to a parent cpuset. @@ -2008,19 +2034,11 @@ static void scan_for_empty_cpusets(struct cpuset *root) { LIST_HEAD(queue); struct cpuset *cp; /* scans cpusets being updated */ - struct cpuset *child; /* scans child cpusets of cp */ - struct cgroup *cont; static nodemask_t oldmems; /* protected by cgroup_mutex */ list_add_tail((struct list_head *)&root->stack_list, &queue); - while (!list_empty(&queue)) { - cp = list_first_entry(&queue, struct cpuset, stack_list); - list_del(queue.next); - list_for_each_entry(cont, &cp->css.cgroup->children, sibling) { - child = cgroup_cs(cont); - list_add_tail(&child->stack_list, &queue); - } + while ((cp = cpuset_next(&queue)) != NULL) { /* Continue past cpusets with all cpus, mems online */ if (cpumask_subset(cp->cpus_allowed, cpu_active_mask) && -- cgit v1.2.3 From 7ddf96b02fe8dd441f452deef879040def5f7b34 Mon Sep 17 00:00:00 2001 From: "Srivatsa S. Bhat" Date: Thu, 24 May 2012 19:46:55 +0530 Subject: cpusets, hotplug: Restructure functions that are invoked during hotplug Separate out the cpuset related handling for CPU/Memory online/offline. This also helps us exploit the most obvious and basic level of optimization that any notification mechanism (CPU/Mem online/offline) has to offer us: "We *know* why we have been invoked. So stop pretending that we are lost, and do only the necessary amount of processing!". And while at it, rename scan_for_empty_cpusets() to scan_cpusets_upon_hotplug(), which is more appropriate considering how it is restructured. Signed-off-by: Srivatsa S. Bhat Signed-off-by: Peter Zijlstra Cc: Linus Torvalds Cc: Andrew Morton Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20120524141650.3692.48637.stgit@srivatsabhat.in.ibm.com Signed-off-by: Ingo Molnar --- include/linux/cpuset.h | 4 +-- kernel/cpuset.c | 88 ++++++++++++++++++++++++++++++++++---------------- kernel/sched/core.c | 4 +-- 3 files changed, 65 insertions(+), 31 deletions(-) (limited to 'kernel') diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index 668f66baac7b..838320fc3d1d 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -20,7 +20,7 @@ extern int number_of_cpusets; /* How many cpusets are defined in system? */ extern int cpuset_init(void); extern void cpuset_init_smp(void); -extern void cpuset_update_active_cpus(void); +extern void cpuset_update_active_cpus(bool cpu_online); extern void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask); extern void cpuset_cpus_allowed_fallback(struct task_struct *p); extern nodemask_t cpuset_mems_allowed(struct task_struct *p); @@ -124,7 +124,7 @@ static inline void set_mems_allowed(nodemask_t nodemask) static inline int cpuset_init(void) { return 0; } static inline void cpuset_init_smp(void) {} -static inline void cpuset_update_active_cpus(void) +static inline void cpuset_update_active_cpus(bool cpu_online) { partition_sched_domains(1, NULL, NULL); } diff --git a/kernel/cpuset.c b/kernel/cpuset.c index ba96349aa522..ba0a4d74d262 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -147,6 +147,12 @@ typedef enum { CS_SPREAD_SLAB, } cpuset_flagbits_t; +/* the type of hotplug event */ +enum hotplug_event { + CPUSET_CPU_OFFLINE, + CPUSET_MEM_OFFLINE, +}; + /* convenient tests for these bits */ static inline int is_cpu_exclusive(const struct cpuset *cs) { @@ -2016,8 +2022,10 @@ static struct cpuset *cpuset_next(struct list_head *queue) /* - * Walk the specified cpuset subtree and look for empty cpusets. - * The tasks of such cpuset must be moved to a parent cpuset. + * Walk the specified cpuset subtree upon a hotplug operation (CPU/Memory + * online/offline) and update the cpusets accordingly. + * For regular CPU/Mem hotplug, look for empty cpusets; the tasks of such + * cpuset must be moved to a parent cpuset. * * Called with cgroup_mutex held. We take callback_mutex to modify * cpus_allowed and mems_allowed. @@ -2030,38 +2038,58 @@ static struct cpuset *cpuset_next(struct list_head *queue) * that has tasks along with an empty 'mems'. But if we did see such * a cpuset, we'd handle it just like we do if its 'cpus' was empty. */ -static void scan_for_empty_cpusets(struct cpuset *root) +static void +scan_cpusets_upon_hotplug(struct cpuset *root, enum hotplug_event event) { LIST_HEAD(queue); - struct cpuset *cp; /* scans cpusets being updated */ + struct cpuset *cp; /* scans cpusets being updated */ static nodemask_t oldmems; /* protected by cgroup_mutex */ list_add_tail((struct list_head *)&root->stack_list, &queue); - while ((cp = cpuset_next(&queue)) != NULL) { + switch (event) { + case CPUSET_CPU_OFFLINE: + while ((cp = cpuset_next(&queue)) != NULL) { + + /* Continue past cpusets with all cpus online */ + if (cpumask_subset(cp->cpus_allowed, cpu_active_mask)) + continue; + + /* Remove offline cpus from this cpuset. */ + mutex_lock(&callback_mutex); + cpumask_and(cp->cpus_allowed, cp->cpus_allowed, + cpu_active_mask); + mutex_unlock(&callback_mutex); + + /* Move tasks from the empty cpuset to a parent */ + if (cpumask_empty(cp->cpus_allowed)) + remove_tasks_in_empty_cpuset(cp); + else + update_tasks_cpumask(cp, NULL); + } + break; - /* Continue past cpusets with all cpus, mems online */ - if (cpumask_subset(cp->cpus_allowed, cpu_active_mask) && - nodes_subset(cp->mems_allowed, node_states[N_HIGH_MEMORY])) - continue; + case CPUSET_MEM_OFFLINE: + while ((cp = cpuset_next(&queue)) != NULL) { - oldmems = cp->mems_allowed; + /* Continue past cpusets with all mems online */ + if (nodes_subset(cp->mems_allowed, + node_states[N_HIGH_MEMORY])) + continue; - /* Remove offline cpus and mems from this cpuset. */ - mutex_lock(&callback_mutex); - cpumask_and(cp->cpus_allowed, cp->cpus_allowed, - cpu_active_mask); - nodes_and(cp->mems_allowed, cp->mems_allowed, + oldmems = cp->mems_allowed; + + /* Remove offline mems from this cpuset. */ + mutex_lock(&callback_mutex); + nodes_and(cp->mems_allowed, cp->mems_allowed, node_states[N_HIGH_MEMORY]); - mutex_unlock(&callback_mutex); + mutex_unlock(&callback_mutex); - /* Move tasks from the empty cpuset to a parent */ - if (cpumask_empty(cp->cpus_allowed) || - nodes_empty(cp->mems_allowed)) - remove_tasks_in_empty_cpuset(cp); - else { - update_tasks_cpumask(cp, NULL); - update_tasks_nodemask(cp, &oldmems, NULL); + /* Move tasks from the empty cpuset to a parent */ + if (nodes_empty(cp->mems_allowed)) + remove_tasks_in_empty_cpuset(cp); + else + update_tasks_nodemask(cp, &oldmems, NULL); } } } @@ -2080,8 +2108,11 @@ static void scan_for_empty_cpusets(struct cpuset *root) * * Called within get_online_cpus(). Needs to call cgroup_lock() * before calling generate_sched_domains(). + * + * @cpu_online: Indicates whether this is a CPU online event (true) or + * a CPU offline event (false). */ -void cpuset_update_active_cpus(void) +void cpuset_update_active_cpus(bool cpu_online) { struct sched_domain_attr *attr; cpumask_var_t *doms; @@ -2091,7 +2122,10 @@ void cpuset_update_active_cpus(void) mutex_lock(&callback_mutex); cpumask_copy(top_cpuset.cpus_allowed, cpu_active_mask); mutex_unlock(&callback_mutex); - scan_for_empty_cpusets(&top_cpuset); + + if (!cpu_online) + scan_cpusets_upon_hotplug(&top_cpuset, CPUSET_CPU_OFFLINE); + ndoms = generate_sched_domains(&doms, &attr); cgroup_unlock(); @@ -2122,9 +2156,9 @@ static int cpuset_track_online_nodes(struct notifier_block *self, case MEM_OFFLINE: /* * needn't update top_cpuset.mems_allowed explicitly because - * scan_for_empty_cpusets() will update it. + * scan_cpusets_upon_hotplug() will update it. */ - scan_for_empty_cpusets(&top_cpuset); + scan_cpusets_upon_hotplug(&top_cpuset, CPUSET_MEM_OFFLINE); break; default: break; diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 4c1d80c6b318..4b4a63d34396 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7134,7 +7134,7 @@ static int cpuset_cpu_active(struct notifier_block *nfb, unsigned long action, case CPU_ONLINE: case CPU_DOWN_FAILED: - cpuset_update_active_cpus(); + cpuset_update_active_cpus(true); break; default: return NOTIFY_DONE; @@ -7147,7 +7147,7 @@ static int cpuset_cpu_inactive(struct notifier_block *nfb, unsigned long action, { switch (action) { case CPU_DOWN_PREPARE: - cpuset_update_active_cpus(); + cpuset_update_active_cpus(false); break; case CPU_DOWN_PREPARE_FROZEN: num_cpus_frozen++; -- cgit v1.2.3 From a1cd2b13f754b2c56fb87b8c4912c015f8f57c0c Mon Sep 17 00:00:00 2001 From: "Srivatsa S. Bhat" Date: Thu, 24 May 2012 19:47:03 +0530 Subject: cpusets: Remove/update outdated comments cpuset_track_online_cpus() is no longer present. So remove the outdated comment and replace it with reference to cpuset_update_active_cpus() which is its equivalent. Also, we don't lack memory hot-unplug anymore. And David Rientjes pointed out how it is dealt with. So update that comment as well. Signed-off-by: Srivatsa S. Bhat Signed-off-by: Peter Zijlstra Cc: Linus Torvalds Cc: Andrew Morton Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20120524141700.3692.98192.stgit@srivatsabhat.in.ibm.com Signed-off-by: Ingo Molnar --- kernel/cpuset.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/cpuset.c b/kernel/cpuset.c index ba0a4d74d262..f33c7153b6d7 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -2034,9 +2034,8 @@ static struct cpuset *cpuset_next(struct list_head *queue) * before dropping down to the next. It always processes a node before * any of its children. * - * For now, since we lack memory hot unplug, we'll never see a cpuset - * that has tasks along with an empty 'mems'. But if we did see such - * a cpuset, we'd handle it just like we do if its 'cpus' was empty. + * In the case of memory hot-unplug, it will remove nodes from N_HIGH_MEMORY + * if all present pages from a node are offlined. */ static void scan_cpusets_upon_hotplug(struct cpuset *root, enum hotplug_event event) @@ -2137,7 +2136,7 @@ void cpuset_update_active_cpus(bool cpu_online) /* * Keep top_cpuset.mems_allowed tracking node_states[N_HIGH_MEMORY]. * Call this routine anytime after node_states[N_HIGH_MEMORY] changes. - * See also the previous routine cpuset_track_online_cpus(). + * See cpuset_update_active_cpus() for CPU hotplug handling. */ static int cpuset_track_online_nodes(struct notifier_block *self, unsigned long action, void *arg) -- cgit v1.2.3 From 970e178985cadbca660feb02f4d2ee3a09f7fdda Mon Sep 17 00:00:00 2001 From: Mike Galbraith Date: Tue, 12 Jun 2012 05:18:32 +0200 Subject: sched: Improve scalability via 'CPU buddies', which withstand random perturbations Traversing an entire package is not only expensive, it also leads to tasks bouncing all over a partially idle and possible quite large package. Fix that up by assigning a 'buddy' CPU to try to motivate. Each buddy may try to motivate that one other CPU, if it's busy, tough, it may then try its SMT sibling, but that's all this optimization is allowed to cost. Sibling cache buddies are cross-wired to prevent bouncing. 4 socket 40 core + SMT Westmere box, single 30 sec tbench runs, higher is better: clients 1 2 4 8 16 32 64 128 .......................................................................... pre 30 41 118 645 3769 6214 12233 14312 post 299 603 1211 2418 4697 6847 11606 14557 A nice increase in performance. Signed-off-by: Mike Galbraith Signed-off-by: Peter Zijlstra Cc: Linus Torvalds Cc: Andrew Morton Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1339471112.7352.32.camel@marge.simpson.net Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 + kernel/sched/core.c | 39 ++++++++++++++++++++++++++++++++++++++- kernel/sched/fair.c | 28 +++++++--------------------- 3 files changed, 46 insertions(+), 22 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 4a1f493e0fef..bc9952991710 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -949,6 +949,7 @@ struct sched_domain { unsigned int smt_gain; int flags; /* See SD_* */ int level; + int idle_buddy; /* cpu assigned to select_idle_sibling() */ /* Runtime fields. */ unsigned long last_balance; /* init to jiffies. units in jiffies */ diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 4b4a63d34396..536b213f0ce5 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -6024,6 +6024,11 @@ static void destroy_sched_domains(struct sched_domain *sd, int cpu) * SD_SHARE_PKG_RESOURCE set (Last Level Cache Domain) for this * allows us to avoid some pointer chasing select_idle_sibling(). * + * Iterate domains and sched_groups downward, assigning CPUs to be + * select_idle_sibling() hw buddy. Cross-wiring hw makes bouncing + * due to random perturbation self canceling, ie sw buddies pull + * their counterpart to their CPU's hw counterpart. + * * Also keep a unique ID per domain (we use the first cpu number in * the cpumask of the domain), this allows us to quickly tell if * two cpus are in the same cache domain, see cpus_share_cache(). @@ -6037,8 +6042,40 @@ static void update_top_cache_domain(int cpu) int id = cpu; sd = highest_flag_domain(cpu, SD_SHARE_PKG_RESOURCES); - if (sd) + if (sd) { + struct sched_domain *tmp = sd; + struct sched_group *sg, *prev; + bool right; + + /* + * Traverse to first CPU in group, and count hops + * to cpu from there, switching direction on each + * hop, never ever pointing the last CPU rightward. + */ + do { + id = cpumask_first(sched_domain_span(tmp)); + prev = sg = tmp->groups; + right = 1; + + while (cpumask_first(sched_group_cpus(sg)) != id) + sg = sg->next; + + while (!cpumask_test_cpu(cpu, sched_group_cpus(sg))) { + prev = sg; + sg = sg->next; + right = !right; + } + + /* A CPU went down, never point back to domain start. */ + if (right && cpumask_first(sched_group_cpus(sg->next)) == id) + right = false; + + sg = right ? sg->next : prev; + tmp->idle_buddy = cpumask_first(sched_group_cpus(sg)); + } while ((tmp = tmp->child)); + id = cpumask_first(sched_domain_span(sd)); + } rcu_assign_pointer(per_cpu(sd_llc, cpu), sd); per_cpu(sd_llc_id, cpu) = id; diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index c099cc6eebe3..dd00aaf44fda 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -2637,8 +2637,6 @@ static int select_idle_sibling(struct task_struct *p, int target) int cpu = smp_processor_id(); int prev_cpu = task_cpu(p); struct sched_domain *sd; - struct sched_group *sg; - int i; /* * If the task is going to be woken-up on this cpu and if it is @@ -2655,29 +2653,17 @@ static int select_idle_sibling(struct task_struct *p, int target) return prev_cpu; /* - * Otherwise, iterate the domains and find an elegible idle cpu. + * Otherwise, check assigned siblings to find an elegible idle cpu. */ sd = rcu_dereference(per_cpu(sd_llc, target)); - for_each_lower_domain(sd) { - sg = sd->groups; - do { - if (!cpumask_intersects(sched_group_cpus(sg), - tsk_cpus_allowed(p))) - goto next; - - for_each_cpu(i, sched_group_cpus(sg)) { - if (!idle_cpu(i)) - goto next; - } - target = cpumask_first_and(sched_group_cpus(sg), - tsk_cpus_allowed(p)); - goto done; -next: - sg = sg->next; - } while (sg != sd->groups); + for_each_lower_domain(sd) { + if (!cpumask_test_cpu(sd->idle_buddy, tsk_cpus_allowed(p))) + continue; + if (idle_cpu(sd->idle_buddy)) + return sd->idle_buddy; } -done: + return target; } -- cgit v1.2.3 From 85c1e7dae165acd004429f81fe52bfbf55b57a98 Mon Sep 17 00:00:00 2001 From: Prashanth Nageshappa Date: Tue, 19 Jun 2012 17:47:34 +0530 Subject: sched: Reorder 'struct lb_env' members to reduce its size Members of 'struct lb_env' are not in appropriate order to reuse compiler added padding on 64bit architectures. In this patch we reorder those struct members and help reduce the size of the structure from 96 bytes to 80 bytes on 64 bit architectures. Suggested-by: Srivatsa Vaddagiri Signed-off-by: Prashanth Nageshappa Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/4FE06DDE.7000403@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- kernel/sched/fair.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index dd00aaf44fda..9361669d4242 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3058,8 +3058,8 @@ static unsigned long __read_mostly max_load_balance_interval = HZ/10; struct lb_env { struct sched_domain *sd; - int src_cpu; struct rq *src_rq; + int src_cpu; int dst_cpu; struct rq *dst_rq; -- cgit v1.2.3 From bbf18b19495942cc730e8ff11fc3ffadf20cbfe1 Mon Sep 17 00:00:00 2001 From: Prashanth Nageshappa Date: Tue, 19 Jun 2012 17:52:07 +0530 Subject: sched: Reset loop counters if all tasks are pinned and we need to redo load balance While load balancing, if all tasks on the source runqueue are pinned, we retry after excluding the corresponding source cpu. However, loop counters env.loop and env.loop_break are not reset before retrying, which can lead to failure in moving the tasks. In this patch we reset env.loop and env.loop_break to their inital values before we retry. Signed-off-by: Prashanth Nageshappa Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/4FE06EEF.2090709@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- kernel/sched/fair.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 9361669d4242..f9f9aa0edf3c 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4288,8 +4288,11 @@ more_balance: /* All tasks on this runqueue were pinned by CPU affinity */ if (unlikely(env.flags & LBF_ALL_PINNED)) { cpumask_clear_cpu(cpu_of(busiest), cpus); - if (!cpumask_empty(cpus)) + if (!cpumask_empty(cpus)) { + env.loop = 0; + env.loop_break = sched_nr_migrate_break; goto redo; + } goto out_balanced; } } -- cgit v1.2.3 From 88b8dac0a14c511ff41486b83a8c3d688936eec0 Mon Sep 17 00:00:00 2001 From: Srivatsa Vaddagiri Date: Tue, 19 Jun 2012 17:43:15 +0530 Subject: sched: Improve balance_cpu() to consider other cpus in its group as target of (pinned) task Current load balance scheme requires only one cpu in a sched_group (balance_cpu) to look at other peer sched_groups for imbalance and pull tasks towards itself from a busy cpu. Tasks thus pulled by balance_cpu could later get picked up by cpus that are in the same sched_group as that of balance_cpu. This scheme however fails to pull tasks that are not allowed to run on balance_cpu (but are allowed to run on other cpus in its sched_group). That can affect fairness and in some worst case scenarios cause starvation. Consider a two core (2 threads/core) system running tasks as below: Core0 Core1 / \ / \ C0 C1 C2 C3 | | | | v v v v F0 T1 F1 [idle] T2 F0 = SCHED_FIFO task (pinned to C0) F1 = SCHED_FIFO task (pinned to C2) T1 = SCHED_OTHER task (pinned to C1) T2 = SCHED_OTHER task (pinned to C1 and C2) F1 could become a cpu hog, which will starve T2 unless C1 pulls it. Between C0 and C1 however, C0 is required to look for imbalance between cores, which will fail to pull T2 towards Core0. T2 will starve eternally in this case. The same scenario can arise in presence of non-rt tasks as well (say we replace F1 with high irq load). We tackle this problem by having balance_cpu move pinned tasks to one of its sibling cpus (where they can run). We first check if load balance goal can be met by ignoring pinned tasks, failing which we retry move_tasks() with a new env->dst_cpu. This patch modifies load balance semantics on who can move load towards a given cpu in a given sched_domain. Before this patch, a given_cpu or a ilb_cpu acting on behalf of an idle given_cpu is responsible for moving load to given_cpu. With this patch applied, balance_cpu can in addition decide on moving some load to a given_cpu. There is a remote possibility that excess load could get moved as a result of this (balance_cpu and given_cpu/ilb_cpu deciding *independently* and at *same* time to move some load to a given_cpu). However we should see less of such conflicting decisions in practice and moreover subsequent load balance cycles should correct the excess load moved to given_cpu. Signed-off-by: Srivatsa Vaddagiri Signed-off-by: Prashanth Nageshappa Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/4FE06CDB.2060605@linux.vnet.ibm.com [ minor edits ] Signed-off-by: Ingo Molnar --- kernel/sched/fair.c | 78 ++++++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 74 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index f9f9aa0edf3c..22321db64952 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3054,6 +3054,7 @@ static unsigned long __read_mostly max_load_balance_interval = HZ/10; #define LBF_ALL_PINNED 0x01 #define LBF_NEED_BREAK 0x02 +#define LBF_SOME_PINNED 0x04 struct lb_env { struct sched_domain *sd; @@ -3064,6 +3065,8 @@ struct lb_env { int dst_cpu; struct rq *dst_rq; + struct cpumask *dst_grpmask; + int new_dst_cpu; enum cpu_idle_type idle; long imbalance; unsigned int flags; @@ -3131,9 +3134,31 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) * 3) are cache-hot on their current CPU. */ if (!cpumask_test_cpu(env->dst_cpu, tsk_cpus_allowed(p))) { + int new_dst_cpu; + schedstat_inc(p, se.statistics.nr_failed_migrations_affine); + + /* + * Remember if this task can be migrated to any other cpu in + * our sched_group. We may want to revisit it if we couldn't + * meet load balance goals by pulling other tasks on src_cpu. + * + * Also avoid computing new_dst_cpu if we have already computed + * one in current iteration. + */ + if (!env->dst_grpmask || (env->flags & LBF_SOME_PINNED)) + return 0; + + new_dst_cpu = cpumask_first_and(env->dst_grpmask, + tsk_cpus_allowed(p)); + if (new_dst_cpu < nr_cpu_ids) { + env->flags |= LBF_SOME_PINNED; + env->new_dst_cpu = new_dst_cpu; + } return 0; } + + /* Record that we found atleast one task that could run on dst_cpu */ env->flags &= ~LBF_ALL_PINNED; if (task_running(env->src_rq, p)) { @@ -4213,7 +4238,8 @@ static int load_balance(int this_cpu, struct rq *this_rq, struct sched_domain *sd, enum cpu_idle_type idle, int *balance) { - int ld_moved, active_balance = 0; + int ld_moved, cur_ld_moved, active_balance = 0; + int lb_iterations, max_lb_iterations; struct sched_group *group; struct rq *busiest; unsigned long flags; @@ -4223,11 +4249,13 @@ static int load_balance(int this_cpu, struct rq *this_rq, .sd = sd, .dst_cpu = this_cpu, .dst_rq = this_rq, + .dst_grpmask = sched_group_cpus(sd->groups), .idle = idle, .loop_break = sched_nr_migrate_break, }; cpumask_copy(cpus, cpu_active_mask); + max_lb_iterations = cpumask_weight(env.dst_grpmask); schedstat_inc(sd, lb_count[idle]); @@ -4253,6 +4281,7 @@ redo: schedstat_add(sd, lb_imbalance[idle], env.imbalance); ld_moved = 0; + lb_iterations = 1; if (busiest->nr_running > 1) { /* * Attempt to move tasks. If find_busiest_group has found @@ -4270,7 +4299,13 @@ more_balance: double_rq_lock(this_rq, busiest); if (!env.loop) update_h_load(env.src_cpu); - ld_moved += move_tasks(&env); + + /* + * cur_ld_moved - load moved in current iteration + * ld_moved - cumulative load moved across iterations + */ + cur_ld_moved = move_tasks(&env); + ld_moved += cur_ld_moved; double_rq_unlock(this_rq, busiest); local_irq_restore(flags); @@ -4282,8 +4317,43 @@ more_balance: /* * some other cpu did the load balance for us. */ - if (ld_moved && this_cpu != smp_processor_id()) - resched_cpu(this_cpu); + if (cur_ld_moved && env.dst_cpu != smp_processor_id()) + resched_cpu(env.dst_cpu); + + /* + * Revisit (affine) tasks on src_cpu that couldn't be moved to + * us and move them to an alternate dst_cpu in our sched_group + * where they can run. The upper limit on how many times we + * iterate on same src_cpu is dependent on number of cpus in our + * sched_group. + * + * This changes load balance semantics a bit on who can move + * load to a given_cpu. In addition to the given_cpu itself + * (or a ilb_cpu acting on its behalf where given_cpu is + * nohz-idle), we now have balance_cpu in a position to move + * load to given_cpu. In rare situations, this may cause + * conflicts (balance_cpu and given_cpu/ilb_cpu deciding + * _independently_ and at _same_ time to move some load to + * given_cpu) causing exceess load to be moved to given_cpu. + * This however should not happen so much in practice and + * moreover subsequent load balance cycles should correct the + * excess load moved. + */ + if ((env.flags & LBF_SOME_PINNED) && env.imbalance > 0 && + lb_iterations++ < max_lb_iterations) { + + this_rq = cpu_rq(env.new_dst_cpu); + env.dst_rq = this_rq; + env.dst_cpu = env.new_dst_cpu; + env.flags &= ~LBF_SOME_PINNED; + env.loop = 0; + env.loop_break = sched_nr_migrate_break; + /* + * Go back to "more_balance" rather than "redo" since we + * need to continue with same src_cpu. + */ + goto more_balance; + } /* All tasks on this runqueue were pinned by CPU affinity */ if (unlikely(env.flags & LBF_ALL_PINNED)) { -- cgit v1.2.3 From 8323f26ce3425460769605a6aece7a174edaa7d1 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 22 Jun 2012 13:36:05 +0200 Subject: sched: Fix race in task_group() Stefan reported a crash on a kernel before a3e5d1091c1 ("sched: Don't call task_group() too many times in set_task_rq()"), he found the reason to be that the multiple task_group() invocations in set_task_rq() returned different values. Looking at all that I found a lack of serialization and plain wrong comments. The below tries to fix it using an extra pointer which is updated under the appropriate scheduler locks. Its not pretty, but I can't really see another way given how all the cgroup stuff works. Reported-and-tested-by: Stefan Bader Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1340364965.18025.71.camel@twins Signed-off-by: Ingo Molnar --- include/linux/init_task.h | 12 +++++++++++- include/linux/sched.h | 5 ++++- kernel/sched/core.c | 9 ++++++++- kernel/sched/sched.h | 23 ++++++++++------------- 4 files changed, 33 insertions(+), 16 deletions(-) (limited to 'kernel') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 9e65eff6af3b..b806b821e735 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -123,8 +123,17 @@ extern struct group_info init_groups; extern struct cred init_cred; +extern struct task_group root_task_group; + +#ifdef CONFIG_CGROUP_SCHED +# define INIT_CGROUP_SCHED(tsk) \ + .sched_task_group = &root_task_group, +#else +# define INIT_CGROUP_SCHED(tsk) +#endif + #ifdef CONFIG_PERF_EVENTS -# define INIT_PERF_EVENTS(tsk) \ +# define INIT_PERF_EVENTS(tsk) \ .perf_event_mutex = \ __MUTEX_INITIALIZER(tsk.perf_event_mutex), \ .perf_event_list = LIST_HEAD_INIT(tsk.perf_event_list), @@ -161,6 +170,7 @@ extern struct cred init_cred; }, \ .tasks = LIST_HEAD_INIT(tsk.tasks), \ INIT_PUSHABLE_TASKS(tsk) \ + INIT_CGROUP_SCHED(tsk) \ .ptraced = LIST_HEAD_INIT(tsk.ptraced), \ .ptrace_entry = LIST_HEAD_INIT(tsk.ptrace_entry), \ .real_parent = &tsk, \ diff --git a/include/linux/sched.h b/include/linux/sched.h index bc9952991710..fd9436a3a545 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1245,6 +1245,9 @@ struct task_struct { const struct sched_class *sched_class; struct sched_entity se; struct sched_rt_entity rt; +#ifdef CONFIG_CGROUP_SCHED + struct task_group *sched_task_group; +#endif #ifdef CONFIG_PREEMPT_NOTIFIERS /* list of struct preempt_notifier: */ @@ -2724,7 +2727,7 @@ extern int sched_group_set_rt_period(struct task_group *tg, extern long sched_group_rt_period(struct task_group *tg); extern int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk); #endif -#endif +#endif /* CONFIG_CGROUP_SCHED */ extern int task_can_switch_user(struct user_struct *up, struct task_struct *tsk); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 536b213f0ce5..5d011ef4c0df 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1096,7 +1096,7 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) * a task's CPU. ->pi_lock for waking tasks, rq->lock for runnable tasks. * * sched_move_task() holds both and thus holding either pins the cgroup, - * see set_task_rq(). + * see task_group(). * * Furthermore, all task_rq users should acquire both locks, see * task_rq_lock(). @@ -7658,6 +7658,7 @@ void sched_destroy_group(struct task_group *tg) */ void sched_move_task(struct task_struct *tsk) { + struct task_group *tg; int on_rq, running; unsigned long flags; struct rq *rq; @@ -7672,6 +7673,12 @@ void sched_move_task(struct task_struct *tsk) if (unlikely(running)) tsk->sched_class->put_prev_task(rq, tsk); + tg = container_of(task_subsys_state_check(tsk, cpu_cgroup_subsys_id, + lockdep_is_held(&tsk->sighand->siglock)), + struct task_group, css); + tg = autogroup_task_group(tsk, tg); + tsk->sched_task_group = tg; + #ifdef CONFIG_FAIR_GROUP_SCHED if (tsk->sched_class->task_move_group) tsk->sched_class->task_move_group(tsk, on_rq); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 55844f24435a..c35a1a7dd4d6 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -538,22 +538,19 @@ extern int group_balance_cpu(struct sched_group *sg); /* * Return the group to which this tasks belongs. * - * We use task_subsys_state_check() and extend the RCU verification with - * pi->lock and rq->lock because cpu_cgroup_attach() holds those locks for each - * task it moves into the cgroup. Therefore by holding either of those locks, - * we pin the task to the current cgroup. + * We cannot use task_subsys_state() and friends because the cgroup + * subsystem changes that value before the cgroup_subsys::attach() method + * is called, therefore we cannot pin it and might observe the wrong value. + * + * The same is true for autogroup's p->signal->autogroup->tg, the autogroup + * core changes this before calling sched_move_task(). + * + * Instead we use a 'copy' which is updated from sched_move_task() while + * holding both task_struct::pi_lock and rq::lock. */ static inline struct task_group *task_group(struct task_struct *p) { - struct task_group *tg; - struct cgroup_subsys_state *css; - - css = task_subsys_state_check(p, cpu_cgroup_subsys_id, - lockdep_is_held(&p->pi_lock) || - lockdep_is_held(&task_rq(p)->lock)); - tg = container_of(css, struct task_group, css); - - return autogroup_task_group(p, tg); + return p->sched_task_group; } /* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */ -- cgit v1.2.3 From b6070a8d9853eda010a549fa9a09eb8d7269b929 Mon Sep 17 00:00:00 2001 From: Darren Hart Date: Fri, 20 Jul 2012 11:53:29 -0700 Subject: futex: Test for pi_mutex on fault in futex_wait_requeue_pi() If fixup_pi_state_owner() faults, pi_mutex may be NULL. Test for pi_mutex != NULL before testing the owner against current and possibly unlocking it. Signed-off-by: Darren Hart Cc: Dave Jones Cc: Dan Carpenter Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/dc59890338fc413606f04e5c5b131530734dae3d.1342809673.git.dvhart@linux.intel.com Signed-off-by: Thomas Gleixner --- kernel/futex.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/futex.c b/kernel/futex.c index e2b0fb9a0b3b..05018bfe21a7 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -2370,7 +2370,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, * fault, unlock the rt_mutex and return the fault to userspace. */ if (ret == -EFAULT) { - if (rt_mutex_owner(pi_mutex) == current) + if (pi_mutex && rt_mutex_owner(pi_mutex) == current) rt_mutex_unlock(pi_mutex); } else if (ret == -EINTR) { /* -- cgit v1.2.3 From f27071cb7fe3e1d37a9dbe6c0dfc5395cd40fa43 Mon Sep 17 00:00:00 2001 From: Darren Hart Date: Fri, 20 Jul 2012 11:53:30 -0700 Subject: futex: Fix bug in WARN_ON for NULL q.pi_state The WARN_ON in futex_wait_requeue_pi() for a NULL q.pi_state was testing the address (&q.pi_state) of the pointer instead of the value (q.pi_state) of the pointer. Correct it accordingly. Signed-off-by: Darren Hart Cc: Dave Jones Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1c85d97f6e5f79ec389a4ead3e367363c74bd09a.1342809673.git.dvhart@linux.intel.com Signed-off-by: Thomas Gleixner --- kernel/futex.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/futex.c b/kernel/futex.c index 05018bfe21a7..5551adaf7cdf 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -2343,7 +2343,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, * signal. futex_unlock_pi() will not destroy the lock_ptr nor * the pi_state. */ - WARN_ON(!&q.pi_state); + WARN_ON(!q.pi_state); pi_mutex = &q.pi_state->pi_mutex; ret = rt_mutex_finish_proxy_lock(pi_mutex, to, &rt_waiter, 1); debug_rt_mutex_free_waiter(&rt_waiter); -- cgit v1.2.3 From 6f7b0a2a5c0fb03be7c25bd1745baa50582348ef Mon Sep 17 00:00:00 2001 From: Darren Hart Date: Fri, 20 Jul 2012 11:53:31 -0700 Subject: futex: Forbid uaddr == uaddr2 in futex_wait_requeue_pi() If uaddr == uaddr2, then we have broken the rule of only requeueing from a non-pi futex to a pi futex with this call. If we attempt this, as the trinity test suite manages to do, we miss early wakeups as q.key is equal to key2 (because they are the same uaddr). We will then attempt to dereference the pi_mutex (which would exist had the futex_q been properly requeued to a pi futex) and trigger a NULL pointer dereference. Signed-off-by: Darren Hart Cc: Dave Jones Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/ad82bfe7f7d130247fbe2b5b4275654807774227.1342809673.git.dvhart@linux.intel.com Signed-off-by: Thomas Gleixner --- kernel/futex.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/futex.c b/kernel/futex.c index 5551adaf7cdf..3717e7b306e0 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -2231,11 +2231,11 @@ int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, * @uaddr2: the pi futex we will take prior to returning to user-space * * The caller will wait on uaddr and will be requeued by futex_requeue() to - * uaddr2 which must be PI aware. Normal wakeup will wake on uaddr2 and - * complete the acquisition of the rt_mutex prior to returning to userspace. - * This ensures the rt_mutex maintains an owner when it has waiters; without - * one, the pi logic wouldn't know which task to boost/deboost, if there was a - * need to. + * uaddr2 which must be PI aware and unique from uaddr. Normal wakeup will wake + * on uaddr2 and complete the acquisition of the rt_mutex prior to returning to + * userspace. This ensures the rt_mutex maintains an owner when it has waiters; + * without one, the pi logic would not know which task to boost/deboost, if + * there was a need to. * * We call schedule in futex_wait_queue_me() when we enqueue and return there * via the following: @@ -2272,6 +2272,9 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, struct futex_q q = futex_q_init; int res, ret; + if (uaddr == uaddr2) + return -EINVAL; + if (!bitset) return -EINVAL; -- cgit v1.2.3 From b44d50dcacea0d485ca2ff9140f8cc28ee22f28d Mon Sep 17 00:00:00 2001 From: John Stultz Date: Mon, 23 Jul 2012 16:22:37 -0400 Subject: time: Fix casting issue in tk_set_xtime and tk_xtime_add commit 1e75fa8b (time: Condense timekeeper.xtime into xtime_sec) introduced helper functions which apply a timespec to the core internal timekeeper data. The internal storage type is u64. The timespec tv_nsec value must be shifted before set or added to the internal value. tv_nsec is a long, which is 32bit on a 32bit system, so without casting tv_nsec to u64 we lose the bits which are shifted over the 32bit boundary. Add the proper typecasts. Reported-by: Konrad Rzeszutek Wilk Tested-by: Konrad Rzeszutek Wilk Signed-off-by: John Stultz Acked-by: Prarit Bhargava Link: http://lkml.kernel.org/r/1343074957-16541-1-git-send-email-john.stultz@linaro.org Signed-off-by: Thomas Gleixner --- kernel/time/timekeeping.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 5980e902978c..8f2aba1246f2 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -108,13 +108,13 @@ static struct timespec tk_xtime(struct timekeeper *tk) static void tk_set_xtime(struct timekeeper *tk, const struct timespec *ts) { tk->xtime_sec = ts->tv_sec; - tk->xtime_nsec = ts->tv_nsec << tk->shift; + tk->xtime_nsec = (u64)ts->tv_nsec << tk->shift; } static void tk_xtime_add(struct timekeeper *tk, const struct timespec *ts) { tk->xtime_sec += ts->tv_sec; - tk->xtime_nsec += ts->tv_nsec << tk->shift; + tk->xtime_nsec += (u64)ts->tv_nsec << tk->shift; } /** -- cgit v1.2.3 From 4c0946c47463defa681b83294383dc996d255bb7 Mon Sep 17 00:00:00 2001 From: Grant Likely Date: Sun, 3 Jun 2012 22:04:39 -0700 Subject: irqdomain: eliminate slow-path revmap lookups With the current state of irq_domain, the reverse map is always updated when new IRQs get mapped. This means that the irq_find_mapping() function can be simplified to execute the revmap lookup functions unconditionally This patch adds lookup functions for the revmaps that don't yet have one and removes the slow path lookup code path. v8: Broke out unrelated changes into separate patches. Rebased on Paul's irq association patches. v7: Rebased to irqdomain/next for v3.4 and applied before the removal of 'hint' v6: Remove the slow path entirely. The only place where the slow path could get called is for a linear mapping if the hwirq number is larger than the linear revmap size. There shouldn't be any interrupt controllers that do that. v5: rewrite to not use a ->revmap() callback. It is simpler, smaller, safer and faster to open code each of the revmap lookups directly into irq_find_mapping() via a switch statement. v4: Fix build failure on incorrect variable reference. Signed-off-by: Grant Likely Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Milton Miller Cc: Paul Mundt Cc: Rob Herring --- kernel/irq/irqdomain.c | 65 +++++++++++++++++++------------------------------- 1 file changed, 25 insertions(+), 40 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index c0e638ba9c2a..170724a07a2c 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -686,16 +686,11 @@ EXPORT_SYMBOL_GPL(irq_dispose_mapping); * irq_find_mapping() - Find a linux irq from an hw irq number. * @domain: domain owning this hardware interrupt * @hwirq: hardware irq number in that domain space - * - * This is a slow path, for use by generic code. It's expected that an - * irq controller implementation directly calls the appropriate low level - * mapping function. */ unsigned int irq_find_mapping(struct irq_domain *domain, irq_hw_number_t hwirq) { - unsigned int i; - unsigned int hint = hwirq % nr_irqs; + struct irq_data *data; /* Look for default domain if nececssary */ if (domain == NULL) @@ -703,22 +698,25 @@ unsigned int irq_find_mapping(struct irq_domain *domain, if (domain == NULL) return 0; - /* legacy -> bail early */ - if (domain->revmap_type == IRQ_DOMAIN_MAP_LEGACY) + switch (domain->revmap_type) { + case IRQ_DOMAIN_MAP_LEGACY: return irq_domain_legacy_revmap(domain, hwirq); - - /* Slow path does a linear search of the map */ - if (hint == 0) - hint = 1; - i = hint; - do { - struct irq_data *data = irq_get_irq_data(i); + case IRQ_DOMAIN_MAP_LINEAR: + return irq_linear_revmap(domain, hwirq); + case IRQ_DOMAIN_MAP_TREE: + rcu_read_lock(); + data = radix_tree_lookup(&domain->revmap_data.tree, hwirq); + rcu_read_unlock(); + if (data) + return data->irq; + break; + case IRQ_DOMAIN_MAP_NOMAP: + data = irq_get_irq_data(hwirq); if (data && (data->domain == domain) && (data->hwirq == hwirq)) - return i; - i++; - if (i >= nr_irqs) - i = 1; - } while(i != hint); + return hwirq; + break; + } + return 0; } EXPORT_SYMBOL_GPL(irq_find_mapping); @@ -728,32 +726,19 @@ EXPORT_SYMBOL_GPL(irq_find_mapping); * @domain: domain owning this hardware interrupt * @hwirq: hardware irq number in that domain space * - * This is a fast path, for use by irq controller code that uses linear - * revmaps. It does fallback to the slow path if the revmap doesn't exist - * yet and will create the revmap entry with appropriate locking + * This is a fast path that can be called directly by irq controller code to + * save a handful of instructions. */ unsigned int irq_linear_revmap(struct irq_domain *domain, irq_hw_number_t hwirq) { - unsigned int *revmap; + BUG_ON(domain->revmap_type != IRQ_DOMAIN_MAP_LINEAR); - if (WARN_ON_ONCE(domain->revmap_type != IRQ_DOMAIN_MAP_LINEAR)) - return irq_find_mapping(domain, hwirq); - - /* Check revmap bounds */ - if (unlikely(hwirq >= domain->revmap_data.linear.size)) - return irq_find_mapping(domain, hwirq); - - /* Check if revmap was allocated */ - revmap = domain->revmap_data.linear.revmap; - if (unlikely(revmap == NULL)) - return irq_find_mapping(domain, hwirq); - - /* Fill up revmap with slow path if no mapping found */ - if (unlikely(!revmap[hwirq])) - revmap[hwirq] = irq_find_mapping(domain, hwirq); + /* Check revmap bounds; complain if exceeded */ + if (WARN_ON(hwirq >= domain->revmap_data.linear.size)) + return 0; - return revmap[hwirq]; + return domain->revmap_data.linear.revmap[hwirq]; } EXPORT_SYMBOL_GPL(irq_linear_revmap); -- cgit v1.2.3 From f5a1ad057e6da5d0fc9c5677ff44797d193d3e62 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 20 Jul 2012 10:33:19 +0100 Subject: irqdomain: Improve diagnostics when a domain mapping fails When the map operation fails log the error code we get and add a WARN_ON() so we get a backtrace (which should help work out which interrupt is the source of the issue). Signed-off-by: Mark Brown Signed-off-by: Grant Likely --- kernel/irq/irqdomain.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 170724a07a2c..49a77727db42 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -415,7 +415,7 @@ int irq_domain_associate_many(struct irq_domain *domain, unsigned int irq_base, { unsigned int virq = irq_base; irq_hw_number_t hwirq = hwirq_base; - int i; + int i, ret; pr_debug("%s(%s, irqbase=%i, hwbase=%i, count=%i)\n", __func__, of_node_full_name(domain->of_node), irq_base, (int)hwirq_base, count); @@ -436,11 +436,16 @@ int irq_domain_associate_many(struct irq_domain *domain, unsigned int irq_base, irq_data->hwirq = hwirq; irq_data->domain = domain; - if (domain->ops->map && domain->ops->map(domain, virq, hwirq)) { - pr_err("irq-%i==>hwirq-0x%lx mapping failed\n", virq, hwirq); - irq_data->domain = NULL; - irq_data->hwirq = 0; - goto err_unmap; + if (domain->ops->map) { + ret = domain->ops->map(domain, virq, hwirq); + if (ret != 0) { + pr_err("irq-%i==>hwirq-0x%lx mapping failed: %d\n", + virq, hwirq, ret); + WARN_ON(1); + irq_data->domain = NULL; + irq_data->hwirq = 0; + goto err_unmap; + } } switch (domain->revmap_type) { -- cgit v1.2.3 From dc9b229a58dc0dfed34272ff26c6d5fd17c674e0 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 13 Jul 2012 19:29:45 +0200 Subject: genirq: Allow irq chips to mark themself oneshot safe Some interrupt chips like MSI are oneshot safe by implementation. For those interrupts we can avoid the mask/unmask sequence for threaded interrupt handlers. Signed-off-by: Thomas Gleixner Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1207132056540.32033@ionos Cc: Linus Torvalds Cc: Avi Kivity Cc: Marcelo Tosatti Cc: Jan Kiszka --- include/linux/irq.h | 1 + kernel/irq/manage.c | 15 ++++++++++++++- 2 files changed, 15 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/include/linux/irq.h b/include/linux/irq.h index 553fb66da130..216b0ba109d7 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -349,6 +349,7 @@ enum { IRQCHIP_MASK_ON_SUSPEND = (1 << 2), IRQCHIP_ONOFFLINE_ENABLED = (1 << 3), IRQCHIP_SKIP_SET_WAKE = (1 << 4), + IRQCHIP_ONESHOT_SAFE = (1 << 5), }; /* This include will go away once we isolated irq_desc usage to core code */ diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 8c548232ba39..2e326d1ebec1 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -959,6 +959,18 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) goto out_thread; } + /* + * Drivers are often written to work w/o knowledge about the + * underlying irq chip implementation, so a request for a + * threaded irq without a primary hard irq context handler + * requires the ONESHOT flag to be set. Some irq chips like + * MSI based interrupts are per se one shot safe. Check the + * chip flags, so we can avoid the unmask dance at the end of + * the threaded handler for those. + */ + if (desc->irq_data.chip->flags & IRQCHIP_ONESHOT_SAFE) + new->flags &= ~IRQF_ONESHOT; + /* * The following block of code has to be executed atomically */ @@ -1033,7 +1045,8 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) */ new->thread_mask = 1 << ffz(thread_mask); - } else if (new->handler == irq_default_primary_handler) { + } else if (new->handler == irq_default_primary_handler && + !(desc->irq_data.chip->flags & IRQCHIP_ONESHOT_SAFE)) { /* * The interrupt was requested with handler = NULL, so * we use the default primary handler for it. But it -- cgit v1.2.3 From 45afb1734fa6323a8ba08bd6c392ee227df67dde Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Sat, 7 Jul 2012 16:49:02 +0900 Subject: sched: Use task_rq_unlock() in __sched_setscheduler() It seems there's no specific reason to open-code it. I guess commit 0122ec5b02f76 ("sched: Add p->pi_lock to task_rq_lock()") simply missed it. Let's be consistent with others. Signed-off-by: Namhyung Kim Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1341647342-6742-1-git-send-email-namhyung@kernel.org Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 5d011ef4c0df..2cb4e7777998 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -4340,9 +4340,7 @@ recheck: */ if (unlikely(policy == p->policy && (!rt_policy(policy) || param->sched_priority == p->rt_priority))) { - - __task_rq_unlock(rq); - raw_spin_unlock_irqrestore(&p->pi_lock, flags); + task_rq_unlock(rq, p, &flags); return 0; } -- cgit v1.2.3 From 014acbf0d5c8445e0ff88ae60edd676dd9cc461c Mon Sep 17 00:00:00 2001 From: Ying Xue Date: Thu, 12 Jul 2012 15:03:42 +0800 Subject: sched: Fix minor code style issues Delete redudant spaces between type name and data name or operators. Signed-off-by: Ying Xue Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1342076622-6606-1-git-send-email-ying.xue0@gmail.com Signed-off-by: Ingo Molnar --- kernel/sched/cpupri.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/cpupri.c b/kernel/sched/cpupri.c index d72586fdf660..23aa789c53ee 100644 --- a/kernel/sched/cpupri.c +++ b/kernel/sched/cpupri.c @@ -65,8 +65,8 @@ static int convert_prio(int prio) int cpupri_find(struct cpupri *cp, struct task_struct *p, struct cpumask *lowest_mask) { - int idx = 0; - int task_pri = convert_prio(p->prio); + int idx = 0; + int task_pri = convert_prio(p->prio); if (task_pri >= MAX_RT_PRIO) return 0; @@ -137,9 +137,9 @@ int cpupri_find(struct cpupri *cp, struct task_struct *p, */ void cpupri_set(struct cpupri *cp, int cpu, int newpri) { - int *currpri = &cp->cpu_to_pri[cpu]; - int oldpri = *currpri; - int do_mb = 0; + int *currpri = &cp->cpu_to_pri[cpu]; + int oldpri = *currpri; + int do_mb = 0; newpri = convert_prio(newpri); -- cgit v1.2.3 From 895dd92c032e1604aa0d7afaef7716e058343b67 Mon Sep 17 00:00:00 2001 From: Andrew Vagin Date: Thu, 12 Jul 2012 14:14:29 +0400 Subject: sched: Deliver sched_switch events to the current task Otherwise they can't be filtered for a defined task: perf record -e sched:sched_switch ./foo This command doesn't report any events without this patch. I think it isn't a security concern if someone knows who will be executed next - this can already be observed by polling /proc state. By default perf is disabled for non-root users in any case. I need these events for profiling sleep times. sched_switch is used for getting callchains and sched_stat_* is used for getting time periods. These events are combined in user space, then it can be analyzed by perf tools. Signed-off-by: Andrew Vagin Signed-off-by: Peter Zijlstra Cc: Steven Rostedt Cc: Arun Sharma Link: http://lkml.kernel.org/r/1342088069-1005148-1-git-send-email-avagin@openvz.org Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 468bdd44c1ba..ad732b56ba70 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1910,12 +1910,12 @@ static inline void prepare_task_switch(struct rq *rq, struct task_struct *prev, struct task_struct *next) { + trace_sched_switch(prev, next); sched_info_switch(prev, next); perf_event_task_sched_out(prev, next); fire_sched_out_preempt_notifiers(prev, next); prepare_lock_switch(rq, next); prepare_arch_switch(next); - trace_sched_switch(prev, next); } /** -- cgit v1.2.3 From 8ded2bbc1845e19c771eb55209aab166ef011243 Mon Sep 17 00:00:00 2001 From: Josh Boyer Date: Wed, 25 Jul 2012 10:40:34 -0400 Subject: posix_types.h: Cleanup stale __NFDBITS and related definitions Recently, glibc made a change to suppress sign-conversion warnings in FD_SET (glibc commit ceb9e56b3d1). This uncovered an issue with the kernel's definition of __NFDBITS if applications #include after including . A build failure would be seen when passing the -Werror=sign-compare and -D_FORTIFY_SOURCE=2 flags to gcc. It was suggested that the kernel should either match the glibc definition of __NFDBITS or remove that entirely. The current in-kernel uses of __NFDBITS can be replaced with BITS_PER_LONG, and there are no uses of the related __FDELT and __FDMASK defines. Given that, we'll continue the cleanup that was started with commit 8b3d1cda4f5f ("posix_types: Remove fd_set macros") and drop the remaining unused macros. Additionally, linux/time.h has similar macros defined that expand to nothing so we'll remove those at the same time. Reported-by: Jeff Law Suggested-by: Linus Torvalds CC: Signed-off-by: Josh Boyer [ .. and fix up whitespace as per akpm ] Signed-off-by: Linus Torvalds --- arch/mips/kernel/kspd.c | 2 +- fs/exec.c | 2 +- fs/select.c | 10 +++++----- include/linux/posix_types.h | 18 +++--------------- include/linux/time.h | 8 -------- kernel/exit.c | 2 +- security/selinux/hooks.c | 2 +- 7 files changed, 12 insertions(+), 32 deletions(-) (limited to 'kernel') diff --git a/arch/mips/kernel/kspd.c b/arch/mips/kernel/kspd.c index 84d0639e4580..b77f56bbb477 100644 --- a/arch/mips/kernel/kspd.c +++ b/arch/mips/kernel/kspd.c @@ -323,7 +323,7 @@ static void sp_cleanup(void) fdt = files_fdtable(files); for (;;) { unsigned long set; - i = j * __NFDBITS; + i = j * BITS_PER_LONG; if (i >= fdt->max_fds) break; set = fdt->open_fds[j++]; diff --git a/fs/exec.c b/fs/exec.c index da27b91ff1e8..e95aeeddd25c 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1020,7 +1020,7 @@ static void flush_old_files(struct files_struct * files) unsigned long set, i; j++; - i = j * __NFDBITS; + i = j * BITS_PER_LONG; fdt = files_fdtable(files); if (i >= fdt->max_fds) break; diff --git a/fs/select.c b/fs/select.c index bae321569dfa..db14c781335e 100644 --- a/fs/select.c +++ b/fs/select.c @@ -345,8 +345,8 @@ static int max_select_fd(unsigned long n, fd_set_bits *fds) struct fdtable *fdt; /* handle last in-complete long-word first */ - set = ~(~0UL << (n & (__NFDBITS-1))); - n /= __NFDBITS; + set = ~(~0UL << (n & (BITS_PER_LONG-1))); + n /= BITS_PER_LONG; fdt = files_fdtable(current->files); open_fds = fdt->open_fds + n; max = 0; @@ -373,7 +373,7 @@ get_max: max++; set >>= 1; } while (set); - max += n * __NFDBITS; + max += n * BITS_PER_LONG; } return max; @@ -435,11 +435,11 @@ int do_select(int n, fd_set_bits *fds, struct timespec *end_time) in = *inp++; out = *outp++; ex = *exp++; all_bits = in | out | ex; if (all_bits == 0) { - i += __NFDBITS; + i += BITS_PER_LONG; continue; } - for (j = 0; j < __NFDBITS; ++j, ++i, bit <<= 1) { + for (j = 0; j < BITS_PER_LONG; ++j, ++i, bit <<= 1) { int fput_needed; if (i >= n) break; diff --git a/include/linux/posix_types.h b/include/linux/posix_types.h index f04c98cf44f3..988f76e636e3 100644 --- a/include/linux/posix_types.h +++ b/include/linux/posix_types.h @@ -15,26 +15,14 @@ */ /* - * Those macros may have been defined in . But we always - * use the ones here. + * This macro may have been defined in . But we always + * use the one here. */ -#undef __NFDBITS -#define __NFDBITS (8 * sizeof(unsigned long)) - #undef __FD_SETSIZE #define __FD_SETSIZE 1024 -#undef __FDSET_LONGS -#define __FDSET_LONGS (__FD_SETSIZE/__NFDBITS) - -#undef __FDELT -#define __FDELT(d) ((d) / __NFDBITS) - -#undef __FDMASK -#define __FDMASK(d) (1UL << ((d) % __NFDBITS)) - typedef struct { - unsigned long fds_bits [__FDSET_LONGS]; + unsigned long fds_bits[__FD_SETSIZE / (8 * sizeof(long))]; } __kernel_fd_set; /* Type of a signal handler. */ diff --git a/include/linux/time.h b/include/linux/time.h index 179f4d6755fc..c81c5e40fcb5 100644 --- a/include/linux/time.h +++ b/include/linux/time.h @@ -257,14 +257,6 @@ static __always_inline void timespec_add_ns(struct timespec *a, u64 ns) #endif /* __KERNEL__ */ -#define NFDBITS __NFDBITS - -#define FD_SETSIZE __FD_SETSIZE -#define FD_SET(fd,fdsetp) __FD_SET(fd,fdsetp) -#define FD_CLR(fd,fdsetp) __FD_CLR(fd,fdsetp) -#define FD_ISSET(fd,fdsetp) __FD_ISSET(fd,fdsetp) -#define FD_ZERO(fdsetp) __FD_ZERO(fdsetp) - /* * Names of the interval timers, and structure * defining a timer setting: diff --git a/kernel/exit.c b/kernel/exit.c index d17f6c4ddfa9..f65345f9e5bb 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -483,7 +483,7 @@ static void close_files(struct files_struct * files) rcu_read_unlock(); for (;;) { unsigned long set; - i = j * __NFDBITS; + i = j * BITS_PER_LONG; if (i >= fdt->max_fds) break; set = fdt->open_fds[j++]; diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 689fe2d22165..94c45a1531a4 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -2129,7 +2129,7 @@ static inline void flush_unauthorized_files(const struct cred *cred, int fd; j++; - i = j * __NFDBITS; + i = j * BITS_PER_LONG; fdt = files_fdtable(files); if (i >= fdt->max_fds) break; -- cgit v1.2.3 From 800179c9b8a1e796e441674776d11cd4c05d61d7 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 25 Jul 2012 17:29:07 -0700 Subject: fs: add link restrictions MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This adds symlink and hardlink restrictions to the Linux VFS. Symlinks: A long-standing class of security issues is the symlink-based time-of-check-time-of-use race, most commonly seen in world-writable directories like /tmp. The common method of exploitation of this flaw is to cross privilege boundaries when following a given symlink (i.e. a root process follows a symlink belonging to another user). For a likely incomplete list of hundreds of examples across the years, please see: http://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=/tmp The solution is to permit symlinks to only be followed when outside a sticky world-writable directory, or when the uid of the symlink and follower match, or when the directory owner matches the symlink's owner. Some pointers to the history of earlier discussion that I could find: 1996 Aug, Zygo Blaxell http://marc.info/?l=bugtraq&m=87602167419830&w=2 1996 Oct, Andrew Tridgell http://lkml.indiana.edu/hypermail/linux/kernel/9610.2/0086.html 1997 Dec, Albert D Cahalan http://lkml.org/lkml/1997/12/16/4 2005 Feb, Lorenzo Hernández García-Hierro http://lkml.indiana.edu/hypermail/linux/kernel/0502.0/1896.html 2010 May, Kees Cook https://lkml.org/lkml/2010/5/30/144 Past objections and rebuttals could be summarized as: - Violates POSIX. - POSIX didn't consider this situation and it's not useful to follow a broken specification at the cost of security. - Might break unknown applications that use this feature. - Applications that break because of the change are easy to spot and fix. Applications that are vulnerable to symlink ToCToU by not having the change aren't. Additionally, no applications have yet been found that rely on this behavior. - Applications should just use mkstemp() or O_CREATE|O_EXCL. - True, but applications are not perfect, and new software is written all the time that makes these mistakes; blocking this flaw at the kernel is a single solution to the entire class of vulnerability. - This should live in the core VFS. - This should live in an LSM. (https://lkml.org/lkml/2010/5/31/135) - This should live in an LSM. - This should live in the core VFS. (https://lkml.org/lkml/2010/8/2/188) Hardlinks: On systems that have user-writable directories on the same partition as system files, a long-standing class of security issues is the hardlink-based time-of-check-time-of-use race, most commonly seen in world-writable directories like /tmp. The common method of exploitation of this flaw is to cross privilege boundaries when following a given hardlink (i.e. a root process follows a hardlink created by another user). Additionally, an issue exists where users can "pin" a potentially vulnerable setuid/setgid file so that an administrator will not actually upgrade a system fully. The solution is to permit hardlinks to only be created when the user is already the existing file's owner, or if they already have read/write access to the existing file. Many Linux users are surprised when they learn they can link to files they have no access to, so this change appears to follow the doctrine of "least surprise". Additionally, this change does not violate POSIX, which states "the implementation may require that the calling process has permission to access the existing file"[1]. This change is known to break some implementations of the "at" daemon, though the version used by Fedora and Ubuntu has been fixed[2] for a while. Otherwise, the change has been undisruptive while in use in Ubuntu for the last 1.5 years. [1] http://pubs.opengroup.org/onlinepubs/9699919799/functions/linkat.html [2] http://anonscm.debian.org/gitweb/?p=collab-maint/at.git;a=commitdiff;h=f4114656c3a6c6f6070e315ffdf940a49eda3279 This patch is based on the patches in Openwall and grsecurity, along with suggestions from Al Viro. I have added a sysctl to enable the protected behavior, and documentation. Signed-off-by: Kees Cook Acked-by: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- Documentation/sysctl/fs.txt | 42 +++++++++++++++ fs/namei.c | 122 ++++++++++++++++++++++++++++++++++++++++++++ include/linux/fs.h | 2 + kernel/sysctl.c | 18 +++++++ 4 files changed, 184 insertions(+) (limited to 'kernel') diff --git a/Documentation/sysctl/fs.txt b/Documentation/sysctl/fs.txt index 13d6166d7a27..d4a372e75750 100644 --- a/Documentation/sysctl/fs.txt +++ b/Documentation/sysctl/fs.txt @@ -32,6 +32,8 @@ Currently, these files are in /proc/sys/fs: - nr_open - overflowuid - overflowgid +- protected_hardlinks +- protected_symlinks - suid_dumpable - super-max - super-nr @@ -157,6 +159,46 @@ The default is 65534. ============================================================== +protected_hardlinks: + +A long-standing class of security issues is the hardlink-based +time-of-check-time-of-use race, most commonly seen in world-writable +directories like /tmp. The common method of exploitation of this flaw +is to cross privilege boundaries when following a given hardlink (i.e. a +root process follows a hardlink created by another user). Additionally, +on systems without separated partitions, this stops unauthorized users +from "pinning" vulnerable setuid/setgid files against being upgraded by +the administrator, or linking to special files. + +When set to "0", hardlink creation behavior is unrestricted. + +When set to "1" hardlinks cannot be created by users if they do not +already own the source file, or do not have read/write access to it. + +This protection is based on the restrictions in Openwall and grsecurity. + +============================================================== + +protected_symlinks: + +A long-standing class of security issues is the symlink-based +time-of-check-time-of-use race, most commonly seen in world-writable +directories like /tmp. The common method of exploitation of this flaw +is to cross privilege boundaries when following a given symlink (i.e. a +root process follows a symlink belonging to another user). For a likely +incomplete list of hundreds of examples across the years, please see: +http://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=/tmp + +When set to "0", symlink following behavior is unrestricted. + +When set to "1" symlinks are permitted to be followed only when outside +a sticky world-writable directory, or when the uid of the symlink and +follower match, or when the directory owner matches the symlink's owner. + +This protection is based on the restrictions in Openwall and grsecurity. + +============================================================== + suid_dumpable: This value can be used to query and set the core dump mode for setuid diff --git a/fs/namei.c b/fs/namei.c index afa087649ddb..3861d85f8488 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -650,6 +650,119 @@ static inline void put_link(struct nameidata *nd, struct path *link, void *cooki path_put(link); } +int sysctl_protected_symlinks __read_mostly = 1; +int sysctl_protected_hardlinks __read_mostly = 1; + +/** + * may_follow_link - Check symlink following for unsafe situations + * @link: The path of the symlink + * + * In the case of the sysctl_protected_symlinks sysctl being enabled, + * CAP_DAC_OVERRIDE needs to be specifically ignored if the symlink is + * in a sticky world-writable directory. This is to protect privileged + * processes from failing races against path names that may change out + * from under them by way of other users creating malicious symlinks. + * It will permit symlinks to be followed only when outside a sticky + * world-writable directory, or when the uid of the symlink and follower + * match, or when the directory owner matches the symlink's owner. + * + * Returns 0 if following the symlink is allowed, -ve on error. + */ +static inline int may_follow_link(struct path *link, struct nameidata *nd) +{ + const struct inode *inode; + const struct inode *parent; + + if (!sysctl_protected_symlinks) + return 0; + + /* Allowed if owner and follower match. */ + inode = link->dentry->d_inode; + if (current_cred()->fsuid == inode->i_uid) + return 0; + + /* Allowed if parent directory not sticky and world-writable. */ + parent = nd->path.dentry->d_inode; + if ((parent->i_mode & (S_ISVTX|S_IWOTH)) != (S_ISVTX|S_IWOTH)) + return 0; + + /* Allowed if parent directory and link owner match. */ + if (parent->i_uid == inode->i_uid) + return 0; + + path_put_conditional(link, nd); + path_put(&nd->path); + return -EACCES; +} + +/** + * safe_hardlink_source - Check for safe hardlink conditions + * @inode: the source inode to hardlink from + * + * Return false if at least one of the following conditions: + * - inode is not a regular file + * - inode is setuid + * - inode is setgid and group-exec + * - access failure for read and write + * + * Otherwise returns true. + */ +static bool safe_hardlink_source(struct inode *inode) +{ + umode_t mode = inode->i_mode; + + /* Special files should not get pinned to the filesystem. */ + if (!S_ISREG(mode)) + return false; + + /* Setuid files should not get pinned to the filesystem. */ + if (mode & S_ISUID) + return false; + + /* Executable setgid files should not get pinned to the filesystem. */ + if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP)) + return false; + + /* Hardlinking to unreadable or unwritable sources is dangerous. */ + if (inode_permission(inode, MAY_READ | MAY_WRITE)) + return false; + + return true; +} + +/** + * may_linkat - Check permissions for creating a hardlink + * @link: the source to hardlink from + * + * Block hardlink when all of: + * - sysctl_protected_hardlinks enabled + * - fsuid does not match inode + * - hardlink source is unsafe (see safe_hardlink_source() above) + * - not CAP_FOWNER + * + * Returns 0 if successful, -ve on error. + */ +static int may_linkat(struct path *link) +{ + const struct cred *cred; + struct inode *inode; + + if (!sysctl_protected_hardlinks) + return 0; + + cred = current_cred(); + inode = link->dentry->d_inode; + + /* Source inode owner (or CAP_FOWNER) can hardlink all they like, + * otherwise, it must be a safe source. + */ + if (cred->fsuid == inode->i_uid || safe_hardlink_source(inode) || + capable(CAP_FOWNER)) + return 0; + + return -EPERM; +} + static __always_inline int follow_link(struct path *link, struct nameidata *nd, void **p) { @@ -1818,6 +1931,9 @@ static int path_lookupat(int dfd, const char *name, while (err > 0) { void *cookie; struct path link = path; + err = may_follow_link(&link, nd); + if (unlikely(err)) + break; nd->flags |= LOOKUP_PARENT; err = follow_link(&link, nd, &cookie); if (err) @@ -2778,6 +2894,9 @@ static struct file *path_openat(int dfd, const char *pathname, error = -ELOOP; break; } + error = may_follow_link(&link, nd); + if (unlikely(error)) + break; nd->flags |= LOOKUP_PARENT; nd->flags &= ~(LOOKUP_OPEN|LOOKUP_CREATE|LOOKUP_EXCL); error = follow_link(&link, nd, &cookie); @@ -3421,6 +3540,9 @@ SYSCALL_DEFINE5(linkat, int, olddfd, const char __user *, oldname, error = -EXDEV; if (old_path.mnt != new_path.mnt) goto out_dput; + error = may_linkat(&old_path); + if (unlikely(error)) + goto out_dput; error = security_path_link(old_path.dentry, &new_path, new_dentry); if (error) goto out_dput; diff --git a/include/linux/fs.h b/include/linux/fs.h index 478237844648..80c819cbe272 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -437,6 +437,8 @@ extern unsigned long get_max_files(void); extern int sysctl_nr_open; extern struct inodes_stat_t inodes_stat; extern int leases_enable, lease_break_time; +extern int sysctl_protected_symlinks; +extern int sysctl_protected_hardlinks; struct buffer_head; typedef int (get_block_t)(struct inode *inode, sector_t iblock, diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 4ab11879aeb4..5d9a1d2b27b4 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1493,6 +1493,24 @@ static struct ctl_table fs_table[] = { }, #endif #endif + { + .procname = "protected_symlinks", + .data = &sysctl_protected_symlinks, + .maxlen = sizeof(int), + .mode = 0600, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, + { + .procname = "protected_hardlinks", + .data = &sysctl_protected_hardlinks, + .maxlen = sizeof(int), + .mode = 0600, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, { .procname = "suid_dumpable", .data = &suid_dumpable, -- cgit v1.2.3 From a51d9eaa41866ab6b4b6ecad7b621f8b66ece0dc Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 25 Jul 2012 17:29:08 -0700 Subject: fs: add link restriction audit reporting Adds audit messages for unexpected link restriction violations so that system owners will have some sort of potentially actionable information about misbehaving processes. Signed-off-by: Kees Cook Signed-off-by: Al Viro --- fs/namei.c | 2 ++ include/linux/audit.h | 4 ++++ kernel/audit.c | 21 +++++++++++++++++++++ 3 files changed, 27 insertions(+) (limited to 'kernel') diff --git a/fs/namei.c b/fs/namei.c index 3861d85f8488..618d3531cf9f 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -692,6 +692,7 @@ static inline int may_follow_link(struct path *link, struct nameidata *nd) path_put_conditional(link, nd); path_put(&nd->path); + audit_log_link_denied("follow_link", link); return -EACCES; } @@ -760,6 +761,7 @@ static int may_linkat(struct path *link) capable(CAP_FOWNER)) return 0; + audit_log_link_denied("linkat", link); return -EPERM; } diff --git a/include/linux/audit.h b/include/linux/audit.h index 22f292a917a3..36abf2aa7e68 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -130,6 +130,7 @@ #define AUDIT_LAST_KERN_ANOM_MSG 1799 #define AUDIT_ANOM_PROMISCUOUS 1700 /* Device changed promiscuous mode */ #define AUDIT_ANOM_ABEND 1701 /* Process ended abnormally */ +#define AUDIT_ANOM_LINK 1702 /* Suspicious use of file links */ #define AUDIT_INTEGRITY_DATA 1800 /* Data integrity verification */ #define AUDIT_INTEGRITY_METADATA 1801 /* Metadata integrity verification */ #define AUDIT_INTEGRITY_STATUS 1802 /* Integrity enable status */ @@ -687,6 +688,8 @@ extern void audit_log_d_path(struct audit_buffer *ab, const struct path *path); extern void audit_log_key(struct audit_buffer *ab, char *key); +extern void audit_log_link_denied(const char *operation, + struct path *link); extern void audit_log_lost(const char *message); #ifdef CONFIG_SECURITY extern void audit_log_secctx(struct audit_buffer *ab, u32 secid); @@ -716,6 +719,7 @@ extern int audit_enabled; #define audit_log_untrustedstring(a,s) do { ; } while (0) #define audit_log_d_path(b, p, d) do { ; } while (0) #define audit_log_key(b, k) do { ; } while (0) +#define audit_log_link_denied(o, l) do { ; } while (0) #define audit_log_secctx(b,s) do { ; } while (0) #define audit_enabled 0 #endif diff --git a/kernel/audit.c b/kernel/audit.c index 1c7f2c61416b..fda8bd9e1d3a 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -1449,6 +1449,27 @@ void audit_log_key(struct audit_buffer *ab, char *key) audit_log_format(ab, "(null)"); } +/** + * audit_log_link_denied - report a link restriction denial + * @operation: specific link opreation + * @link: the path that triggered the restriction + */ +void audit_log_link_denied(const char *operation, struct path *link) +{ + struct audit_buffer *ab; + + ab = audit_log_start(current->audit_context, GFP_KERNEL, + AUDIT_ANOM_LINK); + audit_log_format(ab, "op=%s action=denied", operation); + audit_log_format(ab, " pid=%d comm=", current->pid); + audit_log_untrustedstring(ab, current->comm); + audit_log_d_path(ab, " path=", link); + audit_log_format(ab, " dev="); + audit_log_untrustedstring(ab, link->dentry->d_inode->i_sb->s_id); + audit_log_format(ab, " ino=%lu", link->dentry->d_inode->i_ino); + audit_log_end(ab); +} + /** * audit_log_end - end one audit record * @ab: the audit_buffer -- cgit v1.2.3 From f403072c6108e15f319a4a5036b650c77760522c Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:12 +0200 Subject: uprobes: Don't recheck vma/f_mapping in write_opcode() write_opcode() rechecks valid_vma() and ->f_mapping, this is pointless. The caller, register_for_each_vma() or uprobe_mmap(), has already done these checks under mmap_sem. To clarify, uprobe_mmap() checks valid_vma() only, but we can rely on build_probe_list(vm_file->f_mapping->host). Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182212.GA20304@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 19 +------------------ 1 file changed, 1 insertion(+), 18 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index f93532748bca..a2b32a51d0a2 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -206,33 +206,16 @@ static int write_opcode(struct arch_uprobe *auprobe, struct mm_struct *mm, unsigned long vaddr, uprobe_opcode_t opcode) { struct page *old_page, *new_page; - struct address_space *mapping; void *vaddr_old, *vaddr_new; struct vm_area_struct *vma; - struct uprobe *uprobe; int ret; + retry: /* Read the page with vaddr into memory */ ret = get_user_pages(NULL, mm, vaddr, 1, 0, 0, &old_page, &vma); if (ret <= 0) return ret; - ret = -EINVAL; - - /* - * We are interested in text pages only. Our pages of interest - * should be mapped for read and execute only. We desist from - * adding probes in write mapped pages since the breakpoints - * might end up in the file copy. - */ - if (!valid_vma(vma, is_swbp_insn(&opcode))) - goto put_out; - - uprobe = container_of(auprobe, struct uprobe, arch); - mapping = uprobe->inode->i_mapping; - if (mapping != vma->vm_file->f_mapping) - goto put_out; - ret = -ENOMEM; new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, vaddr); if (!new_page) -- cgit v1.2.3 From c517ee744b96e441d9c731e245f83c6d08dc0a19 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:16 +0200 Subject: uprobes: __replace_page() should not use page_address_in_vma() page_address_in_vma(old_page) in __replace_page() is ugly and wrong. The caller already knows the correct virtual address, this page was found by get_user_pages(vaddr). However, page_address_in_vma() can actually fail if page->mapping was cleared by __delete_from_page_cache() after get_user_pages() returns. But this means the race with page reclaim, write_opcode() should not fail, it should retry and read this page again. Probably the race with remove_mapping() is not possible due to page_freeze_refs() logic, but afaics at least shmem_writepage()->shmem_delete_from_page_cache() can clear ->mapping. We could change __replace_page() to return -EAGAIN in this case, but it would be better to simply use the caller's vaddr and rely on page_check_address(). Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182216.GA20311@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index a2b32a51d0a2..6fda7996892b 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -127,22 +127,19 @@ static loff_t vma_address(struct vm_area_struct *vma, loff_t offset) * based on replace_page in mm/ksm.c * * @vma: vma that holds the pte pointing to page + * @addr: address the old @page is mapped at * @page: the cowed page we are replacing by kpage * @kpage: the modified page we replace page by * * Returns 0 on success, -EFAULT on failure. */ -static int __replace_page(struct vm_area_struct *vma, struct page *page, struct page *kpage) +static int __replace_page(struct vm_area_struct *vma, unsigned long addr, + struct page *page, struct page *kpage) { struct mm_struct *mm = vma->vm_mm; - unsigned long addr; spinlock_t *ptl; pte_t *ptep; - addr = page_address_in_vma(page, vma); - if (addr == -EFAULT) - return -EFAULT; - ptep = page_check_address(page, mm, addr, &ptl, 0); if (!ptep) return -EAGAIN; @@ -243,7 +240,7 @@ retry: goto unlock_out; lock_page(new_page); - ret = __replace_page(vma, old_page, new_page); + ret = __replace_page(vma, vaddr, old_page, new_page); unlock_page(new_page); unlock_out: -- cgit v1.2.3 From 089ba999dc881a7549d97c55ac9e0052d061867d Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:18 +0200 Subject: uprobes: Kill write_opcode()->lock_page(new_page) write_opcode() does lock_page(new_page) for no reason. Nobody can see this page until __replace_page() exposes it under ptl lock, and we do nothing with this page after pte_unmap_unlock(). If nothing else, the similar code in do_wp_page() doesn't lock the new page for page_add_new_anon_rmap/set_pte_at_notify. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182218.GA20315@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 6fda7996892b..23c562b7fc2e 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -239,9 +239,7 @@ retry: if (ret) goto unlock_out; - lock_page(new_page); ret = __replace_page(vma, vaddr, old_page, new_page); - unlock_page(new_page); unlock_out: unlock_page(old_page); -- cgit v1.2.3 From 9f92448ceeea5326db7d114005a7e7ac03904edf Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:20 +0200 Subject: uprobes: Clean up and document write_opcode()->lock_page(old_page) The comment above write_opcode()->lock_page(old_page) tells about the race with do_wp_page(). I don't really understand which exactly race it means, but afaics this lock_page() was not enough to close all races with do_wp_page(). Anyway, since: 77fc4af1b59d uprobes: Change register_for_each_vma() to take mm->mmap_sem for writing this code is always called with ->mmap_sem held for writing, so we can forget about do_wp_page(). However, we can't simply remove this lock_page(), and the only (afaics) reason is __replace_page()->try_to_free_swap(). Nothing in write_opcode() needs it, move it into __replace_page() and fix the comment. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182220.GA20322@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 23c562b7fc2e..5db150b306d2 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -139,10 +139,15 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, struct mm_struct *mm = vma->vm_mm; spinlock_t *ptl; pte_t *ptep; + int err; + /* freeze PageSwapCache() for try_to_free_swap() below */ + lock_page(page); + + err = -EAGAIN; ptep = page_check_address(page, mm, addr, &ptl, 0); if (!ptep) - return -EAGAIN; + goto unlock; get_page(kpage); page_add_new_anon_rmap(kpage, vma, addr); @@ -162,7 +167,10 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, put_page(page); pte_unmap_unlock(ptep, ptl); - return 0; + err = 0; + unlock: + unlock_page(page); + return err; } /** @@ -216,15 +224,10 @@ retry: ret = -ENOMEM; new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, vaddr); if (!new_page) - goto put_out; + goto put_old; __SetPageUptodate(new_page); - /* - * lock page will serialize against do_wp_page()'s - * PageAnon() handling - */ - lock_page(old_page); /* copy the page now that we've got it stable */ vaddr_old = kmap_atomic(old_page); vaddr_new = kmap_atomic(new_page); @@ -237,15 +240,13 @@ retry: ret = anon_vma_prepare(vma); if (ret) - goto unlock_out; + goto put_new; ret = __replace_page(vma, vaddr, old_page, new_page); -unlock_out: - unlock_page(old_page); +put_new: page_cache_release(new_page); - -put_out: +put_old: put_page(old_page); if (unlikely(ret == -EAGAIN)) -- cgit v1.2.3 From 665605a2a207dbe1fa429b474f932d6ea138ba92 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:29 +0200 Subject: uprobes: Uprobe_mmap/munmap needs list_for_each_entry_safe() The bug was introduced by me in 449d0d7c ("uprobes: Simplify the usage of uprobe->pending_list"). Yes, we do not care about uprobe->pending_list after return and nobody can remove the current list entry, but put_uprobe(uprobe) can actually free it and thus we need list_for_each_safe(). Reported-by: Srikar Dronamraju Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Link: http://lkml.kernel.org/r/20120729182229.GA20329@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 5db150b306d2..bed2161620d7 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -1010,7 +1010,7 @@ static void build_probe_list(struct inode *inode, struct list_head *head) int uprobe_mmap(struct vm_area_struct *vma) { struct list_head tmp_list; - struct uprobe *uprobe; + struct uprobe *uprobe, *u; struct inode *inode; int ret, count; @@ -1028,7 +1028,7 @@ int uprobe_mmap(struct vm_area_struct *vma) ret = 0; count = 0; - list_for_each_entry(uprobe, &tmp_list, pending_list) { + list_for_each_entry_safe(uprobe, u, &tmp_list, pending_list) { if (!ret) { loff_t vaddr = vma_address(vma, uprobe->offset); @@ -1076,7 +1076,7 @@ int uprobe_mmap(struct vm_area_struct *vma) void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned long end) { struct list_head tmp_list; - struct uprobe *uprobe; + struct uprobe *uprobe, *u; struct inode *inode; if (!atomic_read(&uprobe_events) || !valid_vma(vma, false)) @@ -1093,7 +1093,7 @@ void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned lon mutex_lock(uprobes_mmap_hash(inode)); build_probe_list(inode, &tmp_list); - list_for_each_entry(uprobe, &tmp_list, pending_list) { + list_for_each_entry_safe(uprobe, u, &tmp_list, pending_list) { loff_t vaddr = vma_address(vma, uprobe->offset); if (vaddr >= start && vaddr < end) { -- cgit v1.2.3 From 2fd611a991391a6050cbd139201a2e12fc306540 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:31 +0200 Subject: uprobes: Suppress uprobe_munmap() from mmput() uprobe_munmap() does get_user_pages() and it is also called from the final mmput()->exit_mmap() path. This slows down exit/mmput() for no reason, and I think it is simply dangerous/wrong to try to fault-in a page into the dying mm. If nothing else, this happens after the last sync_mm_rss(), afaics handle_mm_fault() can change the task->rss_stat and make the subsequent check_mm() unhappy. Change uprobe_munmap() to check mm->mm_users != 0. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182231.GA20336@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index bed2161620d7..9db9cdf8ff34 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -1082,6 +1082,9 @@ void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned lon if (!atomic_read(&uprobe_events) || !valid_vma(vma, false)) return; + if (!atomic_read(&vma->vm_mm->mm_users)) /* called by mmput() ? */ + return; + if (!atomic_read(&vma->vm_mm->uprobes_state.count)) return; -- cgit v1.2.3 From aefd8933d445abf7ff0d4027c624737898827bcd Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:33 +0200 Subject: uprobes: Fix overflow in vma_address()/find_active_uprobe() vma->vm_pgoff is "unsigned long", it should be promoted to loff_t before the multiplication to avoid the overflow. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182233.GA20339@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 9db9cdf8ff34..fb961d5e205c 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -117,7 +117,7 @@ static loff_t vma_address(struct vm_area_struct *vma, loff_t offset) loff_t vaddr; vaddr = vma->vm_start + offset; - vaddr -= vma->vm_pgoff << PAGE_SHIFT; + vaddr -= (loff_t)vma->vm_pgoff << PAGE_SHIFT; return vaddr; } @@ -1450,7 +1450,7 @@ static struct uprobe *find_active_uprobe(unsigned long bp_vaddr, int *is_swbp) inode = vma->vm_file->f_mapping->host; offset = bp_vaddr - vma->vm_start; - offset += (vma->vm_pgoff << PAGE_SHIFT); + offset += (loff_t)vma->vm_pgoff << PAGE_SHIFT; uprobe = find_uprobe(inode, offset); } -- cgit v1.2.3 From 891c39708144bbe401b4aa151bffd0fe41b1dafd Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:40 +0200 Subject: uprobes: Teach build_probe_list() to consider the range Currently build_probe_list() builds the list of all uprobes attached to the given inode, and the caller should filter out those who don't fall into the [start,end) range, this is sub-optimal. This patch turns find_least_offset_node() into find_node_in_range() which returns the first node inside the [min,max] range, and changes build_probe_list() to use this node as a starting point for rb_prev() and rb_next() to find all other nodes the caller needs. The resulting list is no longer sorted but we do not care. This can speed up both build_probe_list() and the callers, but there is another reason to introduce find_node_in_range(). It can be used to figure out whether the given vma has uprobes or not, this will be needed soon. While at it, shift INIT_LIST_HEAD(tmp_list) into build_probe_list(). Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182240.GA20352@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 103 +++++++++++++++++++++++------------------------- 1 file changed, 50 insertions(+), 53 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index fb961d5e205c..2ef123306183 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -939,59 +939,66 @@ void uprobe_unregister(struct inode *inode, loff_t offset, struct uprobe_consume put_uprobe(uprobe); } -/* - * Of all the nodes that correspond to the given inode, return the node - * with the least offset. - */ -static struct rb_node *find_least_offset_node(struct inode *inode) +static struct rb_node * +find_node_in_range(struct inode *inode, loff_t min, loff_t max) { - struct uprobe u = { .inode = inode, .offset = 0}; struct rb_node *n = uprobes_tree.rb_node; - struct rb_node *close_node = NULL; - struct uprobe *uprobe; - int match; while (n) { - uprobe = rb_entry(n, struct uprobe, rb_node); - match = match_uprobe(&u, uprobe); - - if (uprobe->inode == inode) - close_node = n; + struct uprobe *u = rb_entry(n, struct uprobe, rb_node); - if (!match) - return close_node; - - if (match < 0) + if (inode < u->inode) { n = n->rb_left; - else + } else if (inode > u->inode) { n = n->rb_right; + } else { + if (max < u->offset) + n = n->rb_left; + else if (min > u->offset) + n = n->rb_right; + else + break; + } } - return close_node; + return n; } /* - * For a given inode, build a list of probes that need to be inserted. + * For a given range in vma, build a list of probes that need to be inserted. */ -static void build_probe_list(struct inode *inode, struct list_head *head) +static void build_probe_list(struct inode *inode, + struct vm_area_struct *vma, + unsigned long start, unsigned long end, + struct list_head *head) { - struct uprobe *uprobe; + loff_t min, max; unsigned long flags; - struct rb_node *n; - - spin_lock_irqsave(&uprobes_treelock, flags); - - n = find_least_offset_node(inode); + struct rb_node *n, *t; + struct uprobe *u; - for (; n; n = rb_next(n)) { - uprobe = rb_entry(n, struct uprobe, rb_node); - if (uprobe->inode != inode) - break; + INIT_LIST_HEAD(head); + min = ((loff_t)vma->vm_pgoff << PAGE_SHIFT) + start - vma->vm_start; + max = min + (end - start) - 1; - list_add(&uprobe->pending_list, head); - atomic_inc(&uprobe->ref); + spin_lock_irqsave(&uprobes_treelock, flags); + n = find_node_in_range(inode, min, max); + if (n) { + for (t = n; t; t = rb_prev(t)) { + u = rb_entry(t, struct uprobe, rb_node); + if (u->inode != inode || u->offset < min) + break; + list_add(&u->pending_list, head); + atomic_inc(&u->ref); + } + for (t = n; (t = rb_next(t)); ) { + u = rb_entry(t, struct uprobe, rb_node); + if (u->inode != inode || u->offset > max) + break; + list_add(&u->pending_list, head); + atomic_inc(&u->ref); + } } - spin_unlock_irqrestore(&uprobes_treelock, flags); } @@ -1021,9 +1028,8 @@ int uprobe_mmap(struct vm_area_struct *vma) if (!inode) return 0; - INIT_LIST_HEAD(&tmp_list); mutex_lock(uprobes_mmap_hash(inode)); - build_probe_list(inode, &tmp_list); + build_probe_list(inode, vma, vma->vm_start, vma->vm_end, &tmp_list); ret = 0; count = 0; @@ -1032,11 +1038,6 @@ int uprobe_mmap(struct vm_area_struct *vma) if (!ret) { loff_t vaddr = vma_address(vma, uprobe->offset); - if (vaddr < vma->vm_start || vaddr >= vma->vm_end) { - put_uprobe(uprobe); - continue; - } - ret = install_breakpoint(uprobe, vma->vm_mm, vma, vaddr); /* * We can race against uprobe_register(), see the @@ -1092,21 +1093,17 @@ void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned lon if (!inode) return; - INIT_LIST_HEAD(&tmp_list); mutex_lock(uprobes_mmap_hash(inode)); - build_probe_list(inode, &tmp_list); + build_probe_list(inode, vma, start, end, &tmp_list); list_for_each_entry_safe(uprobe, u, &tmp_list, pending_list) { loff_t vaddr = vma_address(vma, uprobe->offset); - - if (vaddr >= start && vaddr < end) { - /* - * An unregister could have removed the probe before - * unmap. So check before we decrement the count. - */ - if (is_swbp_at_addr(vma->vm_mm, vaddr) == 1) - atomic_dec(&vma->vm_mm->uprobes_state.count); - } + /* + * An unregister could have removed the probe before + * unmap. So check before we decrement the count. + */ + if (is_swbp_at_addr(vma->vm_mm, vaddr) == 1) + atomic_dec(&vma->vm_mm->uprobes_state.count); put_uprobe(uprobe); } mutex_unlock(uprobes_mmap_hash(inode)); -- cgit v1.2.3 From cb113b47d098185f3f1f67e8300d05ddce842b66 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:42 +0200 Subject: uprobes: Introduce vaddr_to_offset(vma, vaddr) Add the new helper, vaddr_to_offset(vma, vaddr) which returns the offset in vma->vm_file this vaddr is mapped at. Change build_probe_list() and find_active_uprobe() to use the new helper, the next patch adds another user. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182242.GA20355@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 2ef123306183..b03256cced52 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -122,6 +122,11 @@ static loff_t vma_address(struct vm_area_struct *vma, loff_t offset) return vaddr; } +static loff_t vaddr_to_offset(struct vm_area_struct *vma, unsigned long vaddr) +{ + return ((loff_t)vma->vm_pgoff << PAGE_SHIFT) + (vaddr - vma->vm_start); +} + /** * __replace_page - replace page in vma by new page. * based on replace_page in mm/ksm.c @@ -978,7 +983,7 @@ static void build_probe_list(struct inode *inode, struct uprobe *u; INIT_LIST_HEAD(head); - min = ((loff_t)vma->vm_pgoff << PAGE_SHIFT) + start - vma->vm_start; + min = vaddr_to_offset(vma, start); max = min + (end - start) - 1; spin_lock_irqsave(&uprobes_treelock, flags); @@ -1442,12 +1447,9 @@ static struct uprobe *find_active_uprobe(unsigned long bp_vaddr, int *is_swbp) vma = find_vma(mm, bp_vaddr); if (vma && vma->vm_start <= bp_vaddr) { if (valid_vma(vma, false)) { - struct inode *inode; - loff_t offset; + struct inode *inode = vma->vm_file->f_mapping->host; + loff_t offset = vaddr_to_offset(vma, bp_vaddr); - inode = vma->vm_file->f_mapping->host; - offset = bp_vaddr - vma->vm_start; - offset += (loff_t)vma->vm_pgoff << PAGE_SHIFT; uprobe = find_uprobe(inode, offset); } -- cgit v1.2.3 From f4d6dfe55115efe981b4b5f37183ddccaaa792f0 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:44 +0200 Subject: uprobes: Fix register_for_each_vma()->vma_address() check 1. register_for_each_vma() checks that vma_address() == vaddr, but this is not enough. We should also ensure that vaddr >= vm_start, find_vma() guarantees "vaddr < vm_end" only. 2. After the prevous changes, register_for_each_vma() is the only reason why vma_address() has to return loff_t, all other users know that we have the valid mapping at this offset and thus the overflow is not possible. Change the code to use vaddr_to_offset() instead, imho this looks more clean/understandable and now we can change vma_address(). 3. While at it, remove the unnecessary type-cast. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182244.GA20362@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index b03256cced52..cdc3c951251c 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -823,12 +823,13 @@ static int register_for_each_vma(struct uprobe *uprobe, bool is_register) goto free; down_write(&mm->mmap_sem); - vma = find_vma(mm, (unsigned long)info->vaddr); - if (!vma || !valid_vma(vma, is_register)) + vma = find_vma(mm, info->vaddr); + if (!vma || !valid_vma(vma, is_register) || + vma->vm_file->f_mapping->host != uprobe->inode) goto unlock; - if (vma->vm_file->f_mapping->host != uprobe->inode || - vma_address(vma, uprobe->offset) != info->vaddr) + if (vma->vm_start > info->vaddr || + vaddr_to_offset(vma, info->vaddr) != uprobe->offset) goto unlock; if (is_register) { -- cgit v1.2.3 From 57683f72b8c01c53c85fe13e82fe1feb3a06bcd5 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:47 +0200 Subject: uprobes: Rename vma_address() and make it return "unsigned long" 1. vma_address() returns loff_t, this looks confusing and this is unnecessary after the previous change. Make it return "ulong", all callers truncate the result anyway. 2. Its name conflicts with mm/rmap.c:vma_address(), rename it to offset_to_vaddr(), this matches vaddr_to_offset(). Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182247.GA20365@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index cdc3c951251c..bb30a4fc5050 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -112,14 +112,9 @@ static bool valid_vma(struct vm_area_struct *vma, bool is_register) return false; } -static loff_t vma_address(struct vm_area_struct *vma, loff_t offset) +static unsigned long offset_to_vaddr(struct vm_area_struct *vma, loff_t offset) { - loff_t vaddr; - - vaddr = vma->vm_start + offset; - vaddr -= (loff_t)vma->vm_pgoff << PAGE_SHIFT; - - return vaddr; + return vma->vm_start + offset - ((loff_t)vma->vm_pgoff << PAGE_SHIFT); } static loff_t vaddr_to_offset(struct vm_area_struct *vma, unsigned long vaddr) @@ -775,7 +770,7 @@ build_map_info(struct address_space *mapping, loff_t offset, bool is_register) curr = info; info->mm = vma->vm_mm; - info->vaddr = vma_address(vma, offset); + info->vaddr = offset_to_vaddr(vma, offset); } mutex_unlock(&mapping->i_mmap_mutex); @@ -1042,7 +1037,7 @@ int uprobe_mmap(struct vm_area_struct *vma) list_for_each_entry_safe(uprobe, u, &tmp_list, pending_list) { if (!ret) { - loff_t vaddr = vma_address(vma, uprobe->offset); + unsigned long vaddr = offset_to_vaddr(vma, uprobe->offset); ret = install_breakpoint(uprobe, vma->vm_mm, vma, vaddr); /* @@ -1103,7 +1098,7 @@ void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned lon build_probe_list(inode, vma, start, end, &tmp_list); list_for_each_entry_safe(uprobe, u, &tmp_list, pending_list) { - loff_t vaddr = vma_address(vma, uprobe->offset); + unsigned long vaddr = offset_to_vaddr(vma, uprobe->offset); /* * An unregister could have removed the probe before * unmap. So check before we decrement the count. -- cgit v1.2.3 From 194f8dcbe9629d8e9346cf96345a9c0bbf0e67ae Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 29 Jul 2012 20:22:49 +0200 Subject: uprobes: __replace_page() needs munlock_vma_page() Like do_wp_page(), __replace_page() should do munlock_vma_page() for the case when the old page still has other !VM_LOCKED mappings. Unfortunately this needs mm/internal.h. Also, move put_page() outside of ptl lock. This doesn't really matter but looks a bit better. Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: Anton Arapov Cc: Srikar Dronamraju Link: http://lkml.kernel.org/r/20120729182249.GA20372@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/uprobes.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index bb30a4fc5050..c08a22d02f72 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -32,6 +32,7 @@ #include /* try_to_free_swap */ #include /* user_enable_single_step */ #include /* notifier mechanism */ +#include "../../mm/internal.h" /* munlock_vma_page */ #include @@ -141,7 +142,7 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep; int err; - /* freeze PageSwapCache() for try_to_free_swap() below */ + /* For try_to_free_swap() and munlock_vma_page() below */ lock_page(page); err = -EAGAIN; @@ -164,9 +165,12 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr, page_remove_rmap(page); if (!page_mapped(page)) try_to_free_swap(page); - put_page(page); pte_unmap_unlock(ptep, ptl); + if (vma->vm_flags & VM_LOCKED) + munlock_vma_page(page); + put_page(page); + err = 0; unlock: unlock_page(page); -- cgit v1.2.3 From f1fd75bfa07822b1de314062baff3280419a8bf4 Mon Sep 17 00:00:00 2001 From: Sasikantha babu Date: Mon, 30 Jul 2012 14:39:08 -0700 Subject: prctl: remove redunant assignment of "error" to zero Just setting the "error" to error number is enough on failure and It doesn't require to set "error" variable to zero in each switch case, since it was already initialized with zero. And also removed return 0 in switch case with break statement Signed-off-by: Sasikantha babu Acked-by: Kees Cook Acked-by: Serge E. Hallyn Cc: Cyrill Gorcunov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sys.c | 13 ++----------- 1 file changed, 2 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/kernel/sys.c b/kernel/sys.c index 2d39a84cd857..b04ae0390df3 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -2015,7 +2015,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, break; } me->pdeath_signal = arg2; - error = 0; break; case PR_GET_PDEATHSIG: error = put_user(me->pdeath_signal, (int __user *)arg2); @@ -2029,7 +2028,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, break; } set_dumpable(me->mm, arg2); - error = 0; break; case PR_SET_UNALIGN: @@ -2056,10 +2054,7 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, case PR_SET_TIMING: if (arg2 != PR_TIMING_STATISTICAL) error = -EINVAL; - else - error = 0; break; - case PR_SET_NAME: comm[sizeof(me->comm)-1] = 0; if (strncpy_from_user(comm, (char __user *)arg2, @@ -2067,20 +2062,19 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, return -EFAULT; set_task_comm(me, comm); proc_comm_connector(me); - return 0; + break; case PR_GET_NAME: get_task_comm(comm, me); if (copy_to_user((char __user *)arg2, comm, sizeof(comm))) return -EFAULT; - return 0; + break; case PR_GET_ENDIAN: error = GET_ENDIAN(me, arg2); break; case PR_SET_ENDIAN: error = SET_ENDIAN(me, arg2); break; - case PR_GET_SECCOMP: error = prctl_get_seccomp(); break; @@ -2108,7 +2102,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, current->default_timer_slack_ns; else current->timer_slack_ns = arg2; - error = 0; break; case PR_MCE_KILL: if (arg4 | arg5) @@ -2134,7 +2127,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, default: return -EINVAL; } - error = 0; break; case PR_MCE_KILL_GET: if (arg2 | arg3 | arg4 | arg5) @@ -2153,7 +2145,6 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, break; case PR_SET_CHILD_SUBREAPER: me->signal->is_child_subreaper = !!arg2; - error = 0; break; case PR_GET_CHILD_SUBREAPER: error = put_user(me->signal->is_child_subreaper, -- cgit v1.2.3 From 54b501992dd2a839e94e76aa392c392b55080ce8 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Mon, 30 Jul 2012 14:39:18 -0700 Subject: coredump: warn about unsafe suid_dumpable / core_pattern combo When suid_dumpable=2, detect unsafe core_pattern settings and warn when they are seen. Signed-off-by: Kees Cook Suggested-by: Andrew Morton Cc: Alexander Viro Cc: Alan Cox Cc: "Eric W. Biederman" Cc: Doug Ledford Cc: Serge Hallyn Cc: James Morris Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 10 +++++----- include/linux/sched.h | 5 +++++ kernel/sysctl.c | 37 +++++++++++++++++++++++++++++++++++-- 3 files changed, 45 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/fs/exec.c b/fs/exec.c index 95aae3f9c036..5af8390e0fae 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -2002,17 +2002,17 @@ static void coredump_finish(struct mm_struct *mm) void set_dumpable(struct mm_struct *mm, int value) { switch (value) { - case 0: + case SUID_DUMPABLE_DISABLED: clear_bit(MMF_DUMPABLE, &mm->flags); smp_wmb(); clear_bit(MMF_DUMP_SECURELY, &mm->flags); break; - case 1: + case SUID_DUMPABLE_ENABLED: set_bit(MMF_DUMPABLE, &mm->flags); smp_wmb(); clear_bit(MMF_DUMP_SECURELY, &mm->flags); break; - case 2: + case SUID_DUMPABLE_SAFE: set_bit(MMF_DUMP_SECURELY, &mm->flags); smp_wmb(); set_bit(MMF_DUMPABLE, &mm->flags); @@ -2025,7 +2025,7 @@ static int __get_dumpable(unsigned long mm_flags) int ret; ret = mm_flags & MMF_DUMPABLE_MASK; - return (ret >= 2) ? 2 : ret; + return (ret > SUID_DUMPABLE_ENABLED) ? SUID_DUMPABLE_SAFE : ret; } int get_dumpable(struct mm_struct *mm) @@ -2142,7 +2142,7 @@ void do_coredump(long signr, int exit_code, struct pt_regs *regs) * so we dump it as root in mode 2, and only into a controlled * environment (pipe handler or fully qualified path). */ - if (__get_dumpable(cprm.mm_flags) == 2) { + if (__get_dumpable(cprm.mm_flags) == SUID_DUMPABLE_SAFE) { /* Setuid core dump mode */ flag = O_EXCL; /* Stop rewrite attacks */ cred->fsuid = GLOBAL_ROOT_UID; /* Dump root private */ diff --git a/include/linux/sched.h b/include/linux/sched.h index a721cef7e2d4..1e26a5e45aa6 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -406,6 +406,11 @@ static inline void arch_pick_mmap_layout(struct mm_struct *mm) {} extern void set_dumpable(struct mm_struct *mm, int value); extern int get_dumpable(struct mm_struct *mm); +/* get/set_dumpable() values */ +#define SUID_DUMPABLE_DISABLED 0 +#define SUID_DUMPABLE_ENABLED 1 +#define SUID_DUMPABLE_SAFE 2 + /* mm flags */ /* dumpable bits */ #define MMF_DUMPABLE 0 /* core dump is permitted */ diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 4ab11879aeb4..b46f496405e4 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -174,6 +174,11 @@ static int proc_dointvec_minmax_sysadmin(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos); #endif +static int proc_dointvec_minmax_coredump(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos); +static int proc_dostring_coredump(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos); + #ifdef CONFIG_MAGIC_SYSRQ /* Note: sysrq code uses it's own private copy */ static int __sysrq_enabled = SYSRQ_DEFAULT_ENABLE; @@ -410,7 +415,7 @@ static struct ctl_table kern_table[] = { .data = core_pattern, .maxlen = CORENAME_MAX_SIZE, .mode = 0644, - .proc_handler = proc_dostring, + .proc_handler = proc_dostring_coredump, }, { .procname = "core_pipe_limit", @@ -1498,7 +1503,7 @@ static struct ctl_table fs_table[] = { .data = &suid_dumpable, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = proc_dointvec_minmax, + .proc_handler = proc_dointvec_minmax_coredump, .extra1 = &zero, .extra2 = &two, }, @@ -2009,6 +2014,34 @@ int proc_dointvec_minmax(struct ctl_table *table, int write, do_proc_dointvec_minmax_conv, ¶m); } +static void validate_coredump_safety(void) +{ + if (suid_dumpable == SUID_DUMPABLE_SAFE && + core_pattern[0] != '/' && core_pattern[0] != '|') { + printk(KERN_WARNING "Unsafe core_pattern used with "\ + "suid_dumpable=2. Pipe handler or fully qualified "\ + "core dump path required.\n"); + } +} + +static int proc_dointvec_minmax_coredump(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int error = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + if (!error) + validate_coredump_safety(); + return error; +} + +static int proc_dostring_coredump(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + int error = proc_dostring(table, write, buffer, lenp, ppos); + if (!error) + validate_coredump_safety(); + return error; +} + static int __do_proc_doulongvec_minmax(void *data, struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos, -- cgit v1.2.3 From 190320c3b6640d4104650f55ff69611e050ea06b Mon Sep 17 00:00:00 2001 From: Vikram Mulukutla Date: Mon, 30 Jul 2012 14:39:58 -0700 Subject: panic: fix a possible deadlock in panic() panic_lock is meant to ensure that panic processing takes place only on one cpu; if any of the other cpus encounter a panic, they will spin waiting to be shut down. However, this causes a regression in this scenario: 1. Cpu 0 encounters a panic and acquires the panic_lock and proceeds with the panic processing. 2. There is an interrupt on cpu 0 that also encounters an error condition and invokes panic. 3. This second invocation fails to acquire the panic_lock and enters the infinite while loop in panic_smp_self_stop. Thus all panic processing is stopped, and the cpu is stuck for eternity in the while(1) inside panic_smp_self_stop. To address this, disable local interrupts with local_irq_disable before acquiring the panic_lock. This will prevent interrupt handlers from executing during the panic processing, thus avoiding this particular problem. Signed-off-by: Vikram Mulukutla Reviewed-by: Stephen Boyd Cc: Michael Holzheu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/panic.c | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'kernel') diff --git a/kernel/panic.c b/kernel/panic.c index d2a5f4ecc6dd..e1b2822fff97 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -74,6 +74,14 @@ void panic(const char *fmt, ...) long i, i_next = 0; int state = 0; + /* + * Disable local interrupts. This will prevent panic_smp_self_stop + * from deadlocking the first cpu that invokes the panic, since + * there is nothing to prevent an interrupt handler (that runs + * after the panic_lock is acquired) from invoking panic again. + */ + local_irq_disable(); + /* * It's possible to come here directly from a panic-assertion and * not have preempt disabled. Some functions called from here want -- cgit v1.2.3 From 45226e944ce071d0231949f2fea90969437cd2dc Mon Sep 17 00:00:00 2001 From: Sameer Nanda Date: Mon, 30 Jul 2012 14:40:00 -0700 Subject: NMI watchdog: fix for lockup detector breakage on resume On the suspend/resume path the boot CPU does not go though an offline->online transition. This breaks the NMI detector post-resume since it depends on PMU state that is lost when the system gets suspended. Fix this by forcing a CPU offline->online transition for the lockup detector on the boot CPU during resume. To provide more context, we enable NMI watchdog on Chrome OS. We have seen several reports of systems freezing up completely which indicated that the NMI watchdog was not firing for some reason. Debugging further, we found a simple way of repro'ing system freezes -- issuing the command 'tasket 1 sh -c "echo nmilockup > /proc/breakme"' after the system has been suspended/resumed one or more times. With this patch in place, the system freeze result in panics, as expected. These panics provide a nice stack trace for us to debug the actual issue causing the freeze. [akpm@linux-foundation.org: fiddle with code comment] [akpm@linux-foundation.org: make lockup_detector_bootcpu_resume() conditional on CONFIG_SUSPEND] [akpm@linux-foundation.org: fix section errors] Signed-off-by: Sameer Nanda Cc: Ingo Molnar Cc: Peter Zijlstra Cc: "Rafael J. Wysocki" Cc: Don Zickus Cc: Mandeep Singh Baines Cc: Srivatsa S. Bhat Cc: Anshuman Khandual Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 8 ++++++++ kernel/power/suspend.c | 3 +++ kernel/watchdog.c | 21 +++++++++++++++++++-- 3 files changed, 30 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 1e26a5e45aa6..68dcffaa62a0 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -334,6 +334,14 @@ static inline void lockup_detector_init(void) } #endif +#if defined(CONFIG_LOCKUP_DETECTOR) && defined(CONFIG_SUSPEND) +void lockup_detector_bootcpu_resume(void); +#else +static inline void lockup_detector_bootcpu_resume(void) +{ +} +#endif + #ifdef CONFIG_DETECT_HUNG_TASK extern unsigned int sysctl_hung_task_panic; extern unsigned long sysctl_hung_task_check_count; diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index c8b7446b27df..1da39ea248fd 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -178,6 +178,9 @@ static int suspend_enter(suspend_state_t state, bool *wakeup) arch_suspend_enable_irqs(); BUG_ON(irqs_disabled()); + /* Kick the lockup detector */ + lockup_detector_bootcpu_resume(); + Enable_cpus: enable_nonboot_cpus(); diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 4b1dfba70f7c..69add8a9da68 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -575,7 +575,7 @@ out: /* * Create/destroy watchdog threads as CPUs come and go: */ -static int __cpuinit +static int cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { int hotcpu = (unsigned long)hcpu; @@ -610,10 +610,27 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) return NOTIFY_OK; } -static struct notifier_block __cpuinitdata cpu_nfb = { +static struct notifier_block cpu_nfb = { .notifier_call = cpu_callback }; +#ifdef CONFIG_SUSPEND +/* + * On exit from suspend we force an offline->online transition on the boot CPU + * so that the PMU state that was lost while in suspended state gets set up + * properly for the boot CPU. This information is required for restarting the + * NMI watchdog. + */ +void lockup_detector_bootcpu_resume(void) +{ + void *cpu = (void *)(long)smp_processor_id(); + + cpu_callback(&cpu_nfb, CPU_DEAD_FROZEN, cpu); + cpu_callback(&cpu_nfb, CPU_UP_PREPARE_FROZEN, cpu); + cpu_callback(&cpu_nfb, CPU_ONLINE_FROZEN, cpu); +} +#endif + void __init lockup_detector_init(void) { void *cpu = (void *)(long)smp_processor_id(); -- cgit v1.2.3 From b57b44ae698944ffc6161352b8ff5c9cf9c592e2 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 30 Jul 2012 14:40:03 -0700 Subject: kernel/sys.c: avoid argv_free(NULL) If argv_split() failed, the code will end up calling argv_free(NULL). Fix it up and clean things up a bit. Addresses Coverity report 703573. Cc: Cyrill Gorcunov Cc: Kees Cook Cc: Serge Hallyn Cc: "Eric W. Biederman" Cc: WANG Cong Cc: Alan Cox Cc: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sys.c | 44 +++++++++++++++++++++++++------------------- 1 file changed, 25 insertions(+), 19 deletions(-) (limited to 'kernel') diff --git a/kernel/sys.c b/kernel/sys.c index b04ae0390df3..241507f23eca 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -2186,46 +2186,52 @@ static void argv_cleanup(struct subprocess_info *info) argv_free(info->argv); } -/** - * orderly_poweroff - Trigger an orderly system poweroff - * @force: force poweroff if command execution fails - * - * This may be called from any context to trigger a system shutdown. - * If the orderly shutdown fails, it will force an immediate shutdown. - */ -int orderly_poweroff(bool force) +static int __orderly_poweroff(void) { int argc; - char **argv = argv_split(GFP_ATOMIC, poweroff_cmd, &argc); + char **argv; static char *envp[] = { "HOME=/", "PATH=/sbin:/bin:/usr/sbin:/usr/bin", NULL }; - int ret = -ENOMEM; + int ret; + argv = argv_split(GFP_ATOMIC, poweroff_cmd, &argc); if (argv == NULL) { printk(KERN_WARNING "%s failed to allocate memory for \"%s\"\n", __func__, poweroff_cmd); - goto out; + return -ENOMEM; } ret = call_usermodehelper_fns(argv[0], argv, envp, UMH_NO_WAIT, NULL, argv_cleanup, NULL); -out: - if (likely(!ret)) - return 0; - if (ret == -ENOMEM) argv_free(argv); - if (force) { + return ret; +} + +/** + * orderly_poweroff - Trigger an orderly system poweroff + * @force: force poweroff if command execution fails + * + * This may be called from any context to trigger a system shutdown. + * If the orderly shutdown fails, it will force an immediate shutdown. + */ +int orderly_poweroff(bool force) +{ + int ret = __orderly_poweroff(); + + if (ret && force) { printk(KERN_WARNING "Failed to start orderly shutdown: " "forcing the issue\n"); - /* I guess this should try to kick off some daemon to - sync and poweroff asap. Or not even bother syncing - if we're doing an emergency shutdown? */ + /* + * I guess this should try to kick off some daemon to sync and + * poweroff asap. Or not even bother syncing if we're doing an + * emergency shutdown? + */ emergency_sync(); kernel_power_off(); } -- cgit v1.2.3 From cdf53441368cc02ee4aa8a8343a5dc25132836f0 Mon Sep 17 00:00:00 2001 From: Kay Sievers Date: Mon, 30 Jul 2012 14:40:08 -0700 Subject: kmsg: /dev/kmsg - properly return possible copy_from_user() failure Reported-by: Andrew Morton Signed-off-by: Kay Sievers Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/printk.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index 50c96b5651b6..852269adad25 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -389,8 +389,10 @@ static ssize_t devkmsg_writev(struct kiocb *iocb, const struct iovec *iv, line = buf; for (i = 0; i < count; i++) { - if (copy_from_user(line, iv[i].iov_base, iv[i].iov_len)) + if (copy_from_user(line, iv[i].iov_base, iv[i].iov_len)) { + ret = -EFAULT; goto out; + } line += iv[i].iov_len; } -- cgit v1.2.3 From acc8fa41ad31c576cdbc569cc3e0e443b1b98b44 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:40:09 -0700 Subject: printk: add generic functions to find KERN_ headers The current form of a KERN_ is "<.>". Add printk_get_level and printk_skip_level functions to handle these formats. These functions centralize tests of KERN_ so a future modification can change the KERN_ style and shorten the number of bytes consumed by these headers. [akpm@linux-foundation.org: fix build error and warning] Signed-off-by: Joe Perches Cc: Kay Sievers Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/printk.h | 26 ++++++++++++++++++++++++++ kernel/printk.c | 14 +++++++++----- 2 files changed, 35 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/include/linux/printk.h b/include/linux/printk.h index 1bec2f7a2d42..6e12e1f09047 100644 --- a/include/linux/printk.h +++ b/include/linux/printk.h @@ -24,6 +24,32 @@ extern const char linux_proc_banner[]; */ #define KERN_CONT "" +static inline int printk_get_level(const char *buffer) +{ + if (buffer[0] == '<' && buffer[1] && buffer[2] == '>') { + switch (buffer[1]) { + case '0' ... '7': + case 'd': /* KERN_DEFAULT */ + case 'c': /* KERN_CONT */ + return buffer[1]; + } + } + return 0; +} + +static inline const char *printk_skip_level(const char *buffer) +{ + if (printk_get_level(buffer)) { + switch (buffer[1]) { + case '0' ... '7': + case 'd': /* KERN_DEFAULT */ + case 'c': /* KERN_CONT */ + return buffer + 3; + } + } + return buffer; +} + extern int console_printk[]; #define console_loglevel (console_printk[0]) diff --git a/kernel/printk.c b/kernel/printk.c index 852269adad25..0d882a2f231e 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1487,6 +1487,7 @@ asmlinkage int vprintk_emit(int facility, int level, size_t text_len; enum log_flags lflags = 0; unsigned long flags; + int kern_level; int this_cpu; int printed_len = 0; @@ -1543,17 +1544,20 @@ asmlinkage int vprintk_emit(int facility, int level, } /* strip syslog prefix and extract log level or control flags */ - if (text[0] == '<' && text[1] && text[2] == '>') { - switch (text[1]) { + kern_level = printk_get_level(text); + if (kern_level) { + const char *end_of_header = printk_skip_level(text); + switch (kern_level) { case '0' ... '7': if (level == -1) - level = text[1] - '0'; + level = kern_level - '0'; case 'd': /* KERN_DEFAULT */ lflags |= LOG_PREFIX; case 'c': /* KERN_CONT */ - text += 3; - text_len -= 3; + break; } + text_len -= end_of_header - text; + text = (char *)end_of_header; } if (level == -1) -- cgit v1.2.3 From 088a52aac810655c1db1e40331e4936946701e9c Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 30 Jul 2012 14:40:19 -0700 Subject: printk: only look for prefix levels in kernel messages vprintk_emit() prefix parsing should only be done for internal kernel messages. This allows existing behavior to be kept in all cases. Signed-off-by: Joe Perches Cc: Kay Sievers Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/printk.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index 0d882a2f231e..6a76ab9d4476 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1487,7 +1487,6 @@ asmlinkage int vprintk_emit(int facility, int level, size_t text_len; enum log_flags lflags = 0; unsigned long flags; - int kern_level; int this_cpu; int printed_len = 0; @@ -1543,21 +1542,24 @@ asmlinkage int vprintk_emit(int facility, int level, lflags |= LOG_NEWLINE; } - /* strip syslog prefix and extract log level or control flags */ - kern_level = printk_get_level(text); - if (kern_level) { - const char *end_of_header = printk_skip_level(text); - switch (kern_level) { - case '0' ... '7': - if (level == -1) - level = kern_level - '0'; - case 'd': /* KERN_DEFAULT */ - lflags |= LOG_PREFIX; - case 'c': /* KERN_CONT */ - break; + /* strip kernel syslog prefix and extract log level or control flags */ + if (facility == 0) { + int kern_level = printk_get_level(text); + + if (kern_level) { + const char *end_of_header = printk_skip_level(text); + switch (kern_level) { + case '0' ... '7': + if (level == -1) + level = kern_level - '0'; + case 'd': /* KERN_DEFAULT */ + lflags |= LOG_PREFIX; + case 'c': /* KERN_CONT */ + break; + } + text_len -= end_of_header - text; + text = (char *)end_of_header; } - text_len -= end_of_header - text; - text = (char *)end_of_header; } if (level == -1) -- cgit v1.2.3 From 79c743dd1e8de61c31f484c0a1b48930543044b3 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 30 Jul 2012 14:42:17 -0700 Subject: kernel/kmod.c: document call_usermodehelper_fns() a bit This function's interface is, uh, subtle. Attempt to apologise for it. Cc: WANG Cong Cc: Cyrill Gorcunov Cc: Kees Cook Cc: Serge Hallyn Cc: "Eric W. Biederman" Cc: Alan Cox Cc: Oleg Nesterov Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kmod.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'kernel') diff --git a/kernel/kmod.c b/kernel/kmod.c index ff2c7cb86d77..2a8351516a0e 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -577,6 +577,12 @@ unlock: return retval; } +/* + * call_usermodehelper_fns() will not run the caller-provided cleanup function + * if a memory allocation failure is experienced. So the caller might need to + * check the call_usermodehelper_fns() return value: if it is -ENOMEM, perform + * the necessaary cleanup within the caller. + */ int call_usermodehelper_fns( char *path, char **argv, char **envp, int wait, int (*init)(struct subprocess_info *info, struct cred *new), -- cgit v1.2.3 From 0f20784d4ba3f88ca33b703b23372d8ccf6dbd42 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Mon, 30 Jul 2012 14:42:20 -0700 Subject: kmod: avoid deadlock from recursive kmod call The system deadlocks (at least since 2.6.10) when call_usermodehelper(UMH_WAIT_EXEC) request triggers call_usermodehelper(UMH_WAIT_PROC) request. This is because "khelper thread is waiting for the worker thread at wait_for_completion() in do_fork() since the worker thread was created with CLONE_VFORK flag" and "the worker thread cannot call complete() because do_execve() is blocked at UMH_WAIT_PROC request" and "the khelper thread cannot start processing UMH_WAIT_PROC request because the khelper thread is waiting for the worker thread at wait_for_completion() in do_fork()". The easiest example to observe this deadlock is to use a corrupted /sbin/hotplug binary (like shown below). # : > /tmp/dummy # chmod 755 /tmp/dummy # echo /tmp/dummy > /proc/sys/kernel/hotplug # modprobe whatever call_usermodehelper("/tmp/dummy", UMH_WAIT_EXEC) is called from kobject_uevent_env() in lib/kobject_uevent.c upon loading/unloading a module. do_execve("/tmp/dummy") triggers a call to request_module("binfmt-0000") from search_binary_handler() which in turn calls call_usermodehelper(UMH_WAIT_PROC). In order to avoid deadlock, as a for-now and easy-to-backport solution, do not try to call wait_for_completion() in call_usermodehelper_exec() if the worker thread was created by khelper thread with CLONE_VFORK flag. Future and fundamental solution might be replacing singleton khelper thread with some workqueue so that recursive calls up to max_active dependency loop can be handled without deadlock. [akpm@linux-foundation.org: add comment to kmod_thread_locker] Signed-off-by: Tetsuo Handa Cc: Arjan van de Ven Acked-by: Rusty Russell Cc: Tejun Heo Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kmod.c | 31 +++++++++++++++++++++++++++++-- 1 file changed, 29 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/kmod.c b/kernel/kmod.c index 2a8351516a0e..6f99aead66c6 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -45,6 +45,13 @@ extern int max_threads; static struct workqueue_struct *khelper_wq; +/* + * kmod_thread_locker is used for deadlock avoidance. There is no explicit + * locking to protect this global - it is private to the singleton khelper + * thread and should only ever be modified by that thread. + */ +static const struct task_struct *kmod_thread_locker; + #define CAP_BSET (void *)1 #define CAP_PI (void *)2 @@ -221,6 +228,13 @@ fail: return 0; } +static int call_helper(void *data) +{ + /* Worker thread started blocking khelper thread. */ + kmod_thread_locker = current; + return ____call_usermodehelper(data); +} + static void call_usermodehelper_freeinfo(struct subprocess_info *info) { if (info->cleanup) @@ -295,9 +309,12 @@ static void __call_usermodehelper(struct work_struct *work) if (wait == UMH_WAIT_PROC) pid = kernel_thread(wait_for_helper, sub_info, CLONE_FS | CLONE_FILES | SIGCHLD); - else - pid = kernel_thread(____call_usermodehelper, sub_info, + else { + pid = kernel_thread(call_helper, sub_info, CLONE_VFORK | SIGCHLD); + /* Worker thread stopped blocking khelper thread. */ + kmod_thread_locker = NULL; + } switch (wait) { case UMH_NO_WAIT: @@ -548,6 +565,16 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, int wait) retval = -EBUSY; goto out; } + /* + * Worker thread must not wait for khelper thread at below + * wait_for_completion() if the thread was created with CLONE_VFORK + * flag, for khelper thread is already waiting for the thread at + * wait_for_completion() in do_fork(). + */ + if (wait != UMH_NO_WAIT && current == kmod_thread_locker) { + retval = -EBUSY; + goto out; + } sub_info->complete = &done; sub_info->wait = wait; -- cgit v1.2.3 From b2412b7fa7a3816fa8633dc2ff19f1a90aabe423 Mon Sep 17 00:00:00 2001 From: Huang Shijie Date: Mon, 30 Jul 2012 14:42:30 -0700 Subject: fork: use vma_pages() to simplify the code The current code can be replaced by vma_pages(). So use it to simplify the code. [akpm@linux-foundation.org: initialise `len' at its definition site] Signed-off-by: Huang Shijie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index ff1cad3b7bdc..2c1802948a38 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -391,8 +391,8 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) } charge = 0; if (mpnt->vm_flags & VM_ACCOUNT) { - unsigned long len; - len = (mpnt->vm_end - mpnt->vm_start) >> PAGE_SHIFT; + unsigned long len = vma_pages(mpnt); + if (security_vm_enough_memory_mm(oldmm, len)) /* sic */ goto fail_nomem; charge = len; -- cgit v1.2.3 From 87bec58a52652e2eb2a575692a40f9466c7bd31b Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 30 Jul 2012 14:42:31 -0700 Subject: revert "sched: Fix fork() error path to not crash" To make way for "fork: fix error handling in dup_task()", which fixes the errors more completely. Cc: Salman Qazi Cc: Peter Zijlstra Cc: Ingo Molnar Cc: Akinobu Mita Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index 2c1802948a38..088025bd1925 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -304,17 +304,12 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) } err = arch_dup_task_struct(tsk, orig); - - /* - * We defer looking at err, because we will need this setup - * for the clean up path to work correctly. - */ - tsk->stack = ti; - setup_thread_stack(tsk, orig); - if (err) goto out; + tsk->stack = ti; + + setup_thread_stack(tsk, orig); clear_user_return_notifier(tsk); clear_tsk_need_resched(tsk); stackend = end_of_stack(tsk); -- cgit v1.2.3 From f19b9f74b7ea3b21ddcee55d852a6488239608a4 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 Jul 2012 14:42:33 -0700 Subject: fork: fix error handling in dup_task() The function dup_task() may fail at the following function calls in the following order. 0) alloc_task_struct_node() 1) alloc_thread_info_node() 2) arch_dup_task_struct() Error by 0) is not a matter, it can just return. But error by 1) requires releasing task_struct allocated by 0) before it returns. Likewise, error by 2) requires releasing task_struct and thread_info allocated by 0) and 1). The existing error handling calls free_task_struct() and free_thread_info() which do not only release task_struct and thread_info, but also call architecture specific arch_release_task_struct() and arch_release_thread_info(). The problem is that task_struct and thread_info are not fully initialized yet at this point, but arch_release_task_struct() and arch_release_thread_info() are called with them. For example, x86 defines its own arch_release_task_struct() that releases a task_xstate. If alloc_thread_info_node() fails in dup_task(), arch_release_task_struct() is called with task_struct which is just allocated and filled with garbage in this error handling. This actually happened with tools/testing/fault-injection/failcmd.sh # env FAILCMD_TYPE=fail_page_alloc \ ./tools/testing/fault-injection/failcmd.sh --times=100 \ --min-order=0 --ignore-gfp-wait=0 \ -- make -C tools/testing/selftests/ run_tests In order to fix this issue, make free_{task_struct,thread_info}() not to call arch_release_{task_struct,thread_info}() and call arch_release_{task_struct,thread_info}() implicitly where needed. Default arch_release_task_struct() and arch_release_thread_info() are defined as empty by default. So this change only affects the architectures which implement their own arch_release_task_struct() or arch_release_thread_info() as listed below. arch_release_task_struct(): x86, sh arch_release_thread_info(): mn10300, tile Signed-off-by: Akinobu Mita Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: David Howells Cc: Koichi Yasutake Cc: Paul Mundt Cc: Chris Metcalf Cc: Salman Qazi Cc: Peter Zijlstra Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/fork.c | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index 088025bd1925..8efac1fe56bc 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -114,6 +114,10 @@ int nr_processes(void) return total; } +void __weak arch_release_task_struct(struct task_struct *tsk) +{ +} + #ifndef CONFIG_ARCH_TASK_STRUCT_ALLOCATOR static struct kmem_cache *task_struct_cachep; @@ -122,17 +126,17 @@ static inline struct task_struct *alloc_task_struct_node(int node) return kmem_cache_alloc_node(task_struct_cachep, GFP_KERNEL, node); } -void __weak arch_release_task_struct(struct task_struct *tsk) { } - static inline void free_task_struct(struct task_struct *tsk) { - arch_release_task_struct(tsk); kmem_cache_free(task_struct_cachep, tsk); } #endif +void __weak arch_release_thread_info(struct thread_info *ti) +{ +} + #ifndef CONFIG_ARCH_THREAD_INFO_ALLOCATOR -void __weak arch_release_thread_info(struct thread_info *ti) { } /* * Allocate pages if THREAD_SIZE is >= PAGE_SIZE, otherwise use a @@ -150,7 +154,6 @@ static struct thread_info *alloc_thread_info_node(struct task_struct *tsk, static inline void free_thread_info(struct thread_info *ti) { - arch_release_thread_info(ti); free_pages((unsigned long)ti, THREAD_SIZE_ORDER); } # else @@ -164,7 +167,6 @@ static struct thread_info *alloc_thread_info_node(struct task_struct *tsk, static void free_thread_info(struct thread_info *ti) { - arch_release_thread_info(ti); kmem_cache_free(thread_info_cache, ti); } @@ -205,10 +207,12 @@ static void account_kernel_stack(struct thread_info *ti, int account) void free_task(struct task_struct *tsk) { account_kernel_stack(tsk->stack, -1); + arch_release_thread_info(tsk->stack); free_thread_info(tsk->stack); rt_mutex_debug_task_free(tsk); ftrace_graph_exit_task(tsk); put_seccomp_filter(tsk); + arch_release_task_struct(tsk); free_task_struct(tsk); } EXPORT_SYMBOL(free_task); @@ -298,14 +302,12 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) return NULL; ti = alloc_thread_info_node(tsk, node); - if (!ti) { - free_task_struct(tsk); - return NULL; - } + if (!ti) + goto free_tsk; err = arch_dup_task_struct(tsk, orig); if (err) - goto out; + goto free_ti; tsk->stack = ti; @@ -333,8 +335,9 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) return tsk; -out: +free_ti: free_thread_info(ti); +free_tsk: free_task_struct(tsk); return NULL; } -- cgit v1.2.3 From 63dca8d5b5ef7effb58b8d6892a024125c0fab0b Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Mon, 30 Jul 2012 14:42:36 -0700 Subject: kdump: append newline to the last lien of vmcoreinfo note The last line of vmcoreinfo note does not end with \n. Parsing all the lines in note becomes easier if all lines end with \n instead of trying to special case the last line. I know at least one tool, vmcore-dmesg in kexec-tools tree which made the assumption that all lines end with \n. I think it is a good idea to fix it. Signed-off-by: Vivek Goyal Cc: "Eric W. Biederman" Cc: Atsushi Kumagai Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kexec.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/kexec.c b/kernel/kexec.c index 4e2e472f6aeb..0668d58d6413 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -1424,7 +1424,7 @@ static void update_vmcoreinfo_note(void) void crash_save_vmcoreinfo(void) { - vmcoreinfo_append_str("CRASHTIME=%ld", get_seconds()); + vmcoreinfo_append_str("CRASHTIME=%ld\n", get_seconds()); update_vmcoreinfo_note(); } -- cgit v1.2.3 From fd4b616b0fbb77e3f349e7d60914f2b7c7e39f9c Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 30 Jul 2012 14:42:48 -0700 Subject: sysctl: suppress kmemleak messages register_sysctl_table() is a strange function, as it makes internal allocations (a header) to register a sysctl_table. This header is a handle to the table that is created, and can be used to unregister the table. But if the table is permanent and never unregistered, the header acts the same as a static variable. Unfortunately, this allocation of memory that is never expected to be freed fools kmemleak in thinking that we have leaked memory. For those sysctl tables that are never unregistered, and have no pointer referencing them, kmemleak will think that these are memory leaks: unreferenced object 0xffff880079fb9d40 (size 192): comm "swapper/0", pid 0, jiffies 4294667316 (age 12614.152s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [] kmemleak_alloc+0x73/0x98 [] kmemleak_alloc_recursive.constprop.42+0x16/0x18 [] __kmalloc+0x107/0x153 [] kzalloc.constprop.8+0xe/0x10 [] __register_sysctl_paths+0xe1/0x160 [] register_sysctl_paths+0x1b/0x1d [] register_sysctl_table+0x18/0x1a [] sysctl_init+0x10/0x14 [] proc_sys_init+0x2f/0x31 [] proc_root_init+0xa5/0xa7 [] start_kernel+0x3d0/0x40a [] x86_64_start_reservations+0xae/0xb2 [] x86_64_start_kernel+0x102/0x111 [] 0xffffffffffffffff The sysctl_base_table used by sysctl itself is one such instance that registers the table to never be unregistered. Use kmemleak_not_leak() to suppress the kmemleak false positive. Signed-off-by: Steven Rostedt Acked-by: Catalin Marinas Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index b46f496405e4..97186b99b0e4 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include @@ -1556,7 +1557,10 @@ static struct ctl_table dev_table[] = { int __init sysctl_init(void) { - register_sysctl_table(sysctl_base_table); + struct ctl_table_header *hdr; + + hdr = register_sysctl_table(sysctl_base_table); + kmemleak_not_leak(hdr); return 0; } -- cgit v1.2.3 From 25353b3377d5a75d4b830477bb90a3691155de72 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Mon, 30 Jul 2012 14:42:49 -0700 Subject: taskstats: check nla_reserve() return Addresses https://bugzilla.kernel.org/show_bug.cgi?id=44621 Reported-by: Signed-off-by: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/taskstats.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel') diff --git a/kernel/taskstats.c b/kernel/taskstats.c index e66046456f4f..d0a32796550f 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -436,6 +436,11 @@ static int cgroupstats_user_cmd(struct sk_buff *skb, struct genl_info *info) na = nla_reserve(rep_skb, CGROUPSTATS_TYPE_CGROUP_STATS, sizeof(struct cgroupstats)); + if (na == NULL) { + rc = -EMSGSIZE; + goto err; + } + stats = nla_data(na); memset(stats, 0, sizeof(*stats)); -- cgit v1.2.3 From 65fed8f6f23070b56d0ed3841173ddd410130a89 Mon Sep 17 00:00:00 2001 From: Octavian Purdila Date: Mon, 30 Jul 2012 14:42:58 -0700 Subject: resource: make sure requested range is included in the root range When the requested range is outside of the root range the logic in __reserve_region_with_split will cause an infinite recursion which will overflow the stack as seen in the warning bellow. This particular stack overflow was caused by requesting the (100000000-107ffffff) range while the root range was (0-ffffffff). In this case __request_resource would return the whole root range as conflict range (i.e. 0-ffffffff). Then, the logic in __reserve_region_with_split would continue the recursion requesting the new range as (conflict->end+1, end) which incidentally in this case equals the originally requested range. This patch aborts looking for an usable range when the request does not intersect with the root range. When the request partially overlaps with the root range, it ajust the request to fall in the root range and then continues with the new request. When the request is modified or aborted errors and a stack trace are logged to allow catching the errors in the upper layers. [ 5.968374] WARNING: at kernel/sched.c:4129 sub_preempt_count+0x63/0x89() [ 5.975150] Modules linked in: [ 5.978184] Pid: 1, comm: swapper Not tainted 3.0.22-mid27-00004-gb72c817 #46 [ 5.985324] Call Trace: [ 5.987759] [] ? console_unlock+0x17b/0x18d [ 5.992891] [] warn_slowpath_common+0x48/0x5d [ 5.998194] [] ? sub_preempt_count+0x63/0x89 [ 6.003412] [] warn_slowpath_null+0xf/0x13 [ 6.008453] [] sub_preempt_count+0x63/0x89 [ 6.013499] [] _raw_spin_unlock+0x27/0x3f [ 6.018453] [] add_partial+0x36/0x3b [ 6.022973] [] deactivate_slab+0x96/0xb4 [ 6.027842] [] __slab_alloc.isra.54.constprop.63+0x204/0x241 [ 6.034456] [] ? kzalloc.constprop.5+0x29/0x38 [ 6.039842] [] ? kzalloc.constprop.5+0x29/0x38 [ 6.045232] [] kmem_cache_alloc_trace+0x51/0xb0 [ 6.050710] [] ? kzalloc.constprop.5+0x29/0x38 [ 6.056100] [] kzalloc.constprop.5+0x29/0x38 [ 6.061320] [] __reserve_region_with_split+0x1c/0xd1 [ 6.067230] [] __reserve_region_with_split+0xc6/0xd1 ... [ 7.179057] [] __reserve_region_with_split+0xc6/0xd1 [ 7.184970] [] reserve_region_with_split+0x30/0x42 [ 7.190709] [] e820_reserve_resources_late+0xd1/0xe9 [ 7.196623] [] pcibios_resource_survey+0x23/0x2a [ 7.202184] [] pcibios_init+0x23/0x35 [ 7.206789] [] pci_subsys_init+0x3f/0x44 [ 7.211659] [] do_one_initcall+0x72/0x122 [ 7.216615] [] ? pci_legacy_init+0x3d/0x3d [ 7.221659] [] kernel_init+0xa6/0x118 [ 7.226265] [] ? start_kernel+0x334/0x334 [ 7.231223] [] kernel_thread_helper+0x6/0x10 Signed-off-by: Octavian Purdila Signed-off-by: Ram Pai Cc: Jesse Barnes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/resource.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/resource.c b/kernel/resource.c index dc8b47764443..34d45886ee84 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -7,6 +7,8 @@ * Arbitrary resource management. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include @@ -791,8 +793,28 @@ void __init reserve_region_with_split(struct resource *root, resource_size_t start, resource_size_t end, const char *name) { + int abort = 0; + write_lock(&resource_lock); - __reserve_region_with_split(root, start, end, name); + if (root->start > start || root->end < end) { + pr_err("requested range [0x%llx-0x%llx] not in root %pr\n", + (unsigned long long)start, (unsigned long long)end, + root); + if (start > root->end || end < root->start) + abort = 1; + else { + if (end > root->end) + end = root->end; + if (start < root->start) + start = root->start; + pr_err("fixing request to [0x%llx-0x%llx]\n", + (unsigned long long)start, + (unsigned long long)end); + } + dump_stack(); + } + if (!abort) + __reserve_region_with_split(root, start, end, name); write_unlock(&resource_lock); } -- cgit v1.2.3 From 0f26d0e0a715556270d85b7946b99546a2f92888 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Mon, 30 Jul 2012 22:44:41 -0500 Subject: kdb: Remove unused KDB_FLAG_ONLY_DO_DUMP This code cleanup was missed in the original kdb merge, and this code is simply not used at all. The code that was previously used to set the KDB_FLAG_ONLY_DO_DUMP was removed prior to the initial kdb merge. Signed-off-by: Jason Wessel --- include/linux/kdb.h | 2 -- kernel/debug/kdb/kdb_main.c | 15 +-------------- 2 files changed, 1 insertion(+), 16 deletions(-) (limited to 'kernel') diff --git a/include/linux/kdb.h b/include/linux/kdb.h index 064725854db8..42d9e863a313 100644 --- a/include/linux/kdb.h +++ b/include/linux/kdb.h @@ -75,8 +75,6 @@ extern const char *kdb_diemsg; #define KDB_FLAG_CATASTROPHIC (1 << 1) /* A catastrophic event has occurred */ #define KDB_FLAG_CMD_INTERRUPT (1 << 2) /* Previous command was interrupted */ #define KDB_FLAG_NOIPI (1 << 3) /* Do not send IPIs */ -#define KDB_FLAG_ONLY_DO_DUMP (1 << 4) /* Only do a dump, used when - * kdb is off */ #define KDB_FLAG_NO_CONSOLE (1 << 5) /* No console is available, * kdb is disabled */ #define KDB_FLAG_NO_VT_CONSOLE (1 << 6) /* No VT console is available, do diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c index 1f91413edb87..31df1706b9a9 100644 --- a/kernel/debug/kdb/kdb_main.c +++ b/kernel/debug/kdb/kdb_main.c @@ -139,11 +139,10 @@ static const int __nkdb_err = sizeof(kdbmsgs) / sizeof(kdbmsg_t); static char *__env[] = { #if defined(CONFIG_SMP) "PROMPT=[%d]kdb> ", - "MOREPROMPT=[%d]more> ", #else "PROMPT=kdb> ", - "MOREPROMPT=more> ", #endif + "MOREPROMPT=more> ", "RADIX=16", "MDCOUNT=8", /* lines of md output */ KDB_PLATFORM_ENV, @@ -1236,18 +1235,6 @@ static int kdb_local(kdb_reason_t reason, int error, struct pt_regs *regs, *cmdbuf = '\0'; *(cmd_hist[cmd_head]) = '\0'; - if (KDB_FLAG(ONLY_DO_DUMP)) { - /* kdb is off but a catastrophic error requires a dump. - * Take the dump and reboot. - * Turn on logging so the kdb output appears in the log - * buffer in the dump. - */ - const char *setargs[] = { "set", "LOGGING", "1" }; - kdb_set(2, setargs); - kdb_reboot(0, NULL); - /*NOTREACHED*/ - } - do_full_getstr: #if defined(CONFIG_SMP) snprintf(kdb_prompt_str, CMD_BUFLEN, kdbgetenv("PROMPT"), -- cgit v1.2.3 From 07cd27bbd4d07af6c3e24ae479316a69e7935e1e Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Mon, 30 Jul 2012 22:44:41 -0500 Subject: kdb: Remove cpu from the more prompt Having the CPU in the more prompt is completely redundent vs the standard kdb prompt, and it also wastes 32 bytes on the stack. Signed-off-by: Jason Wessel --- kernel/debug/kdb/kdb_io.c | 11 ----------- 1 file changed, 11 deletions(-) (limited to 'kernel') diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c index bb9520f0f6ff..0a69d2adc4f3 100644 --- a/kernel/debug/kdb/kdb_io.c +++ b/kernel/debug/kdb/kdb_io.c @@ -715,9 +715,6 @@ kdb_printit: /* check for having reached the LINES number of printed lines */ if (kdb_nextline == linecount) { char buf1[16] = ""; -#if defined(CONFIG_SMP) - char buf2[32]; -#endif /* Watch out for recursion here. Any routine that calls * kdb_printf will come back through here. And kdb_read @@ -732,14 +729,6 @@ kdb_printit: if (moreprompt == NULL) moreprompt = "more> "; -#if defined(CONFIG_SMP) - if (strchr(moreprompt, '%')) { - sprintf(buf2, moreprompt, get_cpu()); - put_cpu(); - moreprompt = buf2; - } -#endif - kdb_input_flush(); c = console_drivers; -- cgit v1.2.3 From b10d22d6e8f76b9e94871aebe0fc62aab2748200 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Mon, 30 Jul 2012 04:58:10 -0700 Subject: kernel/debug: Make use of KGDB_REASON_NMI Currently kernel never set KGDB_REASON_NMI. We do now, when we enter KGDB/KDB from an NMI. This is not to be confused with kgdb_nmicallback(), NMI callback is an entry for the slave CPUs during CPUs roundup, but REASON_NMI is the entry for the master CPU. Signed-off-by: Anton Vorontsov Signed-off-by: Jason Wessel --- kernel/debug/kdb/kdb_debugger.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel') diff --git a/kernel/debug/kdb/kdb_debugger.c b/kernel/debug/kdb/kdb_debugger.c index 8b68ce78ff17..be7b33b73d30 100644 --- a/kernel/debug/kdb/kdb_debugger.c +++ b/kernel/debug/kdb/kdb_debugger.c @@ -12,6 +12,7 @@ #include #include #include +#include #include "kdb_private.h" #include "../debug_core.h" @@ -52,6 +53,9 @@ int kdb_stub(struct kgdb_state *ks) if (atomic_read(&kgdb_setting_breakpoint)) reason = KDB_REASON_KEYBOARD; + if (in_nmi()) + reason = KDB_REASON_NMI; + for (i = 0, bp = kdb_breakpoints; i < KDB_MAXBPT; i++, bp++) { if ((bp->bp_enabled) && (bp->bp_addr == addr)) { reason = KDB_REASON_BREAK; -- cgit v1.2.3 From b9403130a5350fca59a50ed11c198cb8c7e54119 Mon Sep 17 00:00:00 2001 From: Michael Wang Date: Thu, 12 Jul 2012 16:10:13 +0800 Subject: sched/cleanups: Add load balance cpumask pointer to 'struct lb_env' With this patch struct ld_env will have a pointer of the load balancing cpumask and we don't need to pass a cpumask around anymore. Signed-off-by: Michael Wang Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/4FFE8665.3080705@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- kernel/sched/fair.c | 29 ++++++++++++++--------------- 1 file changed, 14 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 22321db64952..d0cc03b3e70b 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3069,6 +3069,9 @@ struct lb_env { int new_dst_cpu; enum cpu_idle_type idle; long imbalance; + /* The set of CPUs under consideration for load-balancing */ + struct cpumask *cpus; + unsigned int flags; unsigned int loop; @@ -3653,8 +3656,7 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group) */ static inline void update_sg_lb_stats(struct lb_env *env, struct sched_group *group, int load_idx, - int local_group, const struct cpumask *cpus, - int *balance, struct sg_lb_stats *sgs) + int local_group, int *balance, struct sg_lb_stats *sgs) { unsigned long nr_running, max_nr_running, min_nr_running; unsigned long load, max_cpu_load, min_cpu_load; @@ -3671,7 +3673,7 @@ static inline void update_sg_lb_stats(struct lb_env *env, max_nr_running = 0; min_nr_running = ~0UL; - for_each_cpu_and(i, sched_group_cpus(group), cpus) { + for_each_cpu_and(i, sched_group_cpus(group), env->cpus) { struct rq *rq = cpu_rq(i); nr_running = rq->nr_running; @@ -3800,8 +3802,7 @@ static bool update_sd_pick_busiest(struct lb_env *env, * @sds: variable to hold the statistics for this sched_domain. */ static inline void update_sd_lb_stats(struct lb_env *env, - const struct cpumask *cpus, - int *balance, struct sd_lb_stats *sds) + int *balance, struct sd_lb_stats *sds) { struct sched_domain *child = env->sd->child; struct sched_group *sg = env->sd->groups; @@ -3818,8 +3819,7 @@ static inline void update_sd_lb_stats(struct lb_env *env, local_group = cpumask_test_cpu(env->dst_cpu, sched_group_cpus(sg)); memset(&sgs, 0, sizeof(sgs)); - update_sg_lb_stats(env, sg, load_idx, local_group, - cpus, balance, &sgs); + update_sg_lb_stats(env, sg, load_idx, local_group, balance, &sgs); if (local_group && !(*balance)) return; @@ -4055,7 +4055,6 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s * to restore balance. * * @env: The load balancing environment. - * @cpus: The set of CPUs under consideration for load-balancing. * @balance: Pointer to a variable indicating if this_cpu * is the appropriate cpu to perform load balancing at this_level. * @@ -4065,7 +4064,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s * put to idle by rebalancing its tasks onto our group. */ static struct sched_group * -find_busiest_group(struct lb_env *env, const struct cpumask *cpus, int *balance) +find_busiest_group(struct lb_env *env, int *balance) { struct sd_lb_stats sds; @@ -4075,7 +4074,7 @@ find_busiest_group(struct lb_env *env, const struct cpumask *cpus, int *balance) * Compute the various statistics relavent for load balancing at * this level. */ - update_sd_lb_stats(env, cpus, balance, &sds); + update_sd_lb_stats(env, balance, &sds); /* * this_cpu is not the appropriate cpu to perform load balancing at @@ -4155,8 +4154,7 @@ ret: * find_busiest_queue - find the busiest runqueue among the cpus in group. */ static struct rq *find_busiest_queue(struct lb_env *env, - struct sched_group *group, - const struct cpumask *cpus) + struct sched_group *group) { struct rq *busiest = NULL, *rq; unsigned long max_load = 0; @@ -4171,7 +4169,7 @@ static struct rq *find_busiest_queue(struct lb_env *env, if (!capacity) capacity = fix_small_capacity(env->sd, group); - if (!cpumask_test_cpu(i, cpus)) + if (!cpumask_test_cpu(i, env->cpus)) continue; rq = cpu_rq(i); @@ -4252,6 +4250,7 @@ static int load_balance(int this_cpu, struct rq *this_rq, .dst_grpmask = sched_group_cpus(sd->groups), .idle = idle, .loop_break = sched_nr_migrate_break, + .cpus = cpus, }; cpumask_copy(cpus, cpu_active_mask); @@ -4260,7 +4259,7 @@ static int load_balance(int this_cpu, struct rq *this_rq, schedstat_inc(sd, lb_count[idle]); redo: - group = find_busiest_group(&env, cpus, balance); + group = find_busiest_group(&env, balance); if (*balance == 0) goto out_balanced; @@ -4270,7 +4269,7 @@ redo: goto out_balanced; } - busiest = find_busiest_queue(&env, group, cpus); + busiest = find_busiest_queue(&env, group); if (!busiest) { schedstat_inc(sd, lb_nobusyq[idle]); goto out_balanced; -- cgit v1.2.3 From e6dab5ffab59e910ec0e3355f4a6f29f7a7be474 Mon Sep 17 00:00:00 2001 From: Andrew Vagin Date: Wed, 11 Jul 2012 18:14:58 +0400 Subject: perf/trace: Add ability to set a target task for events A few events are interesting not only for a current task. For example, sched_stat_* events are interesting for a task which wakes up. For this reason, it will be good if such events will be delivered to a target task too. Now a target task can be set by using __perf_task(). The original idea and a draft patch belongs to Peter Zijlstra. I need these events for profiling sleep times. sched_switch is used for getting callchains and sched_stat_* is used for getting time periods. These events are combined in user space, then it can be analyzed by perf tools. Inspired-by: Peter Zijlstra Cc: Steven Rostedt Cc: Paul Mackerras Cc: Arnaldo Carvalho de Melo Cc: Steven Rostedt Cc: Arun Sharma Signed-off-by: Andrew Vagin Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1342016098-213063-1-git-send-email-avagin@openvz.org Signed-off-by: Ingo Molnar --- include/linux/ftrace_event.h | 5 +++-- include/linux/perf_event.h | 3 ++- include/trace/events/sched.h | 4 ++++ include/trace/ftrace.h | 6 +++++- kernel/events/callchain.c | 9 ++++++++- kernel/events/core.c | 30 ++++++++++++++++++++++++++++-- kernel/events/internal.h | 3 ++- kernel/trace/trace_event_perf.c | 2 +- kernel/trace/trace_kprobe.c | 6 ++++-- kernel/trace/trace_syscalls.c | 4 ++-- kernel/trace/trace_uprobe.c | 2 +- 11 files changed, 60 insertions(+), 14 deletions(-) (limited to 'kernel') diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index af961d6f7ab1..642928cf57b4 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -306,9 +306,10 @@ extern void *perf_trace_buf_prepare(int size, unsigned short type, static inline void perf_trace_buf_submit(void *raw_data, int size, int rctx, u64 addr, - u64 count, struct pt_regs *regs, void *head) + u64 count, struct pt_regs *regs, void *head, + struct task_struct *task) { - perf_tp_event(addr, count, raw_data, size, regs, head, rctx); + perf_tp_event(addr, count, raw_data, size, regs, head, rctx, task); } #endif diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 76c5c8b724a7..7602ccb3f40e 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1272,7 +1272,8 @@ static inline bool perf_paranoid_kernel(void) extern void perf_event_init(void); extern void perf_tp_event(u64 addr, u64 count, void *record, int entry_size, struct pt_regs *regs, - struct hlist_head *head, int rctx); + struct hlist_head *head, int rctx, + struct task_struct *task); extern void perf_bp_event(struct perf_event *event, void *data); #ifndef perf_misc_flags diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h index ea7a2035456d..5a8671e8a67f 100644 --- a/include/trace/events/sched.h +++ b/include/trace/events/sched.h @@ -73,6 +73,9 @@ DECLARE_EVENT_CLASS(sched_wakeup_template, __entry->prio = p->prio; __entry->success = success; __entry->target_cpu = task_cpu(p); + ) + TP_perf_assign( + __perf_task(p); ), TP_printk("comm=%s pid=%d prio=%d success=%d target_cpu=%03d", @@ -325,6 +328,7 @@ DECLARE_EVENT_CLASS(sched_stat_template, ) TP_perf_assign( __perf_count(delay); + __perf_task(tsk); ), TP_printk("comm=%s pid=%d delay=%Lu [ns]", diff --git a/include/trace/ftrace.h b/include/trace/ftrace.h index c6bc2faaf261..a763888a36f9 100644 --- a/include/trace/ftrace.h +++ b/include/trace/ftrace.h @@ -712,6 +712,9 @@ __attribute__((section("_ftrace_events"))) *__event_##call = &event_##call #undef __perf_count #define __perf_count(c) __count = (c) +#undef __perf_task +#define __perf_task(t) __task = (t) + #undef TP_perf_assign #define TP_perf_assign(args...) args @@ -725,6 +728,7 @@ perf_trace_##call(void *__data, proto) \ struct ftrace_raw_##call *entry; \ struct pt_regs __regs; \ u64 __addr = 0, __count = 1; \ + struct task_struct *__task = NULL; \ struct hlist_head *head; \ int __entry_size; \ int __data_size; \ @@ -752,7 +756,7 @@ perf_trace_##call(void *__data, proto) \ \ head = this_cpu_ptr(event_call->perf_events); \ perf_trace_buf_submit(entry, __entry_size, rctx, __addr, \ - __count, &__regs, head); \ + __count, &__regs, head, __task); \ } /* diff --git a/kernel/events/callchain.c b/kernel/events/callchain.c index 6581a040f399..98d4597f43d6 100644 --- a/kernel/events/callchain.c +++ b/kernel/events/callchain.c @@ -153,7 +153,8 @@ put_callchain_entry(int rctx) put_recursion_context(__get_cpu_var(callchain_recursion), rctx); } -struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +struct perf_callchain_entry * +perf_callchain(struct perf_event *event, struct pt_regs *regs) { int rctx; struct perf_callchain_entry *entry; @@ -178,6 +179,12 @@ struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) } if (regs) { + /* + * Disallow cross-task user callchains. + */ + if (event->ctx->task && event->ctx->task != current) + goto exit_put; + perf_callchain_store(entry, PERF_CONTEXT_USER); perf_callchain_user(entry, regs); } diff --git a/kernel/events/core.c b/kernel/events/core.c index f1cf0edeb39a..b7935fcec7d9 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -4039,7 +4039,7 @@ void perf_prepare_sample(struct perf_event_header *header, if (sample_type & PERF_SAMPLE_CALLCHAIN) { int size = 1; - data->callchain = perf_callchain(regs); + data->callchain = perf_callchain(event, regs); if (data->callchain) size += data->callchain->nr; @@ -5209,7 +5209,8 @@ static int perf_tp_event_match(struct perf_event *event, } void perf_tp_event(u64 addr, u64 count, void *record, int entry_size, - struct pt_regs *regs, struct hlist_head *head, int rctx) + struct pt_regs *regs, struct hlist_head *head, int rctx, + struct task_struct *task) { struct perf_sample_data data; struct perf_event *event; @@ -5228,6 +5229,31 @@ void perf_tp_event(u64 addr, u64 count, void *record, int entry_size, perf_swevent_event(event, count, &data, regs); } + /* + * If we got specified a target task, also iterate its context and + * deliver this event there too. + */ + if (task && task != current) { + struct perf_event_context *ctx; + struct trace_entry *entry = record; + + rcu_read_lock(); + ctx = rcu_dereference(task->perf_event_ctxp[perf_sw_context]); + if (!ctx) + goto unlock; + + list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { + if (event->attr.type != PERF_TYPE_TRACEPOINT) + continue; + if (event->attr.config != entry->type) + continue; + if (perf_tp_event_match(event, &data, regs)) + perf_swevent_event(event, count, &data, regs); + } +unlock: + rcu_read_unlock(); + } + perf_swevent_put_recursion_context(rctx); } EXPORT_SYMBOL_GPL(perf_tp_event); diff --git a/kernel/events/internal.h b/kernel/events/internal.h index b0b107f90afc..a096c19f2c2a 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -101,7 +101,8 @@ __output_copy(struct perf_output_handle *handle, } /* Callchain handling */ -extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs); +extern struct perf_callchain_entry * +perf_callchain(struct perf_event *event, struct pt_regs *regs); extern int get_callchain_buffers(void); extern void put_callchain_buffers(void); diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index fee3752ae8f6..8a6d2ee2086c 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -281,7 +281,7 @@ perf_ftrace_function_call(unsigned long ip, unsigned long parent_ip) head = this_cpu_ptr(event_function.perf_events); perf_trace_buf_submit(entry, ENTRY_SIZE, rctx, 0, - 1, ®s, head); + 1, ®s, head, NULL); #undef ENTRY_SIZE } diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index b31d3d5699fe..1a2117043bb1 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -1002,7 +1002,8 @@ static __kprobes void kprobe_perf_func(struct kprobe *kp, store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); head = this_cpu_ptr(call->perf_events); - perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, regs, head); + perf_trace_buf_submit(entry, size, rctx, + entry->ip, 1, regs, head, NULL); } /* Kretprobe profile handler */ @@ -1033,7 +1034,8 @@ static __kprobes void kretprobe_perf_func(struct kretprobe_instance *ri, store_trace_args(sizeof(*entry), tp, regs, (u8 *)&entry[1], dsize); head = this_cpu_ptr(call->perf_events); - perf_trace_buf_submit(entry, size, rctx, entry->ret_ip, 1, regs, head); + perf_trace_buf_submit(entry, size, rctx, + entry->ret_ip, 1, regs, head, NULL); } #endif /* CONFIG_PERF_EVENTS */ diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 96fc73369099..60e4d7875672 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -532,7 +532,7 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) (unsigned long *)&rec->args); head = this_cpu_ptr(sys_data->enter_event->perf_events); - perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head); + perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head, NULL); } int perf_sysenter_enable(struct ftrace_event_call *call) @@ -608,7 +608,7 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) rec->ret = syscall_get_return_value(current, regs); head = this_cpu_ptr(sys_data->exit_event->perf_events); - perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head); + perf_trace_buf_submit(rec, size, rctx, 0, 1, regs, head, NULL); } int perf_sysexit_enable(struct ftrace_event_call *call) diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 2b36ac68549e..03003cd7dd96 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -670,7 +670,7 @@ static void uprobe_perf_func(struct trace_uprobe *tu, struct pt_regs *regs) call_fetch(&tu->args[i].fetch, regs, data + tu->args[i].offset); head = this_cpu_ptr(call->perf_events); - perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, regs, head); + perf_trace_buf_submit(entry, size, rctx, entry->ip, 1, regs, head, NULL); out: preempt_enable(); -- cgit v1.2.3 From 02ab20ae38337b99b5c29c81090f594b8fd61283 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Fri, 27 Jul 2012 14:48:10 -0400 Subject: time/jiffies: Rename ACTHZ to SHIFTED_HZ Ingo noted that ACTHZ is a confusing name, and requested it be renamed, so this patch renames ACTHZ to SHIFTED_HZ to better describe it. Signed-off-by: John Stultz Cc: Prarit Bhargava Link: http://lkml.kernel.org/r/1343414893-45779-3-git-send-email-john.stultz@linaro.org Signed-off-by: Ingo Molnar --- include/linux/jiffies.h | 21 +++++++++++++-------- include/linux/timex.h | 2 +- kernel/time/jiffies.c | 2 +- kernel/time/ntp.c | 2 +- 4 files changed, 16 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h index 7d24466fb80b..82680541576d 100644 --- a/include/linux/jiffies.h +++ b/include/linux/jiffies.h @@ -55,21 +55,26 @@ /* LATCH is used in the interval timer and ftape setup. */ # define LATCH ((CLOCK_TICK_RATE + HZ/2) / HZ) /* For divider */ -/* HZ is the requested value. ACTHZ is actual HZ ("<< 8" is for accuracy) */ -# define ACTHZ (SH_DIV(CLOCK_TICK_RATE, LATCH, 8)) +/* + * HZ is the requested value. However the CLOCK_TICK_RATE may not allow + * for exactly HZ. So SHIFTED_HZ is high res HZ ("<< 8" is for accuracy) + */ +# define SHIFTED_HZ (SH_DIV(CLOCK_TICK_RATE, LATCH, 8)) #else -# define ACTHZ (HZ << 8) +# define SHIFTED_HZ (HZ << 8) #endif -/* TICK_NSEC is the time between ticks in nsec assuming real ACTHZ */ -#define TICK_NSEC (SH_DIV (1000000UL * 1000, ACTHZ, 8)) +/* TICK_NSEC is the time between ticks in nsec assuming SHIFTED_HZ */ +#define TICK_NSEC (SH_DIV(1000000UL * 1000, SHIFTED_HZ, 8)) /* TICK_USEC is the time between ticks in usec assuming fake USER_HZ */ #define TICK_USEC ((1000000UL + USER_HZ/2) / USER_HZ) -/* TICK_USEC_TO_NSEC is the time between ticks in nsec assuming real ACTHZ and */ -/* a value TUSEC for TICK_USEC (can be set bij adjtimex) */ -#define TICK_USEC_TO_NSEC(TUSEC) (SH_DIV (TUSEC * USER_HZ * 1000, ACTHZ, 8)) +/* + * TICK_USEC_TO_NSEC is the time between ticks in nsec assuming SHIFTED_HZ and + * a value TUSEC for TICK_USEC (can be set bij adjtimex) + */ +#define TICK_USEC_TO_NSEC(TUSEC) (SH_DIV(TUSEC * USER_HZ * 1000, SHIFTED_HZ, 8)) /* some arch's have a small-data section that can be accessed register-relative * but that can only take up to, say, 4-byte variables. jiffies being part of diff --git a/include/linux/timex.h b/include/linux/timex.h index 99bc88b1fc02..7c5ceb20e03a 100644 --- a/include/linux/timex.h +++ b/include/linux/timex.h @@ -232,7 +232,7 @@ struct timex { * estimated error = NTP dispersion. */ extern unsigned long tick_usec; /* USER_HZ period (usec) */ -extern unsigned long tick_nsec; /* ACTHZ period (nsec) */ +extern unsigned long tick_nsec; /* SHIFTED_HZ period (nsec) */ extern void ntp_init(void); extern void ntp_clear(void); diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c index a470154e0408..46da0537c10b 100644 --- a/kernel/time/jiffies.c +++ b/kernel/time/jiffies.c @@ -37,7 +37,7 @@ * requested HZ value. It is also not recommended * for "tick-less" systems. */ -#define NSEC_PER_JIFFY ((u32)((((u64)NSEC_PER_SEC)<<8)/ACTHZ)) +#define NSEC_PER_JIFFY ((u32)((((u64)NSEC_PER_SEC)<<8)/SHIFTED_HZ)) /* Since jiffies uses a simple NSEC_PER_JIFFY multiplier * conversion, the .shift value could be zero. However diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index b7fbadc5c973..24174b4d669b 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -28,7 +28,7 @@ DEFINE_SPINLOCK(ntp_lock); /* USER_HZ period (usecs): */ unsigned long tick_usec = TICK_USEC; -/* ACTHZ period (nsecs): */ +/* SHIFTED_HZ period (nsecs): */ unsigned long tick_nsec; static u64 tick_length; -- cgit v1.2.3 From d4e3ab384b2343c7074f713ac330f839c38c52ee Mon Sep 17 00:00:00 2001 From: John Stultz Date: Fri, 27 Jul 2012 14:48:11 -0400 Subject: time: Clean up stray newlines Ingo noted inconsistent newline usage between functions. This patch cleans those up. Signed-off-by: John Stultz Cc: Prarit Bhargava Link: http://lkml.kernel.org/r/1343414893-45779-4-git-send-email-john.stultz@linaro.org Signed-off-by: Ingo Molnar --- kernel/time/timekeeping.c | 10 ---------- 1 file changed, 10 deletions(-) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index cf364db5589f..05b37a5ec7fb 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -239,7 +239,6 @@ static void timekeeping_update(struct timekeeper *tk, bool clearntp) update_vsyscall(&xt, &tk->wall_to_monotonic, tk->clock, tk->mult); } - /** * timekeeping_forward_now - update clock to the current time * @@ -436,7 +435,6 @@ int do_settimeofday(const struct timespec *tv) } EXPORT_SYMBOL(do_settimeofday); - /** * timekeeping_inject_offset - Adds or subtracts from the current time. * @tv: pointer to the timespec variable containing the offset @@ -550,7 +548,6 @@ void getrawmonotonic(struct timespec *ts) } EXPORT_SYMBOL(getrawmonotonic); - /** * timekeeping_valid_for_hres - Check if timekeeping is suitable for hres */ @@ -683,7 +680,6 @@ static void __timekeeping_inject_sleeptime(struct timekeeper *tk, update_sleep_time(timespec_add(tk->total_sleep_time, *delta)); } - /** * timekeeping_inject_sleeptime - Adds suspend interval to timeekeeping values * @delta: pointer to a timespec delta value @@ -718,7 +714,6 @@ void timekeeping_inject_sleeptime(struct timespec *delta) clock_was_set(); } - /** * timekeeping_resume - Resumes the generic timekeeping subsystem. * @@ -1003,7 +998,6 @@ static void timekeeping_adjust(struct timekeeper *tk, s64 offset) } - /** * accumulate_nsecs_to_secs - Accumulates nsecs into secs * @@ -1032,7 +1026,6 @@ static inline void accumulate_nsecs_to_secs(struct timekeeper *tk) } } - /** * logarithmic_accumulation - shifted accumulation of cycles * @@ -1076,7 +1069,6 @@ static cycle_t logarithmic_accumulation(struct timekeeper *tk, cycle_t offset, return offset; } - /** * update_wall_time - Uses the current clocksource to increment the wall time * @@ -1177,7 +1169,6 @@ void getboottime(struct timespec *ts) } EXPORT_SYMBOL_GPL(getboottime); - /** * get_monotonic_boottime - Returns monotonic time since boot * @ts: pointer to the timespec to be set @@ -1358,7 +1349,6 @@ ktime_t ktime_get_monotonic_offset(void) } EXPORT_SYMBOL_GPL(ktime_get_monotonic_offset); - /** * xtime_update() - advances the timekeeping infrastructure * @ticks: number of ticks, that have elapsed since the last call. -- cgit v1.2.3 From 6d0ef903e2bda70da124c10d8ad89f2382c87991 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Fri, 27 Jul 2012 14:48:12 -0400 Subject: time: Clean up offs_real/wall_to_mono and offs_boot/total_sleep_time updates For performance reasons, we maintain ktime_t based duplicates of wall_to_monotonic (offs_real) and total_sleep_time (offs_boot). Since large problems could occur (such as the resume regression on 3.5-rc7, or the leapsecond hrtimer issue) if these value pairs were to be inconsistently updated, this patch this cleans up how we modify these value pairs to ensure we are always consistent. As a side-effect this is also more efficient as we only caulculate the duplicate values when they are changed, rather then every update_wall_time call. This also provides WARN_ONs to detect if future changes break the invariants. Signed-off-by: John Stultz Cc: Peter Zijlstra Cc: Richard Cochran Cc: Prarit Bhargava Link: http://lkml.kernel.org/r/1343414893-45779-5-git-send-email-john.stultz@linaro.org [ Cleaned up minor style issues. ] Signed-off-by: Ingo Molnar --- kernel/time/timekeeping.c | 90 ++++++++++++++++++++++++++++------------------- 1 file changed, 54 insertions(+), 36 deletions(-) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 05b37a5ec7fb..4da65592b4d9 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -65,14 +65,14 @@ struct timekeeper { * used instead. */ struct timespec wall_to_monotonic; - /* time spent in suspend */ - struct timespec total_sleep_time; - /* The raw monotonic time for the CLOCK_MONOTONIC_RAW posix clock. */ - struct timespec raw_time; /* Offset clock monotonic -> clock realtime */ ktime_t offs_real; + /* time spent in suspend */ + struct timespec total_sleep_time; /* Offset clock monotonic -> clock boottime */ ktime_t offs_boot; + /* The raw monotonic time for the CLOCK_MONOTONIC_RAW posix clock. */ + struct timespec raw_time; /* Seqlock for all timekeeper values */ seqlock_t lock; }; @@ -117,6 +117,31 @@ static void tk_xtime_add(struct timekeeper *tk, const struct timespec *ts) tk->xtime_nsec += (u64)ts->tv_nsec << tk->shift; } +static void tk_set_wall_to_mono(struct timekeeper *tk, struct timespec wtm) +{ + struct timespec tmp; + + /* + * Verify consistency of: offset_real = -wall_to_monotonic + * before modifying anything + */ + set_normalized_timespec(&tmp, -tk->wall_to_monotonic.tv_sec, + -tk->wall_to_monotonic.tv_nsec); + WARN_ON_ONCE(tk->offs_real.tv64 != timespec_to_ktime(tmp).tv64); + tk->wall_to_monotonic = wtm; + set_normalized_timespec(&tmp, -wtm.tv_sec, -wtm.tv_nsec); + tk->offs_real = timespec_to_ktime(tmp); +} + +static void tk_set_sleep_time(struct timekeeper *tk, struct timespec t) +{ + /* Verify consistency before modifying */ + WARN_ON_ONCE(tk->offs_boot.tv64 != timespec_to_ktime(tk->total_sleep_time).tv64); + + tk->total_sleep_time = t; + tk->offs_boot = timespec_to_ktime(t); +} + /** * timekeeper_setup_internals - Set up internals to use clocksource clock. * @@ -217,14 +242,6 @@ static inline s64 timekeeping_get_ns_raw(struct timekeeper *tk) return nsec + arch_gettimeoffset(); } -static void update_rt_offset(struct timekeeper *tk) -{ - struct timespec tmp, *wtm = &tk->wall_to_monotonic; - - set_normalized_timespec(&tmp, -wtm->tv_sec, -wtm->tv_nsec); - tk->offs_real = timespec_to_ktime(tmp); -} - /* must hold write on timekeeper.lock */ static void timekeeping_update(struct timekeeper *tk, bool clearntp) { @@ -234,7 +251,6 @@ static void timekeeping_update(struct timekeeper *tk, bool clearntp) tk->ntp_error = 0; ntp_clear(); } - update_rt_offset(tk); xt = tk_xtime(tk); update_vsyscall(&xt, &tk->wall_to_monotonic, tk->clock, tk->mult); } @@ -419,8 +435,8 @@ int do_settimeofday(const struct timespec *tv) ts_delta.tv_sec = tv->tv_sec - xt.tv_sec; ts_delta.tv_nsec = tv->tv_nsec - xt.tv_nsec; - timekeeper.wall_to_monotonic = - timespec_sub(timekeeper.wall_to_monotonic, ts_delta); + tk_set_wall_to_mono(&timekeeper, + timespec_sub(timekeeper.wall_to_monotonic, ts_delta)); tk_set_xtime(&timekeeper, tv); @@ -454,8 +470,8 @@ int timekeeping_inject_offset(struct timespec *ts) tk_xtime_add(&timekeeper, ts); - timekeeper.wall_to_monotonic = - timespec_sub(timekeeper.wall_to_monotonic, *ts); + tk_set_wall_to_mono(&timekeeper, + timespec_sub(timekeeper.wall_to_monotonic, *ts)); timekeeping_update(&timekeeper, true); @@ -621,7 +637,7 @@ void __init timekeeping_init(void) { struct clocksource *clock; unsigned long flags; - struct timespec now, boot; + struct timespec now, boot, tmp; read_persistent_clock(&now); read_boot_clock(&boot); @@ -642,23 +658,19 @@ void __init timekeeping_init(void) if (boot.tv_sec == 0 && boot.tv_nsec == 0) boot = tk_xtime(&timekeeper); - set_normalized_timespec(&timekeeper.wall_to_monotonic, - -boot.tv_sec, -boot.tv_nsec); - update_rt_offset(&timekeeper); - timekeeper.total_sleep_time.tv_sec = 0; - timekeeper.total_sleep_time.tv_nsec = 0; + set_normalized_timespec(&tmp, -boot.tv_sec, -boot.tv_nsec); + tk_set_wall_to_mono(&timekeeper, tmp); + + tmp.tv_sec = 0; + tmp.tv_nsec = 0; + tk_set_sleep_time(&timekeeper, tmp); + write_sequnlock_irqrestore(&timekeeper.lock, flags); } /* time in seconds when suspend began */ static struct timespec timekeeping_suspend_time; -static void update_sleep_time(struct timespec t) -{ - timekeeper.total_sleep_time = t; - timekeeper.offs_boot = timespec_to_ktime(t); -} - /** * __timekeeping_inject_sleeptime - Internal function to add sleep interval * @delta: pointer to a timespec delta value @@ -674,10 +686,9 @@ static void __timekeeping_inject_sleeptime(struct timekeeper *tk, "sleep delta value!\n"); return; } - tk_xtime_add(tk, delta); - tk->wall_to_monotonic = timespec_sub(tk->wall_to_monotonic, *delta); - update_sleep_time(timespec_add(tk->total_sleep_time, *delta)); + tk_set_wall_to_mono(tk, timespec_sub(tk->wall_to_monotonic, *delta)); + tk_set_sleep_time(tk, timespec_add(tk->total_sleep_time, *delta)); } /** @@ -1018,11 +1029,18 @@ static inline void accumulate_nsecs_to_secs(struct timekeeper *tk) /* Figure out if its a leap sec and apply if needed */ leap = second_overflow(tk->xtime_sec); - tk->xtime_sec += leap; - tk->wall_to_monotonic.tv_sec -= leap; - if (leap) - clock_was_set_delayed(); + if (unlikely(leap)) { + struct timespec ts; + + tk->xtime_sec += leap; + ts.tv_sec = leap; + ts.tv_nsec = 0; + tk_set_wall_to_mono(tk, + timespec_sub(tk->wall_to_monotonic, ts)); + + clock_was_set_delayed(); + } } } -- cgit v1.2.3 From 4e250fdde9be50581c7dd5fed88c9b9960615314 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Fri, 27 Jul 2012 14:48:13 -0400 Subject: time: Remove all direct references to timekeeper Ingo noted that the numerous timekeeper.value references made the timekeeping code ugly and caused many long lines that had to be broken up. He recommended replacing timekeeper.value references with tk->value. This patch provides a local tk value for all top level time functions and sets it to &timekeeper. Then all timekeeper access is done via a tk pointer. Signed-off-by: John Stultz Cc: Prarit Bhargava Link: http://lkml.kernel.org/r/1343414893-45779-6-git-send-email-john.stultz@linaro.org Signed-off-by: Ingo Molnar --- kernel/time/timekeeping.c | 282 +++++++++++++++++++++++++--------------------- 1 file changed, 154 insertions(+), 128 deletions(-) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 4da65592b4d9..2988bc819187 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -292,18 +292,19 @@ static void timekeeping_forward_now(struct timekeeper *tk) */ void getnstimeofday(struct timespec *ts) { + struct timekeeper *tk = &timekeeper; unsigned long seq; s64 nsecs = 0; WARN_ON(timekeeping_suspended); do { - seq = read_seqbegin(&timekeeper.lock); + seq = read_seqbegin(&tk->lock); - ts->tv_sec = timekeeper.xtime_sec; - ts->tv_nsec = timekeeping_get_ns(&timekeeper); + ts->tv_sec = tk->xtime_sec; + ts->tv_nsec = timekeeping_get_ns(tk); - } while (read_seqretry(&timekeeper.lock, seq)); + } while (read_seqretry(&tk->lock, seq)); timespec_add_ns(ts, nsecs); } @@ -311,19 +312,18 @@ EXPORT_SYMBOL(getnstimeofday); ktime_t ktime_get(void) { + struct timekeeper *tk = &timekeeper; unsigned int seq; s64 secs, nsecs; WARN_ON(timekeeping_suspended); do { - seq = read_seqbegin(&timekeeper.lock); - secs = timekeeper.xtime_sec + - timekeeper.wall_to_monotonic.tv_sec; - nsecs = timekeeping_get_ns(&timekeeper) + - timekeeper.wall_to_monotonic.tv_nsec; + seq = read_seqbegin(&tk->lock); + secs = tk->xtime_sec + tk->wall_to_monotonic.tv_sec; + nsecs = timekeeping_get_ns(tk) + tk->wall_to_monotonic.tv_nsec; - } while (read_seqretry(&timekeeper.lock, seq)); + } while (read_seqretry(&tk->lock, seq)); /* * Use ktime_set/ktime_add_ns to create a proper ktime on * 32-bit architectures without CONFIG_KTIME_SCALAR. @@ -342,18 +342,19 @@ EXPORT_SYMBOL_GPL(ktime_get); */ void ktime_get_ts(struct timespec *ts) { + struct timekeeper *tk = &timekeeper; struct timespec tomono; unsigned int seq; WARN_ON(timekeeping_suspended); do { - seq = read_seqbegin(&timekeeper.lock); - ts->tv_sec = timekeeper.xtime_sec; - ts->tv_nsec = timekeeping_get_ns(&timekeeper); - tomono = timekeeper.wall_to_monotonic; + seq = read_seqbegin(&tk->lock); + ts->tv_sec = tk->xtime_sec; + ts->tv_nsec = timekeeping_get_ns(tk); + tomono = tk->wall_to_monotonic; - } while (read_seqretry(&timekeeper.lock, seq)); + } while (read_seqretry(&tk->lock, seq)); set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec, ts->tv_nsec + tomono.tv_nsec); @@ -373,22 +374,23 @@ EXPORT_SYMBOL_GPL(ktime_get_ts); */ void getnstime_raw_and_real(struct timespec *ts_raw, struct timespec *ts_real) { + struct timekeeper *tk = &timekeeper; unsigned long seq; s64 nsecs_raw, nsecs_real; WARN_ON_ONCE(timekeeping_suspended); do { - seq = read_seqbegin(&timekeeper.lock); + seq = read_seqbegin(&tk->lock); - *ts_raw = timekeeper.raw_time; - ts_real->tv_sec = timekeeper.xtime_sec; + *ts_raw = tk->raw_time; + ts_real->tv_sec = tk->xtime_sec; ts_real->tv_nsec = 0; - nsecs_raw = timekeeping_get_ns_raw(&timekeeper); - nsecs_real = timekeeping_get_ns(&timekeeper); + nsecs_raw = timekeeping_get_ns_raw(tk); + nsecs_real = timekeeping_get_ns(tk); - } while (read_seqretry(&timekeeper.lock, seq)); + } while (read_seqretry(&tk->lock, seq)); timespec_add_ns(ts_raw, nsecs_raw); timespec_add_ns(ts_real, nsecs_real); @@ -421,28 +423,28 @@ EXPORT_SYMBOL(do_gettimeofday); */ int do_settimeofday(const struct timespec *tv) { + struct timekeeper *tk = &timekeeper; struct timespec ts_delta, xt; unsigned long flags; if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) return -EINVAL; - write_seqlock_irqsave(&timekeeper.lock, flags); + write_seqlock_irqsave(&tk->lock, flags); - timekeeping_forward_now(&timekeeper); + timekeeping_forward_now(tk); - xt = tk_xtime(&timekeeper); + xt = tk_xtime(tk); ts_delta.tv_sec = tv->tv_sec - xt.tv_sec; ts_delta.tv_nsec = tv->tv_nsec - xt.tv_nsec; - tk_set_wall_to_mono(&timekeeper, - timespec_sub(timekeeper.wall_to_monotonic, ts_delta)); + tk_set_wall_to_mono(tk, timespec_sub(tk->wall_to_monotonic, ts_delta)); - tk_set_xtime(&timekeeper, tv); + tk_set_xtime(tk, tv); - timekeeping_update(&timekeeper, true); + timekeeping_update(tk, true); - write_sequnlock_irqrestore(&timekeeper.lock, flags); + write_sequnlock_irqrestore(&tk->lock, flags); /* signal hrtimers about time change */ clock_was_set(); @@ -459,23 +461,23 @@ EXPORT_SYMBOL(do_settimeofday); */ int timekeeping_inject_offset(struct timespec *ts) { + struct timekeeper *tk = &timekeeper; unsigned long flags; if ((unsigned long)ts->tv_nsec >= NSEC_PER_SEC) return -EINVAL; - write_seqlock_irqsave(&timekeeper.lock, flags); + write_seqlock_irqsave(&tk->lock, flags); - timekeeping_forward_now(&timekeeper); + timekeeping_forward_now(tk); - tk_xtime_add(&timekeeper, ts); - tk_set_wall_to_mono(&timekeeper, - timespec_sub(timekeeper.wall_to_monotonic, *ts)); + tk_xtime_add(tk, ts); + tk_set_wall_to_mono(tk, timespec_sub(tk->wall_to_monotonic, *ts)); - timekeeping_update(&timekeeper, true); + timekeeping_update(tk, true); - write_sequnlock_irqrestore(&timekeeper.lock, flags); + write_sequnlock_irqrestore(&tk->lock, flags); /* signal hrtimers about time change */ clock_was_set(); @@ -491,23 +493,24 @@ EXPORT_SYMBOL(timekeeping_inject_offset); */ static int change_clocksource(void *data) { + struct timekeeper *tk = &timekeeper; struct clocksource *new, *old; unsigned long flags; new = (struct clocksource *) data; - write_seqlock_irqsave(&timekeeper.lock, flags); + write_seqlock_irqsave(&tk->lock, flags); - timekeeping_forward_now(&timekeeper); + timekeeping_forward_now(tk); if (!new->enable || new->enable(new) == 0) { - old = timekeeper.clock; - tk_setup_internals(&timekeeper, new); + old = tk->clock; + tk_setup_internals(tk, new); if (old->disable) old->disable(old); } - timekeeping_update(&timekeeper, true); + timekeeping_update(tk, true); - write_sequnlock_irqrestore(&timekeeper.lock, flags); + write_sequnlock_irqrestore(&tk->lock, flags); return 0; } @@ -521,7 +524,9 @@ static int change_clocksource(void *data) */ void timekeeping_notify(struct clocksource *clock) { - if (timekeeper.clock == clock) + struct timekeeper *tk = &timekeeper; + + if (tk->clock == clock) return; stop_machine(change_clocksource, clock, NULL); tick_clock_notify(); @@ -550,15 +555,16 @@ EXPORT_SYMBOL_GPL(ktime_get_real); */ void getrawmonotonic(struct timespec *ts) { + struct timekeeper *tk = &timekeeper; unsigned long seq; s64 nsecs; do { - seq = read_seqbegin(&timekeeper.lock); - nsecs = timekeeping_get_ns_raw(&timekeeper); - *ts = timekeeper.raw_time; + seq = read_seqbegin(&tk->lock); + nsecs = timekeeping_get_ns_raw(tk); + *ts = tk->raw_time; - } while (read_seqretry(&timekeeper.lock, seq)); + } while (read_seqretry(&tk->lock, seq)); timespec_add_ns(ts, nsecs); } @@ -569,15 +575,16 @@ EXPORT_SYMBOL(getrawmonotonic); */ int timekeeping_valid_for_hres(void) { + struct timekeeper *tk = &timekeeper; unsigned long seq; int ret; do { - seq = read_seqbegin(&timekeeper.lock); + seq = read_seqbegin(&tk->lock); - ret = timekeeper.clock->flags & CLOCK_SOURCE_VALID_FOR_HRES; + ret = tk->clock->flags & CLOCK_SOURCE_VALID_FOR_HRES; - } while (read_seqretry(&timekeeper.lock, seq)); + } while (read_seqretry(&tk->lock, seq)); return ret; } @@ -587,15 +594,16 @@ int timekeeping_valid_for_hres(void) */ u64 timekeeping_max_deferment(void) { + struct timekeeper *tk = &timekeeper; unsigned long seq; u64 ret; do { - seq = read_seqbegin(&timekeeper.lock); + seq = read_seqbegin(&tk->lock); - ret = timekeeper.clock->max_idle_ns; + ret = tk->clock->max_idle_ns; - } while (read_seqretry(&timekeeper.lock, seq)); + } while (read_seqretry(&tk->lock, seq)); return ret; } @@ -635,6 +643,7 @@ void __attribute__((weak)) read_boot_clock(struct timespec *ts) */ void __init timekeeping_init(void) { + struct timekeeper *tk = &timekeeper; struct clocksource *clock; unsigned long flags; struct timespec now, boot, tmp; @@ -642,30 +651,30 @@ void __init timekeeping_init(void) read_persistent_clock(&now); read_boot_clock(&boot); - seqlock_init(&timekeeper.lock); + seqlock_init(&tk->lock); ntp_init(); - write_seqlock_irqsave(&timekeeper.lock, flags); + write_seqlock_irqsave(&tk->lock, flags); clock = clocksource_default_clock(); if (clock->enable) clock->enable(clock); - tk_setup_internals(&timekeeper, clock); + tk_setup_internals(tk, clock); - tk_set_xtime(&timekeeper, &now); - timekeeper.raw_time.tv_sec = 0; - timekeeper.raw_time.tv_nsec = 0; + tk_set_xtime(tk, &now); + tk->raw_time.tv_sec = 0; + tk->raw_time.tv_nsec = 0; if (boot.tv_sec == 0 && boot.tv_nsec == 0) - boot = tk_xtime(&timekeeper); + boot = tk_xtime(tk); set_normalized_timespec(&tmp, -boot.tv_sec, -boot.tv_nsec); - tk_set_wall_to_mono(&timekeeper, tmp); + tk_set_wall_to_mono(tk, tmp); tmp.tv_sec = 0; tmp.tv_nsec = 0; - tk_set_sleep_time(&timekeeper, tmp); + tk_set_sleep_time(tk, tmp); - write_sequnlock_irqrestore(&timekeeper.lock, flags); + write_sequnlock_irqrestore(&tk->lock, flags); } /* time in seconds when suspend began */ @@ -703,6 +712,7 @@ static void __timekeeping_inject_sleeptime(struct timekeeper *tk, */ void timekeeping_inject_sleeptime(struct timespec *delta) { + struct timekeeper *tk = &timekeeper; unsigned long flags; struct timespec ts; @@ -711,15 +721,15 @@ void timekeeping_inject_sleeptime(struct timespec *delta) if (!(ts.tv_sec == 0 && ts.tv_nsec == 0)) return; - write_seqlock_irqsave(&timekeeper.lock, flags); + write_seqlock_irqsave(&tk->lock, flags); - timekeeping_forward_now(&timekeeper); + timekeeping_forward_now(tk); - __timekeeping_inject_sleeptime(&timekeeper, delta); + __timekeeping_inject_sleeptime(tk, delta); - timekeeping_update(&timekeeper, true); + timekeeping_update(tk, true); - write_sequnlock_irqrestore(&timekeeper.lock, flags); + write_sequnlock_irqrestore(&tk->lock, flags); /* signal hrtimers about time change */ clock_was_set(); @@ -734,6 +744,7 @@ void timekeeping_inject_sleeptime(struct timespec *delta) */ static void timekeeping_resume(void) { + struct timekeeper *tk = &timekeeper; unsigned long flags; struct timespec ts; @@ -741,18 +752,18 @@ static void timekeeping_resume(void) clocksource_resume(); - write_seqlock_irqsave(&timekeeper.lock, flags); + write_seqlock_irqsave(&tk->lock, flags); if (timespec_compare(&ts, &timekeeping_suspend_time) > 0) { ts = timespec_sub(ts, timekeeping_suspend_time); - __timekeeping_inject_sleeptime(&timekeeper, &ts); + __timekeeping_inject_sleeptime(tk, &ts); } /* re-base the last cycle value */ - timekeeper.clock->cycle_last = timekeeper.clock->read(timekeeper.clock); - timekeeper.ntp_error = 0; + tk->clock->cycle_last = tk->clock->read(tk->clock); + tk->ntp_error = 0; timekeeping_suspended = 0; - timekeeping_update(&timekeeper, false); - write_sequnlock_irqrestore(&timekeeper.lock, flags); + timekeeping_update(tk, false); + write_sequnlock_irqrestore(&tk->lock, flags); touch_softlockup_watchdog(); @@ -764,14 +775,15 @@ static void timekeeping_resume(void) static int timekeeping_suspend(void) { + struct timekeeper *tk = &timekeeper; unsigned long flags; struct timespec delta, delta_delta; static struct timespec old_delta; read_persistent_clock(&timekeeping_suspend_time); - write_seqlock_irqsave(&timekeeper.lock, flags); - timekeeping_forward_now(&timekeeper); + write_seqlock_irqsave(&tk->lock, flags); + timekeeping_forward_now(tk); timekeeping_suspended = 1; /* @@ -780,7 +792,7 @@ static int timekeeping_suspend(void) * try to compensate so the difference in system time * and persistent_clock time stays close to constant. */ - delta = timespec_sub(tk_xtime(&timekeeper), timekeeping_suspend_time); + delta = timespec_sub(tk_xtime(tk), timekeeping_suspend_time); delta_delta = timespec_sub(delta, old_delta); if (abs(delta_delta.tv_sec) >= 2) { /* @@ -793,7 +805,7 @@ static int timekeeping_suspend(void) timekeeping_suspend_time = timespec_add(timekeeping_suspend_time, delta_delta); } - write_sequnlock_irqrestore(&timekeeper.lock, flags); + write_sequnlock_irqrestore(&tk->lock, flags); clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL); clocksource_suspend(); @@ -904,7 +916,7 @@ static void timekeeping_adjust(struct timekeeper *tk, s64 offset) * the error. This causes the likely below to be unlikely. * * The proper fix is to avoid rounding up by using - * the high precision timekeeper.xtime_nsec instead of + * the high precision tk->xtime_nsec instead of * xtime.tv_nsec everywhere. Fixing this will take some * time. */ @@ -1094,21 +1106,22 @@ static cycle_t logarithmic_accumulation(struct timekeeper *tk, cycle_t offset, static void update_wall_time(void) { struct clocksource *clock; + struct timekeeper *tk = &timekeeper; cycle_t offset; int shift = 0, maxshift; unsigned long flags; s64 remainder; - write_seqlock_irqsave(&timekeeper.lock, flags); + write_seqlock_irqsave(&tk->lock, flags); /* Make sure we're fully resumed: */ if (unlikely(timekeeping_suspended)) goto out; - clock = timekeeper.clock; + clock = tk->clock; #ifdef CONFIG_ARCH_USES_GETTIMEOFFSET - offset = timekeeper.cycle_interval; + offset = tk->cycle_interval; #else offset = (clock->read(clock) - clock->cycle_last) & clock->mask; #endif @@ -1121,19 +1134,19 @@ static void update_wall_time(void) * chunk in one go, and then try to consume the next smaller * doubled multiple. */ - shift = ilog2(offset) - ilog2(timekeeper.cycle_interval); + shift = ilog2(offset) - ilog2(tk->cycle_interval); shift = max(0, shift); /* Bound shift to one less than what overflows tick_length */ maxshift = (64 - (ilog2(ntp_tick_length())+1)) - 1; shift = min(shift, maxshift); - while (offset >= timekeeper.cycle_interval) { - offset = logarithmic_accumulation(&timekeeper, offset, shift); - if(offset < timekeeper.cycle_interval<= tk->cycle_interval) { + offset = logarithmic_accumulation(tk, offset, shift); + if (offset < tk->cycle_interval<xtime_nsec & ((1 << tk->shift) - 1); + tk->xtime_nsec -= remainder; + tk->xtime_nsec += 1 << tk->shift; + tk->ntp_error += remainder << tk->ntp_error_shift; /* * Finally, make sure that after the rounding * xtime_nsec isn't larger than NSEC_PER_SEC */ - accumulate_nsecs_to_secs(&timekeeper); + accumulate_nsecs_to_secs(tk); - timekeeping_update(&timekeeper, false); + timekeeping_update(tk, false); out: - write_sequnlock_irqrestore(&timekeeper.lock, flags); + write_sequnlock_irqrestore(&tk->lock, flags); } @@ -1176,11 +1189,12 @@ out: */ void getboottime(struct timespec *ts) { + struct timekeeper *tk = &timekeeper; struct timespec boottime = { - .tv_sec = timekeeper.wall_to_monotonic.tv_sec + - timekeeper.total_sleep_time.tv_sec, - .tv_nsec = timekeeper.wall_to_monotonic.tv_nsec + - timekeeper.total_sleep_time.tv_nsec + .tv_sec = tk->wall_to_monotonic.tv_sec + + tk->total_sleep_time.tv_sec, + .tv_nsec = tk->wall_to_monotonic.tv_nsec + + tk->total_sleep_time.tv_nsec }; set_normalized_timespec(ts, -boottime.tv_sec, -boottime.tv_nsec); @@ -1198,19 +1212,20 @@ EXPORT_SYMBOL_GPL(getboottime); */ void get_monotonic_boottime(struct timespec *ts) { + struct timekeeper *tk = &timekeeper; struct timespec tomono, sleep; unsigned int seq; WARN_ON(timekeeping_suspended); do { - seq = read_seqbegin(&timekeeper.lock); - ts->tv_sec = timekeeper.xtime_sec; - ts->tv_nsec = timekeeping_get_ns(&timekeeper); - tomono = timekeeper.wall_to_monotonic; - sleep = timekeeper.total_sleep_time; + seq = read_seqbegin(&tk->lock); + ts->tv_sec = tk->xtime_sec; + ts->tv_nsec = timekeeping_get_ns(tk); + tomono = tk->wall_to_monotonic; + sleep = tk->total_sleep_time; - } while (read_seqretry(&timekeeper.lock, seq)); + } while (read_seqretry(&tk->lock, seq)); set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec + sleep.tv_sec, ts->tv_nsec + tomono.tv_nsec + sleep.tv_nsec); @@ -1240,31 +1255,38 @@ EXPORT_SYMBOL_GPL(ktime_get_boottime); */ void monotonic_to_bootbased(struct timespec *ts) { - *ts = timespec_add(*ts, timekeeper.total_sleep_time); + struct timekeeper *tk = &timekeeper; + + *ts = timespec_add(*ts, tk->total_sleep_time); } EXPORT_SYMBOL_GPL(monotonic_to_bootbased); unsigned long get_seconds(void) { - return timekeeper.xtime_sec; + struct timekeeper *tk = &timekeeper; + + return tk->xtime_sec; } EXPORT_SYMBOL(get_seconds); struct timespec __current_kernel_time(void) { - return tk_xtime(&timekeeper); + struct timekeeper *tk = &timekeeper; + + return tk_xtime(tk); } struct timespec current_kernel_time(void) { + struct timekeeper *tk = &timekeeper; struct timespec now; unsigned long seq; do { - seq = read_seqbegin(&timekeeper.lock); + seq = read_seqbegin(&tk->lock); - now = tk_xtime(&timekeeper); - } while (read_seqretry(&timekeeper.lock, seq)); + now = tk_xtime(tk); + } while (read_seqretry(&tk->lock, seq)); return now; } @@ -1272,15 +1294,16 @@ EXPORT_SYMBOL(current_kernel_time); struct timespec get_monotonic_coarse(void) { + struct timekeeper *tk = &timekeeper; struct timespec now, mono; unsigned long seq; do { - seq = read_seqbegin(&timekeeper.lock); + seq = read_seqbegin(&tk->lock); - now = tk_xtime(&timekeeper); - mono = timekeeper.wall_to_monotonic; - } while (read_seqretry(&timekeeper.lock, seq)); + now = tk_xtime(tk); + mono = tk->wall_to_monotonic; + } while (read_seqretry(&tk->lock, seq)); set_normalized_timespec(&now, now.tv_sec + mono.tv_sec, now.tv_nsec + mono.tv_nsec); @@ -1309,14 +1332,15 @@ void do_timer(unsigned long ticks) void get_xtime_and_monotonic_and_sleep_offset(struct timespec *xtim, struct timespec *wtom, struct timespec *sleep) { + struct timekeeper *tk = &timekeeper; unsigned long seq; do { - seq = read_seqbegin(&timekeeper.lock); - *xtim = tk_xtime(&timekeeper); - *wtom = timekeeper.wall_to_monotonic; - *sleep = timekeeper.total_sleep_time; - } while (read_seqretry(&timekeeper.lock, seq)); + seq = read_seqbegin(&tk->lock); + *xtim = tk_xtime(tk); + *wtom = tk->wall_to_monotonic; + *sleep = tk->total_sleep_time; + } while (read_seqretry(&tk->lock, seq)); } #ifdef CONFIG_HIGH_RES_TIMERS @@ -1330,19 +1354,20 @@ void get_xtime_and_monotonic_and_sleep_offset(struct timespec *xtim, */ ktime_t ktime_get_update_offsets(ktime_t *offs_real, ktime_t *offs_boot) { + struct timekeeper *tk = &timekeeper; ktime_t now; unsigned int seq; u64 secs, nsecs; do { - seq = read_seqbegin(&timekeeper.lock); + seq = read_seqbegin(&tk->lock); - secs = timekeeper.xtime_sec; - nsecs = timekeeping_get_ns(&timekeeper); + secs = tk->xtime_sec; + nsecs = timekeeping_get_ns(tk); - *offs_real = timekeeper.offs_real; - *offs_boot = timekeeper.offs_boot; - } while (read_seqretry(&timekeeper.lock, seq)); + *offs_real = tk->offs_real; + *offs_boot = tk->offs_boot; + } while (read_seqretry(&tk->lock, seq)); now = ktime_add_ns(ktime_set(secs, 0), nsecs); now = ktime_sub(now, *offs_real); @@ -1355,13 +1380,14 @@ ktime_t ktime_get_update_offsets(ktime_t *offs_real, ktime_t *offs_boot) */ ktime_t ktime_get_monotonic_offset(void) { + struct timekeeper *tk = &timekeeper; unsigned long seq; struct timespec wtom; do { - seq = read_seqbegin(&timekeeper.lock); - wtom = timekeeper.wall_to_monotonic; - } while (read_seqretry(&timekeeper.lock, seq)); + seq = read_seqbegin(&tk->lock); + wtom = tk->wall_to_monotonic; + } while (read_seqretry(&tk->lock, seq)); return timespec_to_ktime(wtom); } -- cgit v1.2.3 From 44de9d0cad41f2c51ef26916842be046b582dcc9 Mon Sep 17 00:00:00 2001 From: Huang Shijie Date: Tue, 31 Jul 2012 16:41:49 -0700 Subject: mm: account the total_vm in the vm_stat_account() vm_stat_account() accounts the shared_vm, stack_vm and reserved_vm now. But we can also account for total_vm in the vm_stat_account() which makes the code tidy. Even for mprotect_fixup(), we can get the right result in the end. Signed-off-by: Huang Shijie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/perfmon.c | 1 - include/linux/mm.h | 1 + kernel/fork.c | 4 +--- mm/mmap.c | 5 ++--- mm/mremap.c | 2 -- 5 files changed, 4 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index d7f558c1e711..3fa4bc536953 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -2353,7 +2353,6 @@ pfm_smpl_buffer_alloc(struct task_struct *task, struct file *filp, pfm_context_t */ insert_vm_struct(mm, vma); - mm->total_vm += size >> PAGE_SHIFT; vm_stat_account(vma->vm_mm, vma->vm_flags, vma->vm_file, vma_pages(vma)); up_write(&task->mm->mmap_sem); diff --git a/include/linux/mm.h b/include/linux/mm.h index f9f279cf5b1b..3955bedeeed1 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1528,6 +1528,7 @@ void vm_stat_account(struct mm_struct *, unsigned long, struct file *, long); static inline void vm_stat_account(struct mm_struct *mm, unsigned long flags, struct file *file, long pages) { + mm->total_vm += pages; } #endif /* CONFIG_PROC_FS */ diff --git a/kernel/fork.c b/kernel/fork.c index 8efac1fe56bc..aaa8813c45d1 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -381,10 +381,8 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) struct file *file; if (mpnt->vm_flags & VM_DONTCOPY) { - long pages = vma_pages(mpnt); - mm->total_vm -= pages; vm_stat_account(mm, mpnt->vm_flags, mpnt->vm_file, - -pages); + -vma_pages(mpnt)); continue; } charge = 0; diff --git a/mm/mmap.c b/mm/mmap.c index 3edfcdfa42d9..1ee2fd8bc486 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -943,6 +943,8 @@ void vm_stat_account(struct mm_struct *mm, unsigned long flags, const unsigned long stack_flags = VM_STACK_FLAGS & (VM_GROWSUP|VM_GROWSDOWN); + mm->total_vm += pages; + if (file) { mm->shared_vm += pages; if ((flags & (VM_EXEC|VM_WRITE)) == VM_EXEC) @@ -1347,7 +1349,6 @@ munmap_back: out: perf_event_mmap(vma); - mm->total_vm += len >> PAGE_SHIFT; vm_stat_account(mm, vm_flags, file, len >> PAGE_SHIFT); if (vm_flags & VM_LOCKED) { if (!mlock_vma_pages_range(vma, addr, addr + len)) @@ -1707,7 +1708,6 @@ static int acct_stack_growth(struct vm_area_struct *vma, unsigned long size, uns return -ENOMEM; /* Ok, everything looks good - let it rip */ - mm->total_vm += grow; if (vma->vm_flags & VM_LOCKED) mm->locked_vm += grow; vm_stat_account(mm, vma->vm_flags, vma->vm_file, grow); @@ -1889,7 +1889,6 @@ static void remove_vma_list(struct mm_struct *mm, struct vm_area_struct *vma) if (vma->vm_flags & VM_ACCOUNT) nr_accounted += nrpages; - mm->total_vm -= nrpages; vm_stat_account(mm, vma->vm_flags, vma->vm_file, -nrpages); vma = remove_vma(vma); } while (vma); diff --git a/mm/mremap.c b/mm/mremap.c index 21fed202ddad..cc06d0e48d05 100644 --- a/mm/mremap.c +++ b/mm/mremap.c @@ -260,7 +260,6 @@ static unsigned long move_vma(struct vm_area_struct *vma, * If this were a serious issue, we'd add a flag to do_munmap(). */ hiwater_vm = mm->hiwater_vm; - mm->total_vm += new_len >> PAGE_SHIFT; vm_stat_account(mm, vma->vm_flags, vma->vm_file, new_len>>PAGE_SHIFT); if (do_munmap(mm, old_addr, old_len) < 0) { @@ -497,7 +496,6 @@ SYSCALL_DEFINE5(mremap, unsigned long, addr, unsigned long, old_len, goto out; } - mm->total_vm += pages; vm_stat_account(mm, vma->vm_flags, vma->vm_file, pages); if (vma->vm_flags & VM_LOCKED) { mm->locked_vm += pages; -- cgit v1.2.3 From 3965c9ae47d64aadf6f13b6fcd37767b83c0689a Mon Sep 17 00:00:00 2001 From: Wanpeng Li Date: Tue, 31 Jul 2012 16:41:52 -0700 Subject: mm: prepare for removal of obsolete /proc/sys/vm/nr_pdflush_threads Since per-BDI flusher threads were introduced in 2.6, the pdflush mechanism is not used any more. But the old interface exported through /proc/sys/vm/nr_pdflush_threads still exists and is obviously useless. For back-compatibility, printk warning information and return 2 to notify the users that the interface is removed. Signed-off-by: Wanpeng Li Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- .../ABI/obsolete/proc-sys-vm-nr_pdflush_threads | 5 +++++ Documentation/feature-removal-schedule.txt | 8 ++++++++ Documentation/sysctl/vm.txt | 11 ----------- fs/fs-writeback.c | 5 ----- include/linux/backing-dev.h | 3 +++ include/linux/writeback.h | 5 ----- kernel/sysctl.c | 8 +++----- kernel/sysctl_binary.c | 2 +- mm/backing-dev.c | 20 ++++++++++++++++++++ 9 files changed, 40 insertions(+), 27 deletions(-) create mode 100644 Documentation/ABI/obsolete/proc-sys-vm-nr_pdflush_threads (limited to 'kernel') diff --git a/Documentation/ABI/obsolete/proc-sys-vm-nr_pdflush_threads b/Documentation/ABI/obsolete/proc-sys-vm-nr_pdflush_threads new file mode 100644 index 000000000000..b0b0eeb20fe3 --- /dev/null +++ b/Documentation/ABI/obsolete/proc-sys-vm-nr_pdflush_threads @@ -0,0 +1,5 @@ +What: /proc/sys/vm/nr_pdflush_threads +Date: June 2012 +Contact: Wanpeng Li +Description: Since pdflush is replaced by per-BDI flusher, the interface of old pdflush + exported in /proc/sys/vm/ should be removed. diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index e9237fb71950..88f2fa48bb63 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -13,6 +13,14 @@ Who: Jim Cromie , Jason Baron --------------------------- +What: /proc/sys/vm/nr_pdflush_threads +When: 2012 +Why: Since pdflush is deprecated, the interface exported in /proc/sys/vm/ + should be removed. +Who: Wanpeng Li + +--------------------------- + What: CONFIG_APM_CPU_IDLE, and its ability to call APM BIOS in idle When: 2012 Why: This optional sub-feature of APM is of dubious reliability, diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt index 84eb25cd69aa..06d662b1c5d5 100644 --- a/Documentation/sysctl/vm.txt +++ b/Documentation/sysctl/vm.txt @@ -42,7 +42,6 @@ Currently, these files are in /proc/sys/vm: - mmap_min_addr - nr_hugepages - nr_overcommit_hugepages -- nr_pdflush_threads - nr_trim_pages (only if CONFIG_MMU=n) - numa_zonelist_order - oom_dump_tasks @@ -426,16 +425,6 @@ See Documentation/vm/hugetlbpage.txt ============================================================== -nr_pdflush_threads - -The current number of pdflush threads. This value is read-only. -The value changes according to the number of dirty pages in the system. - -When necessary, additional pdflush threads are created, one per second, up to -nr_pdflush_threads_max. - -============================================================== - nr_trim_pages This is available only on NOMMU kernels. diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 50d0b78130a1..be3efc4f64f4 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -52,11 +52,6 @@ struct wb_writeback_work { struct completion *done; /* set if the caller waits */ }; -/* - * We don't actually have pdflush, but this one is exported though /proc... - */ -int nr_pdflush_threads; - /** * writeback_in_progress - determine whether there is writeback in progress * @bdi: the device's backing_dev_info structure. diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 489de625cd25..c97c6b9cd38e 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -17,6 +17,7 @@ #include #include #include +#include struct page; struct device; @@ -304,6 +305,8 @@ void clear_bdi_congested(struct backing_dev_info *bdi, int sync); void set_bdi_congested(struct backing_dev_info *bdi, int sync); long congestion_wait(int sync, long timeout); long wait_iff_congested(struct zone *zone, int sync, long timeout); +int pdflush_proc_obsolete(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos); static inline bool bdi_cap_writeback_dirty(struct backing_dev_info *bdi) { diff --git a/include/linux/writeback.h b/include/linux/writeback.h index 6d0a0fcd80e7..c66fe3332d83 100644 --- a/include/linux/writeback.h +++ b/include/linux/writeback.h @@ -189,9 +189,4 @@ void tag_pages_for_writeback(struct address_space *mapping, void account_page_redirty(struct page *page); -/* pdflush.c */ -extern int nr_pdflush_threads; /* Global so it can be exported to sysctl - read-only. */ - - #endif /* WRITEBACK_H */ diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 97186b99b0e4..6502d35a25ba 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1101,11 +1101,9 @@ static struct ctl_table vm_table[] = { .extra1 = &zero, }, { - .procname = "nr_pdflush_threads", - .data = &nr_pdflush_threads, - .maxlen = sizeof nr_pdflush_threads, - .mode = 0444 /* read-only*/, - .proc_handler = proc_dointvec, + .procname = "nr_pdflush_threads", + .mode = 0444 /* read-only */, + .proc_handler = pdflush_proc_obsolete, }, { .procname = "swappiness", diff --git a/kernel/sysctl_binary.c b/kernel/sysctl_binary.c index a650694883a1..65bdcf198d4e 100644 --- a/kernel/sysctl_binary.c +++ b/kernel/sysctl_binary.c @@ -147,7 +147,7 @@ static const struct bin_table bin_vm_table[] = { { CTL_INT, VM_DIRTY_RATIO, "dirty_ratio" }, /* VM_DIRTY_WB_CS "dirty_writeback_centisecs" no longer used */ /* VM_DIRTY_EXPIRE_CS "dirty_expire_centisecs" no longer used */ - { CTL_INT, VM_NR_PDFLUSH_THREADS, "nr_pdflush_threads" }, + /* VM_NR_PDFLUSH_THREADS "nr_pdflush_threads" no longer used */ { CTL_INT, VM_OVERCOMMIT_RATIO, "overcommit_ratio" }, /* VM_PAGEBUF unused */ /* VM_HUGETLB_PAGES "nr_hugepages" no longer used */ diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 3387aea11209..6b4718e2ee34 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -886,3 +886,23 @@ out: return ret; } EXPORT_SYMBOL(wait_iff_congested); + +int pdflush_proc_obsolete(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + char kbuf[] = "0\n"; + + if (*ppos) { + *lenp = 0; + return 0; + } + + if (copy_to_user(buffer, kbuf, sizeof(kbuf))) + return -EFAULT; + printk_once(KERN_WARNING "%s exported in /proc is scheduled for removal\n", + table->procname); + + *lenp = 2; + *ppos += *lenp; + return 2; +} -- cgit v1.2.3 From c255a458055e459f65eb7b7f51dc5dbdd0caf1d8 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 31 Jul 2012 16:43:02 -0700 Subject: memcg: rename config variables Sanity: CONFIG_CGROUP_MEM_RES_CTLR -> CONFIG_MEMCG CONFIG_CGROUP_MEM_RES_CTLR_SWAP -> CONFIG_MEMCG_SWAP CONFIG_CGROUP_MEM_RES_CTLR_SWAP_ENABLED -> CONFIG_MEMCG_SWAP_ENABLED CONFIG_CGROUP_MEM_RES_CTLR_KMEM -> CONFIG_MEMCG_KMEM [mhocko@suse.cz: fix missed bits] Cc: Glauber Costa Acked-by: Michal Hocko Cc: Johannes Weiner Cc: KAMEZAWA Hiroyuki Cc: Hugh Dickins Cc: Tejun Heo Cc: Aneesh Kumar K.V Cc: David Rientjes Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/cgroups/memory.txt | 10 +++++----- arch/powerpc/configs/chroma_defconfig | 4 ++-- arch/s390/defconfig | 2 +- arch/sh/configs/apsh4ad0a_defconfig | 2 +- arch/sh/configs/sdk7786_defconfig | 4 ++-- arch/sh/configs/se7206_defconfig | 2 +- arch/sh/configs/shx3_defconfig | 2 +- arch/sh/configs/urquell_defconfig | 4 ++-- arch/tile/configs/tilegx_defconfig | 4 ++-- arch/tile/configs/tilepro_defconfig | 4 ++-- arch/um/defconfig | 8 ++++---- include/linux/cgroup_subsys.h | 2 +- include/linux/memcontrol.h | 14 +++++++------- include/linux/mmzone.h | 8 ++++---- include/linux/page_cgroup.h | 10 +++++----- include/linux/sched.h | 2 +- include/linux/swap.h | 6 +++--- include/net/sock.h | 4 ++-- init/Kconfig | 14 +++++++------- kernel/fork.c | 2 +- mm/Makefile | 2 +- mm/hwpoison-inject.c | 2 +- mm/memcontrol.c | 20 ++++++++++---------- mm/memory-failure.c | 2 +- mm/mmzone.c | 2 +- mm/oom_kill.c | 2 +- mm/page_cgroup.c | 2 +- mm/vmscan.c | 4 ++-- net/core/sock.c | 2 +- net/ipv4/Makefile | 2 +- net/ipv4/sysctl_net_ipv4.c | 4 ++-- net/ipv4/tcp_ipv4.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- 33 files changed, 78 insertions(+), 78 deletions(-) (limited to 'kernel') diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt index dd88540bb995..672676ac9615 100644 --- a/Documentation/cgroups/memory.txt +++ b/Documentation/cgroups/memory.txt @@ -187,12 +187,12 @@ the cgroup that brought it in -- this will happen on memory pressure). But see section 8.2: when moving a task to another cgroup, its pages may be recharged to the new cgroup, if move_charge_at_immigrate has been chosen. -Exception: If CONFIG_CGROUP_CGROUP_MEM_RES_CTLR_SWAP is not used. +Exception: If CONFIG_CGROUP_CGROUP_MEMCG_SWAP is not used. When you do swapoff and make swapped-out pages of shmem(tmpfs) to be backed into memory in force, charges for pages are accounted against the caller of swapoff rather than the users of shmem. -2.4 Swap Extension (CONFIG_CGROUP_MEM_RES_CTLR_SWAP) +2.4 Swap Extension (CONFIG_MEMCG_SWAP) Swap Extension allows you to record charge for swap. A swapped-in page is charged back to original page allocator if possible. @@ -259,7 +259,7 @@ When oom event notifier is registered, event will be delivered. per-zone-per-cgroup LRU (cgroup's private LRU) is just guarded by zone->lru_lock, it has no lock of its own. -2.7 Kernel Memory Extension (CONFIG_CGROUP_MEM_RES_CTLR_KMEM) +2.7 Kernel Memory Extension (CONFIG_MEMCG_KMEM) With the Kernel memory extension, the Memory Controller is able to limit the amount of kernel memory used by the system. Kernel memory is fundamentally @@ -286,8 +286,8 @@ per cgroup, instead of globally. a. Enable CONFIG_CGROUPS b. Enable CONFIG_RESOURCE_COUNTERS -c. Enable CONFIG_CGROUP_MEM_RES_CTLR -d. Enable CONFIG_CGROUP_MEM_RES_CTLR_SWAP (to use swap extension) +c. Enable CONFIG_MEMCG +d. Enable CONFIG_MEMCG_SWAP (to use swap extension) 1. Prepare the cgroups (see cgroups.txt, Why are cgroups needed?) # mount -t tmpfs none /sys/fs/cgroup diff --git a/arch/powerpc/configs/chroma_defconfig b/arch/powerpc/configs/chroma_defconfig index b1f9597fe312..29bb11ec6c64 100644 --- a/arch/powerpc/configs/chroma_defconfig +++ b/arch/powerpc/configs/chroma_defconfig @@ -21,8 +21,8 @@ CONFIG_CGROUP_DEVICE=y CONFIG_CPUSETS=y CONFIG_CGROUP_CPUACCT=y CONFIG_RESOURCE_COUNTERS=y -CONFIG_CGROUP_MEM_RES_CTLR=y -CONFIG_CGROUP_MEM_RES_CTLR_SWAP=y +CONFIG_CGROUP_MEMCG=y +CONFIG_CGROUP_MEMCG_SWAP=y CONFIG_NAMESPACES=y CONFIG_RELAY=y CONFIG_BLK_DEV_INITRD=y diff --git a/arch/s390/defconfig b/arch/s390/defconfig index 37d2bf267964..de57702a3f44 100644 --- a/arch/s390/defconfig +++ b/arch/s390/defconfig @@ -13,7 +13,7 @@ CONFIG_CGROUPS=y CONFIG_CPUSETS=y CONFIG_CGROUP_CPUACCT=y CONFIG_RESOURCE_COUNTERS=y -CONFIG_CGROUP_MEM_RES_CTLR=y +CONFIG_CGROUP_MEMCG=y CONFIG_CGROUP_MEM_RES_CTLR_SWAP=y CONFIG_CGROUP_SCHED=y CONFIG_RT_GROUP_SCHED=y diff --git a/arch/sh/configs/apsh4ad0a_defconfig b/arch/sh/configs/apsh4ad0a_defconfig index e7583484cc07..95ae23fcfdd6 100644 --- a/arch/sh/configs/apsh4ad0a_defconfig +++ b/arch/sh/configs/apsh4ad0a_defconfig @@ -11,7 +11,7 @@ CONFIG_CGROUP_FREEZER=y CONFIG_CGROUP_DEVICE=y CONFIG_CGROUP_CPUACCT=y CONFIG_RESOURCE_COUNTERS=y -CONFIG_CGROUP_MEM_RES_CTLR=y +CONFIG_CGROUP_MEMCG=y CONFIG_BLK_CGROUP=y CONFIG_NAMESPACES=y CONFIG_BLK_DEV_INITRD=y diff --git a/arch/sh/configs/sdk7786_defconfig b/arch/sh/configs/sdk7786_defconfig index 8a7dd7b59c5c..76a76a295d74 100644 --- a/arch/sh/configs/sdk7786_defconfig +++ b/arch/sh/configs/sdk7786_defconfig @@ -18,8 +18,8 @@ CONFIG_CPUSETS=y # CONFIG_PROC_PID_CPUSET is not set CONFIG_CGROUP_CPUACCT=y CONFIG_RESOURCE_COUNTERS=y -CONFIG_CGROUP_MEM_RES_CTLR=y -CONFIG_CGROUP_MEM_RES_CTLR_SWAP=y +CONFIG_CGROUP_MEMCG=y +CONFIG_CGROUP_MEMCG_SWAP=y CONFIG_CGROUP_SCHED=y CONFIG_RT_GROUP_SCHED=y CONFIG_BLK_CGROUP=y diff --git a/arch/sh/configs/se7206_defconfig b/arch/sh/configs/se7206_defconfig index 72c3fad7383f..6bc30ab9fd18 100644 --- a/arch/sh/configs/se7206_defconfig +++ b/arch/sh/configs/se7206_defconfig @@ -11,7 +11,7 @@ CONFIG_CGROUP_DEBUG=y CONFIG_CGROUP_DEVICE=y CONFIG_CGROUP_CPUACCT=y CONFIG_RESOURCE_COUNTERS=y -CONFIG_CGROUP_MEM_RES_CTLR=y +CONFIG_CGROUP_MEMCG=y CONFIG_RELAY=y CONFIG_NAMESPACES=y CONFIG_UTS_NS=y diff --git a/arch/sh/configs/shx3_defconfig b/arch/sh/configs/shx3_defconfig index 6bb413036892..cd6c519f8fad 100644 --- a/arch/sh/configs/shx3_defconfig +++ b/arch/sh/configs/shx3_defconfig @@ -13,7 +13,7 @@ CONFIG_CGROUP_FREEZER=y CONFIG_CGROUP_DEVICE=y CONFIG_CGROUP_CPUACCT=y CONFIG_RESOURCE_COUNTERS=y -CONFIG_CGROUP_MEM_RES_CTLR=y +CONFIG_CGROUP_MEMCG=y CONFIG_RELAY=y CONFIG_NAMESPACES=y CONFIG_UTS_NS=y diff --git a/arch/sh/configs/urquell_defconfig b/arch/sh/configs/urquell_defconfig index 8bfa4d056d7a..d7f89be9f474 100644 --- a/arch/sh/configs/urquell_defconfig +++ b/arch/sh/configs/urquell_defconfig @@ -15,8 +15,8 @@ CONFIG_CPUSETS=y # CONFIG_PROC_PID_CPUSET is not set CONFIG_CGROUP_CPUACCT=y CONFIG_RESOURCE_COUNTERS=y -CONFIG_CGROUP_MEM_RES_CTLR=y -CONFIG_CGROUP_MEM_RES_CTLR_SWAP=y +CONFIG_CGROUP_MEMCG=y +CONFIG_CGROUP_MEMCG_SWAP=y CONFIG_CGROUP_SCHED=y CONFIG_RT_GROUP_SCHED=y CONFIG_BLK_DEV_INITRD=y diff --git a/arch/tile/configs/tilegx_defconfig b/arch/tile/configs/tilegx_defconfig index b8d99aca5431..0270620a1692 100644 --- a/arch/tile/configs/tilegx_defconfig +++ b/arch/tile/configs/tilegx_defconfig @@ -18,8 +18,8 @@ CONFIG_CGROUP_DEVICE=y CONFIG_CPUSETS=y CONFIG_CGROUP_CPUACCT=y CONFIG_RESOURCE_COUNTERS=y -CONFIG_CGROUP_MEM_RES_CTLR=y -CONFIG_CGROUP_MEM_RES_CTLR_SWAP=y +CONFIG_CGROUP_MEMCG=y +CONFIG_CGROUP_MEMCG_SWAP=y CONFIG_CGROUP_SCHED=y CONFIG_RT_GROUP_SCHED=y CONFIG_BLK_CGROUP=y diff --git a/arch/tile/configs/tilepro_defconfig b/arch/tile/configs/tilepro_defconfig index 2b1fd31894f1..c11de27a9bcb 100644 --- a/arch/tile/configs/tilepro_defconfig +++ b/arch/tile/configs/tilepro_defconfig @@ -17,8 +17,8 @@ CONFIG_CGROUP_DEVICE=y CONFIG_CPUSETS=y CONFIG_CGROUP_CPUACCT=y CONFIG_RESOURCE_COUNTERS=y -CONFIG_CGROUP_MEM_RES_CTLR=y -CONFIG_CGROUP_MEM_RES_CTLR_SWAP=y +CONFIG_CGROUP_MEMCG=y +CONFIG_CGROUP_MEMCG_SWAP=y CONFIG_CGROUP_SCHED=y CONFIG_RT_GROUP_SCHED=y CONFIG_BLK_CGROUP=y diff --git a/arch/um/defconfig b/arch/um/defconfig index 7823ab12e6a4..fec0d5d27460 100644 --- a/arch/um/defconfig +++ b/arch/um/defconfig @@ -155,10 +155,10 @@ CONFIG_CPUSETS=y CONFIG_PROC_PID_CPUSET=y CONFIG_CGROUP_CPUACCT=y CONFIG_RESOURCE_COUNTERS=y -CONFIG_CGROUP_MEM_RES_CTLR=y -CONFIG_CGROUP_MEM_RES_CTLR_SWAP=y -# CONFIG_CGROUP_MEM_RES_CTLR_SWAP_ENABLED is not set -# CONFIG_CGROUP_MEM_RES_CTLR_KMEM is not set +CONFIG_CGROUP_MEMCG=y +CONFIG_CGROUP_MEMCG_SWAP=y +# CONFIG_CGROUP_MEMCG_SWAP_ENABLED is not set +# CONFIG_CGROUP_MEMCG_KMEM is not set CONFIG_CGROUP_SCHED=y CONFIG_FAIR_GROUP_SCHED=y # CONFIG_CFS_BANDWIDTH is not set diff --git a/include/linux/cgroup_subsys.h b/include/linux/cgroup_subsys.h index 5b41ce079024..dfae957398c3 100644 --- a/include/linux/cgroup_subsys.h +++ b/include/linux/cgroup_subsys.h @@ -31,7 +31,7 @@ SUBSYS(cpuacct) /* */ -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG SUBSYS(mem_cgroup) #endif diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 83e7ba90d6e5..170076222431 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -38,7 +38,7 @@ struct mem_cgroup_reclaim_cookie { unsigned int generation; }; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG /* * All "charge" functions with gfp_mask should use GFP_KERNEL or * (gfp_mask & GFP_RECLAIM_MASK). In current implementatin, memcg doesn't @@ -124,7 +124,7 @@ extern void mem_cgroup_print_oom_info(struct mem_cgroup *memcg, extern void mem_cgroup_replace_page_cache(struct page *oldpage, struct page *newpage); -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP extern int do_swap_account; #endif @@ -193,7 +193,7 @@ void mem_cgroup_split_huge_fixup(struct page *head); bool mem_cgroup_bad_page_check(struct page *page); void mem_cgroup_print_bad_page(struct page *page); #endif -#else /* CONFIG_CGROUP_MEM_RES_CTLR */ +#else /* CONFIG_MEMCG */ struct mem_cgroup; static inline int mem_cgroup_newpage_charge(struct page *page, @@ -384,9 +384,9 @@ static inline void mem_cgroup_replace_page_cache(struct page *oldpage, struct page *newpage) { } -#endif /* CONFIG_CGROUP_MEM_RES_CTLR */ +#endif /* CONFIG_MEMCG */ -#if !defined(CONFIG_CGROUP_MEM_RES_CTLR) || !defined(CONFIG_DEBUG_VM) +#if !defined(CONFIG_MEMCG) || !defined(CONFIG_DEBUG_VM) static inline bool mem_cgroup_bad_page_check(struct page *page) { @@ -406,7 +406,7 @@ enum { }; struct sock; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +#ifdef CONFIG_MEMCG_KMEM void sock_update_memcg(struct sock *sk); void sock_release_memcg(struct sock *sk); #else @@ -416,6 +416,6 @@ static inline void sock_update_memcg(struct sock *sk) static inline void sock_release_memcg(struct sock *sk) { } -#endif /* CONFIG_CGROUP_MEM_RES_CTLR_KMEM */ +#endif /* CONFIG_MEMCG_KMEM */ #endif /* _LINUX_MEMCONTROL_H */ diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 458988bd55a1..3bdfa15b2012 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -201,7 +201,7 @@ struct zone_reclaim_stat { struct lruvec { struct list_head lists[NR_LRU_LISTS]; struct zone_reclaim_stat reclaim_stat; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG struct zone *zone; #endif }; @@ -671,7 +671,7 @@ typedef struct pglist_data { int nr_zones; #ifdef CONFIG_FLAT_NODE_MEM_MAP /* means !SPARSEMEM */ struct page *node_mem_map; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG struct page_cgroup *node_page_cgroup; #endif #endif @@ -736,7 +736,7 @@ extern void lruvec_init(struct lruvec *lruvec, struct zone *zone); static inline struct zone *lruvec_zone(struct lruvec *lruvec) { -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG return lruvec->zone; #else return container_of(lruvec, struct zone, lruvec); @@ -1052,7 +1052,7 @@ struct mem_section { /* See declaration of similar field in struct zone */ unsigned long *pageblock_flags; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG /* * If !SPARSEMEM, pgdat doesn't have page_cgroup pointer. We use * section. (see memcontrol.h/page_cgroup.h about this.) diff --git a/include/linux/page_cgroup.h b/include/linux/page_cgroup.h index a88cdba27809..777a524716db 100644 --- a/include/linux/page_cgroup.h +++ b/include/linux/page_cgroup.h @@ -12,7 +12,7 @@ enum { #ifndef __GENERATING_BOUNDS_H #include -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG #include /* @@ -82,7 +82,7 @@ static inline void unlock_page_cgroup(struct page_cgroup *pc) bit_spin_unlock(PCG_LOCK, &pc->flags); } -#else /* CONFIG_CGROUP_MEM_RES_CTLR */ +#else /* CONFIG_MEMCG */ struct page_cgroup; static inline void __meminit pgdat_page_cgroup_init(struct pglist_data *pgdat) @@ -102,11 +102,11 @@ static inline void __init page_cgroup_init_flatmem(void) { } -#endif /* CONFIG_CGROUP_MEM_RES_CTLR */ +#endif /* CONFIG_MEMCG */ #include -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP extern unsigned short swap_cgroup_cmpxchg(swp_entry_t ent, unsigned short old, unsigned short new); extern unsigned short swap_cgroup_record(swp_entry_t ent, unsigned short id); @@ -138,7 +138,7 @@ static inline void swap_cgroup_swapoff(int type) return; } -#endif /* CONFIG_CGROUP_MEM_RES_CTLR_SWAP */ +#endif /* CONFIG_MEMCG_SWAP */ #endif /* !__GENERATING_BOUNDS_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 68dcffaa62a0..865725adb9d3 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1584,7 +1584,7 @@ struct task_struct { /* bitmask and counter of trace recursion */ unsigned long trace_recursion; #endif /* CONFIG_TRACING */ -#ifdef CONFIG_CGROUP_MEM_RES_CTLR /* memcg uses this to do batch job */ +#ifdef CONFIG_MEMCG /* memcg uses this to do batch job */ struct memcg_batch_info { int do_batch; /* incremented when batch uncharge started */ struct mem_cgroup *memcg; /* target memcg of uncharge */ diff --git a/include/linux/swap.h b/include/linux/swap.h index c84ec68eaec9..9a16bb1cefd1 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -301,7 +301,7 @@ static inline void scan_unevictable_unregister_node(struct node *node) extern int kswapd_run(int nid); extern void kswapd_stop(int nid); -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG extern int mem_cgroup_swappiness(struct mem_cgroup *mem); #else static inline int mem_cgroup_swappiness(struct mem_cgroup *mem) @@ -309,7 +309,7 @@ static inline int mem_cgroup_swappiness(struct mem_cgroup *mem) return vm_swappiness; } #endif -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP extern void mem_cgroup_uncharge_swap(swp_entry_t ent); #else static inline void mem_cgroup_uncharge_swap(swp_entry_t ent) @@ -360,7 +360,7 @@ extern int reuse_swap_page(struct page *); extern int try_to_free_swap(struct page *); struct backing_dev_info; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG extern void mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent, bool swapout); #else diff --git a/include/net/sock.h b/include/net/sock.h index e067f8c18f88..cee528c119ca 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -913,7 +913,7 @@ struct proto { #ifdef SOCK_REFCNT_DEBUG atomic_t socks; #endif -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +#ifdef CONFIG_MEMCG_KMEM /* * cgroup specific init/deinit functions. Called once for all * protocols that implement it, from cgroups populate function. @@ -994,7 +994,7 @@ inline void sk_refcnt_debug_release(const struct sock *sk) #define sk_refcnt_debug_release(sk) do { } while (0) #endif /* SOCK_REFCNT_DEBUG */ -#if defined(CONFIG_CGROUP_MEM_RES_CTLR_KMEM) && defined(CONFIG_NET) +#if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_NET) extern struct static_key memcg_socket_limit_enabled; static inline struct cg_proto *parent_cg_proto(struct proto *proto, struct cg_proto *cg_proto) diff --git a/init/Kconfig b/init/Kconfig index 72437760e90e..af6c7f8ba019 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -686,7 +686,7 @@ config RESOURCE_COUNTERS This option enables controller independent resource accounting infrastructure that works with cgroups. -config CGROUP_MEM_RES_CTLR +config MEMCG bool "Memory Resource Controller for Control Groups" depends on RESOURCE_COUNTERS select MM_OWNER @@ -709,9 +709,9 @@ config CGROUP_MEM_RES_CTLR This config option also selects MM_OWNER config option, which could in turn add some fork/exit overhead. -config CGROUP_MEM_RES_CTLR_SWAP +config MEMCG_SWAP bool "Memory Resource Controller Swap Extension" - depends on CGROUP_MEM_RES_CTLR && SWAP + depends on MEMCG && SWAP help Add swap management feature to memory resource controller. When you enable this, you can limit mem+swap usage per cgroup. In other words, @@ -726,9 +726,9 @@ config CGROUP_MEM_RES_CTLR_SWAP if boot option "swapaccount=0" is set, swap will not be accounted. Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page size is 4096bytes, 512k per 1Gbytes of swap. -config CGROUP_MEM_RES_CTLR_SWAP_ENABLED +config MEMCG_SWAP_ENABLED bool "Memory Resource Controller Swap Extension enabled by default" - depends on CGROUP_MEM_RES_CTLR_SWAP + depends on MEMCG_SWAP default y help Memory Resource Controller Swap Extension comes with its price in @@ -739,9 +739,9 @@ config CGROUP_MEM_RES_CTLR_SWAP_ENABLED For those who want to have the feature enabled by default should select this option (if, for some reason, they need to disable it then swapaccount=0 does the trick). -config CGROUP_MEM_RES_CTLR_KMEM +config MEMCG_KMEM bool "Memory Resource Controller Kernel Memory accounting (EXPERIMENTAL)" - depends on CGROUP_MEM_RES_CTLR && EXPERIMENTAL + depends on MEMCG && EXPERIMENTAL default n help The Kernel Memory extension for Memory Resource Controller can limit diff --git a/kernel/fork.c b/kernel/fork.c index aaa8813c45d1..3bd2280d79f6 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1306,7 +1306,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, #ifdef CONFIG_DEBUG_MUTEXES p->blocked_on = NULL; /* not blocked yet */ #endif -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG p->memcg_batch.do_batch = 0; p->memcg_batch.memcg = NULL; #endif diff --git a/mm/Makefile b/mm/Makefile index fd6fc1c1966c..290bbfe33698 100644 --- a/mm/Makefile +++ b/mm/Makefile @@ -49,7 +49,7 @@ obj-$(CONFIG_FS_XIP) += filemap_xip.o obj-$(CONFIG_MIGRATION) += migrate.o obj-$(CONFIG_QUICKLIST) += quicklist.o obj-$(CONFIG_TRANSPARENT_HUGEPAGE) += huge_memory.o -obj-$(CONFIG_CGROUP_MEM_RES_CTLR) += memcontrol.o page_cgroup.o +obj-$(CONFIG_MEMCG) += memcontrol.o page_cgroup.o obj-$(CONFIG_CGROUP_HUGETLB) += hugetlb_cgroup.o obj-$(CONFIG_MEMORY_FAILURE) += memory-failure.o obj-$(CONFIG_HWPOISON_INJECT) += hwpoison-inject.o diff --git a/mm/hwpoison-inject.c b/mm/hwpoison-inject.c index cc448bb983ba..3a61efc518d5 100644 --- a/mm/hwpoison-inject.c +++ b/mm/hwpoison-inject.c @@ -123,7 +123,7 @@ static int pfn_inject_init(void) if (!dentry) goto fail; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP dentry = debugfs_create_u64("corrupt-filter-memcg", 0600, hwpoison_dir, &hwpoison_filter_memcg); if (!dentry) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index a2677e0a6387..55a85e1a342f 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -61,12 +61,12 @@ struct cgroup_subsys mem_cgroup_subsys __read_mostly; #define MEM_CGROUP_RECLAIM_RETRIES 5 static struct mem_cgroup *root_mem_cgroup __read_mostly; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP /* Turned on only when memory cgroup is enabled && really_do_swap_account = 1 */ int do_swap_account __read_mostly; /* for remember boot option*/ -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP_ENABLED +#ifdef CONFIG_MEMCG_SWAP_ENABLED static int really_do_swap_account __initdata = 1; #else static int really_do_swap_account __initdata = 0; @@ -407,7 +407,7 @@ static void mem_cgroup_get(struct mem_cgroup *memcg); static void mem_cgroup_put(struct mem_cgroup *memcg); /* Writing them here to avoid exposing memcg's inner layout */ -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +#ifdef CONFIG_MEMCG_KMEM #include #include @@ -466,9 +466,9 @@ struct cg_proto *tcp_proto_cgroup(struct mem_cgroup *memcg) } EXPORT_SYMBOL(tcp_proto_cgroup); #endif /* CONFIG_INET */ -#endif /* CONFIG_CGROUP_MEM_RES_CTLR_KMEM */ +#endif /* CONFIG_MEMCG_KMEM */ -#if defined(CONFIG_INET) && defined(CONFIG_CGROUP_MEM_RES_CTLR_KMEM) +#if defined(CONFIG_INET) && defined(CONFIG_MEMCG_KMEM) static void disarm_sock_keys(struct mem_cgroup *memcg) { if (!memcg_proto_activated(&memcg->tcp_mem.cg_proto)) @@ -3085,7 +3085,7 @@ mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent, bool swapout) } #endif -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP /* * called from swap_entry_free(). remove record in swap_cgroup and * uncharge "memsw" account. @@ -4518,7 +4518,7 @@ static int mem_cgroup_oom_control_write(struct cgroup *cgrp, return 0; } -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +#ifdef CONFIG_MEMCG_KMEM static int memcg_init_kmem(struct mem_cgroup *memcg, struct cgroup_subsys *ss) { return mem_cgroup_sockets_init(memcg, ss); @@ -4608,7 +4608,7 @@ static struct cftype mem_cgroup_files[] = { .read_seq_string = mem_control_numa_stat_show, }, #endif -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP { .name = "memsw.usage_in_bytes", .private = MEMFILE_PRIVATE(_MEMSWAP, RES_USAGE), @@ -4795,7 +4795,7 @@ struct mem_cgroup *parent_mem_cgroup(struct mem_cgroup *memcg) } EXPORT_SYMBOL(parent_mem_cgroup); -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP static void __init enable_swap_cgroup(void) { if (!mem_cgroup_disabled() && really_do_swap_account) @@ -5526,7 +5526,7 @@ struct cgroup_subsys mem_cgroup_subsys = { .__DEPRECATED_clear_css_refs = true, }; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP static int __init enable_swap_account(char *s) { /* consider enabled if no parameter or 1 is given */ diff --git a/mm/memory-failure.c b/mm/memory-failure.c index b04ff2d6f73d..a6e2141a6610 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -128,7 +128,7 @@ static int hwpoison_filter_flags(struct page *p) * can only guarantee that the page either belongs to the memcg tasks, or is * a freed page. */ -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP u64 hwpoison_filter_memcg; EXPORT_SYMBOL_GPL(hwpoison_filter_memcg); static int hwpoison_filter_task(struct page *p) diff --git a/mm/mmzone.c b/mm/mmzone.c index 6830eab5bf09..3cef80f6ac79 100644 --- a/mm/mmzone.c +++ b/mm/mmzone.c @@ -96,7 +96,7 @@ void lruvec_init(struct lruvec *lruvec, struct zone *zone) for_each_lru(lru) INIT_LIST_HEAD(&lruvec->lists[lru]); -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG lruvec->zone = zone; #endif } diff --git a/mm/oom_kill.c b/mm/oom_kill.c index c82ede69bf3f..e6c10640e56b 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -541,7 +541,7 @@ static void check_panic_on_oom(enum oom_constraint constraint, gfp_t gfp_mask, sysctl_panic_on_oom == 2 ? "compulsory" : "system-wide"); } -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG void mem_cgroup_out_of_memory(struct mem_cgroup *memcg, gfp_t gfp_mask, int order) { diff --git a/mm/page_cgroup.c b/mm/page_cgroup.c index eb750f851395..5ddad0c6daa6 100644 --- a/mm/page_cgroup.c +++ b/mm/page_cgroup.c @@ -317,7 +317,7 @@ void __meminit pgdat_page_cgroup_init(struct pglist_data *pgdat) #endif -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP +#ifdef CONFIG_MEMCG_SWAP static DEFINE_MUTEX(swap_cgroup_mutex); struct swap_cgroup_ctrl { diff --git a/mm/vmscan.c b/mm/vmscan.c index 347b3ff2a478..6b1f89a91212 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -133,7 +133,7 @@ long vm_total_pages; /* The total number of pages which the VM controls */ static LIST_HEAD(shrinker_list); static DECLARE_RWSEM(shrinker_rwsem); -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG static bool global_reclaim(struct scan_control *sc) { return !sc->target_mem_cgroup; @@ -2142,7 +2142,7 @@ unsigned long try_to_free_pages(struct zonelist *zonelist, int order, return nr_reclaimed; } -#ifdef CONFIG_CGROUP_MEM_RES_CTLR +#ifdef CONFIG_MEMCG unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *memcg, gfp_t gfp_mask, bool noswap, diff --git a/net/core/sock.c b/net/core/sock.c index 2676a88f533e..a67b06280e4c 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -142,7 +142,7 @@ static DEFINE_MUTEX(proto_list_mutex); static LIST_HEAD(proto_list); -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +#ifdef CONFIG_MEMCG_KMEM int mem_cgroup_sockets_init(struct mem_cgroup *memcg, struct cgroup_subsys *ss) { struct proto *proto; diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index ae2ccf2890e4..15ca63ec604e 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -49,7 +49,7 @@ obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o -obj-$(CONFIG_CGROUP_MEM_RES_CTLR_KMEM) += tcp_memcontrol.o +obj-$(CONFIG_MEMCG_KMEM) += tcp_memcontrol.o obj-$(CONFIG_NETLABEL) += cipso_ipv4.o obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \ diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 5840c3255721..ed7db3f1b6f2 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -184,7 +184,7 @@ static int ipv4_tcp_mem(ctl_table *ctl, int write, int ret; unsigned long vec[3]; struct net *net = current->nsproxy->net_ns; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +#ifdef CONFIG_MEMCG_KMEM struct mem_cgroup *memcg; #endif @@ -203,7 +203,7 @@ static int ipv4_tcp_mem(ctl_table *ctl, int write, if (ret) return ret; -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +#ifdef CONFIG_MEMCG_KMEM rcu_read_lock(); memcg = mem_cgroup_from_task(current); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 2fbd9921253f..4bc8f6769b57 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2633,7 +2633,7 @@ struct proto tcp_prot = { .compat_setsockopt = compat_tcp_setsockopt, .compat_getsockopt = compat_tcp_getsockopt, #endif -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +#ifdef CONFIG_MEMCG_KMEM .init_cgroup = tcp_init_cgroup, .destroy_cgroup = tcp_destroy_cgroup, .proto_cgroup = tcp_proto_cgroup, diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 221224e72507..61c7b6d83176 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2015,7 +2015,7 @@ struct proto tcpv6_prot = { .compat_setsockopt = compat_tcp_setsockopt, .compat_getsockopt = compat_tcp_getsockopt, #endif -#ifdef CONFIG_CGROUP_MEM_RES_CTLR_KMEM +#ifdef CONFIG_MEMCG_KMEM .proto_cgroup = tcp_proto_cgroup, #endif }; -- cgit v1.2.3 From 9adb62a5df9c0fbef7b4665919329f73a34651ed Mon Sep 17 00:00:00 2001 From: Jiang Liu Date: Tue, 31 Jul 2012 16:43:28 -0700 Subject: mm/hotplug: correctly setup fallback zonelists when creating new pgdat When hotadd_new_pgdat() is called to create new pgdat for a new node, a fallback zonelist should be created for the new node. There's code to try to achieve that in hotadd_new_pgdat() as below: /* * The node we allocated has no zone fallback lists. For avoiding * to access not-initialized zonelist, build here. */ mutex_lock(&zonelists_mutex); build_all_zonelists(pgdat, NULL); mutex_unlock(&zonelists_mutex); But it doesn't work as expected. When hotadd_new_pgdat() is called, the new node is still in offline state because node_set_online(nid) hasn't been called yet. And build_all_zonelists() only builds zonelists for online nodes as: for_each_online_node(nid) { pg_data_t *pgdat = NODE_DATA(nid); build_zonelists(pgdat); build_zonelist_cache(pgdat); } Though we hope to create zonelist for the new pgdat, but it doesn't. So add a new parameter "pgdat" the build_all_zonelists() to build pgdat for the new pgdat too. Signed-off-by: Jiang Liu Signed-off-by: Xishi Qiu Cc: Mel Gorman Cc: Michal Hocko Cc: Minchan Kim Cc: Rusty Russell Cc: Yinghai Lu Cc: Tony Luck Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Cc: David Rientjes Cc: Keping Chen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 2 +- init/main.c | 2 +- kernel/cpu.c | 2 +- mm/memory_hotplug.c | 4 ++-- mm/page_alloc.c | 17 ++++++++++++----- 5 files changed, 17 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index f64afa5929fe..98f079bcf399 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -721,7 +721,7 @@ typedef struct pglist_data { #include extern struct mutex zonelists_mutex; -void build_all_zonelists(void *data); +void build_all_zonelists(pg_data_t *pgdat, struct zone *zone); void wakeup_kswapd(struct zone *zone, int order, enum zone_type classzone_idx); bool zone_watermark_ok(struct zone *z, int order, unsigned long mark, int classzone_idx, int alloc_flags); diff --git a/init/main.c b/init/main.c index 95316a1b4a76..e60679de61c3 100644 --- a/init/main.c +++ b/init/main.c @@ -506,7 +506,7 @@ asmlinkage void __init start_kernel(void) setup_per_cpu_areas(); smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */ - build_all_zonelists(NULL); + build_all_zonelists(NULL, NULL); page_alloc_init(); printk(KERN_NOTICE "Kernel command line: %s\n", boot_command_line); diff --git a/kernel/cpu.c b/kernel/cpu.c index a4eb5227a19e..14d32588cccd 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -416,7 +416,7 @@ int __cpuinit cpu_up(unsigned int cpu) if (pgdat->node_zonelists->_zonerefs->zone == NULL) { mutex_lock(&zonelists_mutex); - build_all_zonelists(NULL); + build_all_zonelists(NULL, NULL); mutex_unlock(&zonelists_mutex); } #endif diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 427bb291dd0f..b8731040b9f9 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -513,7 +513,7 @@ int __ref online_pages(unsigned long pfn, unsigned long nr_pages) zone->present_pages += onlined_pages; zone->zone_pgdat->node_present_pages += onlined_pages; if (need_zonelists_rebuild) - build_all_zonelists(zone); + build_all_zonelists(NULL, zone); else zone_pcp_update(zone); @@ -562,7 +562,7 @@ static pg_data_t __ref *hotadd_new_pgdat(int nid, u64 start) * to access not-initialized zonelist, build here. */ mutex_lock(&zonelists_mutex); - build_all_zonelists(NULL); + build_all_zonelists(pgdat, NULL); mutex_unlock(&zonelists_mutex); return pgdat; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 6c7e3bd93a85..9ad6866ac49c 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3032,7 +3032,7 @@ int numa_zonelist_order_handler(ctl_table *table, int write, user_zonelist_order = oldval; } else if (oldval != user_zonelist_order) { mutex_lock(&zonelists_mutex); - build_all_zonelists(NULL); + build_all_zonelists(NULL, NULL); mutex_unlock(&zonelists_mutex); } } @@ -3415,10 +3415,17 @@ static __init_refok int __build_all_zonelists(void *data) { int nid; int cpu; + pg_data_t *self = data; #ifdef CONFIG_NUMA memset(node_load, 0, sizeof(node_load)); #endif + + if (self && !node_online(self->node_id)) { + build_zonelists(self); + build_zonelist_cache(self); + } + for_each_online_node(nid) { pg_data_t *pgdat = NODE_DATA(nid); @@ -3463,7 +3470,7 @@ static __init_refok int __build_all_zonelists(void *data) * Called with zonelists_mutex held always * unless system_state == SYSTEM_BOOTING. */ -void __ref build_all_zonelists(void *data) +void __ref build_all_zonelists(pg_data_t *pgdat, struct zone *zone) { set_zonelist_order(); @@ -3475,10 +3482,10 @@ void __ref build_all_zonelists(void *data) /* we have to stop all cpus to guarantee there is no user of zonelist */ #ifdef CONFIG_MEMORY_HOTPLUG - if (data) - setup_zone_pageset((struct zone *)data); + if (zone) + setup_zone_pageset(zone); #endif - stop_machine(__build_all_zonelists, NULL, NULL); + stop_machine(__build_all_zonelists, pgdat, NULL); /* cpuset refresh routine should be here */ } vm_total_pages = nr_free_pagecache_pages(); -- cgit v1.2.3 From 907aed48f65efeecf91575397e3d79335d93a466 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Tue, 31 Jul 2012 16:44:07 -0700 Subject: mm: allow PF_MEMALLOC from softirq context This is needed to allow network softirq packet processing to make use of PF_MEMALLOC. Currently softirq context cannot use PF_MEMALLOC due to it not being associated with a task, and therefore not having task flags to fiddle with - thus the gfp to alloc flag mapping ignores the task flags when in interrupts (hard or soft) context. Allowing softirqs to make use of PF_MEMALLOC therefore requires some trickery. This patch borrows the task flags from whatever process happens to be preempted by the softirq. It then modifies the gfp to alloc flags mapping to not exclude task flags in softirq context, and modify the softirq code to save, clear and restore the PF_MEMALLOC flag. The save and clear, ensures the preempted task's PF_MEMALLOC flag doesn't leak into the softirq. The restore ensures a softirq's PF_MEMALLOC flag cannot leak back into the preempted process. This should be safe due to the following reasons Softirqs can run on multiple CPUs sure but the same task should not be executing the same softirq code. Neither should the softirq handler be preempted by any other softirq handler so the flags should not leak to an unrelated softirq. Softirqs re-enable hardware interrupts in __do_softirq() so can be preempted by hardware interrupts so PF_MEMALLOC is inherited by the hard IRQ. However, this is similar to a process in reclaim being preempted by a hardirq. While PF_MEMALLOC is set, gfp_to_alloc_flags() distinguishes between hard and soft irqs and avoids giving a hardirq the ALLOC_NO_WATERMARKS flag. If the softirq is deferred to ksoftirq then its flags may be used instead of a normal tasks but as the softirq cannot be preempted, the PF_MEMALLOC flag does not leak to other code by accident. [davem@davemloft.net: Document why PF_MEMALLOC is safe] Signed-off-by: Peter Zijlstra Signed-off-by: Mel Gorman Cc: David Miller Cc: Neil Brown Cc: Mike Christie Cc: Eric B Munson Cc: Eric Dumazet Cc: Sebastian Andrzej Siewior Cc: Mel Gorman Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/sched.h | 7 +++++++ kernel/softirq.c | 9 +++++++++ mm/page_alloc.c | 6 +++++- 3 files changed, 21 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index 865725adb9d3..c147e7024f11 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1894,6 +1894,13 @@ static inline void rcu_copy_process(struct task_struct *p) #endif +static inline void tsk_restore_flags(struct task_struct *task, + unsigned long orig_flags, unsigned long flags) +{ + task->flags &= ~flags; + task->flags |= orig_flags & flags; +} + #ifdef CONFIG_SMP extern void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask); diff --git a/kernel/softirq.c b/kernel/softirq.c index 671f9594e368..b73e681df09e 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -210,6 +210,14 @@ asmlinkage void __do_softirq(void) __u32 pending; int max_restart = MAX_SOFTIRQ_RESTART; int cpu; + unsigned long old_flags = current->flags; + + /* + * Mask out PF_MEMALLOC s current task context is borrowed for the + * softirq. A softirq handled such as network RX might set PF_MEMALLOC + * again if the socket is related to swap + */ + current->flags &= ~PF_MEMALLOC; pending = local_softirq_pending(); account_system_vtime(current); @@ -265,6 +273,7 @@ restart: account_system_vtime(current); __local_bh_enable(SOFTIRQ_OFFSET); + tsk_restore_flags(current, old_flags, PF_MEMALLOC); } #ifndef __ARCH_HAS_DO_SOFTIRQ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8f65abeb9ad6..cd5390f2f18d 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2296,7 +2296,11 @@ gfp_to_alloc_flags(gfp_t gfp_mask) if (likely(!(gfp_mask & __GFP_NOMEMALLOC))) { if (gfp_mask & __GFP_MEMALLOC) alloc_flags |= ALLOC_NO_WATERMARKS; - else if (likely(!(gfp_mask & __GFP_NOMEMALLOC)) && !in_interrupt()) + else if (in_serving_softirq() && (current->flags & PF_MEMALLOC)) + alloc_flags |= ALLOC_NO_WATERMARKS; + else if (!in_interrupt() && + ((current->flags & PF_MEMALLOC) || + unlikely(test_thread_flag(TIF_MEMDIE)))) alloc_flags |= ALLOC_NO_WATERMARKS; } -- cgit v1.2.3 From 1d17d17484d40f2d5b35c79518597a2b25296996 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sat, 4 Aug 2012 21:21:14 +0200 Subject: time: Fix adjustment cleanup bug in timekeeping_adjust() Tetsuo Handa reported that sporadically the system clock starts counting up too quickly which is enough to confuse the hangcheck timer to print a bogus stall warning. Commit 2a8c0883 "time: Move xtime_nsec adjustment underflow handling timekeeping_adjust" overlooked this exit path: } else return; which should really be a proper exit sequence, fixing the bug as a side effect. Also make the flow more readable by properly balancing curly braces. Reported-by: Tetsuo Handa wrote: Tested-by: Tetsuo Handa wrote: Signed-off-by: Ingo Molnar Cc: john.stultz@linaro.org Cc: a.p.zijlstra@chello.nl Cc: richardcochran@gmail.com Cc: prarit@redhat.com Link: http://lkml.kernel.org/r/20120804192114.GA28347@gmail.com Signed-off-by: Ingo Molnar --- kernel/time/timekeeping.c | 31 +++++++++++++++++-------------- 1 file changed, 17 insertions(+), 14 deletions(-) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 2988bc819187..e16af197a2bc 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -923,20 +923,22 @@ static void timekeeping_adjust(struct timekeeper *tk, s64 offset) if (likely(error <= interval)) adj = 1; else - adj = timekeeping_bigadjust(tk, error, &interval, - &offset); - } else if (error < -interval) { - /* See comment above, this is just switched for the negative */ - error >>= 2; - if (likely(error >= -interval)) { - adj = -1; - interval = -interval; - offset = -offset; - } else - adj = timekeeping_bigadjust(tk, error, &interval, - &offset); - } else - return; + adj = timekeeping_bigadjust(tk, error, &interval, &offset); + } else { + if (error < -interval) { + /* See comment above, this is just switched for the negative */ + error >>= 2; + if (likely(error >= -interval)) { + adj = -1; + interval = -interval; + offset = -offset; + } else { + adj = timekeeping_bigadjust(tk, error, &interval, &offset); + } + } else { + goto out_adjust; + } + } if (unlikely(tk->clock->maxadj && (tk->mult + adj > tk->clock->mult + tk->clock->maxadj))) { @@ -999,6 +1001,7 @@ static void timekeeping_adjust(struct timekeeper *tk, s64 offset) tk->xtime_nsec -= offset; tk->ntp_error -= (interval - offset) << tk->ntp_error_shift; +out_adjust: /* * It may be possible that when we entered this function, xtime_nsec * was very small. Further, if we're slightly speeding the clocksource -- cgit v1.2.3 From 300d3739e873d50d4c6e3656f89007a217fb1d29 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 7 Aug 2012 13:50:22 +0200 Subject: Revert "NMI watchdog: fix for lockup detector breakage on resume" Revert commit 45226e9 (NMI watchdog: fix for lockup detector breakage on resume) which breaks resume from system suspend on my SH7372 Mackerel board (by causing a NULL pointer dereference to happen) and is generally wrong, because it abuses the CPU hotplug functionality in a shamelessly blatant way. The original issue should be addressed through appropriate syscore resume callback instead. Signed-off-by: Rafael J. Wysocki --- include/linux/sched.h | 8 -------- kernel/power/suspend.c | 3 --- kernel/watchdog.c | 21 ++------------------- 3 files changed, 2 insertions(+), 30 deletions(-) (limited to 'kernel') diff --git a/include/linux/sched.h b/include/linux/sched.h index c147e7024f11..b8c86648a2f9 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -334,14 +334,6 @@ static inline void lockup_detector_init(void) } #endif -#if defined(CONFIG_LOCKUP_DETECTOR) && defined(CONFIG_SUSPEND) -void lockup_detector_bootcpu_resume(void); -#else -static inline void lockup_detector_bootcpu_resume(void) -{ -} -#endif - #ifdef CONFIG_DETECT_HUNG_TASK extern unsigned int sysctl_hung_task_panic; extern unsigned long sysctl_hung_task_check_count; diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 1da39ea248fd..c8b7446b27df 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -178,9 +178,6 @@ static int suspend_enter(suspend_state_t state, bool *wakeup) arch_suspend_enable_irqs(); BUG_ON(irqs_disabled()); - /* Kick the lockup detector */ - lockup_detector_bootcpu_resume(); - Enable_cpus: enable_nonboot_cpus(); diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 69add8a9da68..4b1dfba70f7c 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -575,7 +575,7 @@ out: /* * Create/destroy watchdog threads as CPUs come and go: */ -static int +static int __cpuinit cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { int hotcpu = (unsigned long)hcpu; @@ -610,27 +610,10 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) return NOTIFY_OK; } -static struct notifier_block cpu_nfb = { +static struct notifier_block __cpuinitdata cpu_nfb = { .notifier_call = cpu_callback }; -#ifdef CONFIG_SUSPEND -/* - * On exit from suspend we force an offline->online transition on the boot CPU - * so that the PMU state that was lost while in suspended state gets set up - * properly for the boot CPU. This information is required for restarting the - * NMI watchdog. - */ -void lockup_detector_bootcpu_resume(void) -{ - void *cpu = (void *)(long)smp_processor_id(); - - cpu_callback(&cpu_nfb, CPU_DEAD_FROZEN, cpu); - cpu_callback(&cpu_nfb, CPU_UP_PREPARE_FROZEN, cpu); - cpu_callback(&cpu_nfb, CPU_ONLINE_FROZEN, cpu); -} -#endif - void __init lockup_detector_init(void) { void *cpu = (void *)(long)smp_processor_id(); -- cgit v1.2.3 From e3756477aec028427fec767957c0d4b6cfb87208 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Fri, 10 Aug 2012 15:07:09 -0400 Subject: printk: Fix calculation of length used to discard records While tracking down a weird buffer overflow issue in a program that looked to be sane, I started double checking the length returned by syslog(SYSLOG_ACTION_READ_ALL, ...) to make sure it wasn't overflowing the buffer. Sure enough, it was. I saw this in strace: 11339 syslog(SYSLOG_ACTION_READ_ALL, "<5>[244017.708129] REISERFS (dev"..., 8192) = 8279 It turns out that the loops that calculate how much space the entries will take when they're copied don't include the newlines and prefixes that will be included in the final output since prev flags is passed as zero. This patch properly accounts for it and fixes the overflow. CC: stable@kernel.org Signed-off-by: Jeff Mahoney Signed-off-by: Linus Torvalds --- kernel/printk.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/printk.c b/kernel/printk.c index 6a76ab9d4476..66a2ea37b576 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1034,6 +1034,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) struct log *msg = log_from_idx(idx); len += msg_print_text(msg, prev, true, NULL, 0); + prev = msg->flags; idx = log_next(idx); seq++; } @@ -1046,6 +1047,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear) struct log *msg = log_from_idx(idx); len -= msg_print_text(msg, prev, true, NULL, 0); + prev = msg->flags; idx = log_next(idx); seq++; } -- cgit v1.2.3 From a35b6466aabb051568b844e8c63f87a356d3d129 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Aug 2012 21:46:40 +0200 Subject: sched, cgroup: Reduce rq->lock hold times for large cgroup hierarchies Peter Portante reported that for large cgroup hierarchies (and or on large CPU counts) we get immense lock contention on rq->lock and stuff stops working properly. His workload was a ton of processes, each in their own cgroup, everybody idling except for a sporadic wakeup once every so often. It was found that: schedule() idle_balance() load_balance() local_irq_save() double_rq_lock() update_h_load() walk_tg_tree(tg_load_down) tg_load_down() Results in an entire cgroup hierarchy walk under rq->lock for every new-idle balance and since new-idle balance isn't throttled this results in a lot of work while holding the rq->lock. This patch does two things, it removes the work from under rq->lock based on the good principle of race and pray which is widely employed in the load-balancer as a whole. And secondly it throttles the update_h_load() calculation to max once per jiffy. I considered excluding update_h_load() for new-idle balance all-together, but purely relying on regular balance passes to update this data might not work out under some rare circumstances where the new-idle busiest isn't the regular busiest for a while (unlikely, but a nightmare to debug if someone hits it and suffers). Cc: pjt@google.com Cc: Larry Woodman Cc: Mike Galbraith Reported-by: Peter Portante Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-aaarrzfpnaam7pqrekofu8a6@git.kernel.org Signed-off-by: Thomas Gleixner --- kernel/sched/fair.c | 11 +++++++++-- kernel/sched/sched.h | 6 +++++- 2 files changed, 14 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index d0cc03b3e70b..c219bf8d704c 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3387,6 +3387,14 @@ static int tg_load_down(struct task_group *tg, void *data) static void update_h_load(long cpu) { + struct rq *rq = cpu_rq(cpu); + unsigned long now = jiffies; + + if (rq->h_load_throttle == now) + return; + + rq->h_load_throttle = now; + rcu_read_lock(); walk_tg_tree(tg_load_down, tg_nop, (void *)cpu); rcu_read_unlock(); @@ -4293,11 +4301,10 @@ redo: env.src_rq = busiest; env.loop_max = min(sysctl_sched_nr_migrate, busiest->nr_running); + update_h_load(env.src_cpu); more_balance: local_irq_save(flags); double_rq_lock(this_rq, busiest); - if (!env.loop) - update_h_load(env.src_cpu); /* * cur_ld_moved - load moved in current iteration diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index c35a1a7dd4d6..531411b1044e 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -374,7 +374,11 @@ struct rq { #ifdef CONFIG_FAIR_GROUP_SCHED /* list of leaf cfs_rq on this cpu: */ struct list_head leaf_cfs_rq_list; -#endif +#ifdef CONFIG_SMP + unsigned long h_load_throttle; +#endif /* CONFIG_SMP */ +#endif /* CONFIG_FAIR_GROUP_SCHED */ + #ifdef CONFIG_RT_GROUP_SCHED struct list_head leaf_rt_rq_list; #endif -- cgit v1.2.3 From bea6832cc8c4a0a9a65dd17da6aaa657fe27bc3e Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Wed, 8 Aug 2012 11:27:15 +0200 Subject: sched: fix divide by zero at {thread_group,task}_times On architectures where cputime_t is 64 bit type, is possible to trigger divide by zero on do_div(temp, (__force u32) total) line, if total is a non zero number but has lower 32 bit's zeroed. Removing casting is not a good solution since some do_div() implementations do cast to u32 internally. This problem can be triggered in practice on very long lived processes: PID: 2331 TASK: ffff880472814b00 CPU: 2 COMMAND: "oraagent.bin" #0 [ffff880472a51b70] machine_kexec at ffffffff8103214b #1 [ffff880472a51bd0] crash_kexec at ffffffff810b91c2 #2 [ffff880472a51ca0] oops_end at ffffffff814f0b00 #3 [ffff880472a51cd0] die at ffffffff8100f26b #4 [ffff880472a51d00] do_trap at ffffffff814f03f4 #5 [ffff880472a51d60] do_divide_error at ffffffff8100cfff #6 [ffff880472a51e00] divide_error at ffffffff8100be7b [exception RIP: thread_group_times+0x56] RIP: ffffffff81056a16 RSP: ffff880472a51eb8 RFLAGS: 00010046 RAX: bc3572c9fe12d194 RBX: ffff880874150800 RCX: 0000000110266fad RDX: 0000000000000000 RSI: ffff880472a51eb8 RDI: 001038ae7d9633dc RBP: ffff880472a51ef8 R8: 00000000b10a3a64 R9: ffff880874150800 R10: 00007fcba27ab680 R11: 0000000000000202 R12: ffff880472a51f08 R13: ffff880472a51f10 R14: 0000000000000000 R15: 0000000000000007 ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 #7 [ffff880472a51f00] do_sys_times at ffffffff8108845d #8 [ffff880472a51f40] sys_times at ffffffff81088524 #9 [ffff880472a51f80] system_call_fastpath at ffffffff8100b0f2 RIP: 0000003808caac3a RSP: 00007fcba27ab6d8 RFLAGS: 00000202 RAX: 0000000000000064 RBX: ffffffff8100b0f2 RCX: 0000000000000000 RDX: 00007fcba27ab6e0 RSI: 000000000076d58e RDI: 00007fcba27ab6e0 RBP: 00007fcba27ab700 R8: 0000000000000020 R9: 000000000000091b R10: 00007fcba27ab680 R11: 0000000000000202 R12: 00007fff9ca41940 R13: 0000000000000000 R14: 00007fcba27ac9c0 R15: 00007fff9ca41940 ORIG_RAX: 0000000000000064 CS: 0033 SS: 002b Cc: stable@vger.kernel.org Signed-off-by: Stanislaw Gruszka Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20120808092714.GA3580@redhat.com Signed-off-by: Thomas Gleixner --- kernel/sched/core.c | 34 ++++++++++++++++++++-------------- 1 file changed, 20 insertions(+), 14 deletions(-) (limited to 'kernel') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 2cb4e7777998..e2c5841c7dfe 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -3142,6 +3142,20 @@ void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) # define nsecs_to_cputime(__nsecs) nsecs_to_jiffies(__nsecs) #endif +static cputime_t scale_utime(cputime_t utime, cputime_t rtime, cputime_t total) +{ + u64 temp = (__force u64) rtime; + + temp *= (__force u64) utime; + + if (sizeof(cputime_t) == 4) + temp = div_u64(temp, (__force u32) total); + else + temp = div64_u64(temp, (__force u64) total); + + return (__force cputime_t) temp; +} + void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) { cputime_t rtime, utime = p->utime, total = utime + p->stime; @@ -3151,13 +3165,9 @@ void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) */ rtime = nsecs_to_cputime(p->se.sum_exec_runtime); - if (total) { - u64 temp = (__force u64) rtime; - - temp *= (__force u64) utime; - do_div(temp, (__force u32) total); - utime = (__force cputime_t) temp; - } else + if (total) + utime = scale_utime(utime, rtime, total); + else utime = rtime; /* @@ -3184,13 +3194,9 @@ void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) total = cputime.utime + cputime.stime; rtime = nsecs_to_cputime(cputime.sum_exec_runtime); - if (total) { - u64 temp = (__force u64) rtime; - - temp *= (__force u64) cputime.utime; - do_div(temp, (__force u32) total); - utime = (__force cputime_t) temp; - } else + if (total) + utime = scale_utime(cputime.utime, rtime, total); + else utime = rtime; sig->prev_utime = max(sig->prev_utime, utime); -- cgit v1.2.3 From 35cf4e50b16331def6cfcbee11e49270b6db07f5 Mon Sep 17 00:00:00 2001 From: Mike Galbraith Date: Tue, 7 Aug 2012 05:00:13 +0200 Subject: sched,cgroup: Fix up task_groups list With multiple instances of task_groups, for_each_rt_rq() is a noop, no task groups having been added to the rt.c list instance. This renders __enable/disable_runtime() and print_rt_stats() noop, the user (non) visible effect being that rt task groups are missing in /proc/sched_debug. Signed-off-by: Mike Galbraith Cc: stable@kernel.org # v3.3+ Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1344308413.6846.7.camel@marge.simpson.net Signed-off-by: Thomas Gleixner --- kernel/sched/core.c | 1 + kernel/sched/sched.h | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index e2c5841c7dfe..6aa212f3c581 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7252,6 +7252,7 @@ int in_sched_functions(unsigned long addr) #ifdef CONFIG_CGROUP_SCHED struct task_group root_task_group; +LIST_HEAD(task_groups); #endif DECLARE_PER_CPU(cpumask_var_t, load_balance_tmpmask); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 531411b1044e..f6714d009e77 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -80,7 +80,7 @@ extern struct mutex sched_domains_mutex; struct cfs_rq; struct rt_rq; -static LIST_HEAD(task_groups); +extern struct list_head task_groups; struct cfs_bandwidth { #ifdef CONFIG_CFS_BANDWIDTH -- cgit v1.2.3 From e221d028bb08b47e624c5f0a31732c642db9d19a Mon Sep 17 00:00:00 2001 From: Mike Galbraith Date: Tue, 7 Aug 2012 10:02:38 +0200 Subject: sched,rt: fix isolated CPUs leaving root_task_group indefinitely throttled Root task group bandwidth replenishment must service all CPUs, regardless of where the timer was last started, and regardless of the isolation mechanism, lest 'Quoth the Raven, "Nevermore"' become rt scheduling policy. Signed-off-by: Mike Galbraith Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1344326558.6968.25.camel@marge.simpson.net Signed-off-by: Thomas Gleixner --- kernel/sched/rt.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'kernel') diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 573e1ca01102..944cb68420e9 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -788,6 +788,19 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) const struct cpumask *span; span = sched_rt_period_mask(); +#ifdef CONFIG_RT_GROUP_SCHED + /* + * FIXME: isolated CPUs should really leave the root task group, + * whether they are isolcpus or were isolated via cpusets, lest + * the timer run on a CPU which does not service all runqueues, + * potentially leaving other CPUs indefinitely throttled. If + * isolation is really required, the user will turn the throttle + * off to kill the perturbations it causes anyway. Meanwhile, + * this maintains functionality for boot and/or troubleshooting. + */ + if (rt_b == &root_task_group.rt_bandwidth) + span = cpu_online_mask; +#endif for_each_cpu(i, span) { int enqueue = 0; struct rt_rq *rt_rq = sched_rt_period_rt_rq(rt_b, i); -- cgit v1.2.3 From 8f6189684eb4e85e6c593cd710693f09c944450a Mon Sep 17 00:00:00 2001 From: Mike Galbraith Date: Sat, 4 Aug 2012 05:44:14 +0200 Subject: sched: Fix migration thread runtime bogosity Make stop scheduler class do the same accounting as other classes, Migration threads can be caught in the act while doing exec balancing, leading to the below due to use of unmaintained ->se.exec_start. The load that triggered this particular instance was an apparently out of control heavily threaded application that does system monitoring in what equated to an exec bomb, with one of the VERY frequently migrated tasks being ps. %CPU PID USER CMD 99.3 45 root [migration/10] 97.7 53 root [migration/12] 97.0 57 root [migration/13] 90.1 49 root [migration/11] 89.6 65 root [migration/15] 88.7 17 root [migration/3] 80.4 37 root [migration/8] 78.1 41 root [migration/9] 44.2 13 root [migration/2] Signed-off-by: Mike Galbraith Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1344051854.6739.19.camel@marge.simpson.net Signed-off-by: Thomas Gleixner --- kernel/sched/stop_task.c | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c index 7b386e86fd23..da5eb5bed84a 100644 --- a/kernel/sched/stop_task.c +++ b/kernel/sched/stop_task.c @@ -27,8 +27,10 @@ static struct task_struct *pick_next_task_stop(struct rq *rq) { struct task_struct *stop = rq->stop; - if (stop && stop->on_rq) + if (stop && stop->on_rq) { + stop->se.exec_start = rq->clock_task; return stop; + } return NULL; } @@ -52,6 +54,21 @@ static void yield_task_stop(struct rq *rq) static void put_prev_task_stop(struct rq *rq, struct task_struct *prev) { + struct task_struct *curr = rq->curr; + u64 delta_exec; + + delta_exec = rq->clock_task - curr->se.exec_start; + if (unlikely((s64)delta_exec < 0)) + delta_exec = 0; + + schedstat_set(curr->se.statistics.exec_max, + max(curr->se.statistics.exec_max, delta_exec)); + + curr->se.sum_exec_runtime += delta_exec; + account_group_exec_runtime(curr, delta_exec); + + curr->se.exec_start = rq->clock_task; + cpuacct_charge(curr, delta_exec); } static void task_tick_stop(struct rq *rq, struct task_struct *curr, int queued) @@ -60,6 +77,9 @@ static void task_tick_stop(struct rq *rq, struct task_struct *curr, int queued) static void set_curr_task_stop(struct rq *rq) { + struct task_struct *stop = rq->stop; + + stop->se.exec_start = rq->clock_task; } static void switched_to_stop(struct rq *rq, struct task_struct *p) -- cgit v1.2.3 From 0fe33aae0e94b4097dd433c9399e16e17d638cd8 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 15 Aug 2012 12:55:22 +0200 Subject: audit: don't free_chunk() after fsnotify_add_mark() Don't do free_chunk() after fsnotify_add_mark(). That one does a delayed unref via the destroy list and this results in use-after-free. Signed-off-by: Miklos Szeredi Acked-by: Eric Paris CC: stable@vger.kernel.org --- kernel/audit_tree.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 3a5ca582ba1e..69a58515f43f 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -259,7 +259,7 @@ static void untag_chunk(struct node *p) fsnotify_duplicate_mark(&new->mark, entry); if (fsnotify_add_mark(&new->mark, new->mark.group, new->mark.i.inode, NULL, 1)) { - free_chunk(new); + fsnotify_put_mark(&new->mark); goto Fallback; } @@ -322,7 +322,7 @@ static int create_chunk(struct inode *inode, struct audit_tree *tree) entry = &chunk->mark; if (fsnotify_add_mark(entry, audit_tree_group, inode, NULL, 0)) { - free_chunk(chunk); + fsnotify_put_mark(entry); return -ENOSPC; } @@ -396,7 +396,7 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) fsnotify_duplicate_mark(chunk_entry, old_entry); if (fsnotify_add_mark(chunk_entry, chunk_entry->group, chunk_entry->i.inode, NULL, 1)) { spin_unlock(&old_entry->lock); - free_chunk(chunk); + fsnotify_put_mark(chunk_entry); fsnotify_put_mark(old_entry); return -ENOSPC; } -- cgit v1.2.3 From a2140fc0cb0325bb6384e788edd27b9a568714e2 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 15 Aug 2012 12:55:22 +0200 Subject: audit: fix refcounting in audit-tree Refcounting of fsnotify_mark in audit tree is broken. E.g: refcount create_chunk alloc_chunk 1 fsnotify_add_mark 2 untag_chunk fsnotify_get_mark 3 fsnotify_destroy_mark audit_tree_freeing_mark 2 fsnotify_put_mark 1 fsnotify_put_mark 0 via destroy_list fsnotify_mark_destroy -1 This was reported by various people as triggering Oops when stopping auditd. We could just remove the put_mark from audit_tree_freeing_mark() but that would break freeing via inode destruction. So this patch simply omits a put_mark after calling destroy_mark or adds a get_mark before. The additional get_mark is necessary where there's no other put_mark after fsnotify_destroy_mark() since it assumes that the caller is holding a reference (or the inode is keeping the mark pinned, not the case here AFAICS). Signed-off-by: Miklos Szeredi Reported-by: Valentin Avram Reported-by: Peter Moody Acked-by: Eric Paris CC: stable@vger.kernel.org --- kernel/audit_tree.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 69a58515f43f..2b2ffff9eca2 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -250,7 +250,6 @@ static void untag_chunk(struct node *p) spin_unlock(&hash_lock); spin_unlock(&entry->lock); fsnotify_destroy_mark(entry); - fsnotify_put_mark(entry); goto out; } @@ -293,7 +292,6 @@ static void untag_chunk(struct node *p) spin_unlock(&hash_lock); spin_unlock(&entry->lock); fsnotify_destroy_mark(entry); - fsnotify_put_mark(entry); goto out; Fallback: @@ -332,6 +330,7 @@ static int create_chunk(struct inode *inode, struct audit_tree *tree) spin_unlock(&hash_lock); chunk->dead = 1; spin_unlock(&entry->lock); + fsnotify_get_mark(entry); fsnotify_destroy_mark(entry); fsnotify_put_mark(entry); return 0; @@ -412,6 +411,7 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) spin_unlock(&chunk_entry->lock); spin_unlock(&old_entry->lock); + fsnotify_get_mark(chunk_entry); fsnotify_destroy_mark(chunk_entry); fsnotify_put_mark(chunk_entry); @@ -445,7 +445,6 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) spin_unlock(&old_entry->lock); fsnotify_destroy_mark(old_entry); fsnotify_put_mark(old_entry); /* pair to fsnotify_find mark_entry */ - fsnotify_put_mark(old_entry); /* and kill it */ return 0; } -- cgit v1.2.3 From b3e8692b4dde5cf5fc60e4b95385229a72623182 Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Wed, 15 Aug 2012 12:55:22 +0200 Subject: audit: clean up refcounting in audit-tree Drop the initial reference by fsnotify_init_mark early instead of audit_tree_freeing_mark() at destroy time. In the cases we destroy the mark before we drop the initial reference we need to get rid of the get_mark that balances the put_mark in audit_tree_freeing_mark(). Signed-off-by: Miklos Szeredi --- kernel/audit_tree.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c index 2b2ffff9eca2..ed206fd88cca 100644 --- a/kernel/audit_tree.c +++ b/kernel/audit_tree.c @@ -292,6 +292,7 @@ static void untag_chunk(struct node *p) spin_unlock(&hash_lock); spin_unlock(&entry->lock); fsnotify_destroy_mark(entry); + fsnotify_put_mark(&new->mark); /* drop initial reference */ goto out; Fallback: @@ -330,7 +331,6 @@ static int create_chunk(struct inode *inode, struct audit_tree *tree) spin_unlock(&hash_lock); chunk->dead = 1; spin_unlock(&entry->lock); - fsnotify_get_mark(entry); fsnotify_destroy_mark(entry); fsnotify_put_mark(entry); return 0; @@ -346,6 +346,7 @@ static int create_chunk(struct inode *inode, struct audit_tree *tree) insert_hash(chunk); spin_unlock(&hash_lock); spin_unlock(&entry->lock); + fsnotify_put_mark(entry); /* drop initial reference */ return 0; } @@ -411,7 +412,6 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) spin_unlock(&chunk_entry->lock); spin_unlock(&old_entry->lock); - fsnotify_get_mark(chunk_entry); fsnotify_destroy_mark(chunk_entry); fsnotify_put_mark(chunk_entry); @@ -444,6 +444,7 @@ static int tag_chunk(struct inode *inode, struct audit_tree *tree) spin_unlock(&chunk_entry->lock); spin_unlock(&old_entry->lock); fsnotify_destroy_mark(old_entry); + fsnotify_put_mark(chunk_entry); /* drop initial reference */ fsnotify_put_mark(old_entry); /* pair to fsnotify_find mark_entry */ return 0; } @@ -915,7 +916,12 @@ static void audit_tree_freeing_mark(struct fsnotify_mark *entry, struct fsnotify struct audit_chunk *chunk = container_of(entry, struct audit_chunk, mark); evict_chunk(chunk); - fsnotify_put_mark(entry); + + /* + * We are guaranteed to have at least one reference to the mark from + * either the inode or the caller of fsnotify_destroy_mark(). + */ + BUG_ON(atomic_read(&entry->refcnt) < 1); } static bool audit_tree_send_event(struct fsnotify_group *group, struct inode *inode, -- cgit v1.2.3 From 4e8b14526ca7fb046a81c94002c1c43b6fdf0e9b Mon Sep 17 00:00:00 2001 From: John Stultz Date: Wed, 8 Aug 2012 15:36:20 -0400 Subject: time: Improve sanity checking of timekeeping inputs Unexpected behavior could occur if the time is set to a value large enough to overflow a 64bit ktime_t (which is something larger then the year 2262). Also unexpected behavior could occur if large negative offsets are injected via adjtimex. So this patch improves the sanity check timekeeping inputs by improving the timespec_valid() check, and then makes better use of timespec_valid() to make sure we don't set the time to an invalid negative value or one that overflows ktime_t. Note: This does not protect from setting the time close to overflowing ktime_t and then letting natural accumulation cause the overflow. Reported-by: CAI Qian Reported-by: Sasha Levin Signed-off-by: John Stultz Cc: Peter Zijlstra Cc: Prarit Bhargava Cc: Zhouping Liu Cc: Ingo Molnar Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1344454580-17031-1-git-send-email-john.stultz@linaro.org Signed-off-by: Thomas Gleixner --- include/linux/ktime.h | 7 ------- include/linux/time.h | 22 ++++++++++++++++++++-- kernel/time/timekeeping.c | 26 ++++++++++++++++++++++++-- 3 files changed, 44 insertions(+), 11 deletions(-) (limited to 'kernel') diff --git a/include/linux/ktime.h b/include/linux/ktime.h index 603bec2913b0..06177ba10a16 100644 --- a/include/linux/ktime.h +++ b/include/linux/ktime.h @@ -58,13 +58,6 @@ union ktime { typedef union ktime ktime_t; /* Kill this */ -#define KTIME_MAX ((s64)~((u64)1 << 63)) -#if (BITS_PER_LONG == 64) -# define KTIME_SEC_MAX (KTIME_MAX / NSEC_PER_SEC) -#else -# define KTIME_SEC_MAX LONG_MAX -#endif - /* * ktime_t definitions when using the 64-bit scalar representation: */ diff --git a/include/linux/time.h b/include/linux/time.h index c81c5e40fcb5..b0bbd8f0130d 100644 --- a/include/linux/time.h +++ b/include/linux/time.h @@ -107,11 +107,29 @@ static inline struct timespec timespec_sub(struct timespec lhs, return ts_delta; } +#define KTIME_MAX ((s64)~((u64)1 << 63)) +#if (BITS_PER_LONG == 64) +# define KTIME_SEC_MAX (KTIME_MAX / NSEC_PER_SEC) +#else +# define KTIME_SEC_MAX LONG_MAX +#endif + /* * Returns true if the timespec is norm, false if denorm: */ -#define timespec_valid(ts) \ - (((ts)->tv_sec >= 0) && (((unsigned long) (ts)->tv_nsec) < NSEC_PER_SEC)) +static inline bool timespec_valid(const struct timespec *ts) +{ + /* Dates before 1970 are bogus */ + if (ts->tv_sec < 0) + return false; + /* Can't have more nanoseconds then a second */ + if ((unsigned long)ts->tv_nsec >= NSEC_PER_SEC) + return false; + /* Disallow values that could overflow ktime_t */ + if ((unsigned long long)ts->tv_sec >= KTIME_SEC_MAX) + return false; + return true; +} extern void read_persistent_clock(struct timespec *ts); extern void read_boot_clock(struct timespec *ts); diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index e16af197a2bc..898bef066a44 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -427,7 +427,7 @@ int do_settimeofday(const struct timespec *tv) struct timespec ts_delta, xt; unsigned long flags; - if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC) + if (!timespec_valid(tv)) return -EINVAL; write_seqlock_irqsave(&tk->lock, flags); @@ -463,6 +463,8 @@ int timekeeping_inject_offset(struct timespec *ts) { struct timekeeper *tk = &timekeeper; unsigned long flags; + struct timespec tmp; + int ret = 0; if ((unsigned long)ts->tv_nsec >= NSEC_PER_SEC) return -EINVAL; @@ -471,10 +473,17 @@ int timekeeping_inject_offset(struct timespec *ts) timekeeping_forward_now(tk); + /* Make sure the proposed value is valid */ + tmp = timespec_add(tk_xtime(tk), *ts); + if (!timespec_valid(&tmp)) { + ret = -EINVAL; + goto error; + } tk_xtime_add(tk, ts); tk_set_wall_to_mono(tk, timespec_sub(tk->wall_to_monotonic, *ts)); +error: /* even if we error out, we forwarded the time, so call update */ timekeeping_update(tk, true); write_sequnlock_irqrestore(&tk->lock, flags); @@ -482,7 +491,7 @@ int timekeeping_inject_offset(struct timespec *ts) /* signal hrtimers about time change */ clock_was_set(); - return 0; + return ret; } EXPORT_SYMBOL(timekeeping_inject_offset); @@ -649,7 +658,20 @@ void __init timekeeping_init(void) struct timespec now, boot, tmp; read_persistent_clock(&now); + if (!timespec_valid(&now)) { + pr_warn("WARNING: Persistent clock returned invalid value!\n" + " Check your CMOS/BIOS settings.\n"); + now.tv_sec = 0; + now.tv_nsec = 0; + } + read_boot_clock(&boot); + if (!timespec_valid(&boot)) { + pr_warn("WARNING: Boot clock returned invalid value!\n" + " Check your CMOS/BIOS settings.\n"); + boot.tv_sec = 0; + boot.tv_nsec = 0; + } seqlock_init(&tk->lock); -- cgit v1.2.3 From 60916a9382e88fbf5e54fd36a3e658efd7ab7bed Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 16 Aug 2012 18:14:14 +0100 Subject: tracing/syscalls: Fix perf syscall tracing when syscall_nr == -1 syscall_get_nr can return -1 in the case that the task is not executing a system call. This patch fixes perf_syscall_{enter,exit} to check that the syscall number is valid before using it as an index into a bitmap. Link: http://lkml.kernel.org/r/1345137254-7377-1-git-send-email-will.deacon@arm.com Cc: Jason Baron Cc: Wade Farnsworth Cc: Frederic Weisbecker Signed-off-by: Will Deacon Signed-off-by: Steven Rostedt --- kernel/trace/trace_syscalls.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel') diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 60e4d7875672..6b245f64c8dd 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -506,6 +506,8 @@ static void perf_syscall_enter(void *ignore, struct pt_regs *regs, long id) int size; syscall_nr = syscall_get_nr(current, regs); + if (syscall_nr < 0) + return; if (!test_bit(syscall_nr, enabled_perf_enter_syscalls)) return; @@ -580,6 +582,8 @@ static void perf_syscall_exit(void *ignore, struct pt_regs *regs, long ret) int size; syscall_nr = syscall_get_nr(current, regs); + if (syscall_nr < 0) + return; if (!test_bit(syscall_nr, enabled_perf_exit_syscalls)) return; -- cgit v1.2.3 From be53db6e4edd9dc013b21a929ad2b142dea8b9c0 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 19 Aug 2012 14:40:59 +1200 Subject: alpha: take a bunch of syscalls into osf_sys.c New helper: current_thread_info(). Allows to do a bunch of odd syscalls in C. While we are at it, there had never been a reason to do osf_getpriority() in assembler. We also get "namespace"-aware (read: consistent with getuid(2), etc.) behaviour from getx?id() syscalls now. Signed-off-by: Al Viro Signed-off-by: Michael Cree Acked-by: Matt Turner Signed-off-by: Linus Torvalds --- arch/alpha/include/asm/ptrace.h | 5 +- arch/alpha/kernel/entry.S | 109 ---------------------------------------- arch/alpha/kernel/osf_sys.c | 49 ++++++++++++++++++ arch/alpha/kernel/systbls.S | 2 +- kernel/timer.c | 9 ---- 5 files changed, 54 insertions(+), 120 deletions(-) (limited to 'kernel') diff --git a/arch/alpha/include/asm/ptrace.h b/arch/alpha/include/asm/ptrace.h index fd698a174f26..b87755a19554 100644 --- a/arch/alpha/include/asm/ptrace.h +++ b/arch/alpha/include/asm/ptrace.h @@ -76,7 +76,10 @@ struct switch_stack { #define task_pt_regs(task) \ ((struct pt_regs *) (task_stack_page(task) + 2*PAGE_SIZE) - 1) -#define force_successful_syscall_return() (task_pt_regs(current)->r0 = 0) +#define current_pt_regs() \ + ((struct pt_regs *) ((char *)current_thread_info() + 2*PAGE_SIZE) - 1) + +#define force_successful_syscall_return() (current_pt_regs()->r0 = 0) #endif diff --git a/arch/alpha/kernel/entry.S b/arch/alpha/kernel/entry.S index 6d159cee5f2f..22b0c4d414a7 100644 --- a/arch/alpha/kernel/entry.S +++ b/arch/alpha/kernel/entry.S @@ -796,115 +796,6 @@ sys_rt_sigreturn: br ret_from_sys_call .end sys_rt_sigreturn - .align 4 - .globl sys_sethae - .ent sys_sethae -sys_sethae: - .prologue 0 - stq $16, 152($sp) - ret -.end sys_sethae - - .align 4 - .globl osf_getpriority - .ent osf_getpriority -osf_getpriority: - lda $sp, -16($sp) - stq $26, 0($sp) - .prologue 0 - - jsr $26, sys_getpriority - - ldq $26, 0($sp) - blt $0, 1f - - /* Return value is the unbiased priority, i.e. 20 - prio. - This does result in negative return values, so signal - no error by writing into the R0 slot. */ - lda $1, 20 - stq $31, 16($sp) - subl $1, $0, $0 - unop - -1: lda $sp, 16($sp) - ret -.end osf_getpriority - - .align 4 - .globl sys_getxuid - .ent sys_getxuid -sys_getxuid: - .prologue 0 - ldq $2, TI_TASK($8) - ldq $3, TASK_CRED($2) - ldl $0, CRED_UID($3) - ldl $1, CRED_EUID($3) - stq $1, 80($sp) - ret -.end sys_getxuid - - .align 4 - .globl sys_getxgid - .ent sys_getxgid -sys_getxgid: - .prologue 0 - ldq $2, TI_TASK($8) - ldq $3, TASK_CRED($2) - ldl $0, CRED_GID($3) - ldl $1, CRED_EGID($3) - stq $1, 80($sp) - ret -.end sys_getxgid - - .align 4 - .globl sys_getxpid - .ent sys_getxpid -sys_getxpid: - .prologue 0 - ldq $2, TI_TASK($8) - - /* See linux/kernel/timer.c sys_getppid for discussion - about this loop. */ - ldq $3, TASK_GROUP_LEADER($2) - ldq $4, TASK_REAL_PARENT($3) - ldl $0, TASK_TGID($2) -1: ldl $1, TASK_TGID($4) -#ifdef CONFIG_SMP - mov $4, $5 - mb - ldq $3, TASK_GROUP_LEADER($2) - ldq $4, TASK_REAL_PARENT($3) - cmpeq $4, $5, $5 - beq $5, 1b -#endif - stq $1, 80($sp) - ret -.end sys_getxpid - - .align 4 - .globl sys_alpha_pipe - .ent sys_alpha_pipe -sys_alpha_pipe: - lda $sp, -16($sp) - stq $26, 0($sp) - .prologue 0 - - mov $31, $17 - lda $16, 8($sp) - jsr $26, do_pipe_flags - - ldq $26, 0($sp) - bne $0, 1f - - /* The return values are in $0 and $20. */ - ldl $1, 12($sp) - ldl $0, 8($sp) - - stq $1, 80+16($sp) -1: lda $sp, 16($sp) - ret -.end sys_alpha_pipe - .align 4 .globl sys_execve .ent sys_execve diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index 98a103621af6..bc1acdda7a5e 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -1404,3 +1404,52 @@ SYSCALL_DEFINE3(osf_writev, unsigned long, fd, } #endif + +SYSCALL_DEFINE2(osf_getpriority, int, which, int, who) +{ + int prio = sys_getpriority(which, who); + if (prio >= 0) { + /* Return value is the unbiased priority, i.e. 20 - prio. + This does result in negative return values, so signal + no error */ + force_successful_syscall_return(); + prio = 20 - prio; + } + return prio; +} + +SYSCALL_DEFINE0(getxuid) +{ + current_pt_regs()->r20 = sys_geteuid(); + return sys_getuid(); +} + +SYSCALL_DEFINE0(getxgid) +{ + current_pt_regs()->r20 = sys_getegid(); + return sys_getgid(); +} + +SYSCALL_DEFINE0(getxpid) +{ + current_pt_regs()->r20 = sys_getppid(); + return sys_getpid(); +} + +SYSCALL_DEFINE0(alpha_pipe) +{ + int fd[2]; + int res = do_pipe_flags(fd, 0); + if (!res) { + /* The return values are in $0 and $20. */ + current_pt_regs()->r20 = fd[1]; + res = fd[0]; + } + return res; +} + +SYSCALL_DEFINE1(sethae, unsigned long, val) +{ + current_pt_regs()->hae = val; + return 0; +} diff --git a/arch/alpha/kernel/systbls.S b/arch/alpha/kernel/systbls.S index b64157c98e02..2ac6b45c3e00 100644 --- a/arch/alpha/kernel/systbls.S +++ b/arch/alpha/kernel/systbls.S @@ -111,7 +111,7 @@ sys_call_table: .quad sys_socket .quad sys_connect .quad sys_accept - .quad osf_getpriority /* 100 */ + .quad sys_osf_getpriority /* 100 */ .quad sys_send .quad sys_recv .quad sys_sigreturn diff --git a/kernel/timer.c b/kernel/timer.c index a61c09374eba..8c5e7b908c68 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -1407,13 +1407,6 @@ SYSCALL_DEFINE1(alarm, unsigned int, seconds) #endif -#ifndef __alpha__ - -/* - * The Alpha uses getxpid, getxuid, and getxgid instead. Maybe this - * should be moved into arch/i386 instead? - */ - /** * sys_getpid - return the thread group id of the current process * @@ -1469,8 +1462,6 @@ SYSCALL_DEFINE0(getegid) return from_kgid_munged(current_user_ns(), current_egid()); } -#endif - static void process_timeout(unsigned long __data) { wake_up_process((struct task_struct *)__data); -- cgit v1.2.3 From c7a3a88c938fbe3d70c2278e082b80eb830d1c58 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Sun, 19 Aug 2012 19:10:42 +0200 Subject: uprobes: Fix mmap_region()'s mm->mm_rb corruption if uprobe_mmap() fails This patch fixes: https://bugzilla.redhat.com/show_bug.cgi?id=843640 If mmap_region()->uprobe_mmap() fails, unmap_and_free_vma path does unmap_region() but does not remove the soon-to-be-freed vma from rb tree. Actually there are more problems but this is how William noticed this bug. Perhaps we could do do_munmap() + return in this case, but in fact it is simply wrong to abort if uprobe_mmap() fails. Until at least we move the !UPROBE_COPY_INSN code from install_breakpoint() to uprobe_register(). For example, uprobe_mmap()->install_breakpoint() can fail if the probed insn is not supported (remember, uprobe_register() succeeds if nobody mmaps inode/offset), mmap() should not fail in this case. dup_mmap()->uprobe_mmap() is wrong too by the same reason, fork() can race with uprobe_register() and fail for no reason if it wins the race and does install_breakpoint() first. And, if nothing else, both mmap_region() and dup_mmap() return success if uprobe_mmap() fails. Change them to ignore the error code from uprobe_mmap(). Reported-and-tested-by: William Cohen Signed-off-by: Oleg Nesterov Acked-by: Srikar Dronamraju Cc: # v3.5 Cc: Anton Arapov Cc: William Cohen Cc: Linus Torvalds Link: http://lkml.kernel.org/r/20120819171042.GB26957@redhat.com Signed-off-by: Ingo Molnar --- kernel/fork.c | 4 ++-- mm/mmap.c | 5 ++--- 2 files changed, 4 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index 3bd2280d79f6..2c8857e12855 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -455,8 +455,8 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) if (retval) goto out; - if (file && uprobe_mmap(tmp)) - goto out; + if (file) + uprobe_mmap(tmp); } /* a new mm has just been created */ arch_dup_mmap(oldmm, mm); diff --git a/mm/mmap.c b/mm/mmap.c index e3e86914f11a..52e08fc9186d 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1356,9 +1356,8 @@ out: } else if ((flags & MAP_POPULATE) && !(flags & MAP_NONBLOCK)) make_pages_present(addr, addr + len); - if (file && uprobe_mmap(vma)) - /* matching probes but cannot insert */ - goto unmap_and_free_vma; + if (file) + uprobe_mmap(vma); return addr; -- cgit v1.2.3 From f341861fb0b701139849f8a85c2d3cdff466e8e8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 21 Aug 2012 15:05:14 +0200 Subject: task_work: add a scheduling point in task_work_run() It seems commit 4a9d4b024a31 ("switch fput to task_work_add") re- introduced the problem addressed in 944be0b22472 ("close_files(): add scheduling point") If a server process with a lot of files (say 2 million tcp sockets) is killed, we can spend a lot of time in task_work_run() and trigger a soft lockup. Signed-off-by: Eric Dumazet Signed-off-by: Linus Torvalds --- kernel/task_work.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/task_work.c b/kernel/task_work.c index 91d4e1742a0c..d320d44903bd 100644 --- a/kernel/task_work.c +++ b/kernel/task_work.c @@ -75,6 +75,7 @@ void task_work_run(void) p = q->next; q->func(q); q = p; + cond_resched(); } } } -- cgit v1.2.3 From 784ffcbb96c3a97b4c64fd48b1dfe12ef3fcbcda Mon Sep 17 00:00:00 2001 From: John Stultz Date: Tue, 21 Aug 2012 20:30:46 -0400 Subject: time: Ensure we normalize the timekeeper in tk_xtime_add Andreas noticed problems with resume on specific hardware after commit 1e75fa8b (time: Condense timekeeper.xtime into xtime_sec) combined with commit b44d50dca (time: Fix casting issue in tk_set_xtime and tk_xtime_add) After some digging I realized we aren't normalizing the timekeeper after the add. Add the missing normalize call. Reported-by: Andreas Schwab Tested-by: Andreas Schwab Signed-off-by: John Stultz Cc: Prarit Bhargava Cc: Ingo Molnar Link: http://lkml.kernel.org/r/1345595449-34965-2-git-send-email-john.stultz@linaro.org Signed-off-by: Thomas Gleixner --- kernel/time/timekeeping.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 898bef066a44..258164ae45cc 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -115,6 +115,7 @@ static void tk_xtime_add(struct timekeeper *tk, const struct timespec *ts) { tk->xtime_sec += ts->tv_sec; tk->xtime_nsec += (u64)ts->tv_nsec << tk->shift; + tk_normalize_xtime(tk); } static void tk_set_wall_to_mono(struct timekeeper *tk, struct timespec wtm) -- cgit v1.2.3 From 85dc8f05c93c8105987de9d7e7cebf15a72ff4ec Mon Sep 17 00:00:00 2001 From: Andreas Schwab Date: Tue, 21 Aug 2012 20:30:47 -0400 Subject: time: Fix casting issue in timekeeping_forward_now arch_gettimeoffset returns a u32 value which when shifted by tk->shift can overflow. This issue was introduced with 1e75fa8be (time: Condense timekeeper.xtime into xtime_sec) Cast it to u64 first. Signed-off-by: Andreas Schwab Signed-off-by: John Stultz Cc: Prarit Bhargava Cc: Ingo Molnar Link: http://lkml.kernel.org/r/1345595449-34965-3-git-send-email-john.stultz@linaro.org Signed-off-by: Thomas Gleixner --- kernel/time/timekeeping.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 258164ae45cc..1dbf80ec7696 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -277,7 +277,7 @@ static void timekeeping_forward_now(struct timekeeper *tk) tk->xtime_nsec += cycle_delta * tk->mult; /* If arch requires, add in gettimeoffset() */ - tk->xtime_nsec += arch_gettimeoffset() << tk->shift; + tk->xtime_nsec += (u64)arch_gettimeoffset() << tk->shift; tk_normalize_xtime(tk); -- cgit v1.2.3 From 6ea565a9be32a3c8d1092017686f183b6d8c4514 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Tue, 21 Aug 2012 20:30:48 -0400 Subject: time: Avoid potential shift overflow with large shift values Andreas Schwab noticed that the 1 << tk->shift could overflow if the shift value was greater than 30, since 1 would be a 32bit long on 32bit architectures. This issue was introduced by 1e75fa8be (time: Condense timekeeper.xtime into xtime_sec) Use 1ULL instead to ensure we don't overflow on the shift. Reported-by: Andreas Schwab Signed-off-by: John Stultz Cc: Prarit Bhargava Cc: Ingo Molnar Link: http://lkml.kernel.org/r/1345595449-34965-4-git-send-email-john.stultz@linaro.org Signed-off-by: Thomas Gleixner --- kernel/time/timekeeping.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 1dbf80ec7696..a5a9389c4c30 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -1184,9 +1184,9 @@ static void update_wall_time(void) * the vsyscall implementations are converted to use xtime_nsec * (shifted nanoseconds), this can be killed. */ - remainder = tk->xtime_nsec & ((1 << tk->shift) - 1); + remainder = tk->xtime_nsec & ((1ULL << tk->shift) - 1); tk->xtime_nsec -= remainder; - tk->xtime_nsec += 1 << tk->shift; + tk->xtime_nsec += 1ULL << tk->shift; tk->ntp_error += remainder << tk->ntp_error_shift; /* -- cgit v1.2.3 From bf2ac312195155511a0f79325515cbb61929898a Mon Sep 17 00:00:00 2001 From: John Stultz Date: Tue, 21 Aug 2012 20:30:49 -0400 Subject: time: Avoid making adjustments if we haven't accumulated anything If update_wall_time() is called and the current offset isn't large enough to accumulate, avoid re-calling timekeeping_adjust which may change the clock freq and can cause 1ns inconsistencies with CLOCK_REALTIME_COARSE/CLOCK_MONOTONIC_COARSE. Signed-off-by: John Stultz Cc: Prarit Bhargava Cc: Ingo Molnar Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/1345595449-34965-5-git-send-email-john.stultz@linaro.org Signed-off-by: Thomas Gleixner --- kernel/time/timekeeping.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index a5a9389c4c30..0c1485e42be6 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -1152,6 +1152,10 @@ static void update_wall_time(void) offset = (clock->read(clock) - clock->cycle_last) & clock->mask; #endif + /* Check if there's really nothing to do */ + if (offset < tk->cycle_interval) + goto out; + /* * With NO_HZ we may have to accumulate many cycle_intervals * (think "ticks") worth of time at once. To do this efficiently, -- cgit v1.2.3 From cee58483cf56e0ba355fdd97ff5e8925329aa936 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Fri, 31 Aug 2012 13:30:06 -0400 Subject: time: Move ktime_t overflow checking into timespec_valid_strict Andreas Bombe reported that the added ktime_t overflow checking added to timespec_valid in commit 4e8b14526ca7 ("time: Improve sanity checking of timekeeping inputs") was causing problems with X.org because it caused timeouts larger then KTIME_T to be invalid. Previously, these large timeouts would be clamped to KTIME_MAX and would never expire, which is valid. This patch splits the ktime_t overflow checking into a new timespec_valid_strict function, and converts the timekeeping codes internal checking to use this more strict function. Reported-and-tested-by: Andreas Bombe Cc: Zhouping Liu Cc: Ingo Molnar Cc: Prarit Bhargava Cc: Thomas Gleixner Cc: stable@vger.kernel.org Signed-off-by: John Stultz Signed-off-by: Linus Torvalds --- include/linux/time.h | 7 +++++++ kernel/time/timekeeping.c | 10 +++++----- 2 files changed, 12 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/include/linux/time.h b/include/linux/time.h index b0bbd8f0130d..b51e664c83e7 100644 --- a/include/linux/time.h +++ b/include/linux/time.h @@ -125,6 +125,13 @@ static inline bool timespec_valid(const struct timespec *ts) /* Can't have more nanoseconds then a second */ if ((unsigned long)ts->tv_nsec >= NSEC_PER_SEC) return false; + return true; +} + +static inline bool timespec_valid_strict(const struct timespec *ts) +{ + if (!timespec_valid(ts)) + return false; /* Disallow values that could overflow ktime_t */ if ((unsigned long long)ts->tv_sec >= KTIME_SEC_MAX) return false; diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 0c1485e42be6..34e5eac81424 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -428,7 +428,7 @@ int do_settimeofday(const struct timespec *tv) struct timespec ts_delta, xt; unsigned long flags; - if (!timespec_valid(tv)) + if (!timespec_valid_strict(tv)) return -EINVAL; write_seqlock_irqsave(&tk->lock, flags); @@ -476,7 +476,7 @@ int timekeeping_inject_offset(struct timespec *ts) /* Make sure the proposed value is valid */ tmp = timespec_add(tk_xtime(tk), *ts); - if (!timespec_valid(&tmp)) { + if (!timespec_valid_strict(&tmp)) { ret = -EINVAL; goto error; } @@ -659,7 +659,7 @@ void __init timekeeping_init(void) struct timespec now, boot, tmp; read_persistent_clock(&now); - if (!timespec_valid(&now)) { + if (!timespec_valid_strict(&now)) { pr_warn("WARNING: Persistent clock returned invalid value!\n" " Check your CMOS/BIOS settings.\n"); now.tv_sec = 0; @@ -667,7 +667,7 @@ void __init timekeeping_init(void) } read_boot_clock(&boot); - if (!timespec_valid(&boot)) { + if (!timespec_valid_strict(&boot)) { pr_warn("WARNING: Boot clock returned invalid value!\n" " Check your CMOS/BIOS settings.\n"); boot.tv_sec = 0; @@ -713,7 +713,7 @@ static struct timespec timekeeping_suspend_time; static void __timekeeping_inject_sleeptime(struct timekeeper *tk, struct timespec *delta) { - if (!timespec_valid(delta)) { + if (!timespec_valid_strict(delta)) { printk(KERN_WARNING "__timekeeping_inject_sleeptime: Invalid " "sleep delta value!\n"); return; -- cgit v1.2.3 From 96e65306b81351b656835c15931d1d237b252f27 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Sun, 2 Sep 2012 00:28:19 +0800 Subject: workqueue: UNBOUND -> REBIND morphing in rebind_workers() should be atomic The compiler may compile the following code into TWO write/modify instructions. worker->flags &= ~WORKER_UNBOUND; worker->flags |= WORKER_REBIND; so the other CPU may temporarily see worker->flags which doesn't have either WORKER_UNBOUND or WORKER_REBIND set and perform local wakeup prematurely. Fix it by using single explicit assignment via ACCESS_ONCE(). Because idle workers have another WORKER_NOT_RUNNING flag, this bug doesn't exist for them; however, update it to use the same pattern for consistency. tj: Applied the change to idle workers too and updated comments and patch description a bit. Signed-off-by: Lai Jiangshan Signed-off-by: Tejun Heo Cc: stable@vger.kernel.org --- kernel/workqueue.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 692d97628a10..c462cd60c374 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1396,12 +1396,15 @@ retry: /* set REBIND and kick idle ones, we'll wait for these later */ for_each_worker_pool(pool, gcwq) { list_for_each_entry(worker, &pool->idle_list, entry) { + unsigned long worker_flags = worker->flags; + if (worker->flags & WORKER_REBIND) continue; - /* morph UNBOUND to REBIND */ - worker->flags &= ~WORKER_UNBOUND; - worker->flags |= WORKER_REBIND; + /* morph UNBOUND to REBIND atomically */ + worker_flags &= ~WORKER_UNBOUND; + worker_flags |= WORKER_REBIND; + ACCESS_ONCE(worker->flags) = worker_flags; idle_rebind.cnt++; worker->idle_rebind = &idle_rebind; @@ -1434,10 +1437,12 @@ retry: /* rebind busy workers */ for_each_busy_worker(worker, i, pos, gcwq) { struct work_struct *rebind_work = &worker->rebind_work; + unsigned long worker_flags = worker->flags; - /* morph UNBOUND to REBIND */ - worker->flags &= ~WORKER_UNBOUND; - worker->flags |= WORKER_REBIND; + /* morph UNBOUND to REBIND atomically */ + worker_flags &= ~WORKER_UNBOUND; + worker_flags |= WORKER_REBIND; + ACCESS_ONCE(worker->flags) = worker_flags; if (test_and_set_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(rebind_work))) -- cgit v1.2.3 From 90beca5de591e12482a812f23a7f10690962ed4a Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 4 Sep 2012 23:12:33 -0700 Subject: workqueue: move WORKER_REBIND clearing in rebind_workers() to the end of the function This doesn't make any functional difference and is purely to help the next patch to be simpler. Signed-off-by: Tejun Heo Cc: Lai Jiangshan --- kernel/workqueue.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index c462cd60c374..d79a18d0c42e 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1422,19 +1422,7 @@ retry: goto retry; } - /* - * All idle workers are rebound and waiting for %WORKER_REBIND to - * be cleared inside idle_worker_rebind(). Clear and release. - * Clearing %WORKER_REBIND from this foreign context is safe - * because these workers are still guaranteed to be idle. - */ - for_each_worker_pool(pool, gcwq) - list_for_each_entry(worker, &pool->idle_list, entry) - worker->flags &= ~WORKER_REBIND; - - wake_up_all(&gcwq->rebind_hold); - - /* rebind busy workers */ + /* all idle workers are rebound, rebind busy workers */ for_each_busy_worker(worker, i, pos, gcwq) { struct work_struct *rebind_work = &worker->rebind_work; unsigned long worker_flags = worker->flags; @@ -1454,6 +1442,18 @@ retry: worker->scheduled.next, work_color_to_flags(WORK_NO_COLOR)); } + + /* + * All idle workers are rebound and waiting for %WORKER_REBIND to + * be cleared inside idle_worker_rebind(). Clear and release. + * Clearing %WORKER_REBIND from this foreign context is safe + * because these workers are still guaranteed to be idle. + */ + for_each_worker_pool(pool, gcwq) + list_for_each_entry(worker, &pool->idle_list, entry) + worker->flags &= ~WORKER_REBIND; + + wake_up_all(&gcwq->rebind_hold); } static struct worker *alloc_worker(void) -- cgit v1.2.3 From ec58815ab0409a921a7c9744eb4ca44866b14d71 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 4 Sep 2012 23:16:32 -0700 Subject: workqueue: fix possible deadlock in idle worker rebinding Currently, rebind_workers() and idle_worker_rebind() are two-way interlocked. rebind_workers() waits for idle workers to finish rebinding and rebound idle workers wait for rebind_workers() to finish rebinding busy workers before proceeding. Unfortunately, this isn't enough. The second wait from idle workers is implemented as follows. wait_event(gcwq->rebind_hold, !(worker->flags & WORKER_REBIND)); rebind_workers() clears WORKER_REBIND, wakes up the idle workers and then returns. If CPU hotplug cycle happens again before one of the idle workers finishes the above wait_event(), rebind_workers() will repeat the first part of the handshake - set WORKER_REBIND again and wait for the idle worker to finish rebinding - and this leads to deadlock because the idle worker would be waiting for WORKER_REBIND to clear. This is fixed by adding another interlocking step at the end - rebind_workers() now waits for all the idle workers to finish the above WORKER_REBIND wait before returning. This ensures that all rebinding steps are complete on all idle workers before the next hotplug cycle can happen. This problem was diagnosed by Lai Jiangshan who also posted a patch to fix the issue, upon which this patch is based. This is the minimal fix and further patches are scheduled for the next merge window to simplify the CPU hotplug path. Signed-off-by: Tejun Heo Original-patch-by: Lai Jiangshan LKML-Reference: <1346516916-1991-3-git-send-email-laijs@cn.fujitsu.com> --- kernel/workqueue.c | 29 +++++++++++++++++++++++++++-- 1 file changed, 27 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index d79a18d0c42e..dc7b8458e275 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1326,6 +1326,15 @@ static void idle_worker_rebind(struct worker *worker) /* we did our part, wait for rebind_workers() to finish up */ wait_event(gcwq->rebind_hold, !(worker->flags & WORKER_REBIND)); + + /* + * rebind_workers() shouldn't finish until all workers passed the + * above WORKER_REBIND wait. Tell it when done. + */ + spin_lock_irq(&worker->pool->gcwq->lock); + if (!--worker->idle_rebind->cnt) + complete(&worker->idle_rebind->done); + spin_unlock_irq(&worker->pool->gcwq->lock); } /* @@ -1448,12 +1457,28 @@ retry: * be cleared inside idle_worker_rebind(). Clear and release. * Clearing %WORKER_REBIND from this foreign context is safe * because these workers are still guaranteed to be idle. + * + * We need to make sure all idle workers passed WORKER_REBIND wait + * in idle_worker_rebind() before returning; otherwise, workers can + * get stuck at the wait if hotplug cycle repeats. */ - for_each_worker_pool(pool, gcwq) - list_for_each_entry(worker, &pool->idle_list, entry) + idle_rebind.cnt = 1; + INIT_COMPLETION(idle_rebind.done); + + for_each_worker_pool(pool, gcwq) { + list_for_each_entry(worker, &pool->idle_list, entry) { worker->flags &= ~WORKER_REBIND; + idle_rebind.cnt++; + } + } wake_up_all(&gcwq->rebind_hold); + + if (--idle_rebind.cnt) { + spin_unlock_irq(&gcwq->lock); + wait_for_completion(&idle_rebind.done); + spin_lock_irq(&gcwq->lock); + } } static struct worker *alloc_worker(void) -- cgit v1.2.3 From 552a37e9360a293cd20e7f8ff1fb326a244c5f1e Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Mon, 10 Sep 2012 10:03:33 -0700 Subject: workqueue: restore POOL_MANAGING_WORKERS This patch restores POOL_MANAGING_WORKERS which was replaced by pool->manager_mutex by 6037315269 "workqueue: use mutex for global_cwq manager exclusion". There's a subtle idle worker depletion bug across CPU hotplug events and we need to distinguish an actual manager and CPU hotplug preventing management. POOL_MANAGING_WORKERS will be used for the former and manager_mutex the later. This patch just lays POOL_MANAGING_WORKERS on top of the existing manager_mutex and doesn't introduce any synchronization changes. The next patch will update it. Note that this patch fixes a non-critical anomaly where too_many_workers() may return %true spuriously while CPU hotplug is in progress. While the issue could schedule idle timer spuriously, it didn't trigger any actual misbehavior. tj: Rewrote patch description. Signed-off-by: Lai Jiangshan Signed-off-by: Tejun Heo --- kernel/workqueue.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index dc7b8458e275..383548ed0b54 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -66,6 +66,7 @@ enum { /* pool flags */ POOL_MANAGE_WORKERS = 1 << 0, /* need to manage workers */ + POOL_MANAGING_WORKERS = 1 << 1, /* managing workers */ /* worker flags */ WORKER_STARTED = 1 << 0, /* started */ @@ -652,7 +653,7 @@ static bool need_to_manage_workers(struct worker_pool *pool) /* Do we have too many workers and should some go away? */ static bool too_many_workers(struct worker_pool *pool) { - bool managing = mutex_is_locked(&pool->manager_mutex); + bool managing = pool->flags & POOL_MANAGING_WORKERS; int nr_idle = pool->nr_idle + managing; /* manager is considered idle */ int nr_busy = pool->nr_workers - nr_idle; @@ -1827,6 +1828,7 @@ static bool manage_workers(struct worker *worker) if (!mutex_trylock(&pool->manager_mutex)) return ret; + pool->flags |= POOL_MANAGING_WORKERS; pool->flags &= ~POOL_MANAGE_WORKERS; /* @@ -1836,6 +1838,7 @@ static bool manage_workers(struct worker *worker) ret |= maybe_destroy_workers(pool); ret |= maybe_create_worker(pool); + pool->flags &= ~POOL_MANAGING_WORKERS; mutex_unlock(&pool->manager_mutex); return ret; } -- cgit v1.2.3 From ee378aa49b594da9bda6a2c768cc5b2ad585f911 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Mon, 10 Sep 2012 10:03:44 -0700 Subject: workqueue: fix possible idle worker depletion across CPU hotplug To simplify both normal and CPU hotplug paths, worker management is prevented while CPU hoplug is in progress. This is achieved by CPU hotplug holding the same exclusion mechanism used by workers to ensure there's only one manager per pool. If someone else seems to be performing the manager role, workers proceed to execute work items. CPU hotplug using the same mechanism can lead to idle worker depletion because all workers could proceed to execute work items while CPU hotplug is in progress and CPU hotplug itself wouldn't actually perform the worker management duty - it doesn't guarantee that there's an idle worker left when it releases management. This idle worker depletion, under extreme circumstances, can break forward-progress guarantee and thus lead to deadlock. This patch fixes the bug by using separate mechanisms for manager exclusion among workers and hotplug exclusion. For manager exclusion, POOL_MANAGING_WORKERS which was restored by the previous patch is used. pool->manager_mutex is now only used for exclusion between the elected manager and CPU hotplug. The elected manager won't proceed without holding pool->manager_mutex. This ensures that the worker which won the manager position can't skip managing while CPU hotplug is in progress. It will block on manager_mutex and perform management after CPU hotplug is complete. Note that hotplug may happen while waiting for manager_mutex. A manager isn't either on idle or busy list and thus the hoplug code can't unbind/rebind it. Make the manager handle its own un/rebinding. tj: Updated comment and description. Signed-off-by: Lai Jiangshan Signed-off-by: Tejun Heo --- kernel/workqueue.c | 37 ++++++++++++++++++++++++++++++++++++- 1 file changed, 36 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 383548ed0b54..1e1373bcb3e3 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1825,10 +1825,45 @@ static bool manage_workers(struct worker *worker) struct worker_pool *pool = worker->pool; bool ret = false; - if (!mutex_trylock(&pool->manager_mutex)) + if (pool->flags & POOL_MANAGING_WORKERS) return ret; pool->flags |= POOL_MANAGING_WORKERS; + + /* + * To simplify both worker management and CPU hotplug, hold off + * management while hotplug is in progress. CPU hotplug path can't + * grab %POOL_MANAGING_WORKERS to achieve this because that can + * lead to idle worker depletion (all become busy thinking someone + * else is managing) which in turn can result in deadlock under + * extreme circumstances. Use @pool->manager_mutex to synchronize + * manager against CPU hotplug. + * + * manager_mutex would always be free unless CPU hotplug is in + * progress. trylock first without dropping @gcwq->lock. + */ + if (unlikely(!mutex_trylock(&pool->manager_mutex))) { + spin_unlock_irq(&pool->gcwq->lock); + mutex_lock(&pool->manager_mutex); + /* + * CPU hotplug could have happened while we were waiting + * for manager_mutex. Hotplug itself can't handle us + * because manager isn't either on idle or busy list, and + * @gcwq's state and ours could have deviated. + * + * As hotplug is now excluded via manager_mutex, we can + * simply try to bind. It will succeed or fail depending + * on @gcwq's current state. Try it and adjust + * %WORKER_UNBOUND accordingly. + */ + if (worker_maybe_bind_and_lock(worker)) + worker->flags &= ~WORKER_UNBOUND; + else + worker->flags |= WORKER_UNBOUND; + + ret = true; + } + pool->flags &= ~POOL_MANAGE_WORKERS; /* -- cgit v1.2.3 From ec145babe754f9ea1079034a108104b6001e001c Mon Sep 17 00:00:00 2001 From: John Stultz Date: Tue, 11 Sep 2012 19:26:03 -0400 Subject: time: Fix timeekeping_get_ns overflow on 32bit systems Daniel Lezcano reported seeing multi-second stalls from keyboard input on his T61 laptop when NOHZ and CPU_IDLE were enabled on a 32bit kernel. He bisected the problem down to commit 1e75fa8be9fb6 ("time: Condense timekeeper.xtime into xtime_sec"). After reproducing this issue, I narrowed the problem down to the fact that timekeeping_get_ns() returns a 64bit nsec value that hasn't been accumulated. In some cases this value was being then stored in timespec.tv_nsec (which is a long). On 32bit systems, with idle times larger then 4 seconds (or less, depending on the value of xtime_nsec), the returned nsec value would overflow 32bits. This limited kept time from increasing, causing timers to not expire. The fix is to make sure we don't directly store the result of timekeeping_get_ns() into a tv_nsec field, instead using a 64bit nsec value which can then be added into the timespec via timespec_add_ns(). Reported-and-bisected-by: Daniel Lezcano Tested-by: Daniel Lezcano Signed-off-by: John Stultz Acked-by: Prarit Bhargava Cc: Richard Cochran Link: http://lkml.kernel.org/r/1347405963-35715-1-git-send-email-john.stultz@linaro.org Signed-off-by: Ingo Molnar --- kernel/time/timekeeping.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 34e5eac81424..d3b91e75cecd 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -303,10 +303,11 @@ void getnstimeofday(struct timespec *ts) seq = read_seqbegin(&tk->lock); ts->tv_sec = tk->xtime_sec; - ts->tv_nsec = timekeeping_get_ns(tk); + nsecs = timekeeping_get_ns(tk); } while (read_seqretry(&tk->lock, seq)); + ts->tv_nsec = 0; timespec_add_ns(ts, nsecs); } EXPORT_SYMBOL(getnstimeofday); @@ -345,6 +346,7 @@ void ktime_get_ts(struct timespec *ts) { struct timekeeper *tk = &timekeeper; struct timespec tomono; + s64 nsec; unsigned int seq; WARN_ON(timekeeping_suspended); @@ -352,13 +354,14 @@ void ktime_get_ts(struct timespec *ts) do { seq = read_seqbegin(&tk->lock); ts->tv_sec = tk->xtime_sec; - ts->tv_nsec = timekeeping_get_ns(tk); + nsec = timekeeping_get_ns(tk); tomono = tk->wall_to_monotonic; } while (read_seqretry(&tk->lock, seq)); - set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec, - ts->tv_nsec + tomono.tv_nsec); + ts->tv_sec += tomono.tv_sec; + ts->tv_nsec = 0; + timespec_add_ns(ts, nsec + tomono.tv_nsec); } EXPORT_SYMBOL_GPL(ktime_get_ts); @@ -1244,6 +1247,7 @@ void get_monotonic_boottime(struct timespec *ts) { struct timekeeper *tk = &timekeeper; struct timespec tomono, sleep; + s64 nsec; unsigned int seq; WARN_ON(timekeeping_suspended); @@ -1251,14 +1255,15 @@ void get_monotonic_boottime(struct timespec *ts) do { seq = read_seqbegin(&tk->lock); ts->tv_sec = tk->xtime_sec; - ts->tv_nsec = timekeeping_get_ns(tk); + nsec = timekeeping_get_ns(tk); tomono = tk->wall_to_monotonic; sleep = tk->total_sleep_time; } while (read_seqretry(&tk->lock, seq)); - set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec + sleep.tv_sec, - ts->tv_nsec + tomono.tv_nsec + sleep.tv_nsec); + ts->tv_sec += tomono.tv_sec + sleep.tv_sec; + ts->tv_nsec = 0; + timespec_add_ns(ts, nsec + tomono.tv_nsec + sleep.tv_nsec); } EXPORT_SYMBOL_GPL(get_monotonic_boottime); -- cgit v1.2.3 From 59a93c27c4892f04dfd8f91f8b64d0d6eae43e6e Mon Sep 17 00:00:00 2001 From: Todd Poynor Date: Thu, 9 Aug 2012 00:37:27 -0700 Subject: alarmtimer: Implement minimum alarm interval for allowing suspend alarmtimer suspend return -EBUSY if the next alarm will fire in less than 2 seconds. This allows one RTC seconds tick to occur subsequent to this check before the alarm wakeup time is set, ensuring the wakeup time is still in the future (assuming the RTC does not tick one more second prior to setting the alarm). If suspend is rejected due to an imminent alarm, hold a wakeup source for 2 seconds to process the alarm prior to reattempting suspend. If setting the alarm incurs an -ETIME for an alarm set in the past, or any other problem setting the alarm, abort suspend and hold a wakelock for 1 second while the alarm is allowed to be serviced or other hopefully transient conditions preventing the alarm clear up. Signed-off-by: Todd Poynor Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index aa27d391bfc8..54e7145c5414 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -46,6 +46,8 @@ static struct alarm_base { static ktime_t freezer_delta; static DEFINE_SPINLOCK(freezer_delta_lock); +static struct wakeup_source *ws; + #ifdef CONFIG_RTC_CLASS /* rtc timer and device for setting alarm wakeups at suspend */ static struct rtc_timer rtctimer; @@ -250,6 +252,7 @@ static int alarmtimer_suspend(struct device *dev) unsigned long flags; struct rtc_device *rtc; int i; + int ret; spin_lock_irqsave(&freezer_delta_lock, flags); min = freezer_delta; @@ -279,8 +282,10 @@ static int alarmtimer_suspend(struct device *dev) if (min.tv64 == 0) return 0; - /* XXX - Should we enforce a minimum sleep time? */ - WARN_ON(min.tv64 < NSEC_PER_SEC); + if (ktime_to_ns(min) < 2 * NSEC_PER_SEC) { + __pm_wakeup_event(ws, 2 * MSEC_PER_SEC); + return -EBUSY; + } /* Setup an rtc timer to fire that far in the future */ rtc_timer_cancel(rtc, &rtctimer); @@ -288,9 +293,11 @@ static int alarmtimer_suspend(struct device *dev) now = rtc_tm_to_ktime(tm); now = ktime_add(now, min); - rtc_timer_start(rtc, &rtctimer, now, ktime_set(0, 0)); - - return 0; + /* Set alarm, if in the past reject suspend briefly to handle */ + ret = rtc_timer_start(rtc, &rtctimer, now, ktime_set(0, 0)); + if (ret < 0) + __pm_wakeup_event(ws, MSEC_PER_SEC); + return ret; } #else static int alarmtimer_suspend(struct device *dev) @@ -821,6 +828,7 @@ static int __init alarmtimer_init(void) error = PTR_ERR(pdev); goto out_drv; } + ws = wakeup_source_register("alarmtimer"); return 0; out_drv: -- cgit v1.2.3 From dae373be9fec6f850159a05af3a1c36236a70d43 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 13 Sep 2012 19:12:16 -0400 Subject: alarmtimer: Use hrtimer per-alarm instead of per-base MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Arve Hjønnevåg reported numerous crashes from the "BUG_ON(timer->state != HRTIMER_STATE_CALLBACK)" check in __run_hrtimer after it called alarmtimer_fired. It ends up the alarmtimer code was not properly handling possible failures of hrtimer_try_to_cancel, and because these faulres occur when the underlying base hrtimer is being run, this limits the ability to properly handle modifications to any alarmtimers on that base. Because much of the logic duplicates the hrtimer logic, it seems that we might as well have a per-alarmtimer hrtimer, and avoid the extra complextity of trying to multiplex many alarmtimers off of one hrtimer. Thus this patch moves the hrtimer to the alarm structure and simplifies the management logic. Changelog: v2: * Includes a fix for double alarm_start calls found by Arve Cc: Arve Hjønnevåg Cc: Colin Cross Cc: Thomas Gleixner Reported-by: Arve Hjønnevåg Tested-by: Arve Hjønnevåg Signed-off-by: John Stultz --- include/linux/alarmtimer.h | 3 +- kernel/time/alarmtimer.c | 94 +++++++++++++--------------------------------- 2 files changed, 29 insertions(+), 68 deletions(-) (limited to 'kernel') diff --git a/include/linux/alarmtimer.h b/include/linux/alarmtimer.h index 96c5c249b086..f122c9fbf8c7 100644 --- a/include/linux/alarmtimer.h +++ b/include/linux/alarmtimer.h @@ -35,6 +35,7 @@ enum alarmtimer_restart { */ struct alarm { struct timerqueue_node node; + struct hrtimer timer; enum alarmtimer_restart (*function)(struct alarm *, ktime_t now); enum alarmtimer_type type; int state; @@ -43,7 +44,7 @@ struct alarm { void alarm_init(struct alarm *alarm, enum alarmtimer_type type, enum alarmtimer_restart (*function)(struct alarm *, ktime_t)); -void alarm_start(struct alarm *alarm, ktime_t start); +int alarm_start(struct alarm *alarm, ktime_t start); int alarm_try_to_cancel(struct alarm *alarm); int alarm_cancel(struct alarm *alarm); diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 54e7145c5414..b1560ebe759f 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -37,7 +37,6 @@ static struct alarm_base { spinlock_t lock; struct timerqueue_head timerqueue; - struct hrtimer timer; ktime_t (*gettime)(void); clockid_t base_clockid; } alarm_bases[ALARM_NUMTYPE]; @@ -132,21 +131,17 @@ static inline void alarmtimer_rtc_timer_init(void) { } * @base: pointer to the base where the timer is being run * @alarm: pointer to alarm being enqueued. * - * Adds alarm to a alarm_base timerqueue and if necessary sets - * an hrtimer to run. + * Adds alarm to a alarm_base timerqueue * * Must hold base->lock when calling. */ static void alarmtimer_enqueue(struct alarm_base *base, struct alarm *alarm) { + if (alarm->state & ALARMTIMER_STATE_ENQUEUED) + timerqueue_del(&base->timerqueue, &alarm->node); + timerqueue_add(&base->timerqueue, &alarm->node); alarm->state |= ALARMTIMER_STATE_ENQUEUED; - - if (&alarm->node == timerqueue_getnext(&base->timerqueue)) { - hrtimer_try_to_cancel(&base->timer); - hrtimer_start(&base->timer, alarm->node.expires, - HRTIMER_MODE_ABS); - } } /** @@ -154,28 +149,17 @@ static void alarmtimer_enqueue(struct alarm_base *base, struct alarm *alarm) * @base: pointer to the base where the timer is running * @alarm: pointer to alarm being removed * - * Removes alarm to a alarm_base timerqueue and if necessary sets - * a new timer to run. + * Removes alarm to a alarm_base timerqueue * * Must hold base->lock when calling. */ static void alarmtimer_remove(struct alarm_base *base, struct alarm *alarm) { - struct timerqueue_node *next = timerqueue_getnext(&base->timerqueue); - if (!(alarm->state & ALARMTIMER_STATE_ENQUEUED)) return; timerqueue_del(&base->timerqueue, &alarm->node); alarm->state &= ~ALARMTIMER_STATE_ENQUEUED; - - if (next == &alarm->node) { - hrtimer_try_to_cancel(&base->timer); - next = timerqueue_getnext(&base->timerqueue); - if (!next) - return; - hrtimer_start(&base->timer, next->expires, HRTIMER_MODE_ABS); - } } @@ -190,42 +174,23 @@ static void alarmtimer_remove(struct alarm_base *base, struct alarm *alarm) */ static enum hrtimer_restart alarmtimer_fired(struct hrtimer *timer) { - struct alarm_base *base = container_of(timer, struct alarm_base, timer); - struct timerqueue_node *next; + struct alarm *alarm = container_of(timer, struct alarm, timer); + struct alarm_base *base = &alarm_bases[alarm->type]; unsigned long flags; - ktime_t now; int ret = HRTIMER_NORESTART; int restart = ALARMTIMER_NORESTART; spin_lock_irqsave(&base->lock, flags); - now = base->gettime(); - while ((next = timerqueue_getnext(&base->timerqueue))) { - struct alarm *alarm; - ktime_t expired = next->expires; - - if (expired.tv64 > now.tv64) - break; - - alarm = container_of(next, struct alarm, node); - - timerqueue_del(&base->timerqueue, &alarm->node); - alarm->state &= ~ALARMTIMER_STATE_ENQUEUED; - - alarm->state |= ALARMTIMER_STATE_CALLBACK; - spin_unlock_irqrestore(&base->lock, flags); - if (alarm->function) - restart = alarm->function(alarm, now); - spin_lock_irqsave(&base->lock, flags); - alarm->state &= ~ALARMTIMER_STATE_CALLBACK; + alarmtimer_remove(base, alarm); + spin_unlock_irqrestore(&base->lock, flags); - if (restart != ALARMTIMER_NORESTART) { - timerqueue_add(&base->timerqueue, &alarm->node); - alarm->state |= ALARMTIMER_STATE_ENQUEUED; - } - } + if (alarm->function) + restart = alarm->function(alarm, base->gettime()); - if (next) { - hrtimer_set_expires(&base->timer, next->expires); + spin_lock_irqsave(&base->lock, flags); + if (restart != ALARMTIMER_NORESTART) { + hrtimer_set_expires(&alarm->timer, alarm->node.expires); + alarmtimer_enqueue(base, alarm); ret = HRTIMER_RESTART; } spin_unlock_irqrestore(&base->lock, flags); @@ -331,6 +296,9 @@ void alarm_init(struct alarm *alarm, enum alarmtimer_type type, enum alarmtimer_restart (*function)(struct alarm *, ktime_t)) { timerqueue_init(&alarm->node); + hrtimer_init(&alarm->timer, alarm_bases[type].base_clockid, + HRTIMER_MODE_ABS); + alarm->timer.function = alarmtimer_fired; alarm->function = function; alarm->type = type; alarm->state = ALARMTIMER_STATE_INACTIVE; @@ -341,17 +309,19 @@ void alarm_init(struct alarm *alarm, enum alarmtimer_type type, * @alarm: ptr to alarm to set * @start: time to run the alarm */ -void alarm_start(struct alarm *alarm, ktime_t start) +int alarm_start(struct alarm *alarm, ktime_t start) { struct alarm_base *base = &alarm_bases[alarm->type]; unsigned long flags; + int ret; spin_lock_irqsave(&base->lock, flags); - if (alarmtimer_active(alarm)) - alarmtimer_remove(base, alarm); alarm->node.expires = start; alarmtimer_enqueue(base, alarm); + ret = hrtimer_start(&alarm->timer, alarm->node.expires, + HRTIMER_MODE_ABS); spin_unlock_irqrestore(&base->lock, flags); + return ret; } /** @@ -365,18 +335,12 @@ int alarm_try_to_cancel(struct alarm *alarm) { struct alarm_base *base = &alarm_bases[alarm->type]; unsigned long flags; - int ret = -1; - spin_lock_irqsave(&base->lock, flags); - - if (alarmtimer_callback_running(alarm)) - goto out; + int ret; - if (alarmtimer_is_queued(alarm)) { + spin_lock_irqsave(&base->lock, flags); + ret = hrtimer_try_to_cancel(&alarm->timer); + if (ret >= 0) alarmtimer_remove(base, alarm); - ret = 1; - } else - ret = 0; -out: spin_unlock_irqrestore(&base->lock, flags); return ret; } @@ -809,10 +773,6 @@ static int __init alarmtimer_init(void) for (i = 0; i < ALARM_NUMTYPE; i++) { timerqueue_init_head(&alarm_bases[i].timerqueue); spin_lock_init(&alarm_bases[i].lock); - hrtimer_init(&alarm_bases[i].timer, - alarm_bases[i].base_clockid, - HRTIMER_MODE_ABS); - alarm_bases[i].timer.function = alarmtimer_fired; } error = alarmtimer_rtc_interface_setup(); -- cgit v1.2.3 From a65bcc12ad74b3efe78847945a1e36cfcbcbc4e6 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 13 Sep 2012 19:25:22 -0400 Subject: alarmtimer: Rename alarmtimer_remove to alarmtimer_dequeue MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Now that alarmtimer_remove has been simplified, change its name to _dequeue to better match its paired _enqueue function. Cc: Arve Hjønnevåg Cc: Colin Cross Cc: Thomas Gleixner Signed-off-by: John Stultz --- kernel/time/alarmtimer.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index b1560ebe759f..f11d83b12949 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -145,7 +145,7 @@ static void alarmtimer_enqueue(struct alarm_base *base, struct alarm *alarm) } /** - * alarmtimer_remove - Removes an alarm timer from an alarm_base timerqueue + * alarmtimer_dequeue - Removes an alarm timer from an alarm_base timerqueue * @base: pointer to the base where the timer is running * @alarm: pointer to alarm being removed * @@ -153,7 +153,7 @@ static void alarmtimer_enqueue(struct alarm_base *base, struct alarm *alarm) * * Must hold base->lock when calling. */ -static void alarmtimer_remove(struct alarm_base *base, struct alarm *alarm) +static void alarmtimer_dequeue(struct alarm_base *base, struct alarm *alarm) { if (!(alarm->state & ALARMTIMER_STATE_ENQUEUED)) return; @@ -181,7 +181,7 @@ static enum hrtimer_restart alarmtimer_fired(struct hrtimer *timer) int restart = ALARMTIMER_NORESTART; spin_lock_irqsave(&base->lock, flags); - alarmtimer_remove(base, alarm); + alarmtimer_dequeue(base, alarm); spin_unlock_irqrestore(&base->lock, flags); if (alarm->function) @@ -340,7 +340,7 @@ int alarm_try_to_cancel(struct alarm *alarm) spin_lock_irqsave(&base->lock, flags); ret = hrtimer_try_to_cancel(&alarm->timer); if (ret >= 0) - alarmtimer_remove(base, alarm); + alarmtimer_dequeue(base, alarm); spin_unlock_irqrestore(&base->lock, flags); return ret; } -- cgit v1.2.3 From b3c869d35b9b014f63ac0beacd31c57372084d01 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Tue, 4 Sep 2012 12:42:27 -0400 Subject: jiffies: Remove compile time assumptions about CLOCK_TICK_RATE CLOCK_TICK_RATE is used to accurately caclulate exactly how a tick will be at a given HZ. This is useful, because while we'd expect NSEC_PER_SEC/HZ, the underlying hardware will have some granularity limit, so we won't be able to have exactly HZ ticks per second. This slight error can cause timekeeping quality problems when using the jiffies or other jiffies driven clocksources. Thus we currently use compile time CLOCK_TICK_RATE value to generate SHIFTED_HZ and NSEC_PER_JIFFIES, which we then use to adjust the jiffies clocksource to correct this error. Unfortunately though, since CLOCK_TICK_RATE is a compile time value, and the jiffies clocksource is registered very early during boot, there are a number of cases where there are different possible hardware timers that have different tick rates. This causes problems in cases like ARM where there are numerous different types of hardware, each having their own compile-time CLOCK_TICK_RATE, making it hard to accurately support different hardware with a single kernel. For the most part, this doesn't matter all that much, as not too many systems actually utilize the jiffies or jiffies driven clocksource. Usually there are other highres clocksources who's granularity error is negligable. Even so, we have some complicated calcualtions that we do everywhere to handle these edge cases. This patch removes the compile time SHIFTED_HZ value, and introduces a register_refined_jiffies() function. This results in the default jiffies clock as being assumed a perfect HZ freq, and allows archtectures that care about jiffies accuracy to call register_refined_jiffies() with the tick rate, specified dynamically at boot. This allows us, where necessary, to not have a compile time CLOCK_TICK_RATE constant, simplifies the jiffies code, and still provides a way to have an accurate jiffies clock. NOTE: Since this patch does not add register_refinied_jiffies() calls for every arch, it may cause time quality regressions in some cases. Its likely these will not be noticable, but if they are an issue, adding the following to the end of setup_arch() should resolve the regression: register_refinied_jiffies(CLOCK_TICK_RATE) Cc: Catalin Marinas Cc: Arnd Bergmann Cc: Richard Cochran Cc: Prarit Bhargava Cc: Thomas Gleixner Signed-off-by: John Stultz --- arch/x86/kernel/setup.c | 3 +++ include/linux/jiffies.h | 15 ++------------- kernel/time/jiffies.c | 32 +++++++++++++++++++++++++++++++- 3 files changed, 36 insertions(+), 14 deletions(-) (limited to 'kernel') diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index f4b9b80e1b95..4062f15bfbdd 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -68,6 +68,7 @@ #include #include #include +#include #include