From 7b9b816f4b9a3513602454b52c77f371388a2485 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Sat, 11 Jan 2020 01:05:06 +0900 Subject: Documentation: bootconfig: Add a doc for extended boot config Add a documentation for extended boot config under admin-guide, since it is including the syntax of boot config. Link: http://lkml.kernel.org/r/157867230658.17873.9309879174829924324.stgit@devnote2 Signed-off-by: Masami Hiramatsu Signed-off-by: Steven Rostedt (VMware) --- Documentation/admin-guide/bootconfig.rst | 184 +++++++++++++++++++++++++++++++ Documentation/admin-guide/index.rst | 1 + 2 files changed, 185 insertions(+) create mode 100644 Documentation/admin-guide/bootconfig.rst (limited to 'Documentation/admin-guide') diff --git a/Documentation/admin-guide/bootconfig.rst b/Documentation/admin-guide/bootconfig.rst new file mode 100644 index 000000000000..f7475df2a718 --- /dev/null +++ b/Documentation/admin-guide/bootconfig.rst @@ -0,0 +1,184 @@ +.. SPDX-License-Identifier: GPL-2.0 + +================== +Boot Configuration +================== + +:Author: Masami Hiramatsu + +Overview +======== + +The boot configuration is expanding current kernel cmdline to support +additional key-value data when boot the kernel in an efficient way. +This allows adoministrators to pass a structured-Key config file. + +Config File Syntax +================== + +The boot config syntax is a simple structured key-value. Each key consists +of dot-connected-words, and key and value are connected by "=". The value +has to be terminated by semi-colon (``;``) or newline (``\n``). +For array value, array entries are separated by comma (``,``). :: + +KEY[.WORD[...]] = VALUE[, VALUE2[...]][;] + +Each key word must contain only alphabets, numbers, dash (``-``) or underscore +(``_``). And each value only contains printable characters or spaces except +for delimiters such as semi-colon (``;``), new-line (``\n``), comma (``,``), +hash (``#``) and closing brace (``}``). + +If you want to use those delimiters in a value, you can use either double- +quotes (``"VALUE"``) or single-quotes (``'VALUE'``) to quote it. Note that +you can not escape these quotes. + +There can be a key which doesn't have value or has an empty value. Those keys +are used for checking the key exists or not (like a boolean). + +Key-Value Syntax +---------------- + +The boot config file syntax allows user to merge partially same word keys +by brace. For example:: + + foo.bar.baz = value1 + foo.bar.qux.quux = value2 + +These can be written also in:: + + foo.bar { + baz = value1 + qux.quux = value2 + } + +Or more shorter, written as following:: + + foo.bar { baz = value1; qux.quux = value2 } + +In both styles, same key words are automatically merged when parsing it +at boot time. So you can append similar trees or key-values. + +Comments +-------- + +The config syntax accepts shell-script style comments. The comments start +with hash ("#") until newline ("\n") will be ignored. + +:: + + # comment line + foo = value # value is set to foo. + bar = 1, # 1st element + 2, # 2nd element + 3 # 3rd element + +This is parsed as below:: + + foo = value + bar = 1, 2, 3 + +Note that you can not put a comment between value and delimiter(``,`` or +``;``). This means following config has a syntax error :: + + key = 1 # comment + ,2 + + +/proc/bootconfig +================ + +/proc/bootconfig is a user-space interface of the boot config. +Unlike /proc/cmdline, this file shows the key-value style list. +Each key-value pair is shown in each line with following style:: + + KEY[.WORDS...] = "[VALUE]"[,"VALUE2"...] + + +Boot Kernel With a Boot Config +============================== + +Since the boot configuration file is loaded with initrd, it will be added +to the end of the initrd (initramfs) image file. The Linux kernel decodes +the last part of the initrd image in memory to get the boot configuration +data. +Because of this "piggyback" method, there is no need to change or +update the boot loader and the kernel image itself. + +To do this operation, Linux kernel provides "bootconfig" command under +tools/bootconfig, which allows admin to apply or delete the config file +to/from initrd image. You can build it by follwoing command:: + + # make -C tools/bootconfig + +To add your boot config file to initrd image, run bootconfig as below +(Old data is removed automatically if exists):: + + # tools/bootconfig/bootconfig -a your-config /boot/initrd.img-X.Y.Z + +To remove the config from the image, you can use -d option as below:: + + # tools/bootconfig/bootconfig -d /boot/initrd.img-X.Y.Z + + +C onfig File Limitation +====================== + +Currently the maximum config size size is 32KB and the total key-words (not +key-value entries) must be under 1024 nodes. +Note: this is not the number of entries but nodes, an entry must consume +more than 2 nodes (a key-word and a value). So theoretically, it will be +up to 512 key-value pairs. If keys contains 3 words in average, it can +contain 256 key-value pairs. In most cases, the number of config items +will be under 100 entries and smaller than 8KB, so it would be enough. +If the node number exceeds 1024, parser returns an error even if the file +size is smaller than 32KB. +Anyway, since bootconfig command verifies it when appending a boot config +to initrd image, user can notice it before boot. + + +Bootconfig APIs +=============== + +User can query or loop on key-value pairs, also it is possible to find +a root (prefix) key node and find key-values under that node. + +If you have a key string, you can query the value directly with the key +using xbc_find_value(). If you want to know what keys exist in the SKC +tree, you can use xbc_for_each_key_value() to iterate key-value pairs. +Note that you need to use xbc_array_for_each_value() for accessing +each arraies value, e.g.:: + + vnode = NULL; + xbc_find_value("key.word", &vnode); + if (vnode && xbc_node_is_array(vnode)) + xbc_array_for_each_value(vnode, value) { + printk("%s ", value); + } + +If you want to focus on keys which has a prefix string, you can use +xbc_find_node() to find a node which prefix key words, and iterate +keys under the prefix node with xbc_node_for_each_key_value(). + +But the most typical usage is to get the named value under prefix +or get the named array under prefix as below:: + + root = xbc_find_node("key.prefix"); + value = xbc_node_find_value(root, "option", &vnode); + ... + xbc_node_for_each_array_value(root, "array-option", value, anode) { + ... + } + +This accesses a value of "key.prefix.option" and an array of +"key.prefix.array-option". + +Locking is not needed, since after initialized, the config becomes readonly. +All data and keys must be copied if you need to modify it. + + +Functions and structures +======================== + +.. kernel-doc:: include/linux/bootconfig.h +.. kernel-doc:: lib/bootconfig.c + diff --git a/Documentation/admin-guide/index.rst b/Documentation/admin-guide/index.rst index 4405b7485312..9e0f1e3fd152 100644 --- a/Documentation/admin-guide/index.rst +++ b/Documentation/admin-guide/index.rst @@ -64,6 +64,7 @@ configure specific aspects of kernel behavior to your liking. binderfs binfmt-misc blockdev/index + bootconfig braille-console btmrvl cgroup-v1/index -- cgit v1.2.3 From 47781947947a8f5e5e1205714fda64563d2b8f05 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Sat, 11 Jan 2020 01:07:40 +0900 Subject: Documentation: tracing: Add boot-time tracing document Add a documentation about boot-time tracing options in boot config. Link: http://lkml.kernel.org/r/157867246028.17873.8047384554383977870.stgit@devnote2 Signed-off-by: Masami Hiramatsu Signed-off-by: Steven Rostedt (VMware) --- Documentation/admin-guide/bootconfig.rst | 2 + Documentation/trace/boottime-trace.rst | 184 +++++++++++++++++++++++++++++++ Documentation/trace/index.rst | 1 + 3 files changed, 187 insertions(+) create mode 100644 Documentation/trace/boottime-trace.rst (limited to 'Documentation/admin-guide') diff --git a/Documentation/admin-guide/bootconfig.rst b/Documentation/admin-guide/bootconfig.rst index f7475df2a718..c8f7cd4cf44e 100644 --- a/Documentation/admin-guide/bootconfig.rst +++ b/Documentation/admin-guide/bootconfig.rst @@ -1,5 +1,7 @@ .. SPDX-License-Identifier: GPL-2.0 +.. _bootconfig: + ================== Boot Configuration ================== diff --git a/Documentation/trace/boottime-trace.rst b/Documentation/trace/boottime-trace.rst new file mode 100644 index 000000000000..1d10fdebf1b2 --- /dev/null +++ b/Documentation/trace/boottime-trace.rst @@ -0,0 +1,184 @@ +.. SPDX-License-Identifier: GPL-2.0 + +================= +Boot-time tracing +================= + +:Author: Masami Hiramatsu + +Overview +======== + +Boot-time tracing allows users to trace boot-time process including +device initialization with full features of ftrace including per-event +filter and actions, histograms, kprobe-events and synthetic-events, +and trace instances. +Since kernel cmdline is not enough to control these complex features, +this uses bootconfig file to describe tracing feature programming. + +Options in the Boot Config +========================== + +Here is the list of available options list for boot time tracing in +boot config file [1]_. All options are under "ftrace." or "kernel." +refix. See kernel parameters for the options which starts +with "kernel." prefix [2]_. + +.. [1] See :ref:`Documentation/admin-guide/bootconfig.rst ` +.. [2] See :ref:`Documentation/admin-guide/kernel-parameters.rst ` + +Ftrace Global Options +--------------------- + +Ftrace global options have "kernel." prefix in boot config, which means +these options are passed as a part of kernel legacy command line. + +kernel.tp_printk + Output trace-event data on printk buffer too. + +kernel.dump_on_oops [= MODE] + Dump ftrace on Oops. If MODE = 1 or omitted, dump trace buffer + on all CPUs. If MODE = 2, dump a buffer on a CPU which kicks Oops. + +kernel.traceoff_on_warning + Stop tracing if WARN_ON() occurs. + +kernel.fgraph_max_depth = MAX_DEPTH + Set MAX_DEPTH to maximum depth of fgraph tracer. + +kernel.fgraph_filters = FILTER[, FILTER2...] + Add fgraph tracing function filters. + +kernel.fgraph_notraces = FILTER[, FILTER2...] + Add fgraph non tracing function filters. + + +Ftrace Per-instance Options +--------------------------- + +These options can be used for each instance including global ftrace node. + +ftrace.[instance.INSTANCE.]options = OPT1[, OPT2[...]] + Enable given ftrace options. + +ftrace.[instance.INSTANCE.]trace_clock = CLOCK + Set given CLOCK to ftrace's trace_clock. + +ftrace.[instance.INSTANCE.]buffer_size = SIZE + Configure ftrace buffer size to SIZE. You can use "KB" or "MB" + for that SIZE. + +ftrace.[instance.INSTANCE.]alloc_snapshot + Allocate snapshot buffer. + +ftrace.[instance.INSTANCE.]cpumask = CPUMASK + Set CPUMASK as trace cpu-mask. + +ftrace.[instance.INSTANCE.]events = EVENT[, EVENT2[...]] + Enable given events on boot. You can use a wild card in EVENT. + +ftrace.[instance.INSTANCE.]tracer = TRACER + Set TRACER to current tracer on boot. (e.g. function) + +ftrace.[instance.INSTANCE.]ftrace.filters + This will take an array of tracing function filter rules + +ftrace.[instance.INSTANCE.]ftrace.notraces + This will take an array of NON-tracing function filter rules + + +Ftrace Per-Event Options +------------------------ + +These options are setting per-event options. + +ftrace.[instance.INSTANCE.]event.GROUP.EVENT.enable + Enables GROUP:EVENT tracing. + +ftrace.[instance.INSTANCE.]event.GROUP.EVENT.filter = FILTER + Set FILTER rule to the GROUP:EVENT. + +ftrace.[instance.INSTANCE.]event.GROUP.EVENT.actions = ACTION[, ACTION2[...]] + Set ACTIONs to the GROUP:EVENT. + +ftrace.[instance.INSTANCE.]event.kprobes.EVENT.probes = PROBE[, PROBE2[...]] + Defines new kprobe event based on PROBEs. It is able to define + multiple probes on one event, but those must have same type of + arguments. This option is available only for the event which + group name is "kprobes". + +ftrace.[instance.INSTANCE.]event.synthetic.EVENT.fields = FIELD[, FIELD2[...]] + Defines new synthetic event with FIELDs. Each field should be + "type varname". + +Note that kprobe and synthetic event definitions can be written under +instance node, but those are also visible from other instances. So please +take care for event name conflict. + + +Examples +======== + +For example, to add filter and actions for each event, define kprobe +events, and synthetic events with histogram, write a boot config like +below:: + + ftrace.event { + task.task_newtask { + filter = "pid < 128" + enable + } + kprobes.vfs_read { + probes = "vfs_read $arg1 $arg2" + filter = "common_pid < 200" + enable + } + synthetic.initcall_latency { + fields = "unsigned long func", "u64 lat" + actions = "hist:keys=func.sym,lat:vals=lat:sort=lat" + } + initcall.initcall_start { + actions = "hist:keys=func:ts0=common_timestamp.usecs" + } + initcall.initcall_finish { + actions = "hist:keys=func:lat=common_timestamp.usecs-$ts0:onmatch(initcall.initcall_start).initcall_latency(func,$lat)" + } + } + +Also, boottime tracing supports "instance" node, which allows us to run +several tracers for different purpose at once. For example, one tracer +is for tracing functions start with "user\_", and others tracing "kernel\_" +functions, you can write boot config as below:: + + ftrace.instance { + foo { + tracer = "function" + ftrace.filters = "user_*" + } + bar { + tracer = "function" + ftrace.filters = "kernel_*" + } + } + +The instance node also accepts event nodes so that each instance +can customize its event tracing. + +This boot-time tracing also supports ftrace kernel parameters via boot +config. +For example, following kernel parameters:: + + trace_options=sym-addr trace_event=initcall:* tp_printk trace_buf_size=1M ftrace=function ftrace_filter="vfs*" + +This can be written in boot config like below:: + + kernel { + trace_options = sym-addr + trace_event = "initcall:*" + tp_printk + trace_buf_size = 1M + ftrace = function + ftrace_filter = "vfs*" + } + +Note that parameters start with "kernel" prefix instead of "ftrace". diff --git a/Documentation/trace/index.rst b/Documentation/trace/index.rst index 04acd277c5f6..fa9e1c730f6a 100644 --- a/Documentation/trace/index.rst +++ b/Documentation/trace/index.rst @@ -19,6 +19,7 @@ Linux Tracing Technologies events-msr mmiotrace histogram + boottime-trace hwlat_detector intel_th stm -- cgit v1.2.3 From a4798eb42a261ff39d991b198a09b840c11010d5 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Mon, 20 Jan 2020 12:23:12 +0900 Subject: Documentation: bootconfig: Fix typos in bootconfig documentation Fix typos in bootconfig.rst according to Randy's suggestions. Link: http://lkml.kernel.org/r/157949059219.25888.16939971423610233631.stgit@devnote2 Suggested-by: Randy Dunlap Acked-by: Randy Dunlap Signed-off-by: Masami Hiramatsu Signed-off-by: Steven Rostedt (VMware) --- Documentation/admin-guide/bootconfig.rst | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) (limited to 'Documentation/admin-guide') diff --git a/Documentation/admin-guide/bootconfig.rst b/Documentation/admin-guide/bootconfig.rst index c8f7cd4cf44e..4d617693c0c8 100644 --- a/Documentation/admin-guide/bootconfig.rst +++ b/Documentation/admin-guide/bootconfig.rst @@ -11,20 +11,22 @@ Boot Configuration Overview ======== -The boot configuration is expanding current kernel cmdline to support -additional key-value data when boot the kernel in an efficient way. -This allows adoministrators to pass a structured-Key config file. +The boot configuration expands the current kernel command line to support +additional key-value data when booting the kernel in an efficient way. +This allows administrators to pass a structured-Key config file. Config File Syntax ================== The boot config syntax is a simple structured key-value. Each key consists -of dot-connected-words, and key and value are connected by "=". The value +of dot-connected-words, and key and value are connected by ``=``. The value has to be terminated by semi-colon (``;``) or newline (``\n``). For array value, array entries are separated by comma (``,``). :: KEY[.WORD[...]] = VALUE[, VALUE2[...]][;] +Unlike the kernel command line syntax, spaces are OK around the comma and ``=``. + Each key word must contain only alphabets, numbers, dash (``-``) or underscore (``_``). And each value only contains printable characters or spaces except for delimiters such as semi-colon (``;``), new-line (``\n``), comma (``,``), @@ -35,7 +37,7 @@ quotes (``"VALUE"``) or single-quotes (``'VALUE'``) to quote it. Note that you can not escape these quotes. There can be a key which doesn't have value or has an empty value. Those keys -are used for checking the key exists or not (like a boolean). +are used for checking if the key exists or not (like a boolean). Key-Value Syntax ---------------- @@ -63,7 +65,7 @@ at boot time. So you can append similar trees or key-values. Comments -------- -The config syntax accepts shell-script style comments. The comments start +The config syntax accepts shell-script style comments. The comments starting with hash ("#") until newline ("\n") will be ignored. :: @@ -108,7 +110,7 @@ update the boot loader and the kernel image itself. To do this operation, Linux kernel provides "bootconfig" command under tools/bootconfig, which allows admin to apply or delete the config file -to/from initrd image. You can build it by follwoing command:: +to/from initrd image. You can build it by the following command:: # make -C tools/bootconfig @@ -122,7 +124,7 @@ To remove the config from the image, you can use -d option as below:: # tools/bootconfig/bootconfig -d /boot/initrd.img-X.Y.Z -C onfig File Limitation +Config File Limitation ====================== Currently the maximum config size size is 32KB and the total key-words (not @@ -145,10 +147,10 @@ User can query or loop on key-value pairs, also it is possible to find a root (prefix) key node and find key-values under that node. If you have a key string, you can query the value directly with the key -using xbc_find_value(). If you want to know what keys exist in the SKC -tree, you can use xbc_for_each_key_value() to iterate key-value pairs. +using xbc_find_value(). If you want to know what keys exist in the boot +config, you can use xbc_for_each_key_value() to iterate key-value pairs. Note that you need to use xbc_array_for_each_value() for accessing -each arraies value, e.g.:: +each array's value, e.g.:: vnode = NULL; xbc_find_value("key.word", &vnode); @@ -157,8 +159,8 @@ each arraies value, e.g.:: printk("%s ", value); } -If you want to focus on keys which has a prefix string, you can use -xbc_find_node() to find a node which prefix key words, and iterate +If you want to focus on keys which have a prefix string, you can use +xbc_find_node() to find a node by the prefix string, and iterate keys under the prefix node with xbc_node_for_each_key_value(). But the most typical usage is to get the named value under prefix @@ -174,8 +176,8 @@ or get the named array under prefix as below:: This accesses a value of "key.prefix.option" and an array of "key.prefix.array-option". -Locking is not needed, since after initialized, the config becomes readonly. -All data and keys must be copied if you need to modify it. +Locking is not needed, since after initialization, the config becomes +read-only. All data and keys must be copied if you need to modify it. Functions and structures -- cgit v1.2.3 From 7495e0926fdf302cb9e62a49f7c22198815624cd Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Tue, 4 Feb 2020 07:33:53 -0500 Subject: bootconfig: Only load bootconfig if "bootconfig" is on the kernel cmdline As the bootconfig is appended to the initrd it is not as easy to modify as the kernel command line. If there's some issue with the kernel, and the developer wants to boot a pristine kernel, it should not be needed to modify the initrd to remove the bootconfig for a single boot. As bootconfig is silently added (if the admin does not know where to look they may not know it's being loaded). It should be explicitly added to the kernel cmdline. The loading of the bootconfig is only done if "bootconfig" is on the kernel command line. This will let admins know that the kernel command line is extended. Note, after adding printk()s for when the size is too great or the checksum is wrong, exposed that the current method always looked for the boot config, and if this size and checksum matched, it would parse it (as if either is wrong a printk has been added to show this). It's better to only check this if the boot config is asked to be looked for. Link: https://lore.kernel.org/r/CAHk-=wjfjO+h6bQzrTf=YCZA53Y3EDyAs3Z4gEsT7icA3u_Psw@mail.gmail.com Acked-by: Masami Hiramatsu Suggested-by: Linus Torvalds Signed-off-by: Steven Rostedt (VMware) --- Documentation/admin-guide/bootconfig.rst | 2 ++ Documentation/admin-guide/kernel-parameters.txt | 6 ++++++ init/main.c | 28 ++++++++++++++++++------- 3 files changed, 29 insertions(+), 7 deletions(-) (limited to 'Documentation/admin-guide') diff --git a/Documentation/admin-guide/bootconfig.rst b/Documentation/admin-guide/bootconfig.rst index 4d617693c0c8..b342a6796392 100644 --- a/Documentation/admin-guide/bootconfig.rst +++ b/Documentation/admin-guide/bootconfig.rst @@ -123,6 +123,8 @@ To remove the config from the image, you can use -d option as below:: # tools/bootconfig/bootconfig -d /boot/initrd.img-X.Y.Z +Then add "bootconfig" on the normal kernel command line to tell the +kernel to look for the bootconfig at the end of the initrd file. Config File Limitation ====================== diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index ade4e6ec23e0..b48c70ba9841 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -437,6 +437,12 @@ no delay (0). Format: integer + bootconfig [KNL] + Extended command line options can be added to an initrd + and this will cause the kernel to look for it. + + See Documentation/admin-guide/bootconfig.rst + bert_disable [ACPI] Disable BERT OS support on buggy BIOSes. diff --git a/init/main.c b/init/main.c index dd7da62d99a5..f174a59d3903 100644 --- a/init/main.c +++ b/init/main.c @@ -336,28 +336,39 @@ u32 boot_config_checksum(unsigned char *p, u32 size) return ret; } -static void __init setup_boot_config(void) +static void __init setup_boot_config(const char *cmdline) { u32 size, csum; char *data, *copy; + const char *p; u32 *hdr; - if (!initrd_end) + p = strstr(cmdline, "bootconfig"); + if (!p || (p != cmdline && !isspace(*(p-1))) || + (p[10] && !isspace(p[10]))) return; + if (!initrd_end) + goto not_found; + hdr = (u32 *)(initrd_end - 8); size = hdr[0]; csum = hdr[1]; - if (size >= XBC_DATA_MAX) + if (size >= XBC_DATA_MAX) { + pr_err("bootconfig size %d greater than max size %d\n", + size, XBC_DATA_MAX); return; + } data = ((void *)hdr) - size; if ((unsigned long)data < initrd_start) - return; + goto not_found; - if (boot_config_checksum((unsigned char *)data, size) != csum) + if (boot_config_checksum((unsigned char *)data, size) != csum) { + pr_err("bootconfig checksum failed\n"); return; + } copy = memblock_alloc(size + 1, SMP_CACHE_BYTES); if (!copy) { @@ -377,9 +388,12 @@ static void __init setup_boot_config(void) /* Also, "init." keys are init arguments */ extra_init_args = xbc_make_cmdline("init"); } + return; +not_found: + pr_err("'bootconfig' found on command line, but no bootconfig found\n"); } #else -#define setup_boot_config() do { } while (0) +#define setup_boot_config(cmdline) do { } while (0) #endif /* Change NUL term back to "=", to make "param" the whole string. */ @@ -760,7 +774,7 @@ asmlinkage __visible void __init start_kernel(void) pr_notice("%s", linux_banner); early_security_init(); setup_arch(&command_line); - setup_boot_config(); + setup_boot_config(command_line); setup_command_line(command_line); setup_nr_cpu_ids(); setup_per_cpu_areas(); -- cgit v1.2.3