Borislav reported the following splat:
=============================
WARNING: suspicious RCU usage
4.19.0-rc1+ #1 Not tainted
-----------------------------
./include/linux/rcupdate.h:631 rcu_read_lock() used illegally while idle!
other info that might help us debug this:
RCU used illegally from idle CPU!
rcu_scheduler_active = 2, debug_locks = 1
RCU used illegally from extended quiescent state!
1 lock held by swapper/0/0:
#0: 000000004557ee0e (rcu_read_lock){....}, at: perf_event_output_forward+0x0/0x130
stack backtrace:
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.19.0-rc1+ #1
Hardware name: LENOVO 2320CTO/2320CTO, BIOS G2ET86WW (2.06 ) 11/13/2012
Call Trace:
dump_stack+0x85/0xcb
perf_event_output_forward+0xf6/0x130
__perf_event_overflow+0x52/0xe0
perf_swevent_overflow+0x91/0xb0
perf_tp_event+0x11a/0x350
? find_held_lock+0x2d/0x90
? __lock_acquire+0x2ce/0x1350
? __lock_acquire+0x2ce/0x1350
? retint_kernel+0x2d/0x2d
? find_held_lock+0x2d/0x90
? tick_nohz_get_sleep_length+0x83/0xb0
? perf_trace_cpu+0xbb/0xd0
? perf_trace_buf_alloc+0x5a/0xa0
perf_trace_cpu+0xbb/0xd0
cpuidle_enter_state+0x185/0x340
do_idle+0x1eb/0x260
cpu_startup_entry+0x5f/0x70
start_kernel+0x49b/0x4a6
secondary_startup_64+0xa4/0xb0
This is due to the tracepoints moving to SRCU usage which does not require
RCU to be "watching". But perf uses these tracepoints with RCU and expects
it to be. Hence, we still need to add in the rcu_irq_enter/exit_irqson()
calls for "rcuidle" tracepoints. This is a temporary fix until we have SRCU
working in NMI context, and then perf can be converted to use that instead
of normal RCU.
Link: http://lkml.kernel.org/r/20180904162611.6a120068@gandalf.local.home
Cc: x86-ml <x86@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Reported-by: Borislav Petkov <bp@alien8.de>
Tested-by: Borislav Petkov <bp@alien8.de>
Reviewed-by: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Fixes: e6753f23d9 ("tracepoint: Make rcuidle tracepoint callers use SRCU")
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
To avoid the need for relocating absolute references to tracepoint
structures at boot time when running relocatable kernels (which may
take a disproportionate amount of space), add the option to emit
these tables as relative references instead.
Link: http://lkml.kernel.org/r/20180704083651.24360-7-ard.biesheuvel@linaro.org
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Acked-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morris <james.morris@microsoft.com>
Cc: James Morris <jmorris@namei.org>
Cc: Jessica Yu <jeyu@kernel.org>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Nicolas Pitre <nico@linaro.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Russell King <linux@armlinux.org.uk>
Cc: "Serge E. Hallyn" <serge@hallyn.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: Thomas Garnier <thgarnie@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In recent tests with IRQ on/off tracepoints, a large performance
overhead ~10% is noticed when running hackbench. This is root caused to
calls to rcu_irq_enter_irqson and rcu_irq_exit_irqson from the
tracepoint code. Following a long discussion on the list [1] about this,
we concluded that srcu is a better alternative for use during rcu idle.
Although it does involve extra barriers, its lighter than the sched-rcu
version which has to do additional RCU calls to notify RCU idle about
entry into RCU sections.
In this patch, we change the underlying implementation of the
trace_*_rcuidle API to use SRCU. This has shown to improve performance
alot for the high frequency irq enable/disable tracepoints.
Test: Tested idle and preempt/irq tracepoints.
Here are some performance numbers:
With a run of the following 30 times on a single core x86 Qemu instance
with 1GB memory:
hackbench -g 4 -f 2 -l 3000
Completion times in seconds. CONFIG_PROVE_LOCKING=y.
No patches (without this series)
Mean: 3.048
Median: 3.025
Std Dev: 0.064
With Lockdep using irq tracepoints with RCU implementation:
Mean: 3.451 (-11.66 %)
Median: 3.447 (-12.22%)
Std Dev: 0.049
With Lockdep using irq tracepoints with SRCU implementation (this series):
Mean: 3.020 (I would consider the improvement against the "without
this series" case as just noise).
Median: 3.013
Std Dev: 0.033
[1] https://patchwork.kernel.org/patch/10344297/
[remove rcu_read_lock_sched_notrace as its the equivalent of
preempt_disable_notrace and is unnecessary to call in tracepoint code]
Link: http://lkml.kernel.org/r/20180730222423.196630-3-joel@joelfernandes.org
Cleaned-up-by: Peter Zijlstra <peterz@infradead.org>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
[ Simplified WARN_ON_ONCE() ]
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
The script:
./scripts/documentation-file-ref-check --fix
Gives multiple hints for broken references on some files.
Manually use the one that applies for some files.
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Acked-by: James Morris <james.morris@microsoft.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Acked-by: Jonathan Corbet <corbet@lwn.net>
Now that the irq path uses the rcu_nmi_{enter,exit}() algorithm,
rcu_irq_enter() and rcu_irq_exit() may be used from any context. There is
thus no need for rcu_irq_enter_disabled() and for the checks using it.
This commit therefore eliminates rcu_irq_enter_disabled().
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Perf has a problem that if sizeof() macros are used within TRACE_EVENT()
macro's they end up in userspace as "sizeof(kernel structure)" which
cannot properly be parsed. Add a macro which can forward this data
through the eval_map for userspace utilization.
Link: http://lkml.kernel.org/r/20170531215653.3240-10-jeremy.linton@arm.com
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Each enum is loaded into the trace_enum_map, as we
are now using this for more than enums rename it.
Link: http://lkml.kernel.org/r/20170531215653.3240-3-jeremy.linton@arm.com
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Stack tracing discovered that there's a small location inside the RCU
infrastructure where calling rcu_irq_enter() does not work. As trace events
use rcu_irq_enter() it must make sure that it is functionable. A check
against rcu_irq_enter_disabled() is added with a WARN_ON_ONCE() as no trace
event should ever be used in that part of RCU. If the warning is triggered,
then the trace event is ignored.
Restructure the __DO_TRACE() a bit to get rid of the prercu and postrcu,
and just have an rcucheck that does the work from within the _DO_TRACE()
macro. gcc optimization will compile out the rcucheck=0 case.
Link: http://lkml.kernel.org/r/20170405093207.404f8deb@gandalf.local.home
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Some tracepoints have a registration function that gets enabled when the
tracepoint is enabled. There may be cases that the registraction function
must fail (for example, can't allocate enough memory). In this case, the
tracepoint should also fail to register, otherwise the user would not know
why the tracepoint is not working.
Cc: David Howells <dhowells@redhat.com>
Cc: Seiji Aguchi <seiji.aguchi@hds.com>
Cc: Anton Blanchard <anton@samba.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Commit f37755490f ("tracepoints: Do not trace when cpu is offline") added
a check to make sure that tracepoints only get called when the cpu is
online, as it uses rcu_read_lock_sched() for protection.
Commit 3a630178fd ("tracing: generate RCU warnings even when tracepoints
are disabled") added lockdep checks (including rcu checks) for events that
are not enabled to catch possible RCU issues that would only be triggered if
a trace event was enabled. Commit f37755490f only stopped the warnings
when the trace event was enabled but did not prevent warnings if the trace
event was called when disabled.
To fix this, the cpu online check is moved to where the condition is added
to the trace event. This will place the cpu online check in all places that
it may be used now and in the future.
Cc: stable@vger.kernel.org # v3.18+
Fixes: f37755490f ("tracepoints: Do not trace when cpu is offline")
Fixes: 3a630178fd ("tracing: generate RCU warnings even when tracepoints are disabled")
Reported-by: Sudeep Holla <sudeep.holla@arm.com>
Tested-by: Sudeep Holla <sudeep.holla@arm.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The tracepoint infrastructure uses RCU sched protection to enable and
disable tracepoints safely. There are some instances where tracepoints are
used in infrastructure code (like kfree()) that get called after a CPU is
going offline, and perhaps when it is coming back online but hasn't been
registered yet.
This can probuce the following warning:
[ INFO: suspicious RCU usage. ]
4.4.0-00006-g0fe53e8-dirty #34 Tainted: G S
-------------------------------
include/trace/events/kmem.h:141 suspicious rcu_dereference_check() usage!
other info that might help us debug this:
RCU used illegally from offline CPU! rcu_scheduler_active = 1, debug_locks = 1
no locks held by swapper/8/0.
stack backtrace:
CPU: 8 PID: 0 Comm: swapper/8 Tainted: G S 4.4.0-00006-g0fe53e8-dirty #34
Call Trace:
[c0000005b76c78d0] [c0000000008b9540] .dump_stack+0x98/0xd4 (unreliable)
[c0000005b76c7950] [c00000000010c898] .lockdep_rcu_suspicious+0x108/0x170
[c0000005b76c79e0] [c00000000029adc0] .kfree+0x390/0x440
[c0000005b76c7a80] [c000000000055f74] .destroy_context+0x44/0x100
[c0000005b76c7b00] [c0000000000934a0] .__mmdrop+0x60/0x150
[c0000005b76c7b90] [c0000000000e3ff0] .idle_task_exit+0x130/0x140
[c0000005b76c7c20] [c000000000075804] .pseries_mach_cpu_die+0x64/0x310
[c0000005b76c7cd0] [c000000000043e7c] .cpu_die+0x3c/0x60
[c0000005b76c7d40] [c0000000000188d8] .arch_cpu_idle_dead+0x28/0x40
[c0000005b76c7db0] [c000000000101e6c] .cpu_startup_entry+0x50c/0x560
[c0000005b76c7ed0] [c000000000043bd8] .start_secondary+0x328/0x360
[c0000005b76c7f90] [c000000000008a6c] start_secondary_prolog+0x10/0x14
This warning is not a false positive either. RCU is not protecting code that
is being executed while the CPU is offline.
Instead of playing "whack-a-mole(TM)" and adding conditional statements to
the tracepoints we find that are used in this instance, simply add a
cpu_online() test to the tracepoint code where the tracepoint will be
ignored if the CPU is offline.
Use of raw_smp_processor_id() is fine, as there should never be a case where
the tracepoint code goes from running on a CPU that is online and suddenly
gets migrated to a CPU that is offline.
Link: http://lkml.kernel.org/r/1455387773-4245-1-git-send-email-kda@linux-powerpc.org
Reported-by: Denis Kirjanov <kda@linux-powerpc.org>
Fixes: 97e1c18e8d ("tracing: Kernel Tracepoints")
Cc: stable@vger.kernel.org # v2.6.28+
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
minor fixes.
Here's what else is new:
o A new TRACE_EVENT_FN_COND macro, combining both _FN and _COND for
those that want both.
o New selftest to test the instance create and delete
o Better debug output when ftrace fails
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJWlU8tAAoJEKKk/i67LK/8JckH/2XIhjwMunm35uCg1308sDqy
d44G3+p0pm8ztjBf8iD8wH2nP3m7z+nC8JBmSPIUgAHsKOYHWsBy2A/36OVWv5lK
1hVXvBwOuZXnyWXr7bC2RO9S9f9acSFaabZXWDi1BCJRJSgEcknz32V7ZAL4jOCO
SfBWBNrWJfUsURbfbElfVxPLArvyUg9Bb5dW5B+QFf6PuoJaORYzNLYXHlbsq++T
WlrlnD+mFZ/DKFZ/gl3FMSGMPaGimw09/3eqMzv/tLQobp6PbCWlJTwjUoxJ/9dO
XOY4sWUrUUZilU8qCk0i0ZSEumWmE+SWS3eq+Ef18B/5haIj/LkoM4UQD3h2Rc4=
=FDR+
-----END PGP SIGNATURE-----
Merge tag 'trace-v4.5' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing updates from Steven Rostedt:
"Not much new with tracing for this release. Mostly just clean ups and
minor fixes.
Here's what else is new:
- A new TRACE_EVENT_FN_COND macro, combining both _FN and _COND for
those that want both.
- New selftest to test the instance create and delete
- Better debug output when ftrace fails"
* tag 'trace-v4.5' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (24 commits)
ftrace: Fix the race between ftrace and insmod
ftrace: Add infrastructure for delayed enabling of module functions
x86: ftrace: Fix the comments for ftrace_modify_code_direct()
tracing: Fix comment to use tracing_on over tracing_enable
metag: ftrace: Fix the comments for ftrace_modify_code
sh: ftrace: Fix the comments for ftrace_modify_code()
ia64: ftrace: Fix the comments for ftrace_modify_code()
ftrace: Clean up ftrace_module_init() code
ftrace: Join functions ftrace_module_init() and ftrace_init_module()
tracing: Introduce TRACE_EVENT_FN_COND macro
tracing: Use seq_buf_used() in seq_buf_to_user() instead of len
bpf: Constify bpf_verifier_ops structure
ftrace: Have ftrace_ops_get_func() handle RCU and PER_CPU flags too
ftrace: Remove use of control list and ops
ftrace: Fix output of enabled_functions for showing tramp
ftrace: Fix a typo in comment
ftrace: Show all tramps registered to a record on ftrace_bug()
ftrace: Add variable ftrace_expected for archs to show expected code
ftrace: Add new type to distinguish what kind of ftrace_bug()
tracing: Update cond flag when enabling or disabling a trigger
...
Pull perf updates from Ingo Molnar:
"Kernel side changes:
- Intel Knights Landing support. (Harish Chegondi)
- Intel Broadwell-EP uncore PMU support. (Kan Liang)
- Core code improvements. (Peter Zijlstra.)
- Event filter, LBR and PEBS fixes. (Stephane Eranian)
- Enable cycles:pp on Intel Atom. (Stephane Eranian)
- Add cycles:ppp support for Skylake. (Andi Kleen)
- Various x86 NMI overhead optimizations. (Andi Kleen)
- Intel PT enhancements. (Takao Indoh)
- AMD cache events fix. (Vince Weaver)
Tons of tooling changes:
- Show random perf tool tips in the 'perf report' bottom line
(Namhyung Kim)
- perf report now defaults to --group if the perf.data file has
grouped events, try it with:
# perf record -e '{cycles,instructions}' -a sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 1.093 MB perf.data (1247 samples) ]
# perf report
# Samples: 1K of event 'anon group { cycles, instructions }'
# Event count (approx.): 1955219195
#
# Overhead Command Shared Object Symbol
2.86% 0.22% swapper [kernel.kallsyms] [k] intel_idle
1.05% 0.33% firefox libxul.so [.] js::SetObjectElement
1.05% 0.00% kworker/0:3 [kernel.kallsyms] [k] gen6_ring_get_seqno
0.88% 0.17% chrome chrome [.] 0x0000000000ee27ab
0.65% 0.86% firefox libxul.so [.] js::ValueToId<(js::AllowGC)1>
0.64% 0.23% JS Helper libxul.so [.] js::SplayTree<js::jit::LiveRange*, js::jit::LiveRange>::splay
0.62% 1.27% firefox libxul.so [.] js::GetIterator
0.61% 1.74% firefox libxul.so [.] js::NativeSetProperty
0.61% 0.31% firefox libxul.so [.] js::SetPropertyByDefining
- Introduce the 'perf stat record/report' workflow:
Generate perf.data files from 'perf stat', to tap into the
scripting capabilities perf has instead of defining a 'perf stat'
specific scripting support to calculate event ratios, etc.
Simple example:
$ perf stat record -e cycles usleep 1
Performance counter stats for 'usleep 1':
1,134,996 cycles
0.000670644 seconds time elapsed
$ perf stat report
Performance counter stats for '/home/acme/bin/perf stat record -e cycles usleep 1':
1,134,996 cycles
0.000670644 seconds time elapsed
$
It generates PERF_RECORD_ userspace records to store the details:
$ perf report -D | grep PERF_RECORD
0xf0 [0x28]: PERF_RECORD_THREAD_MAP nr: 1 thread: 27637
0x118 [0x12]: PERF_RECORD_CPU_MAP nr: 1 cpu: 65535
0x12a [0x40]: PERF_RECORD_STAT_CONFIG
0x16a [0x30]: PERF_RECORD_STAT
-1 -1 0x19a [0x40]: PERF_RECORD_MMAP -1/0: [0xffffffff81000000(0x1f000000) @ 0xffffffff81000000]: x [kernel.kallsyms]_text
0x1da [0x18]: PERF_RECORD_STAT_ROUND
[acme@ssdandy linux]$
An effort was made to make perf.data files generated like this to
not generate cryptic messages when processed by older tools.
The 'perf script' bits need rebasing, will go up later.
- Make command line options always available, even when they depend
on some feature being enabled, warning the user about use of such
options (Wang Nan)
- Support hw breakpoint events (mem:0xAddress) in the default output
mode in 'perf script' (Wang Nan)
- Fixes and improvements for supporting annotating ARM binaries,
support ARM call and jump instructions, more work needed to have
arch specific stuff separated into tools/perf/arch/*/annotate/
(Russell King)
- Add initial 'perf config' command, for now just with a --list
command to the contents of the configuration file in use and a
basic man page describing its format, commands for doing edits and
detailed documentation are being reviewed and proof-read. (Taeung
Song)
- Allows BPF scriptlets specify arguments to be fetched using DWARF
info, using a prologue generated at compile/build time (He Kuang,
Wang Nan)
- Allow attaching BPF scriptlets to module symbols (Wang Nan)
- Allow attaching BPF scriptlets to userspace code using uprobe (Wang
Nan)
- BPF programs now can specify 'perf probe' tunables via its section
name, separating key=val values using semicolons (Wang Nan)
Testing some of these new BPF features:
Use case: get callchains when receiving SSL packets, filter then in the
kernel, at arbitrary place.
# cat ssl.bpf.c
#define SEC(NAME) __attribute__((section(NAME), used))
struct pt_regs;
SEC("func=__inet_lookup_established hnum")
int func(struct pt_regs *ctx, int err, unsigned short port)
{
return err == 0 && port == 443;
}
char _license[] SEC("license") = "GPL";
int _version SEC("version") = LINUX_VERSION_CODE;
#
# perf record -a -g -e ssl.bpf.c
^C[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.787 MB perf.data (3 samples) ]
# perf script | head -30
swapper 0 [000] 58783.268118: perf_bpf_probe:func: (ffffffff816a0f60) hnum=0x1bb
8a0f61 __inet_lookup_established (/lib/modules/4.3.0+/build/vmlinux)
896def ip_rcv_finish (/lib/modules/4.3.0+/build/vmlinux)
8976c2 ip_rcv (/lib/modules/4.3.0+/build/vmlinux)
855eba __netif_receive_skb_core (/lib/modules/4.3.0+/build/vmlinux)
8565d8 __netif_receive_skb (/lib/modules/4.3.0+/build/vmlinux)
8572a8 process_backlog (/lib/modules/4.3.0+/build/vmlinux)
856b11 net_rx_action (/lib/modules/4.3.0+/build/vmlinux)
2a284b __do_softirq (/lib/modules/4.3.0+/build/vmlinux)
2a2ba3 irq_exit (/lib/modules/4.3.0+/build/vmlinux)
96b7a4 do_IRQ (/lib/modules/4.3.0+/build/vmlinux)
969807 ret_from_intr (/lib/modules/4.3.0+/build/vmlinux)
2dede5 cpu_startup_entry (/lib/modules/4.3.0+/build/vmlinux)
95d5bc rest_init (/lib/modules/4.3.0+/build/vmlinux)
1163ffa start_kernel ([kernel.vmlinux].init.text)
11634d7 x86_64_start_reservations ([kernel.vmlinux].init.text)
1163623 x86_64_start_kernel ([kernel.vmlinux].init.text)
qemu-system-x86 9178 [003] 58785.792417: perf_bpf_probe:func: (ffffffff816a0f60) hnum=0x1bb
8a0f61 __inet_lookup_established (/lib/modules/4.3.0+/build/vmlinux)
896def ip_rcv_finish (/lib/modules/4.3.0+/build/vmlinux)
8976c2 ip_rcv (/lib/modules/4.3.0+/build/vmlinux)
855eba __netif_receive_skb_core (/lib/modules/4.3.0+/build/vmlinux)
8565d8 __netif_receive_skb (/lib/modules/4.3.0+/build/vmlinux)
856660 netif_receive_skb_internal (/lib/modules/4.3.0+/build/vmlinux)
8566ec netif_receive_skb_sk (/lib/modules/4.3.0+/build/vmlinux)
430a br_handle_frame_finish ([bridge])
48bc br_handle_frame ([bridge])
855f44 __netif_receive_skb_core (/lib/modules/4.3.0+/build/vmlinux)
8565d8 __netif_receive_skb (/lib/modules/4.3.0+/build/vmlinux)
#
- Use 'perf probe' various options to list functions, see what
variables can be collected at any given point, experiment first
collecting without a filter, then filter, use it together with
'perf trace', 'perf top', with or without callchains, if it
explodes, please tell us!
- Introduce a new callchain mode: "folded", that will list per line
representations of all callchains for a give histogram entry,
facilitating 'perf report' output processing by other tools, such
as Brendan Gregg's flamegraph tools (Namhyung Kim)
E.g:
# perf report | grep -v ^# | head
18.37% 0.00% swapper [kernel.kallsyms] [k] cpu_startup_entry
|
---cpu_startup_entry
|
|--12.07%--start_secondary
|
--6.30%--rest_init
start_kernel
x86_64_start_reservations
x86_64_start_kernel
#
Becomes, in "folded" mode:
# perf report -g folded | grep -v ^# | head -5
18.37% 0.00% swapper [kernel.kallsyms] [k] cpu_startup_entry
12.07% cpu_startup_entry;start_secondary
6.30% cpu_startup_entry;rest_init;start_kernel;x86_64_start_reservations;x86_64_start_kernel
16.90% 0.00% swapper [kernel.kallsyms] [k] call_cpuidle
11.23% call_cpuidle;cpu_startup_entry;start_secondary
5.67% call_cpuidle;cpu_startup_entry;rest_init;start_kernel;x86_64_start_reservations;x86_64_start_kernel
16.90% 0.00% swapper [kernel.kallsyms] [k] cpuidle_enter
11.23% cpuidle_enter;call_cpuidle;cpu_startup_entry;start_secondary
5.67% cpuidle_enter;call_cpuidle;cpu_startup_entry;rest_init;start_kernel;x86_64_start_reservations;x86_64_start_kernel
15.12% 0.00% swapper [kernel.kallsyms] [k] cpuidle_enter_state
#
The user can also select one of "count", "period" or "percent" as
the first column.
... and lots of infrastructure enhancements, plus fixes and other
changes, features I failed to list - see the shortlog and the git log
for details"
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (271 commits)
perf evlist: Add --trace-fields option to show trace fields
perf record: Store data mmaps for dwarf unwind
perf libdw: Check for mmaps also in MAP__VARIABLE tree
perf unwind: Check for mmaps also in MAP__VARIABLE tree
perf unwind: Use find_map function in access_dso_mem
perf evlist: Remove perf_evlist__(enable|disable)_event functions
perf evlist: Make perf_evlist__open() open evsels with their cpus and threads (like perf record does)
perf report: Show random usage tip on the help line
perf hists: Export a couple of hist functions
perf diff: Use perf_hpp__register_sort_field interface
perf tools: Add overhead/overhead_children keys defaults via string
perf tools: Remove list entry from struct sort_entry
perf tools: Include all tools/lib directory for tags/cscope/TAGS targets
perf script: Align event name properly
perf tools: Add missing headers in perf's MANIFEST
perf tools: Do not show trace command if it's not compiled in
perf report: Change default to use event group view
perf top: Decay periods in callchains
tools lib: Move bitmap.[ch] from tools/perf/ to tools/{lib,include}/
tools lib: Sync tools/lib/find_bit.c with the kernel
...
TRACE_EVENT_FN can't be used in some circumstances
like invoking trace functions from offlined CPU due
to RCU usage.
This patch adds the TRACE_EVENT_FN_COND macro
to make such trace points conditional.
Link: http://lkml.kernel.org/r/1450124286-4822-1-git-send-email-kda@linux-powerpc.org
Signed-off-by: Denis Kirjanov <kda@linux-powerpc.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
This commit replaces a local_irq_save()/local_irq_restore() pair with
a lockdep assertion that interrupts are already disabled. This should
remove the corresponding overhead from the interrupt entry/exit fastpaths.
This change was inspired by the fact that Iftekhar Ahmed's mutation
testing showed that removing rcu_irq_enter()'s call to local_ird_restore()
had no effect, which might indicate that interrupts were always enabled
anyway.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Steven recommended open coding access to tracepoint->key to add
trace points to headers. Unfortunately this is difficult for some
headers (such as x86 asm/msr.h) because including tracepoint.h
includes so many other headers that it causes include loops.
The main problem is the include of linux/rcupdate.h, which
pulls in a lot of other headers. The rcu header is only needed
when actually defining trace points.
Move the struct tracepoint into a separate tracepoint-defs.h
header that can be included without pulling in all of RCU.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Link: http://lkml.kernel.org/r/1449018060-1742-2-git-send-email-andi@firstfloor.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The documentation on top of __DECLARE_TRACE() does not match its
implementation since the condition check has been added to the
RCU lockdep checks. Update the documentation to match its
implementation.
Link: http://lkml.kernel.org/r/1446504164-21563-1-git-send-email-mathieu.desnoyers@efficios.com
CC: Dave Hansen <dave@sr71.net>
Fixes: a05d59a567 "tracing: Add condition check to RCU lockdep checks"
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
In order to guarantee that a probe will be called before other probes that
are attached to a tracepoint, there needs to be a mechanism to provide
priority of one probe over the others.
Adding a prio field to the struct tracepoint_func, which lets the probes be
sorted by the priority set in the structure. If no priority is specified,
then a priority of 10 is given (this is a macro, and perhaps may be changed
in the future).
Now probes may be added to affect other probes that are attached to a
tracepoint with a guaranteed order.
One use case would be to allow tracing of tracepoints be able to filter by
pid. A special (higher priority probe) may be added to the sched_switch
tracepoint and set the necessary flags of the other tracepoints to notify
them if they should be traced or not. In case a tracepoint is enabled at the
sched_switch tracepoint too, the order of the two are not random.
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Allow a trace events header file to disable compilation of its
trace events by defining the preprocessor macro NOTRACE.
This could be done, for example, according to a Kconfig option.
Link: http://lkml.kernel.org/r/1438432079-11704-3-git-send-email-tal.shorer@gmail.com
Signed-off-by: Tal Shorer <tal.shorer@gmail.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Several tracepoints use the helper functions __print_symbolic() or
__print_flags() and pass in enums that do the mapping between the
binary data stored and the value to print. This works well for reading
the ASCII trace files, but when the data is read via userspace tools
such as perf and trace-cmd, the conversion of the binary value to a
human string format is lost if an enum is used, as userspace does not
have access to what the ENUM is.
For example, the tracepoint trace_tlb_flush() has:
__print_symbolic(REC->reason,
{ TLB_FLUSH_ON_TASK_SWITCH, "flush on task switch" },
{ TLB_REMOTE_SHOOTDOWN, "remote shootdown" },
{ TLB_LOCAL_SHOOTDOWN, "local shootdown" },
{ TLB_LOCAL_MM_SHOOTDOWN, "local mm shootdown" })
Which maps the enum values to the strings they represent. But perf and
trace-cmd do no know what value TLB_LOCAL_MM_SHOOTDOWN is, and would
not be able to map it.
With TRACE_DEFINE_ENUM(), developers can place these in the event header
files and ftrace will convert the enums to their values:
By adding:
TRACE_DEFINE_ENUM(TLB_FLUSH_ON_TASK_SWITCH);
TRACE_DEFINE_ENUM(TLB_REMOTE_SHOOTDOWN);
TRACE_DEFINE_ENUM(TLB_LOCAL_SHOOTDOWN);
TRACE_DEFINE_ENUM(TLB_LOCAL_MM_SHOOTDOWN);
$ cat /sys/kernel/debug/tracing/events/tlb/tlb_flush/format
[...]
__print_symbolic(REC->reason,
{ 0, "flush on task switch" },
{ 1, "remote shootdown" },
{ 2, "local shootdown" },
{ 3, "local mm shootdown" })
The above is what userspace expects to see, and tools do not need to
be modified to parse them.
Link: http://lkml.kernel.org/r/20150403013802.220157513@goodmis.org
Cc: Guilherme Cox <cox@computer.org>
Cc: Tony Luck <tony.luck@gmail.com>
Cc: Xie XiuQi <xiexiuqi@huawei.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Reviewed-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Tested-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The trace_tlb_flush() tracepoint can be called when a CPU is going offline.
When a CPU is offline, RCU is no longer watching that CPU and since the
tracepoint is protected by RCU, it must not be called. To prevent the
tlb_flush tracepoint from being called when the CPU is offline, it was
converted to a TRACE_EVENT_CONDITION where the condition checks if the
CPU is online before calling the tracepoint.
Unfortunately, this was not enough to stop lockdep from complaining about
it. Even though the RCU protected code of the tracepoint will never be
called, the condition is hidden within the tracepoint, and even though the
condition prevents RCU code from being called, the lockdep checks are
outside the tracepoint (this is to test tracepoints even when they are not
enabled).
Even though tracepoints should be checked to be RCU safe when they are not
enabled, the condition should still be considered when checking RCU.
Link: http://lkml.kernel.org/r/CA+icZUUGiGDoL5NU8RuxKzFjoLjEKRtUWx=JB8B9a0EQv-eGzQ@mail.gmail.com
Fixes: 3a630178fd "tracing: generate RCU warnings even when tracepoints are disabled"
Cc: stable@vger.kernel.org # 3.18+
Acked-by: Dave Hansen <dave@sr71.net>
Reported-by: Sedat Dilek <sedat.dilek@gmail.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Dave Jones reported seeing a bug from one of my TLB tracepoints:
http://lkml.kernel.org/r/20140806181801.GA4605@redhat.com
I've been running these patches for months and never saw this.
But, a big chunk of my testing, especially with all the debugging
enabled, was in a vm where intel_idle doesn't work. On the
systems where I was using intel_idle, I never had lockdep enabled
and this tracepoint on at the same time.
This patch ensures that whenever we have lockdep available, we do
_some_ RCU activity at the site of the tracepoint, despite
whether the tracepoint's condition matches or even if the
tracepoint itself is completely disabled. This is a bit of a
hack, but it is pretty self-contained.
I confirmed that with this patch plus lockdep I get the same
splat as Dave Jones did, but without enabling the tracepoint
explicitly.
Link: http://lkml.kernel.org/p/20140807175204.C257CAC5@viggo.jf.intel.com
Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Dave Hansen <dave@sr71.net>
Cc: Dave Jones <davej@redhat.com>,
Cc: paulmck@linux.vnet.ibm.com
Cc: Ingo Molnar <mingo@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
When CONFIG_TRACING is not enabled, there's no reason to save the trace
strings either by the linker or as a static variable that can be
referenced later. Simply pass back the string that is given to
tracepoint_string().
Had to move the define to include/linux/tracepoint.h so that it is still
visible when CONFIG_TRACING is not set.
Link: http://lkml.kernel.org/p/1406318733-26754-2-git-send-email-nicolas.pitre@linaro.org
Suggested-by: Nicolas Pitre <nico@linaro.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
There are some code paths in the kernel that need to do some preparations
before it calls a tracepoint. As that code is worthless overhead when
the tracepoint is not enabled, it would be prudent to have that code
only run when the tracepoint is active. To accomplish this, all tracepoints
now get a static inline function called "trace_<tracepoint-name>_enabled()"
which returns true when the tracepoint is enabled and false otherwise.
As an added bonus, that function uses the static_key of the tracepoint
such that no branch is needed.
if (trace_mytracepoint_enabled()) {
arg = process_tp_arg();
trace_mytracepoint(arg);
}
Will keep the "process_tp_arg()" (which may be expensive to run) from
being executed when the tracepoint isn't enabled.
It's best to encapsulate the tracepoint itself in the if statement
just to keep races. For example, if you had:
if (trace_mytracepoint_enabled())
arg = process_tp_arg();
trace_mytracepoint(arg);
There's a chance that the tracepoint could be enabled just after the
if statement, and arg will be undefined when calling the tracepoint.
Link: http://lkml.kernel.org/r/20140506094407.507b6435@gandalf.local.home
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Fix the following sparse warnings:
CHECK kernel/tracepoint.c
kernel/tracepoint.c:184:18: warning: incorrect type in assignment (different address spaces)
kernel/tracepoint.c:184:18: expected struct tracepoint_func *tp_funcs
kernel/tracepoint.c:184:18: got struct tracepoint_func [noderef] <asn:4>*funcs
kernel/tracepoint.c:216:18: warning: incorrect type in assignment (different address spaces)
kernel/tracepoint.c:216:18: expected struct tracepoint_func *tp_funcs
kernel/tracepoint.c:216:18: got struct tracepoint_func [noderef] <asn:4>*funcs
kernel/tracepoint.c:392:24: error: return expression in void function
CC kernel/tracepoint.o
kernel/tracepoint.c: In function tracepoint_module_going:
kernel/tracepoint.c:491:6: warning: symbol 'syscall_regfunc' was not declared. Should it be static?
kernel/tracepoint.c:508:6: warning: symbol 'syscall_unregfunc' was not declared. Should it be static?
Link: http://lkml.kernel.org/r/1397049883-28692-1-git-send-email-mathieu.desnoyers@efficios.com
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Instead of copying the num_tracepoints and tracepoints_ptrs from
the module structure to the tp_mod structure, which only uses it to
find the module associated to tracepoints of modules that are coming
and going, simply copy the pointer to the module struct to the tracepoint
tp_module structure.
Also removed un-needed brackets around an if statement.
Link: http://lkml.kernel.org/r/20140408201705.4dad2c4a@gandalf.local.home
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Register/unregister tracepoint probes with struct tracepoint pointer
rather than tracepoint name.
This change, which vastly simplifies tracepoint.c, has been proposed by
Steven Rostedt. It also removes 8.8kB (mostly of text) to the vmlinux
size.
From this point on, the tracers need to pass a struct tracepoint pointer
to probe register/unregister. A probe can now only be connected to a
tracepoint that exists. Moreover, tracers are responsible for
unregistering the probe before the module containing its associated
tracepoint is unloaded.
text data bss dec hex filename
10443444 4282528 10391552 25117524 17f4354 vmlinux.orig
10434930 4282848 10391552 25109330 17f2352 vmlinux
Link: http://lkml.kernel.org/r/1396992381-23785-2-git-send-email-mathieu.desnoyers@efficios.com
CC: Ingo Molnar <mingo@kernel.org>
CC: Frederic Weisbecker <fweisbec@gmail.com>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: Frank Ch. Eigler <fche@redhat.com>
CC: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
[ SDR - fixed return val in void func in tracepoint_module_going() ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
But there were a few features that were added.
Uprobes now work with event triggers and multi buffers.
Uprobes have support under ftrace and perf.
The big feature is that the function tracer can now be used within the
multi buffer instances. That is, you can now trace some functions
in one buffer, others in another buffer, all functions in a third buffer
and so on. They are basically agnostic from each other. This only
works for the function tracer and not for the function graph trace,
although you can have the function graph tracer running in the top level
buffer (or any tracer for that matter) and have different function tracing
going on in the sub buffers.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJTOthtAAoJEKQekfcNnQGu5c8H/Ana/U+0tmksp1dbHkRHsKSH
+Fsv4Jeu8gf1NaFKHEhkUTcFtnzE6qAPV2VCrcJwXbhAhhwZm+LjrnWdoy3215S3
cQW4LftLEonh2cM36Cos74TulMEYN6XmL6dQZV+CILKQkDrWU4qJjQ64okXEkqrd
9iG3p/mSXyvJcmnyg61ALnMOhZDLsXY3djBhWBPhiTPGS6BRb9zh4Pmw6Zv0n2rJ
U93Gt/3AQrv1ybu73dUxqP0abp60oXOiWoF/R2jcbKqIM+K9RPJX79unCV3jq3u9
f+6jMlB9PgAMqQj6ihJdwxKDDuzwyrVdEPnsgvl4jarCBCtVVwhKedBaKN/KS8k=
=HdXY
-----END PGP SIGNATURE-----
Merge tag 'trace-3.15' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing updates from Steven Rostedt:
"Most of the changes were largely clean ups, and some documentation.
But there were a few features that were added:
Uprobes now work with event triggers and multi buffers and have
support under ftrace and perf.
The big feature is that the function tracer can now be used within the
multi buffer instances. That is, you can now trace some functions in
one buffer, others in another buffer, all functions in a third buffer
and so on. They are basically agnostic from each other. This only
works for the function tracer and not for the function graph trace,
although you can have the function graph tracer running in the top
level buffer (or any tracer for that matter) and have different
function tracing going on in the sub buffers"
* tag 'trace-3.15' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (45 commits)
tracing: Add BUG_ON when stack end location is over written
tracepoint: Remove unused API functions
Revert "tracing: Move event storage for array from macro to standalone function"
ftrace: Constify ftrace_text_reserved
tracepoints: API doc update to tracepoint_probe_register() return value
tracepoints: API doc update to data argument
ftrace: Fix compilation warning about control_ops_free
ftrace/x86: BUG when ftrace recovery fails
ftrace: Warn on error when modifying ftrace function
ftrace: Remove freelist from struct dyn_ftrace
ftrace: Do not pass data to ftrace_dyn_arch_init
ftrace: Pass retval through return in ftrace_dyn_arch_init()
ftrace: Inline the code from ftrace_dyn_table_alloc()
ftrace: Cleanup of global variables ftrace_new_pgs and ftrace_update_cnt
tracing: Evaluate len expression only once in __dynamic_array macro
tracing: Correctly expand len expressions from __dynamic_array macro
tracing/module: Replace include of tracepoint.h with jump_label.h in module.h
tracing: Fix event header migrate.h to include tracepoint.h
tracing: Fix event header writeback.h to include tracepoint.h
tracing: Warn if a tracepoint is not set via debugfs
...
After the following commit:
commit b75ef8b44b
Author: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Date: Wed Aug 10 15:18:39 2011 -0400
Tracepoint: Dissociate from module mutex
The following functions became unnecessary:
- tracepoint_probe_register_noupdate,
- tracepoint_probe_unregister_noupdate,
- tracepoint_probe_update_all.
In fact, none of the in-kernel tracers, nor LTTng, nor SystemTAP use
them. Remove those.
Moreover, the functions:
- tracepoint_iter_start,
- tracepoint_iter_next,
- tracepoint_iter_stop,
- tracepoint_iter_reset.
are unused by in-kernel tracers, LTTng and SystemTAP. Remove those too.
Link: http://lkml.kernel.org/r/1395379142-2118-2-git-send-email-mathieu.desnoyers@efficios.com
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
If a module fails to add its tracepoints due to module tainting, do not
create the module event infrastructure in the debugfs directory. As the events
will not work and worse yet, they will silently fail, making the user wonder
why the events they enable do not display anything.
Having a warning on module load and the events not visible to the users
will make the cause of the problem much clearer.
Link: http://lkml.kernel.org/r/20140227154923.265882695@goodmis.org
Fixes: 6d723736e4 "tracing/events: add support for modules to TRACE_EVENT"
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: stable@vger.kernel.org # 2.6.31+
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Binary was written as binay, probably by mistake. Fix it.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Vince's perf-trinity fuzzer found yet another 'interesting' problem.
When we sample the irq_work_exit tracepoint with period==1 (or
PERF_SAMPLE_PERIOD) and we add an fasync SIGNAL handler we create an
infinite event generation loop:
,-> <IPI>
| irq_work_exit() ->
| trace_irq_work_exit() ->
| ...
| __perf_event_overflow() -> (due to fasync)
| irq_work_queue() -> (irq_work_list must be empty)
'--------- arch_irq_work_raise()
Similar things can happen due to regular poll() wakeups if we exceed
the ring-buffer wakeup watermark, or have an event_limit.
To avoid this, dis-allow sampling this particular tracepoint.
In order to achieve this, create a special perf_perm function pointer
for each event and call this (when set) on trying to create a
tracepoint perf event.
[ roasted: use expr... to allow for ',' in your expression ]
Reported-by: Vince Weaver <vincent.weaver@maine.edu>
Tested-by: Vince Weaver <vincent.weaver@maine.edu>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Dave Jones <davej@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/20131114152304.GC5364@laptop.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Each TRACE_EVENT() adds several helper functions. If two or more trace events
share the same structure and print format, they can also share most of these
helper functions and save a lot of space from duplicate code. This is why the
DECLARE_EVENT_CLASS() and DEFINE_EVENT() were created.
Some events require a trigger to be called at registering and unregistering of
the event and to do so they use TRACE_EVENT_FN().
If multiple events require a trigger, they currently have no choice but to use
TRACE_EVENT_FN() as there's no DEFINE_EVENT_FN() available. This unfortunately
causes a lot of wasted duplicate code created.
By adding a DEFINE_EVENT_FN(), these events can still use a
DECLARE_EVENT_CLASS() and then define their own triggers.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/51C3236C.8030508@hds.com
Signed-off-by: Seiji Aguchi <seiji.aguchi@hds.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
__DECLARE_TRACE_RCU() currently creates an _rcuidle() tracepoint which
may safely be invoked from what RCU considers to be an idle CPU.
However, these _rcuidle() tracepoints may -not- be invoked from the
handler of an irq taken from idle, because rcu_idle_enter() zeroes
RCU's nesting-level counter, so that the rcu_irq_exit() returning to
idle will trigger a WARN_ON_ONCE().
This commit therefore substitutes rcu_irq_enter() for rcu_idle_exit()
and rcu_irq_exit() for rcu_idle_enter() in order to make the _rcuidle()
tracepoints usable from irq handlers as well as from process context.
Reported-by: Dave Jones <davej@redhat.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Tracepoints declare a static inline trace_*_rcuidle variant of the trace
function, to support safely generating trace events from the idle loop.
Module code never actually uses that variant of trace functions, because
modules don't run code that needs tracing with RCU idled. However, the
declaration of those otherwise unused functions causes the module to
reference rcu_idle_exit and rcu_idle_enter, which RCU does not export to
modules.
To avoid this, don't generate trace_*_rcuidle functions for tracepoints
declared in module code.
Link: http://lkml.kernel.org/r/20120905062306.GA14756@leaf
Reported-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
So here's a boot tested patch on top of Jason's series that does
all the cleanups I talked about and turns jump labels into a
more intuitive to use facility. It should also address the
various misconceptions and confusions that surround jump labels.
Typical usage scenarios:
#include <linux/static_key.h>
struct static_key key = STATIC_KEY_INIT_TRUE;
if (static_key_false(&key))
do unlikely code
else
do likely code
Or:
if (static_key_true(&key))
do likely code
else
do unlikely code
The static key is modified via:
static_key_slow_inc(&key);
...
static_key_slow_dec(&key);
The 'slow' prefix makes it abundantly clear that this is an
expensive operation.
I've updated all in-kernel code to use this everywhere. Note
that I (intentionally) have not pushed through the rename
blindly through to the lowest levels: the actual jump-label
patching arch facility should be named like that, so we want to
decouple jump labels from the static-key facility a bit.
On non-jump-label enabled architectures static keys default to
likely()/unlikely() branches.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Jason Baron <jbaron@redhat.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: a.p.zijlstra@chello.nl
Cc: mathieu.desnoyers@efficios.com
Cc: davem@davemloft.net
Cc: ddaney.cavm@gmail.com
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/20120222085809.GA26397@elte.hu
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Added is a new static inline function that lets *any* tracepoint be used
inside a rcu_idle_exit() section. And this also solves the problem where
the same tracepoint may be used inside a rcu_idle_exit() section as well
as outside of one.
I added a new tracepoint function with a "_rcuidle" extension. All
tracepoints can be used with either the normal "trace_foobar()"
function, or the "trace_foobar_rcuidle()" function when inside a
rcu_idle_exit() section.
All tracepoints defined by TRACE_EVENT() or any of the derivatives
will have a "_rcuidle()" function also defined. When a tracepoint is
used within an rcu_idle_exit() section, the "_rcuidle()" version must
be used. This denotes that the tracepoint is within rcu_idle_exit()
and it allows the rcu read locks within the tracepoint to still
be valid, as this version takes us out of rcu_idle_exit().
Another nice aspect about this patch is that "static inline"s are not
compiled into text when not used. So only the tracepoints that actually
use the _rcuidle() version will have them defined in the actual text
that is booted.
Link: http://lkml.kernel.org/r/1328563113.2200.39.camel@gandalf.stny.rr.com>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Copy the information needed from struct module into a local module list
held within tracepoint.c from within the module coming/going notifier.
This vastly simplifies locking of tracepoint registration /
unregistration, because we don't have to take the module mutex to
register and unregister tracepoints anymore. Steven Rostedt ran into
dependency problems related to modules mutex vs kprobes mutex vs ftrace
mutex vs tracepoint mutex that seems to be hard to fix without removing
this dependency between tracepoint and module mutex. (note: it should be
investigated whether kprobes could benefit of being dissociated from the
modules mutex too.)
This also fixes module handling of tracepoint list iterators, because it
was expecting the list to be sorted by pointer address. Given we have
control on our own list now, it's OK to sort this list which has
tracepoints as its only purpose. The reason why this sorting is required
is to handle the fact that seq files (and any read() operation from
user-space) cannot hold the tracepoint mutex across multiple calls, so
list entries may vanish between calls. With sorting, the tracepoint
iterator becomes usable even if the list don't contain the exact item
pointed to by the iterator anymore.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Jason Baron <jbaron@redhat.com>
CC: Ingo Molnar <mingo@elte.hu>
CC: Lai Jiangshan <laijs@cn.fujitsu.com>
CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Link: http://lkml.kernel.org/r/20110810191839.GC8525@Krystal
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Introduce:
static __always_inline bool static_branch(struct jump_label_key *key);
instead of the old JUMP_LABEL(key, label) macro.
In this way, jump labels become really easy to use:
Define:
struct jump_label_key jump_key;
Can be used as:
if (static_branch(&jump_key))
do unlikely code
enable/disale via:
jump_label_inc(&jump_key);
jump_label_dec(&jump_key);
that's it!
For the jump labels disabled case, the static_branch() becomes an
atomic_read(), and jump_label_inc()/dec() are simply atomic_inc(),
atomic_dec() operations. We show testing results for this change below.
Thanks to H. Peter Anvin for suggesting the 'static_branch()' construct.
Since we now require a 'struct jump_label_key *key', we can store a pointer into
the jump table addresses. In this way, we can enable/disable jump labels, in
basically constant time. This change allows us to completely remove the previous
hashtable scheme. Thanks to Peter Zijlstra for this re-write.
Testing:
I ran a series of 'tbench 20' runs 5 times (with reboots) for 3
configurations, where tracepoints were disabled.
jump label configured in
avg: 815.6
jump label *not* configured in (using atomic reads)
avg: 800.1
jump label *not* configured in (regular reads)
avg: 803.4
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20110316212947.GA8792@redhat.com>
Signed-off-by: Jason Baron <jbaron@redhat.com>
Suggested-by: H. Peter Anvin <hpa@linux.intel.com>
Tested-by: David Daney <ddaney@caviumnetworks.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Make the tracepoints more robust, making them solid enough to handle compiler
changes by not relying on anything based on compiler-specific behavior with
respect to structure alignment. Implement an approach proposed by David Miller:
use an array of const pointers to refer to the individual structures, and export
this pointer array through the linker script rather than the structures per se.
It will consume 32 extra bytes per tracepoint (24 for structure padding and 8
for the pointers), but are less likely to break due to compiler changes.
History:
commit 7e066fb8 tracepoints: add DECLARE_TRACE() and DEFINE_TRACE()
added the aligned(32) type and variable attribute to the tracepoint structures
to deal with gcc happily aligning statically defined structures on 32-byte
multiples.
One attempt was to use a 8-byte alignment for tracepoint structures by applying
both the variable and type attribute to tracepoint structures definitions and
declarations. It worked fine with gcc 4.5.1, but broke with gcc 4.4.4 and 4.4.5.
The reason is that the "aligned" attribute only specify the _minimum_ alignment
for a structure, leaving both the compiler and the linker free to align on
larger multiples. Because tracepoint.c expects the structures to be placed as an
array within each section, up-alignment cause NULL-pointer exceptions due to the
extra unexpected padding.
(this patch applies on top of -tip)
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: David S. Miller <davem@davemloft.net>
LKML-Reference: <20110126222622.GA10794@Krystal>
CC: Frederic Weisbecker <fweisbec@gmail.com>
CC: Ingo Molnar <mingo@elte.hu>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: Peter Zijlstra <peterz@infradead.org>
CC: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
There are instances in the kernel that we only want to trace
a tracepoint when a certain condition is set. But we do not
want to test for that condition in the core kernel.
If we test for that condition before calling the tracepoin, then
we will be performing that test even when tracing is not enabled.
This is 99.99% of the time.
We currently can just filter out on that condition, but that happens
after we write to the trace buffer. We just wasted time writing to
the ring buffer for an event we never cared about.
This patch adds:
TRACE_EVENT_CONDITION() and DEFINE_EVENT_CONDITION()
These have a new TP_CONDITION() argument that comes right after
the TP_ARGS(). This condition can use the parameters of TP_ARGS()
in the TRACE_EVENT() to determine if the tracepoint should be traced
or not. The TP_CONDITION() will be placed in a if (cond) trace;
For example, for the tracepoint sched_wakeup, it is useless to
trace a wakeup event where the caller never actually wakes
anything up (where success == 0). So adding:
TP_CONDITION(success),
which uses the "success" parameter of the wakeup tracepoint
will have it only trace when we have successfully woken up a
task.
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
This introduces the new TRACE_EVENT_FLAGS() macro in order
to set up initial event flags value.
This macro must simply follow the definition of a trace event
and take the event name and the flag value as parameters:
TRACE_EVENT(my_event, .....
....
);
TRACE_EVENT_FLAGS(my_event, 1)
This will set up 1 as the initial my_event->flags value.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Jason Baron <jbaron@redhat.com>
Make use of the jump label infrastructure for tracepoints.
Signed-off-by: Jason Baron <jbaron@redhat.com>
LKML-Reference: <a9ba2056e2c9cf332c3c300b577463ce66ff23a8.1284733808.git.jbaron@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The header file include/linux/tracepoint.h may be included without
include/linux/errno.h and then the compiler will fail on building for
undelcared ENOSYS. This patch fixes this problem via including <linux/errno.h>
to include/linux/tracepoint.h.
Signed-off-by: Wu Zhangjin <wuzhangjin@gmail.com>
LKML-Reference: <1277118549-622-1-git-send-email-wuzhangjin@gmail.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
This patch adds data to be passed to tracepoint callbacks.
The created functions from DECLARE_TRACE() now need a mandatory data
parameter. For example:
DECLARE_TRACE(mytracepoint, int value, value)
Will create the register function:
int register_trace_mytracepoint((void(*)(void *data, int value))probe,
void *data);
As the first argument, all callbacks (probes) must take a (void *data)
parameter. So a callback for the above tracepoint will look like:
void myprobe(void *data, int value)
{
}
The callback may choose to ignore the data parameter.
This change allows callbacks to register a private data pointer along
with the function probe.
void mycallback(void *data, int value);
register_trace_mytracepoint(mycallback, mydata);
Then the mycallback() will receive the "mydata" as the first parameter
before the args.
A more detailed example:
DECLARE_TRACE(mytracepoint, TP_PROTO(int status), TP_ARGS(status));
/* In the C file */
DEFINE_TRACE(mytracepoint, TP_PROTO(int status), TP_ARGS(status));
[...]
trace_mytracepoint(status);
/* In a file registering this tracepoint */
int my_callback(void *data, int status)
{
struct my_struct my_data = data;
[...]
}
[...]
my_data = kmalloc(sizeof(*my_data), GFP_KERNEL);
init_my_data(my_data);
register_trace_mytracepoint(my_callback, my_data);
The same callback can also be registered to the same tracepoint as long
as the data registered is different. Note, the data must also be used
to unregister the callback:
unregister_trace_mytracepoint(my_callback, my_data);
Because of the data parameter, tracepoints declared this way can not have
no args. That is:
DECLARE_TRACE(mytracepoint, TP_PROTO(void), TP_ARGS());
will cause an error.
If no arguments are needed, a new macro can be used instead:
DECLARE_TRACE_NOARGS(mytracepoint);
Since there are no arguments, the proto and args fields are left out.
This is part of a series to make the tracepoint footprint smaller:
text data bss dec hex filename
4913961 1088356 861512 6863829 68bbd5 vmlinux.orig
4914025 1088868 861512 6864405 68be15 vmlinux.class
4918492 1084612 861512 6864616 68bee8 vmlinux.tracepoint
Again, this patch also increases the size of the kernel, but
lays the ground work for decreasing it.
v5: Fixed net/core/drop_monitor.c to handle these updates.
v4: Moved the DECLARE_TRACE() DECLARE_TRACE_NOARGS out of the
#ifdef CONFIG_TRACE_POINTS, since the two are the same in both
cases. The __DECLARE_TRACE() is what changes.
Thanks to Frederic Weisbecker for pointing this out.
v3: Made all register_* functions require data to be passed and
all callbacks to take a void * parameter as its first argument.
This makes the calling functions comply with C standards.
Also added more comments to the modifications of DECLARE_TRACE().
v2: Made the DECLARE_TRACE() have the ability to pass arguments
and added a new DECLARE_TRACE_NOARGS() for tracepoints that
do not need any arguments.
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Masami Hiramatsu <mhiramat@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
This check is meant to be used by tracepoint users which do a direct cast of
callbacks to (void *) for direct registration, thus bypassing the
register_trace_##name and unregister_trace_##name checks.
This permits to ensure that the callback type matches the function type at the
call site, but without generating any code.
Acked-by: Masami Hiramatsu <mhiramat@redhat.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
LKML-Reference: <20100430165959.GA25605@Krystal>
CC: Ingo Molnar <mingo@elte.hu>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Peter Zijlstra <peterz@infradead.org>
CC: Arnaldo Carvalho de Melo <acme@redhat.com>
CC: Lai Jiangshan <laijs@cn.fujitsu.com>
CC: Li Zefan <lizf@cn.fujitsu.com>
CC: Christoph Hellwig <hch@lst.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
When more than one header is included under CREATE_TRACE_POINTS
the DECLARE_TRACE() macro is not defined back to its original meaning
and the second include will fail to initialize the TRACE_EVENT()
and DECLARE_TRACE() correctly.
To fix this the tracepoint.h file moves the define of DECLARE_TRACE()
out of the #ifdef _LINUX_TRACEPOINT_H protection (just like the
define of the TRACE_EVENT()). This way the define_trace.h will undef
the DECLARE_TRACE() at the end and allow new headers to start
from scratch.
This patch also requires fixing the include/events/napi.h
It currently uses DECLARE_TRACE() and should be converted to a TRACE_EVENT()
format. But I'll leave that change to the authors of that file.
But since the napi.h file depends on using the CREATE_TRACE_POINTS
and does not define its own DEFINE_TRACE() it must use the define_trace.h
method instead.
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
It is not quite obvious at first sight what TRACE_EVENT_TEMPLATE
does: does it define an event as well beyond defining a template?
To clarify this, rename it to DECLARE_EVENT_CLASS, which follows
the various 'DECLARE_*()' idioms we already have in the kernel:
DECLARE_EVENT_CLASS(class)
DEFINE_EVENT(class, event1)
DEFINE_EVENT(class, event2)
DEFINE_EVENT(class, event3)
To complete this logic we should also rename TRACE_EVENT() to:
DEFINE_SINGLE_EVENT(single_event)
... but in a more quiet moment of the kernel cycle.
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4B0E286A.2000405@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
After creating the TRACE_EVENT_TEMPLATE I started to look at other
trace points to see what duplication was made. I noticed that there
are several trace points where they are almost identical except for
the name and the output format. Since TRACE_EVENT_TEMPLATE was successful
in bringing down the size of trace events, I added a DEFINE_EVENT_PRINT.
DEFINE_EVENT_PRINT is used just like DEFINE_EVENT is. That is, the
DEFINE_EVENT_PRINT also uses a TRACE_EVENT_TEMPLATE, but it allows the
developer to overwrite the print format. If there are two or more
TRACE_EVENTS that are identical except for the name and print, then
they can be converted to use a TRACE_EVENT_TEMPLATE. Since the
TRACE_EVENT_TEMPLATE already does the print output, the first trace event
would have its print format held in the TRACE_EVENT_TEMPLATE and
be defined with a DEFINE_EVENT. The rest will use the DEFINE_EVENT_PRINT
and override the print format.
Converting the sched trace points to both DEFINE_EVENT and
DEFINE_EVENT_PRINT. Five were converted to DEFINE_EVENT and two were
converted to DEFINE_EVENT_PRINT.
I was able to get the following:
$ size kernel/sched.o-*
text data bss dec hex filename
79299 6776 2520 88595 15a13 kernel/sched.o-notrace
101941 11896 2584 116421 1c6c5 kernel/sched.o-templ
104779 11896 2584 119259 1d1db kernel/sched.o-trace
sched.o-notrace is the scheduler compiled with no trace points.
sched.o-templ is with the use of DEFINE_EVENT and DEFINE_EVENT_PRINT
sched.o-trace is the current trace events.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
There are some places in the kernel that define several tracepoints and
they are all identical besides the name. The code to enable, disable and
record is created for every trace point even if most of the code is
identical.
This patch adds TRACE_EVENT_TEMPLATE that lets the developer create
a template TRACE_EVENT and create trace points with DEFINE_EVENT, which
is based off of a given template. Each trace point used by this
will share most of the code, and bring down the size of the kernel
when there are several duplicate events.
Usage is:
TRACE_EVENT_TEMPLATE(name, proto, args, tstruct, assign, print);
Which would be the same as defining a normal TRACE_EVENT.
To create the trace events that the trace points will use:
DEFINE_EVENT(template, name, proto, args) is done. The template
is the name of the TRACE_EVENT_TEMPLATE to use. The name is the
name of the trace point. The parameters proto and args must be the same
as the proto and args of the template. If they are not the same,
then a compile error will result. I tried hard removing this duplication
but the C preprocessor is not powerful enough (or my CPP magic
experience points is not at a high enough level) to not need them.
A lot of trace events are coming in with new XFS development. Most of
the trace points are identical except for the name. The following shows
the advantage of having TRACE_EVENT_TEMPLATE:
$ size fs/xfs/xfs.o.*
text data bss dec hex filename
452114 2788 3520 458422 6feb6 fs/xfs/xfs.o.old
638482 38116 3744 680342 a6196 fs/xfs/xfs.o.template
996954 38116 4480 1039550 fdcbe fs/xfs/xfs.o.trace
xfs.o.old is without any tracepoints.
xfs.o.template uses the new TRACE_EVENT_TEMPLATE.
xfs.o.trace uses the current TRACE_EVENT macros.
Requested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Fix the tracepoint documentation path in tracepoints headers and
a misaligned tabulation.
Signed-off-by: Li Hong <lihong.hi@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Ingo Molnar <mingo@redhat.com>
LKML-Reference: <3a3680030909220300h7cf18849q4d4702b9d4feaa67@mail.gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
The commit:
commit 5ac35daa93
Author: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
tracing/events: fix the include file dependencies
Moved the TRACE_EVENT out of the ifdef protection of tracepoints.h
but uses the define of TRACE_EVENT itself as protection. This patch
adds comments to explain why.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The TRACE_EVENT depends on the include/linux/tracepoint.h first
and include/trace/ftrace.h later, if we include the ftrace.h early,
a building error will occur.
Both define TRACE_EVENT in trace_a.h and trace_b.h, if we include
those in .c file, like this:
#define CREATE_TRACE_POINTS
include <trace/events/trace_a.h>
include <trace/events/trace_b.h>
The above will not work, because the TRACE_EVENT was re-defined by
the previous .h file.
Reported-by: Wei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
LKML-Reference: <4A937F5E.3020802@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
It's not strictly correct for the tracepoint reg/unreg callbacks to
occur when a client is hooking up, because the actual tracepoint may not
be present yet. This happens to be fine for syscall, since that's in
the core kernel, but it would cause problems for tracepoints defined in
a module that hasn't been loaded yet. It also means the reg/unreg has
to be EXPORTed for any modules to use the tracepoint (as in SystemTap).
This patch removes DECLARE_TRACE_WITH_CALLBACK, and instead introduces
DEFINE_TRACE_FN which stores the callbacks in struct tracepoint. The
callbacks are used now when the active state of the tracepoint changes
in set_tracepoint & disable_tracepoint.
This also introduces TRACE_EVENT_FN, so ftrace events can also provide
registration callbacks if needed.
Signed-off-by: Josh Stone <jistone@redhat.com>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
LKML-Reference: <1251150194-1713-4-git-send-email-jistone@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Introduce a new 'DECLARE_TRACE_WITH_CALLBACK()' macro, so that
tracepoints can associate an external register/unregister function.
This prepares for the syscalls tracer conversion to trace events. We
will need to perform arch level operations once a syscall event is
turned on/off, such as TIF flags setting, hence the need of such
specific callbacks.
Signed-off-by: Jason Baron <jbaron@redhat.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jiaying Zhang <jiayingz@google.com>
Cc: Martin Bligh <mbligh@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Many developers use "/debug/" or "/debugfs/" or "/sys/kernel/debug/"
directory name to mount debugfs filesystem for ftrace according to
./Documentation/tracers/ftrace.txt file.
And, three directory names(ex:/debug/, /debugfs/, /sys/kernel/debug/) is
existed in kernel source like ftrace, DRM, Wireless, Documentation,
Network[sky2]files to mount debugfs filesystem.
debugfs means debug filesystem for debugging easy to use by greg kroah
hartman. "/sys/kernel/debug/" name is suitable as directory name
of debugfs filesystem.
- debugfs related reference: http://lwn.net/Articles/334546/
Fix inconsistency of directory name to mount debugfs filesystem.
* From Steven Rostedt
- find_debugfs() and tracing_files() in this patch.
Signed-off-by: GeunSik Lim <geunsik.lim@samsung.com>
Acked-by : Inaky Perez-Gonzalez <inaky@linux.intel.com>
Reviewed-by : Steven Rostedt <rostedt@goodmis.org>
Reviewed-by : James Smart <james.smart@emulex.com>
CC: Jiri Kosina <trivial@kernel.org>
CC: David Airlie <airlied@linux.ie>
CC: Peter Osterlund <petero2@telia.com>
CC: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
CC: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
CC: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The TRACE_FORMAT macro has been deprecated by the TRACE_EVENT macro.
There are no more users. All new users must use the TRACE_EVENT macro.
[ Impact: remove old functionality ]
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Impact: clean up
Neil Horman (et. al.) criticized the way the trace events were broken up
into two files. The reason for that was that ftrace needed to separate out
the declarations from where the #include <linux/tracepoint.h> was used.
It then dawned on me that the tracepoint.h header only needs to define the
TRACE_EVENT macro if it is not already defined.
The solution is simply to test if TRACE_EVENT is defined, and if it is not
then the linux/tracepoint.h header can define it. This change consolidates
all the <traces>.h and <traces>_event_types.h into the <traces>.h file.
Reported-by: Neil Horman <nhorman@tuxdriver.com>
Reported-by: Theodore Tso <tytso@mit.edu>
Reported-by: Jiaying Zhang <jiayingz@google.com>
Cc: Zhaolei <zhaolei@cn.fujitsu.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Jason Baron <jbaron@redhat.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Impact: clean up
There existed a lot of <space><tab>'s in the tracing code. This
patch removes them.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Impact: clean up / comments
Kosaki Motohiro asked about an explanation to the TRACE_EVENT macro.
Ingo Molnar replied with a nice description.
This patch takes the description that Ingo wrote (with some slight
modifications) and adds it to the tracepoint.h file.
Reported-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Impact: clean up
In trying to stay consistant with the C style format in the TRACE_EVENT
macro, it makes more sense to do the printk after the assigning of
the variables.
Reported-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Impact: clean up
The TRACE_EVENT_FORMAT macro is no longer used by trace points
and only the DECLARE_TRACE, TRACE_FORMAT or TRACE_EVENT macros should
be used by them. Although the TRACE_EVENT_FORMAT macro is still used
by the internal tracing utility, it should not be used in core
kernel code.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Impact: clean up and enhancement
The TRACE_EVENT_FORMAT macro looks quite ugly and is limited in its
ability to save data as well as to print the record out. Working with
Ingo Molnar, we came up with a new format that is much more pleasing to
the eye of C developers. This new macro is more C style than the old
macro, and is more obvious to what it does.
Here's the example. The only updated macro in this patch is the
sched_switch trace point.
The old method looked like this:
TRACE_EVENT_FORMAT(sched_switch,
TP_PROTO(struct rq *rq, struct task_struct *prev,
struct task_struct *next),
TP_ARGS(rq, prev, next),
TP_FMT("task %s:%d ==> %s:%d",
prev->comm, prev->pid, next->comm, next->pid),
TRACE_STRUCT(
TRACE_FIELD(pid_t, prev_pid, prev->pid)
TRACE_FIELD(int, prev_prio, prev->prio)
TRACE_FIELD_SPECIAL(char next_comm[TASK_COMM_LEN],
next_comm,
TP_CMD(memcpy(TRACE_ENTRY->next_comm,
next->comm,
TASK_COMM_LEN)))
TRACE_FIELD(pid_t, next_pid, next->pid)
TRACE_FIELD(int, next_prio, next->prio)
),
TP_RAW_FMT("prev %d:%d ==> next %s:%d:%d")
);
The above method is hard to read and requires two format fields.
The new method:
/*
* Tracepoint for task switches, performed by the scheduler:
*
* (NOTE: the 'rq' argument is not used by generic trace events,
* but used by the latency tracer plugin. )
*/
TRACE_EVENT(sched_switch,
TP_PROTO(struct rq *rq, struct task_struct *prev,
struct task_struct *next),
TP_ARGS(rq, prev, next),
TP_STRUCT__entry(
__array( char, prev_comm, TASK_COMM_LEN )
__field( pid_t, prev_pid )
__field( int, prev_prio )
__array( char, next_comm, TASK_COMM_LEN )
__field( pid_t, next_pid )
__field( int, next_prio )
),
TP_printk("task %s:%d [%d] ==> %s:%d [%d]",
__entry->prev_comm, __entry->prev_pid, __entry->prev_prio,
__entry->next_comm, __entry->next_pid, __entry->next_prio),
TP_fast_assign(
memcpy(__entry->next_comm, next->comm, TASK_COMM_LEN);
__entry->prev_pid = prev->pid;
__entry->prev_prio = prev->prio;
memcpy(__entry->prev_comm, prev->comm, TASK_COMM_LEN);
__entry->next_pid = next->pid;
__entry->next_prio = next->prio;
)
);
This macro is called TRACE_EVENT, it is broken up into 5 parts:
TP_PROTO: the proto type of the trace point
TP_ARGS: the arguments of the trace point
TP_STRUCT_entry: the structure layout of the entry in the ring buffer
TP_printk: the printk format
TP_fast_assign: the method used to write the entry into the ring buffer
The structure is the definition of how the event will be saved in the
ring buffer. The printk is used by the internal tracing in case of
an oops, and the kernel needs to print out the format of the record
to the console. This the TP_printk gives a means to show the records
in a human readable format. It is also used to print out the data
from the trace file.
The TP_fast_assign is executed directly. It is basically like a C function,
where the __entry is the handle to the record.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Impact: clean up
The macros TPPROTO, TPARGS, TPFMT, TPRAWFMT, and TPCMD all look a bit
ugly. This patch adds an underscore to their names.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
This patch utilizes the TRACE_EVENT_FORMAT macro to enable the C style
faster tracing for the sched subsystem trace points.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Peter Zijlstra warned that TPPROTO and TPARGS might become something
other than a simple copy of itself. To prevent this from having
side effects in the TRACE_FORMAT macro in tracepoint.h, we add a
PARAMS() macro to be defined as just a wrapper.
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
There's been a bit confusion to whether DEFINE/DECLARE_TRACE_FMT should
be a DEFINE or a DECLARE. Ingo Molnar suggested simply calling it
TRACE_FORMAT.
Reported-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
This patch creates a DEFINE_TRACE_FMT to map to DECLARE_TRACE.
This allows for the developers to place format strings and
args in with their tracepoint declaration. A tracer may now
override the DEFINE_TRACE_FMT macro and use it to record
a default format.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Impact: API *CHANGE*. Must update all tracepoint users.
Add DEFINE_TRACE() to tracepoints to let them declare the tracepoint
structure in a single spot for all the kernel. It helps reducing memory
consumption, especially when declaring a lot of tracepoints, e.g. for
kmalloc tracing.
*API CHANGE WARNING*: now, DECLARE_TRACE() must be used in headers for
tracepoint declarations rather than DEFINE_TRACE(). This is the sane way
to do it. The name previously used was misleading.
Updates scheduler instrumentation to follow this API change.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: cleanup
That's overkill, takes space. We have a global tracepoint registery in
header files anyway.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Make sure tracepoints can be called within ftrace callbacks.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Impact: add new tracepoint APIs to allow the batched registration of probes
new APIs separate tracepoint_probe_register(),
tracepoint_probe_unregister() into 2 steps. The first step of them
is just update tracepoint_entry, not connect or disconnect.
this patch introduces tracepoint_probe_update_all() for update all.
these APIs are very useful for registering lots of probes
but just updating once. Another very important thing is that
*_noupdate APIs do not require module_mutex.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Turn tracepoint synchronize unregister into a static inline. There is no
reason to keep it as a macro over a static inline.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Create tracepoint_synchronize_unregister() which must be called before the end
of exit() to make sure every probe callers have exited the non preemptible
section and thus are not executing the probe code anymore.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Implementation of kernel tracepoints. Inspired from the Linux Kernel
Markers. Allows complete typing verification by declaring both tracing
statement inline functions and probe registration/unregistration static
inline functions within the same macro "DEFINE_TRACE". No format string
is required. See the tracepoint Documentation and Samples patches for
usage examples.
Taken from the documentation patch :
"A tracepoint placed in code provides a hook to call a function (probe)
that you can provide at runtime. A tracepoint can be "on" (a probe is
connected to it) or "off" (no probe is attached). When a tracepoint is
"off" it has no effect, except for adding a tiny time penalty (checking
a condition for a branch) and space penalty (adding a few bytes for the
function call at the end of the instrumented function and adds a data
structure in a separate section). When a tracepoint is "on", the
function you provide is called each time the tracepoint is executed, in
the execution context of the caller. When the function provided ends its
execution, it returns to the caller (continuing from the tracepoint
site).
You can put tracepoints at important locations in the code. They are
lightweight hooks that can pass an arbitrary number of parameters, which
prototypes are described in a tracepoint declaration placed in a header
file."
Addition and removal of tracepoints is synchronized by RCU using the
scheduler (and preempt_disable) as guarantees to find a quiescent state
(this is really RCU "classic"). The update side uses rcu_barrier_sched()
with call_rcu_sched() and the read/execute side uses
"preempt_disable()/preempt_enable()".
We make sure the previous array containing probes, which has been
scheduled for deletion by the rcu callback, is indeed freed before we
proceed to the next update. It therefore limits the rate of modification
of a single tracepoint to one update per RCU period. The objective here
is to permit fast batch add/removal of probes on _different_
tracepoints.
Changelog :
- Use #name ":" #proto as string to identify the tracepoint in the
tracepoint table. This will make sure not type mismatch happens due to
connexion of a probe with the wrong type to a tracepoint declared with
the same name in a different header.
- Add tracepoint_entry_free_old.
- Change __TO_TRACE to get rid of the 'i' iterator.
Masami Hiramatsu <mhiramat@redhat.com> :
Tested on x86-64.
Performance impact of a tracepoint : same as markers, except that it
adds about 70 bytes of instructions in an unlikely branch of each
instrumented function (the for loop, the stack setup and the function
call). It currently adds a memory read, a test and a conditional branch
at the instrumentation site (in the hot path). Immediate values will
eventually change this into a load immediate, test and branch, which
removes the memory read which will make the i-cache impact smaller
(changing the memory read for a load immediate removes 3-4 bytes per
site on x86_32 (depending on mov prefixes), or 7-8 bytes on x86_64, it
also saves the d-cache hit).
About the performance impact of tracepoints (which is comparable to
markers), even without immediate values optimizations, tests done by
Hideo Aoki on ia64 show no regression. His test case was using hackbench
on a kernel where scheduler instrumentation (about 5 events in code
scheduler code) was added.
Quoting Hideo Aoki about Markers :
I evaluated overhead of kernel marker using linux-2.6-sched-fixes git
tree, which includes several markers for LTTng, using an ia64 server.
While the immediate trace mark feature isn't implemented on ia64, there
is no major performance regression. So, I think that we don't have any
issues to propose merging marker point patches into Linus's tree from
the viewpoint of performance impact.
I prepared two kernels to evaluate. The first one was compiled without
CONFIG_MARKERS. The second one was enabled CONFIG_MARKERS.
I downloaded the original hackbench from the following URL:
http://devresources.linux-foundation.org/craiger/hackbench/src/hackbench.c
I ran hackbench 5 times in each condition and calculated the average and
difference between the kernels.
The parameter of hackbench: every 50 from 50 to 800
The number of CPUs of the server: 2, 4, and 8
Below is the results. As you can see, major performance regression
wasn't found in any case. Even if number of processes increases,
differences between marker-enabled kernel and marker- disabled kernel
doesn't increase. Moreover, if number of CPUs increases, the differences
doesn't increase either.
Curiously, marker-enabled kernel is better than marker-disabled kernel
in more than half cases, although I guess it comes from the difference
of memory access pattern.
* 2 CPUs
Number of | without | with | diff | diff |
processes | Marker [Sec] | Marker [Sec] | [Sec] | [%] |
--------------------------------------------------------------
50 | 4.811 | 4.872 | +0.061 | +1.27 |
100 | 9.854 | 10.309 | +0.454 | +4.61 |
150 | 15.602 | 15.040 | -0.562 | -3.6 |
200 | 20.489 | 20.380 | -0.109 | -0.53 |
250 | 25.798 | 25.652 | -0.146 | -0.56 |
300 | 31.260 | 30.797 | -0.463 | -1.48 |
350 | 36.121 | 35.770 | -0.351 | -0.97 |
400 | 42.288 | 42.102 | -0.186 | -0.44 |
450 | 47.778 | 47.253 | -0.526 | -1.1 |
500 | 51.953 | 52.278 | +0.325 | +0.63 |
550 | 58.401 | 57.700 | -0.701 | -1.2 |
600 | 63.334 | 63.222 | -0.112 | -0.18 |
650 | 68.816 | 68.511 | -0.306 | -0.44 |
700 | 74.667 | 74.088 | -0.579 | -0.78 |
750 | 78.612 | 79.582 | +0.970 | +1.23 |
800 | 85.431 | 85.263 | -0.168 | -0.2 |
--------------------------------------------------------------
* 4 CPUs
Number of | without | with | diff | diff |
processes | Marker [Sec] | Marker [Sec] | [Sec] | [%] |
--------------------------------------------------------------
50 | 2.586 | 2.584 | -0.003 | -0.1 |
100 | 5.254 | 5.283 | +0.030 | +0.56 |
150 | 8.012 | 8.074 | +0.061 | +0.76 |
200 | 11.172 | 11.000 | -0.172 | -1.54 |
250 | 13.917 | 14.036 | +0.119 | +0.86 |
300 | 16.905 | 16.543 | -0.362 | -2.14 |
350 | 19.901 | 20.036 | +0.135 | +0.68 |
400 | 22.908 | 23.094 | +0.186 | +0.81 |
450 | 26.273 | 26.101 | -0.172 | -0.66 |
500 | 29.554 | 29.092 | -0.461 | -1.56 |
550 | 32.377 | 32.274 | -0.103 | -0.32 |
600 | 35.855 | 35.322 | -0.533 | -1.49 |
650 | 39.192 | 38.388 | -0.804 | -2.05 |
700 | 41.744 | 41.719 | -0.025 | -0.06 |
750 | 45.016 | 44.496 | -0.520 | -1.16 |
800 | 48.212 | 47.603 | -0.609 | -1.26 |
--------------------------------------------------------------
* 8 CPUs
Number of | without | with | diff | diff |
processes | Marker [Sec] | Marker [Sec] | [Sec] | [%] |
--------------------------------------------------------------
50 | 2.094 | 2.072 | -0.022 | -1.07 |
100 | 4.162 | 4.273 | +0.111 | +2.66 |
150 | 6.485 | 6.540 | +0.055 | +0.84 |
200 | 8.556 | 8.478 | -0.078 | -0.91 |
250 | 10.458 | 10.258 | -0.200 | -1.91 |
300 | 12.425 | 12.750 | +0.325 | +2.62 |
350 | 14.807 | 14.839 | +0.032 | +0.22 |
400 | 16.801 | 16.959 | +0.158 | +0.94 |
450 | 19.478 | 19.009 | -0.470 | -2.41 |
500 | 21.296 | 21.504 | +0.208 | +0.98 |
550 | 23.842 | 23.979 | +0.137 | +0.57 |
600 | 26.309 | 26.111 | -0.198 | -0.75 |
650 | 28.705 | 28.446 | -0.259 | -0.9 |
700 | 31.233 | 31.394 | +0.161 | +0.52 |
750 | 34.064 | 33.720 | -0.344 | -1.01 |
800 | 36.320 | 36.114 | -0.206 | -0.57 |
--------------------------------------------------------------
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Acked-by: Masami Hiramatsu <mhiramat@redhat.com>
Acked-by: 'Peter Zijlstra' <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
