Allows a way of measuring low level kernel implementation of FUTEX_LOCK_PI and
FUTEX_UNLOCK_PI.
The program comes in two flavors:
(i) single futex (default), all threads contend on the same uaddr. For the
sake of the benchmark, we call into kernel space even when the lock is
uncontended. The kernel will set it to TID, any waters that come in and
contend for the pi futex will be handled respectively by the kernel.
(ii) -M option for multiple futexes, each thread deals with its own futex. This
is a trivial scenario and only measures kernel handling of 0->TID transition.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Cc: Mel Gorman <mgorman@suse.de>
Link: http://lkml.kernel.org/r/1436259353.12255.78.camel@stgolabs.net
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The futex-wake benchmark only measures wakeups done within a single
process. While this has value in its own, it does not really generate
any hb->lock contention.
A new benchmark 'wake-parallel' is added, by extending the futex-wake
code such that we can measure parallel waker threads. The program output
shows the avg per-thread latency in order to complete its share of
wakeups:
Run summary [PID 13474]: blocking on 512 threads (at [private] futex 0xa88668), 8 threads waking up 64 at a time.
[Run 1]: Avg per-thread latency (waking 64/512 threads) in 0.6230 ms (+-15.31%)
[Run 2]: Avg per-thread latency (waking 64/512 threads) in 0.5175 ms (+-29.95%)
[Run 3]: Avg per-thread latency (waking 64/512 threads) in 0.7578 ms (+-18.03%)
[Run 4]: Avg per-thread latency (waking 64/512 threads) in 0.8944 ms (+-12.54%)
[Run 5]: Avg per-thread latency (waking 64/512 threads) in 1.1204 ms (+-23.85%)
Avg per-thread latency (waking 64/512 threads) in 0.7826 ms (+-9.91%)
Naturally, different combinations of numbers of blocking and waker
threads will exhibit different information.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Davidlohr Bueso <dbueso@suse.de>
Link: http://lkml.kernel.org/r/1431110280-20231-1-git-send-email-dave@stgolabs.net
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
There are a number of benchmarks that do single runs and as a result
does not really help users gain a general idea of how the workload
performs. So the user must either manually do multiple runs or just use
single bogus results.
This option will enable users to specify the amount of runs (arbitrarily
defaulted to 10, to use the existing benchmarks default) through the
'--repeat' option. Add it to perf-bench instead of implementing it
always in each specific benchmark.
Signed-off-by: Davidlohr Bueso <davidlohr@hp.com>
Cc: Aswin Chandramouleeswaran <aswin@hp.com>
Cc: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Jiri Olsa <jolsa@kernel.org>
Link: http://lkml.kernel.org/r/1402942467-10671-2-git-send-email-davidlohr@hp.com
[ Kept the existing default of 10, changing it to something else should
be done on separate patch ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Block a bunch of threads on a futex and requeue them on another, N at a
time.
This program is particularly useful to measure the latency of nthread
requeues without waking up any tasks -- thus mimicking a regular
futex_wait.
An example run:
$ perf bench futex requeue -r 100 -t 64
Run summary [PID 151011]: Requeuing 64 threads (from 0x7d15c4 to 0x7d15c8), 1 at a time.
[Run 1]: Requeued 64 of 64 threads in 0.0400 ms
[Run 2]: Requeued 64 of 64 threads in 0.0390 ms
[Run 3]: Requeued 64 of 64 threads in 0.0400 ms
...
[Run 100]: Requeued 64 of 64 threads in 0.0390 ms
Requeued 64 of 64 threads in 0.0399 ms (+-0.37%)
Signed-off-by: Davidlohr Bueso <davidlohr@hp.com>
Acked-by: Darren Hart <dvhart@linux.intel.com>
Cc: Aswin Chandramouleeswaran <aswin@hp.com>
Cc: Darren Hart <dvhart@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jason Low <jason.low2@hp.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Scott J Norton <scott.norton@hp.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Waiman Long <Waiman.Long@hp.com>
Link: http://lkml.kernel.org/r/1387081917-9102-4-git-send-email-davidlohr@hp.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Block a bunch of threads on a futex and wake them up, N at a time.
This program is particularly useful to measure the latency of nthread
wakeups in non-error situations: all waiters are queued and all wake
calls wakeup one or more tasks.
An example run:
$ perf bench futex wake -t 512 -r 100
Run summary [PID 27823]: blocking on 512 threads (at futex 0x7e10d4), waking up 1 at a time.
[Run 1]: Wokeup 512 of 512 threads in 6.0080 ms
[Run 2]: Wokeup 512 of 512 threads in 5.2280 ms
[Run 3]: Wokeup 512 of 512 threads in 4.8300 ms
...
[Run 100]: Wokeup 512 of 512 threads in 5.0100 ms
Wokeup 512 of 512 threads in 5.0109 ms (+-2.25%)
Signed-off-by: Davidlohr Bueso <davidlohr@hp.com>
Acked-by: Darren Hart <dvhart@linux.intel.com>
Cc: Aswin Chandramouleeswaran <aswin@hp.com>
Cc: Darren Hart <dvhart@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jason Low <jason.low2@hp.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Scott J Norton <scott.norton@hp.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Waiman Long <Waiman.Long@hp.com>
Link: http://lkml.kernel.org/r/1387081917-9102-3-git-send-email-davidlohr@hp.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Introduce futexes to perf-bench and add a program that stresses and
measures the kernel's implementation of the hash table.
This is a multi-threaded program that simply measures the amount of
failed futex wait calls - we only want to deal with the hashing
overhead, so a negative return of futex_wait_setup() is enough to do the
trick.
An example run:
$ perf bench futex hash -t 32
Run summary [PID 10989]: 32 threads, each operating on 1024 [private] futexes for 10 secs.
[thread 0] futexes: 0x19d9b10 ... 0x19dab0c [ 418713 ops/sec ]
[thread 1] futexes: 0x19daca0 ... 0x19dbc9c [ 469913 ops/sec ]
[thread 2] futexes: 0x19dbe30 ... 0x19dce2c [ 479744 ops/sec ]
...
[thread 31] futexes: 0x19fbb80 ... 0x19fcb7c [ 464179 ops/sec ]
Averaged 454310 operations/sec (+- 0.84%), total secs = 10
Signed-off-by: Davidlohr Bueso <davidlohr@hp.com>
Acked-by: Darren Hart <dvhart@linux.intel.com>
Cc: Aswin Chandramouleeswaran <aswin@hp.com>
Cc: Darren Hart <dvhart@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jason Low <jason.low2@hp.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Scott J Norton <scott.norton@hp.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Waiman Long <Waiman.Long@hp.com>
Link: http://lkml.kernel.org/r/1387081917-9102-2-git-send-email-davidlohr@hp.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The tokens MADV_HUGEPAGE and MADV_NOHUGEPAGE are not available with
glibc 2.12 and older. Define these tokens if they are not already
defined.
This patch fixes these build errors with older versions of glibc.
CC bench/numa.o
bench/numa.c: In function ‘alloc_data’:
bench/numa.c:334: error: ‘MADV_HUGEPAGE’ undeclared (first use in this function)
bench/numa.c:334: error: (Each undeclared identifier is reported only once
bench/numa.c:334: error: for each function it appears in.)
bench/numa.c:341: error: ‘MADV_NOHUGEPAGE’ undeclared (first use in this function)
make: *** [bench/numa.o] Error 1
Signed-off-by: Vinson Lee <vlee@twitter.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Irina Tirdea <irina.tirdea@intel.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1363214064-4671-2-git-send-email-vlee@twitter.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add a suite of NUMA performance benchmarks.
The goal was simulate the behavior and access patterns of real NUMA
workloads, via a wide range of parameters, so this tool goes well
beyond simple bzero() measurements that most NUMA micro-benchmarks use:
- It processes the data and creates a chain of data dependencies,
like a real workload would. Neither the compiler, nor the
kernel (via KSM and other optimizations) nor the CPU can
eliminate parts of the workload.
- It randomizes the initial state and also randomizes the target
addresses of the processing - it's not a simple forward scan
of addresses.
- It provides flexible options to set process, thread and memory
relationship information: -G sets "global" memory shared between
all test processes, -P sets "process" memory shared by all
threads of a process and -T sets "thread" private memory.
- There's a NUMA convergence monitoring and convergence latency
measurement option via -c and -m.
- Micro-sleeps and synchronization can be injected to provoke lock
contention and scheduling, via the -u and -S options. This simulates
IO and contention.
- The -x option instructs the workload to 'perturb' itself artificially
every N seconds, by moving to the first and last CPU of the system
periodically. This way the stability of convergence equilibrium and
the number of steps taken for the scheduler to reach equilibrium again
can be measured.
- The amount of work can be specified via the -l loop count, and/or
via a -s seconds-timeout value.
- CPU and node memory binding options, to test hard binding scenarios.
THP can be turned on and off via madvise() calls.
- Live reporting of convergence progress in an 'at glance' output format.
Printing of convergence and deconvergence events.
The 'perf bench numa mem -a' option will start an array of about 30
individual tests that will each output such measurements:
# Running 5x5-bw-thread, "perf bench numa mem -p 5 -t 5 -P 512 -s 20 -zZ0q --thp 1"
5x5-bw-thread, 20.276, secs, runtime-max/thread
5x5-bw-thread, 20.004, secs, runtime-min/thread
5x5-bw-thread, 20.155, secs, runtime-avg/thread
5x5-bw-thread, 0.671, %, spread-runtime/thread
5x5-bw-thread, 21.153, GB, data/thread
5x5-bw-thread, 528.818, GB, data-total
5x5-bw-thread, 0.959, nsecs, runtime/byte/thread
5x5-bw-thread, 1.043, GB/sec, thread-speed
5x5-bw-thread, 26.081, GB/sec, total-speed
See the help text and the code for more details.
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
perf defines both __used and __unused variables to use for marking
unused variables. The variable __used is defined to
__attribute__((__unused__)), which contradicts the kernel definition to
__attribute__((__used__)) for new gcc versions. On Android, __used is
also defined in system headers and this leads to warnings like: warning:
'__used__' attribute ignored
__unused is not defined in the kernel and is not a standard definition.
If __unused is included everywhere instead of __used, this leads to
conflicts with glibc headers, since glibc has a variables with this name
in its headers.
The best approach is to use __maybe_unused, the definition used in the
kernel for __attribute__((unused)). In this way there is only one
definition in perf sources (instead of 2 definitions that point to the
same thing: __used and __unused) and it works on both Linux and Android.
This patch simply replaces all instances of __used and __unused with
__maybe_unused.
Signed-off-by: Irina Tirdea <irina.tirdea@intel.com>
Acked-by: Pekka Enberg <penberg@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/1347315303-29906-7-git-send-email-irina.tirdea@intel.com
[ committer note: fixed up conflict with a116e05 in builtin-sched.c ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
'perf bench mem memcpy' is a benchmark suite for measuring memcpy()
performance.
Example on a Intel(R) Core(TM)2 Duo CPU E6850 @ 3.00GHz:
| % perf bench mem memcpy -l 1GB
| # Running mem/memcpy benchmark...
| # Copying 1MB Bytes from 0xb7d98008 to 0xb7e99008 ...
|
| 726.216412 MB/Sec
Signed-off-by: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <1258471212-30281-1-git-send-email-mitake@dcl.info.waseda.ac.jp>
[ v2: updated changelog, clarified history of builtin-bench.c ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Clean up initializers in bench.h:
- No need to break the line for function prototypes, they are more
readable in a single line. (even if checkpatch complains about it
- We try to align definitions / structure fields vertically,
to make it all a bit more readable.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1257853855-28934-2-git-send-email-mitake@dcl.info.waseda.ac.jp>
This patch adds some constants and extern declaration to
bench.h. These are used for unified output formatting
of 'perf bench'.
Signed-off-by: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1257808802-9420-2-git-send-email-mitake@dcl.info.waseda.ac.jp>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Davidlohr Bueso