From ca86a5d2f9feed16e11023891c8c70aa57ced24e Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Thu, 16 May 2024 00:10:44 +0200 Subject: tmpfs: don't interrupt fallocate with EINTR I have a program that sets up a periodic timer with 10ms interval. When the program attempts to call fallocate(2) on tmpfs, it goes into an infinite loop. fallocate(2) takes longer than 10ms, so it gets interrupted by a signal and it returns EINTR. On EINTR, the fallocate call is restarted, going into the same loop again. Let's change the signal_pending() check in shmem_fallocate() loop to fatal_signal_pending(). This solves the problem of shmem_fallocate() constantly restarting. Since most other filesystem's fallocate methods don't react to signals, it is unlikely userspace really relies on timely delivery of non-fatal signals while fallocate is running. Also the comment before the signal check: /* * Good, the fallocate(2) manpage permits EINTR: we may have * been interrupted because we are using up too much memory. */ indicates that the check was mainly added for OOM situations in which case the process will be sent a fatal signal so this change preserves the behavior in OOM situations. [JK: Update changelog and comment based on upstream discussion] Signed-off-by: Mikulas Patocka Signed-off-by: Jan Kara Link: https://lore.kernel.org/r/20240515221044.590-1-jack@suse.cz Reviewed-by: Matthew Wilcox (Oracle) Signed-off-by: Christian Brauner --- mm/shmem.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'mm') diff --git a/mm/shmem.c b/mm/shmem.c index f5d60436b604..ff7c756a7d02 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -3166,10 +3166,13 @@ static long shmem_fallocate(struct file *file, int mode, loff_t offset, struct folio *folio; /* - * Good, the fallocate(2) manpage permits EINTR: we may have - * been interrupted because we are using up too much memory. + * Check for fatal signal so that we abort early in OOM + * situations. We don't want to abort in case of non-fatal + * signals as large fallocate can take noticeable time and + * e.g. periodic timers may result in fallocate constantly + * restarting. */ - if (signal_pending(current)) + if (fatal_signal_pending(current)) error = -EINTR; else if (shmem_falloc.nr_unswapped > shmem_falloc.nr_falloced) error = -ENOMEM; -- cgit v1.2.3 From 8e3447822d7d8c0f562c6851a7a31e24d1ede55e Mon Sep 17 00:00:00 2001 From: Mateusz Guzik Date: Mon, 24 Jun 2024 10:54:02 +0200 Subject: vfs: remove redundant smp_mb for thp handling in do_dentry_open opening for write performs: if (f->f_mode & FMODE_WRITE) { [snip] smp_mb(); if (filemap_nr_thps(inode->i_mapping)) { [snip] } } filemap_nr_thps on kernels built without CONFIG_READ_ONLY_THP_FOR expands to 0, allowing the compiler to eliminate the entire thing, with exception of the fence (and the branch leading there). So happens required synchronisation between i_writecount and nr_thps changes is already provided by the full fence coming from get_write_access -> atomic_inc_unless_negative, thus the smp_mb instance above can be removed regardless of CONFIG_READ_ONLY_THP_FOR. While I updated commentary in places claiming to match the now-removed fence, I did not try to patch them to act on the compile option. I did not bother benchmarking it, not issuing a spurious full fence in the fast path does not warrant justification from perf standpoint. Signed-off-by: Mateusz Guzik Link: https://lore.kernel.org/r/20240624085402.493630-1-mjguzik@gmail.com Signed-off-by: Christian Brauner --- fs/open.c | 9 ++++----- mm/khugepaged.c | 10 +++++----- 2 files changed, 9 insertions(+), 10 deletions(-) (limited to 'mm') diff --git a/fs/open.c b/fs/open.c index a5c4f8a0f143..c4e9b01aafd8 100644 --- a/fs/open.c +++ b/fs/open.c @@ -986,12 +986,11 @@ static int do_dentry_open(struct file *f, */ if (f->f_mode & FMODE_WRITE) { /* - * Paired with smp_mb() in collapse_file() to ensure nr_thps - * is up to date and the update to i_writecount by - * get_write_access() is visible. Ensures subsequent insertion - * of THPs into the page cache will fail. + * Depends on full fence from get_write_access() to synchronize + * against collapse_file() regarding i_writecount and nr_thps + * updates. Ensures subsequent insertion of THPs into the page + * cache will fail. */ - smp_mb(); if (filemap_nr_thps(inode->i_mapping)) { struct address_space *mapping = inode->i_mapping; diff --git a/mm/khugepaged.c b/mm/khugepaged.c index 774a97e6e2da..aab471791bd9 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -2000,9 +2000,9 @@ out_unlock: if (!is_shmem) { filemap_nr_thps_inc(mapping); /* - * Paired with smp_mb() in do_dentry_open() to ensure - * i_writecount is up to date and the update to nr_thps is - * visible. Ensures the page cache will be truncated if the + * Paired with the fence in do_dentry_open() -> get_write_access() + * to ensure i_writecount is up to date and the update to nr_thps + * is visible. Ensures the page cache will be truncated if the * file is opened writable. */ smp_mb(); @@ -2190,8 +2190,8 @@ rollback: if (!is_shmem && result == SCAN_COPY_MC) { filemap_nr_thps_dec(mapping); /* - * Paired with smp_mb() in do_dentry_open() to - * ensure the update to nr_thps is visible. + * Paired with the fence in do_dentry_open() -> get_write_access() + * to ensure the update to nr_thps is visible. */ smp_mb(); } -- cgit v1.2.3