These workload traces cover the whole "long double" range.
This patch was prepared with the help of Adhemerval Zanella.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
This patch adds workload traces for all double format functions where such
files are missing. For each function, a set of 1000 random values is
generated at random using SageMath, such that the output values are
meaningful (for example avoiding too large inputs for exp10 where the
output would be +Inf). More details about the generated values are
given at the beginning of each file.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Since commit 2682695e5c, `make bench-build' with `--enable-static-pie'
fails due to bench-timing-type being incorrectly built with MODULE_NAME
set to `libc'. This commit sets MODULE_NAME to nonlib, thus fixing the
build failure.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
I used these shell commands:
../glibc/scripts/update-copyrights $PWD/../gnulib/build-aux/update-copyright
(cd ../glibc && git commit -am"[this commit message]")
and then ignored the output, which consisted lines saying "FOO: warning:
copyright statement not found" for each of 6694 files FOO.
I then removed trailing white space from benchtests/bench-pthread-locks.c
and iconvdata/tst-iconv-big5-hkscs-to-2ucs4.c, to work around this
diagnostic from Savannah:
remote: *** pre-commit check failed ...
remote: *** error: lines with trailing whitespace found
remote: error: hook declined to update refs/heads/master
Performance benchmarks for various posix locks: mutex, rwlock,
spinlock, condvar, and semaphore. Each test is performed with
an empty loop body or with a computationally "interesting" (i.e.
difficult to optimize away, and used just to allow lock code to
be "hidden" in the filler's CPU cycles).
__float128 is a non-standard name and is not available on some architectures
(like aarch64 or s390x) even though they may support the standard _Float128
type. Other architectures (like armv7) don't support quad-precision
floating-point operations at all.
This commit replaces benchtests references to __float128 with _Float128 and
runs the corresponding tests only on architectures that support it.
commit 7621e38bf3
Author: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Date: Tue Jan 29 17:43:45 2019 +0000
Add generic hp-timing support
removed the clock_gettime option. Restore the clock_gettime option for
some x86 CPUs on which value from RDTSC may not be incremented at a fixed
rate.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* benchtests/Makefile (bench-math): Add logb.
* benchtests/logb-inputs: New file.
* benchtests/logbf-inputs: New file.
Reviewed-by: Gabriel F. T. Gomes <gabrielftg@linux.ibm.com>
Inputs are based on argument reductions from generic and powerpc
implementation.
* benchtests/Makefile (bench-math): Add hypot.
* benchtests/hypot-inputs: New file.
Reviewed-by: Gabriel F. T. Gomes <gabrielftg@linux.ibm.com>
* benchtests/Makefile (bench-math): Add isnan, isinf, and isfinite.
(CFLAGS-bench-isnan.c, CFLAGS-bench-isinf.c,
CFLAGS-bench-isfinite.c): New rule.
* benchtests/isnan-input: New file.
* benchtests/isinf-input: New file.
* benchtests/isfinite-input: New file.
Reviewed-by: Gabriel F. T. Gomes <gabrielftg@linux.ibm.com>
Benchmarks should reflect distribution build policies, so it makes
sense to honor the BIND_NOW configuration for them.
This commit keeps using $(+link-tests), so that the benchmarks are
linked according to the --enable-hardcoded-path-in-tests configure
option.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Reduce the total time taken by benchtests. The malloc thread test takes 4
minutes to run which is significantly more than most other tests. Reduce
this to a more reasonable 40 seconds. The math tests take 10 seconds each,
eventhough all they do is loop on the same input. Anything more than 1
second runtime is way overkill, so set the limit to 1 second.
* benchtests/Makefile (BENCH_DURATION): Set to 1 second.
* benchtests/bench-malloc-thread.c (BENCH_DURATION): Set to 10 seconds.
Add missing generic hp_timing support. It uses clock_gettime (CLOCK_MONOTONIC)
which has unspecified starting time, nano-second accuracy, and should faster on
architectures that implementes the symbol as vDSO.
Checked on aarch64-linux-gnu, x86_64-linux-gnu, and i686-linux-gnu. I also
checked the builds for all afected ABIs.
* benchtests/Makefile (USE_CLOCK_GETTIME) Remove.
* benchtests/README: Update description.
* benchtests/bench-timing.h: Default to hp-timing.
* sysdeps/generic/hp-timing.h (HP_TIMING_DIFF, HP_TIMING_ACCUM_NT,
HP_TIMING_PRINT): Remove.
(HP_TIMING_NOW): Add generic implementation.
(hp_timing_t): Change to uint64_t.
Add a malloc micro benchmark to enable accurate testing of the
various paths in malloc and free. The benchmark does a varying
number of allocations of a given block size, then frees them again.
It tests 3 different scenarios: single-threaded using main arena,
multi-threaded using thread-arena, main arena with SINGLE_THREAD_P
false.
* benchtests/Makefile: Add malloc-simple benchmark.
* benchtests/bench-malloc-simple.c: New benchmark.
This patch makes Python 3.4 or later a required tool for building
glibc, so allowing changes of awk, perl etc. code used in the build
and test to Python code without any such changes needing makefile
conditionals or to handle older Python versions.
This patch makes the configure test for Python check the version and
give an error if Python is missing or too old, and removes makefile
conditionals that are no longer needed. It does not itself convert
any code from another language to Python, and does not remove any
compatibility with older Python versions from existing scripts.
Tested for x86_64.
* configure.ac (PYTHON_PROG): Use AC_CHECK_PROG_VER. Set
critic_missing for versions before 3.4.
* configure: Regenerated.
* manual/install.texi (Tools for Compilation): Document
requirement for Python to build glibc.
* INSTALL: Regenerated.
* Rules [PYTHON]: Make code unconditional.
* benchtests/Makefile [PYTHON]: Likewise.
* conform/Makefile [PYTHON]: Likewise.
* manual/Makefile [PYTHON]: Likewise.
* math/Makefile [PYTHON]: Likewise.
RDTSCP waits until all previous instructions have executed and all
previous loads are globally visible before reading the counter. RDTSC
doesn't wait until all previous instructions have been executed before
reading the counter. All x86 processors since 2010 support RDTSCP
instruction. This patch adds RDTSCP support to benchtests.
* benchtests/Makefile (CPPFLAGS-nonlib): Add -DUSE_RDTSCP if
USE_RDTSCP is defined.
* sysdeps/x86/hp-timing.h (HP_TIMING_NOW): Use RDTSCP if
USE_RDTSCP is defined.
Currently the benchtests are run with internal GLIBC headers, which is incorrect.
Defining _ISOMAC in the makefile ensures the internal headers are bypassed.
Fix all tests which were relying on internal defines or includes.
* benchtests/Makefile: Define _ISOMAC.
* benchtests/bench-strcoll.c: Add missing sys/stat.h include.
* benchtests/bench-string.h: Define inhibit_loop_to_libcall macro.
* benchtests/bench-strstr.c: Define empty libc_hidden_builtin_def.
* benchtests/bench-strtok.c (oldstrtok): Use rawmemchr.
* benchtests/bench-timing.h: Define attribute_hidden.
This patch adds BENCHSET variable to benchtests/Makefile in order to
provide the capability to run a list of subsets of benchmark tests, ie;
make bench BENCHSET="bench-pthread bench-math malloc-thread"
This helps users to benchmark specific glibc area
ChangeLog:
* benchtests/Makefile:Add BENCHSET to allow subsets of
benchmarks to be run.
* benchtests/README: Add documentation for: Running subsets of
benchmarks.
Signed-off-by: Victor Rodriguez <victor.rodriguez.bahena@intel.com>
Signed-off-by: Icarus Sparry <icarus.w.sparry@intel.com>
Reviewed-By: Siddhesh Poyarekar <siddhesh@sourceware.org>
This benchmark is an attempt to eliminate cache effects from string
benchmarks. The benchmark walks both ways through a large memory area
and copies different sizes of memory and alignments one at a time
instead of looping around in the same memory area. This is a good
metric to have alongside the simple memmove benchmark (which is only
really useful for smaller sizes) especially for larger sizes where the
likelihood of the call being done only once is pretty high.
This benchmark is different from memcpy in that it also tests
overlapping copies.
* benchtests/bench-memmove-walk.c: New file.
* benchtests/Makefile (string-benchset): Add it.
This benchmark is an attempt to eliminate cache effects from string
benchmarks. The benchmark walks backward through a large memory area
and sets different sizes of memory and alignments one at a time
instead of looping around in the same memory area. This is a good
metric to have alongside the simple memset benchmark (which is only
really useful for smaller sizes) especially for larger sizes where the
likelihood of the call being done only once is pretty high.
* benchtests/bench-memset-walk.c: New file.
* benchtests/Makefile (string-benchset): Add it.
This benchmark is an attempt to eliminate cache effects from string
benchmarks. The benchmark walks both ways through a large memory area
and copies different sizes of memory and alignments one at a time
instead of looping around in the same memory area. This is a good
metric to have alongside the other memcpy benchmarks, especially for
larger sizes where the likelihood of the call being done only once is
pretty high.
* benchtests/bench-memcpy-walk.c: New file.
* benchtests/Makefile (string-benchset): Add it.
exp2f and log2f benchmark traces are just copies of the existing
expf and logf traces from wrf_r.
* benchtests/Makefile: Add exp2f and log2f benchmarks.
* benchtests/exp2f-inputs: Copy of expf-inputs.
* benchtests/log2f-inputs: Copy of logf-inputs.
Add a trace for logf. This is a reduced trace based on 2.8 billion
samples extracted from wrf_r.
* benchtests/Makefile: Add logf benchmark.
* benchtests/logf-inputs: Add reduced trace from wrf_r.
Add a trace for expf. This is a reduced trace based on 2.4 billion
samples extracted from wrf_r.
* benchtests/Makefile: Add expf benchmark.
* benchtests/expf-inputs: Add reduced trace from wrf_r.
This patch adds benchtests for the trunc and truncf functions. The
inputs listed are fairly arbitrary; I do not assert they are
representative of any particular application.
* benchtests/Makefile (bench-math): Add trunc and truncf.
(CFLAGS-bench-trunc.c): New variable.
(CFLAGS-bench-truncf.c): Likewise.
* benchtests/trunc-inputs: New file.
* benchtests/truncf-inputs: Likewise.
Add powf() bench test with input which covers these cases:
- positive base to positive exponent
- exponent 0
- negative base to even exponent
- exponent 1
- exponent -1
- squared
- squareroot
- 1 to negative exponent
- -1 to negative exponent
- base 0
- -1 to even exponent
- small base
- small exponent
* benchtests/Makefile (bench-math): Add powf.
* benchtests/powf-inputs: New file.
Current allocate_stack logic for create stacks is to first mmap all
the required memory with the desirable memory and then mprotect the
guard area with PROT_NONE if required. Although it works as expected,
it pessimizes the allocation because it requires the kernel to actually
increase commit charge (it counts against the available physical/swap
memory available for the system).
The only issue is to actually check this change since side-effects are
really Linux specific and to actually account them it would require a
kernel specific tests to parse the system wide information. On the kernel
I checked /proc/self/statm does not show any meaningful difference for
vmm and/or rss before and after thread creation. I could only see
really meaningful information checking on system wide /proc/meminfo
between thread creation: MemFree, MemAvailable, and Committed_AS shows
large difference without the patch. I think trying to use these
kind of information on a testcase is fragile.
The BZ#18988 reports shows that the commit pages are easily seen with
mlockall (MCL_FUTURE) (with lock all pages that become mapped in the
process) however a more straighfoward testcase shows that pthread_create
could be faster using this patch:
--
static const int inner_count = 256;
static const int outer_count = 128;
static
void *thread1(void *arg)
{
return NULL;
}
static
void *sleeper(void *arg)
{
pthread_t ts[inner_count];
for (int i = 0; i < inner_count; i++)
pthread_create (&ts[i], &a, thread1, NULL);
for (int i = 0; i < inner_count; i++)
pthread_join (ts[i], NULL);
return NULL;
}
int main(void)
{
pthread_attr_init(&a);
pthread_attr_setguardsize(&a, 1<<20);
pthread_attr_setstacksize(&a, 1134592);
pthread_t ts[outer_count];
for (int i = 0; i < outer_count; i++)
pthread_create(&ts[i], &a, sleeper, NULL);
for (int i = 0; i < outer_count; i++)
pthread_join(ts[i], NULL);
assert(r == 0);
}
return 0;
}
--
On x86_64 (4.4.0-45-generic, gcc 5.4.0) running the small benchtests
I see:
$ time ./test
real 0m3.647s
user 0m0.080s
sys 0m11.836s
While with the patch I see:
$ time ./test
real 0m0.696s
user 0m0.040s
sys 0m1.152s
So I added a pthread_create benchtest (thread_create) which check
the thread creation latency. As for the simple benchtests, I saw
improvements in thread creation on all architectures I tested the
change.
Checked on x86_64-linux-gnu, i686-linux-gnu, aarch64-linux-gnu,
arm-linux-gnueabihf, powerpc64le-linux-gnu, sparc64-linux-gnu,
and sparcv9-linux-gnu.
[BZ #18988]
* benchtests/thread_create-inputs: New file.
* benchtests/thread_create-source.c: Likewise.
* support/xpthread_attr_setguardsize.c: Likewise.
* support/Makefile (libsupport-routines): Add
xpthread_attr_setguardsize object.
* support/xthread.h: Add xpthread_attr_setguardsize prototype.
* benchtests/Makefile (bench-pthread): Add thread_create.
* nptl/allocatestack.c (allocate_stack): Call mmap with PROT_NONE and
then mprotect the required area.
cppflags-iterator.mk no longer has anything to do with CPPFLAGS; all
it does is set libof-$(foo) for a list of files. extra-modules.mk
does the same thing, but with a different input variable, and doesn't
let the caller control the module. Therefore, this patch gives
cppflags-iterator.mk a better name, removes extra-modules.mk, and
updates all uses of both.
* extra-modules.mk: Delete file.
* cppflags-iterator.mk: Rename to ...
* libof-iterator.mk: ...this. Adjust comments.
* Makerules, extra-lib.mk, benchtests/Makefile, elf/Makefile
* elf/rtld-Rules, iconv/Makefile, locale/Makefile, malloc/Makefile
* nscd/Makefile, sunrpc/Makefile, sysdeps/s390/Makefile:
Use libof-iterator.mk instead of cppflags-iterator.mk or
extra-modules.mk.
* benchtests/strcoll-inputs/filelist#en_US.UTF-8: Remove
extra-modules.mk and cppflags-iterator.mk, add libof-iterator.mk.
of the size and alignment is based on a trace of SPEC2006. Instead of
repeating the same copy over and over again like the existing tests, it times
several thousand different copies to more accurately estimate the overhead of
branch prediction.
* benchtests/Makefile (string-benchset): Add memcpy-random.
* benchtests/bench-memcpy-random.c: New file.
Add a configure check that looks for python3 and python in that order
since we had agreed in the past to prefer python3 over python in all
our code. The patch also adjusts invocations through the various
Makefiles to use the set variable.
* configure.ac: Check for python3 or python.
* configure: Regenerated.
* config.make.in (PYTHON): New variable.
* benchtests/Makefile: Don't define PYTHON.
(bench): Define target only if PYTHON was defined.
* Rules: Don't define PYTHON.
Define pretty printer targets only if PYTHON was defined.
(tests-printers): Add to tests-unsupported if PYTHON is not
found.
(python-flags, python-invoke): Remove.
(tests-printers-out): Use PYTHON instead of python-invoke.
This patch adds fmaxf and fminf benchtests. It is based on
math/s_fmax_template.c implementation which checks for basically four
different classes:
1. if x is greater or equal than y.
2. if x is less than y.
3. if x or y is signaling.
4. if y is nan.
Cases 1 and 2 are used for default input number (by mixing normal double
numbers and infinity), while case 3 and 4 are used each for on for a
benchmark class.
Checked on x86_64-linux-gnu and powerpc64-linux-gnu.
* benchtests/Makefile (bench-math): Add fminf and fmaxf.
(CFLAGS-bench-fmaxf.c): New rule.
(CFLAGS-bench-fminf.c): Likewise.
* benchtests/fmaxf-inputs: New file.
* benchtests/fminf-inputs: Likewise.
This patch adds fmax and fmin benchtests. It is based math/s_fmax_template.c
implementation which checks for basically four different classes:
1. if x is greater or equal than y.
2. if x is less than y.
3. if x or y is signaling.
4. if y is nan.
Cases 1 and 2 are used for default input number (by mixing normal double
numbers and infinity), while case 3 and 4 are used each for on for a
benchmark class.
Checked on x86_64-linux-gnu and powerpc64-linux-gnu.
* benchtests/Makefile (bench-math): Add fmin and fmax.
(CFLAGS-bench-fmax.c): New rule.
(CFLAGS-bench-fmin.c): New rule.
* benchtests/fmax-inputs: New file.
* benchtests/fmin-inputs: Likewise.
Benchsets in benchtests use test-skeleton, so they too need to be
linked against the new libsupport DSO.
* benchtests/Makefile (binaries-benchset): Depend on libsupport
DSO.
This patch makes the sqrt benchmark use -fno-builtin, as already done
for benchmarks of ffs and ffsll, so that it actually benchmarks the
glibc function as (presumably) intended even in the presence of the
compiler inlining sqrt.
Tested for x86_64 and also used for benchmarking my ARM sqrt patch.
* benchtests/Makefile (CFLAGS-bench-sqrt.c): New variable.
This patch adds full support for cross-building benchmarks. Some
benchmarks like those that need locales to be generated cannot be
built and are hence skipped for cross builds.
Tested by cross building for aarch64 on x86_64 and then running the
generated benchmark on aarch64.
* benchtests/Makefile (wcsmbs-benchset): Include only for
native builds and runs.
(LOCALES): Likewise.
(bench-build): Build timing-type here instead of the bench
target. Generate locale only for native builds.
* benchtests/README: Add note for cross-building.
