You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

Steven Rostedt c32e827b25 tracing: add raw trace point recording infrastructure ... Impact: lower overhead tracing The current event tracer can automatically pick up trace points that are registered with the TRACE_FORMAT macro. But it required a printf format string and parsing. Although, this adds the ability to get guaranteed information like task names and such, it took a hit in overhead processing. This processing can add about 500-1000 nanoseconds overhead, but in some cases that too is considered too much and we want to shave off as much from this overhead as possible. Tom Zanussi recently posted tracing patches to lkml that are based on a nice idea about capturing the data via C structs using STRUCT_ENTER, STRUCT_EXIT type of macros. I liked that method very much, but did not like the implementation that required a developer to add data/code in several disjoint locations. This patch extends the event_tracer macros to do a similar "raw C" approach that Tom Zanussi did. But instead of having the developers needing to tweak a bunch of code all over the place, they can do it all in one macro - preferably placed near the code that it is tracing. That makes it much more likely that tracepoints will be maintained on an ongoing basis by the code they modify. The new macro TRACE_EVENT_FORMAT is created for this approach. (Note, a developer may still utilize the more low level DECLARE_TRACE macros if they don't care about getting their traces automatically in the event tracer.) They can also use the existing TRACE_FORMAT if they don't need to code the tracepoint in C, but just want to use the convenience of printf. So if the developer wants to "hardwire" a tracepoint in the fastest possible way, and wants to acquire their data via a user space utility in a raw binary format, or wants to see it in the trace output but not sacrifice any performance, then they can implement the faster but more complex TRACE_EVENT_FORMAT macro. Here's what usage looks like: TRACE_EVENT_FORMAT(name, TPPROTO(proto), TPARGS(args), TPFMT(fmt, fmt_args), TRACE_STUCT( TRACE_FIELD(type1, item1, assign1) TRACE_FIELD(type2, item2, assign2) [...] ), TPRAWFMT(raw_fmt) ); Note name, proto, args, and fmt, are all identical to what TRACE_FORMAT uses. name: is the unique identifier of the trace point proto: The proto type that the trace point uses args: the args in the proto type fmt: printf format to use with the event printf tracer fmt_args: the printf argments to match fmt TRACE_STRUCT starts the ability to create a structure. Each item in the structure is defined with a TRACE_FIELD TRACE_FIELD(type, item, assign) type: the C type of item. item: the name of the item in the stucture assign: what to assign the item in the trace point callback raw_fmt is a way to pretty print the struct. It must match the order of the items are added in TRACE_STUCT An example of this would be: TRACE_EVENT_FORMAT(sched_wakeup, TPPROTO(struct rq *rq, struct task_struct *p, int success), TPARGS(rq, p, success), TPFMT("task %s:%d %s", p->comm, p->pid, success?"succeeded":"failed"), TRACE_STRUCT( TRACE_FIELD(pid_t, pid, p->pid) TRACE_FIELD(int, success, success) ), TPRAWFMT("task %d success=%d") ); This creates us a unique struct of: struct { pid_t pid; int success; }; And the way the call back would assign these values would be: entry->pid = p->pid; entry->success = success; The nice part about this is that the creation of the assignent is done via macro magic in the event tracer. Once the TRACE_EVENT_FORMAT is created, the developer will then have a faster method to record into the ring buffer. They do not need to worry about the tracer itself. The developer would only need to touch the files in include/trace/*.h Again, I would like to give special thanks to Tom Zanussi for this nice idea. Idea-from: Tom Zanussi <tzanussi@gmail.com> Signed-off-by: Steven Rostedt <srostedt@redhat.com>		16 years ago
..
irq	tracing, genirq: add irq enter and exit trace events	16 years ago
power	PM: Split up sysdev_[suspend|resume] from device_power_[down|up]	16 years ago
time	hrtimers: allow the hot-unplugging of all cpus	16 years ago
trace	tracing: add raw trace point recording infrastructure	16 years ago
.gitignore	…
Kconfig.freezer	container freezer: implement freezer cgroup subsystem	17 years ago
Kconfig.hz	…
Kconfig.preempt	rcu: provide RCU options on non-preempt architectures too	16 years ago
Makefile	PM: fix build for CONFIG_PM unset	16 years ago
acct.c	[CVE-2009-0029] System call wrappers part 04	16 years ago
async.c	async: use list_move_tail	16 years ago
audit.c	[PATCH] fix broken timestamps in AVC generated by kernel threads	16 years ago
audit.h	fixing audit rule ordering mess, part 1	16 years ago
audit_tree.c	audit: validate comparison operations, store them in sane form	16 years ago
auditfilter.c	audit: validate comparison operations, store them in sane form	16 years ago
auditsc.c	make sure that filterkey of task,always rules is reported	16 years ago
backtracetest.c	…
bounds.c	…
capability.c	[CVE-2009-0029] System call wrappers part 04	16 years ago
cgroup.c	cgroups: fix possible use after free	16 years ago
cgroup_debug.c	cgroups: fix probable race with put_css_set[_taskexit] and find_css_set	17 years ago
cgroup_freezer.c	freezer_cg: disable writing freezer.state of root cgroup	16 years ago
compat.c	Allow times and time system calls to return small negative values	16 years ago
configs.c	kernel/configs.c: remove useless comments	17 years ago
cpu.c	stop_machine/cpu hotplug: fix disable_nonboot_cpus	16 years ago
cpuset.c	cpuset: fix possible deadlock in async_rebuild_sched_domains	16 years ago
cred-internals.h	CRED: Inaugurate COW credentials	16 years ago
cred.c	Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6	16 years ago
delayacct.c	schedstat: consolidate per-task cpu runtime stats	16 years ago
dma-coherent.c	dma-coherent: Restore dma_alloc_from_coherent() large alloc fall back policy.	16 years ago
dma.c	kernel/dma.c: remove a CVS keyword	17 years ago
exec_domain.c	[CVE-2009-0029] System call wrappers part 04	16 years ago
exit.c	signal: re-add dead task accumulation stats.	16 years ago
extable.c	tracing/function-graph-tracer: drop the kernel_text_address check	16 years ago
fork.c	Merge branch 'timers-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip	16 years ago
freezer.c	freezer_cg: use thaw_process() in unfreeze_cgroup()	16 years ago
futex.c	futex: fix reference leak	16 years ago
futex_compat.c	CRED: Use RCU to access another task's creds and to release a task's own creds	16 years ago
hrtimer.c	hrtimer: prevent negative expiry value after clock_was_set()	16 years ago
itimer.c	timers: split process wide cpu clocks/timers	16 years ago
kallsyms.c	Revert "kbuild: strip generated symbols from *.ko"	16 years ago
kexec.c	PM: Split up sysdev_[suspend|resume] from device_power_[down|up]	16 years ago
kfifo.c	…
kgdb.c	kgdb: call touch_softlockup_watchdog on resume	17 years ago
kmod.c	kmod: fix varargs kernel-doc	16 years ago
kprobes.c	kprobes: check CONFIG_FREEZER instead of CONFIG_PM	16 years ago
ksysfs.c	kernel/ksysfs.c:fix dependence on CONFIG_NET	16 years ago
kthread.c	tracepoints: add DECLARE_TRACE() and DEFINE_TRACE()	16 years ago
latencytop.c	KSYM_SYMBOL_LEN fixes	16 years ago
lockdep.c	Merge branch 'core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip	16 years ago
lockdep_internals.h	…
lockdep_proc.c	lockstat: contend with points	17 years ago
marker.c	markers/tracpoints: fix non-modular build	16 years ago
module.c	tracing/function-graph-tracer: drop the kernel_text_address check	16 years ago
mutex-debug.c	…
mutex-debug.h	…
mutex.c	mutex: __used is needed for function referenced only from inline asm	16 years ago
mutex.h	…
notifier.c	Merge commit 'v2.6.28-rc6' into core/debug	16 years ago
ns_cgroup.c	ns_cgroup: remove unused spinlock	16 years ago
nsproxy.c	User namespaces: set of cleanups (v2)	16 years ago
panic.c	oops: increment the oops UUID every time we oops	16 years ago
params.c	Fix compile warning in kernel/params.c	17 years ago
pid.c	pid: generalize task_active_pid_ns	16 years ago
pid_namespace.c	…
pm_qos_params.c	…
posix-cpu-timers.c	timers: more consistently use clock vs timer	16 years ago
posix-timers.c	[CVE-2009-0029] System call wrappers part 05	16 years ago
printk.c	PM: Fix suspend_console and resume_console to use only one semaphore	16 years ago
profile.c	profiling: fix broken profiling regression	16 years ago
ptrace.c	[CVE-2009-0029] System call wrappers part 27	16 years ago
rcuclassic.c	rcu: add __cpuinit to rcu_init_percpu_data()	16 years ago
rcupdate.c	rcu: eliminate synchronize_rcu_xxx macro	16 years ago
rcupreempt.c	rcu: eliminate synchronize_rcu_xxx macro	16 years ago
rcupreempt_trace.c	"Tree RCU": scalable classic RCU implementation	16 years ago
rcutorture.c	rcu: fix bug in rcutorture system-shutdown code	16 years ago
rcutree.c	rcu: add __cpuinit to rcu_init_percpu_data()	16 years ago
rcutree_trace.c	"Tree RCU": scalable classic RCU implementation	16 years ago
relay.c	Merge branches 'tracing/ftrace', 'tracing/kmemtrace' and 'linus' into tracing/core	16 years ago
res_counter.c	memcg: memory cgroup resource counters for hierarchy	16 years ago
resource.c	resources: fix parameter name and kernel-doc	16 years ago
rtmutex-debug.c	…
rtmutex-debug.h	…
rtmutex-tester.c	…
rtmutex.c	hrtimer: convert kernel/* to the new hrtimer apis	17 years ago
rtmutex.h	…
rtmutex_common.h	…
rwsem.c	…
sched.c	Merge branch 'linus' into tracing/blktrace	16 years ago
sched_clock.c	Merge branch 'sched/clock' into tracing/ftrace	16 years ago
sched_cpupri.c	sched: fix section mismatch	16 years ago
sched_cpupri.h	sched: convert struct cpupri_vec cpumask_var_t.	16 years ago
sched_debug.c	sched: partly revert "sched debug: remove NULL checking in print_cfs_rt_rq()"	16 years ago
sched_fair.c	sched: revert recent sync wakeup changes	16 years ago
sched_features.h	sched: backward looking buddy	16 years ago
sched_idletask.c	sched: add CONFIG_SMP consistency	17 years ago
sched_rt.c	sched_rt: don't use first_cpu on cpumask created with cpumask_and	16 years ago
sched_stats.h	timers: split process wide cpu clocks/timers	16 years ago
seccomp.c	…
semaphore.c	…
signal.c	signal: re-add dead task accumulation stats.	16 years ago
smp.c	generic-ipi: use per cpu data for single cpu ipi calls	16 years ago
softirq.c	trace, lockdep: manual preempt count adding for local_bh_disable	16 years ago
softlockup.c	softlock: fix false panic which can occur if softlockup_thresh is reduced	16 years ago
spinlock.c	…
srcu.c	…
stacktrace.c	stacktrace: provide save_stack_trace_tsk() weak alias	16 years ago
stop_machine.c	stop_machine: introduce stop_machine_create/destroy.	16 years ago
sys.c	revert "rlimit: permit setting RLIMIT_NOFILE to RLIM_INFINITY"	16 years ago
sys_ni.c	[CVE-2009-0029] Make sys_syslog a conditional system call	16 years ago
sysctl.c	mm: fix dirty_bytes/dirty_background_bytes sysctls on 64bit arches	16 years ago
sysctl_check.c	[XFS] remove restricted chown parameter from xfs linux	16 years ago
taskstats.c	cpumask: convert rest of files in kernel/	16 years ago
test_kprobes.c	kprobes: add tests for register_kprobes	16 years ago
time.c	[CVE-2009-0029] System call wrappers part 01	16 years ago
timeconst.pl	…
timer.c	[CVE-2009-0029] System call wrappers part 27	16 years ago
tracepoint.c	markers/tracpoints: fix non-modular build	16 years ago
tsacct.c	mm: introduce get_mm_hiwater_xxx(), fix taskstats->hiwater_xxx accounting	16 years ago
uid16.c	[CVE-2009-0029] System call wrappers part 19	16 years ago
up.c	smp_call_function_single(): be slightly less stupid, fix #2	16 years ago
user.c	User namespaces: Only put the userns when we unhash the uid	16 years ago
user_namespace.c	User namespaces: set of cleanups (v2)	16 years ago
utsname.c	…
utsname_sysctl.c	sysctl: simplify ->strategy	17 years ago
wait.c	wait: prevent exclusive waiter starvation	16 years ago
workqueue.c	Merge branches 'tracing/ftrace', 'tracing/kmemtrace' and 'linus' into tracing/core	16 years ago