The CORE-MATH implementation is correctly rounded (for any rounding mode).
This can be checked by exhaustive tests in a few minutes since there are
less than 2^32 values to check against for example GNU MPFR.
This patch also adds some bench values for tgammaf.
Tested on x86_64 and x86 (cfarm26).
With the initial GNU libc code it gave on an Intel(R) Core(TM) i7-8700:
"tgammaf": {
"": {
"duration": 3.50188e+09,
"iterations": 2e+07,
"max": 602.891,
"min": 65.1415,
"mean": 175.094
}
}
With the new code:
"tgammaf": {
"": {
"duration": 3.30825e+09,
"iterations": 5e+07,
"max": 211.592,
"min": 32.0325,
"mean": 66.1649
}
}
With the initial GNU libc code it gave on cfarm26 (i686):
"tgammaf": {
"": {
"duration": 3.70505e+09,
"iterations": 6e+06,
"max": 2420.23,
"min": 243.154,
"mean": 617.509
}
}
With the new code:
"tgammaf": {
"": {
"duration": 3.24497e+09,
"iterations": 1.8e+07,
"max": 1238.15,
"min": 101.155,
"mean": 180.276
}
}
Signed-off-by: Alexei Sibidanov <sibid@uvic.ca>
Signed-off-by: Paul Zimmermann <Paul.Zimmermann@inria.fr>
Changes in v2:
- include <math.h> (fix the linknamespace failures)
- restored original benchtests/strcoll-inputs/filelist#en_US.UTF-8 file
- restored original wrapper code (math/w_tgammaf_compat.c),
except for the dealing with the sign
- removed the tgammaf/float entries in all libm-test-ulps files
- address other comments from Joseph Myers
(https://sourceware.org/pipermail/libc-alpha/2024-July/158736.html)
Changes in v3:
- pass NULL argument for signgam from w_tgammaf_compat.c
- use of math_narrow_eval
- added more comments
Changes in v4:
- initialize local_signgam to 0 in math/w_tgamma_template.c
- replace sysdeps/ieee754/dbl-64/gamma_productf.c by dummy file
Changes in v5:
- do not mention local_signgam any more in math/w_tgammaf_compat.c
- initialize local_signgam to 1 instead of 0 in w_tgamma_template.c
and added comment
Changes in v6:
- pass NULL as 2nd argument of __ieee754_gammaf_r in
w_tgammaf_compat.c, and check for NULL in e_gammaf_r.c
Changes in v7:
- added Signed-off-by line for Alexei Sibidanov (author of the code)
Changes in v8:
- added Signed-off-by line for Paul Zimmermann (submitted of the patch)
Changes in v9:
- address comments from review by Adhemerval Zanella
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Add a new randomized memset test similar to bench-random-memcpy. Instead of
repeating the same call to memset over and over again, it times a large number
of different inputs. The distribution of memset length and alignment is based
on SPEC2017 (length up to 4096 and alignment up to 64).
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
The walk benchmarks don't measure anything useful - memory is not initialized
properly so doing a single walk in 32MB just measures reading the 4KB zero
page for reads and clear_page overhead for writes. The memset variants don't
even manage to do a walk in the 32MB region due to using incorrect pointer
increments... Neither is it clear why it is walking backwards since this
won't confuse modern prefetchers. If you fix the benchmark and print the
bandwidth, the results are identical for all sizes larger than ~1KB since it
is just testing memory bandwidth of a single 32MB block. This case is already
tested by the large benchmark, so overall it doesn't seem useful to keep these.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
POSIX.1-2024 (now official) specifies tm_gmtoff and tm_zone.
This is a good time to update the manual’s “Date and Time”
chapter so I went through it, fixed some outdated
stuff that had been in there for decades, and improved it to match
POSIX.1-2024 better and to clarify some implementation-defined
behavior. Glibc already conforms to POSIX.1-2024 in these matters, so
this is merely a documentation change.
* manual/examples/strftim.c: Use snprintf instead of now-deprecated
function asctime. Check for localtime failure. Simplify by using
puts instead of fputs. Prefer ‘buf, sizeof buf’ to less-obvious
‘buffer, SIZE’.
* manual/examples/timespec_subtract.c: Modernize to use struct
timespec not struct timeval, and rename from timeval_subtract.c.
All uses changed. Check for overflow. Do not check for negative
return value, which ought to be OK since negative time_t is OK.
Use GNU indenting style.
* manual/time.texi:
Document CLOCKS_PER_SEC, TIME_UTC, timespec_get, timespec_getres,
strftime_l.
Document the storage lifetime of tm_zone and of tzname.
Caution against use of tzname, timezone and daylight, saying that
these variables have unspecified values when TZ is geographic.
This is what glibc actually does (contrary to what the manual said
before this patch), and POSIX is planned to say the same thing
<https://austingroupbugs.net/view.php?id=1816>.
Also say that directly accessing the variables is not thread-safe.
Say that localtime_r and ctime_r don’t necessarily set time zone
state. Similarly, in the tzset documentation, say that it is called
by ctime, localtime, mktime, strftime, not that it is called by all
time conversion functions that depend on the time zone.
Say that tm_isdst is useful mostly just for mktime, and that
other uses should prefer tm_gmtoff and tm_zone instead.
Do not say that strftime ignores tm_gmtoff and tm_zone, because
it doesn’t do that.
Document what gmtime does to tm_gmtoff and tm_zone.
Say that the asctime, asctime_r, ctime, and ctime_r are now deprecated
and/or obsolescent, and that behavior is undefined if the year is <
1000 or > 9999. Document strftime before these now-obsolescent
functions, so that readers see the useful function first.
Coin the terms “geographical format” and “proleptic format” for the
two main formats of TZ settings, to simplify exposition. Use this
wording consistently.
Update top-level proleptic syntax to match POSIX.1-2024, which glibc
already implements. Document the angle-bracket quoted forms of time
zone abbreviations in proleptic TZ. Say that time zone abbreviations
can contain only ASCII alphanumerics, ‘+’, and ‘-’.
Document what happens if the proleptic form specifies a DST
abbreviation and offset but omits the rules. POSIX says this is
implementation-defined so we need to document it. Although this
documentation mentions ‘posixrules’ tersely, we need to rethink
‘posixrules’ since I think it stops working after 2038.
Clarify wording about TZ settings beginning with ‘;’.
Say that timegm is in ISO C (as of C23).
Say that POSIX.1-2024 removed gettimeofday.
Say that tm_gmtoff and tm_zone are extensions to ISO C, which is
clearer than saying they are invisible in a struct ISO C enviroment,
and gives us more wiggle room if we want to make them visible in
strict ISO C, something that ISO C allows.
Drop mention of old standards like POSIX.1c and POSIX.2-1992 in the
text when the history is so old that it’s no longer useful in a
general-purpose manual.
Define Coordinated Universal Time (UTC), time zone, time zone ruleset,
and POSIX Epoch, and use these phrases more consistently.
Improve TZ examples to show more variety, and to reflect current
practice and timestamps. Remove obsolete example about Argentina.
Add an example for Ireland.
Don’t rely on GCC extensions when explaining ctime_r.
Do not say that difftime produces the mathematically correct result,
since it might be inexact.
For clock_t don’t say “as in the example above” when there is no
such example, and don’t say that casting to double works “properly
and consistently no matter what”, as it suffers from rounding and
overflow.
Don’t say broken-down time is not useful for calculations; it’s
merely painful.
Say that UTC is not defined before 1960.
Rename Time Zone Functions to Time Zone State. All uses changed.
Update Internet RFC 822 → 5322, 1305 → 5905. Drop specific years of
ISO 8601 as they don’t matter.
Minor style changes: @code{"..."} → @t{"..."} to avoid overquoting in
info files, @code → @env for environment variables, Daylight Saving
Time → daylight saving time, white space → whitespace, prime meridian
→ Prime Meridian.
Measure duration of 100 fclose calls after opening 1 million FILEs.
Signed-off-by: H.J. Lu <hjl.tools@gmail.com>
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Add another difficult needle to strstr that clearly shows the quadratic
complexity of bruteforce algorithms.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Add a simple benchmark to measure the overhead of internal libc locks in
the random() implementation on both single- and multi-threaded cases.
This relies on the implementation of random using internal locks to
access shared global data, and that the runtime uses multi-threaded
locking once a thread has been created (even after it finishes).
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Use same strategy as bench-strstr.c (93eebae516 and 80b2bfb535)
and use json_ctx for output to help standardize format across all
benchtests.
Reviewed-by: Arjun Shankar <arjun@redhat.com>
This patch adds more benchtests for rounding functions.
The double inputs are copied from trunc-inputs, the float inputs are copied from truncf-inputs. and the rintf is copied from rint-inputs.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Increase benchmark iterations for math and vector math functions to improve
timing accuracy. Vector math benchmarks now take 1-3 seconds on a modern CPU.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Few tests needed to properly check for asprintf and system calls return
values with _FORTIFY_SOURCE enabled.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reflow all long lines adding comment terminators.
Sort all reflowed text using scripts/sort-makefile-lines.py.
No regressions running microbenchmarks.
No code generation changes observed in binary artifacts.
No regressions on x86_64 and i686.
This patch enables libmvec on AArch64. The proposed change is mainly
implementing build infrastructure to add the new routines to ABI,
tests and benchmarks. I have demonstrated how this all fits together
by adding implementations for vector cos, in both single and double
precision, targeting both Advanced SIMD and SVE.
The implementations of the routines themselves are just loops over the
scalar routine from libm for now, as we are more concerned with
getting the plumbing right at this point. We plan to contribute vector
routines from the Arm Optimized Routines repo that are compliant with
requirements described in the libmvec wiki.
Building libmvec requires minimum GCC 10 for SVE ACLE. To avoid raising
the minimum GCC by such a big jump, we allow users to disable libmvec
if their compiler is too old.
Note that at this point users have to manually call the vector math
functions. This seems to be acceptable to some downstream users.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Adjust iteration counts so benchmarks don't run too slowly or quickly.
Ensure benchmarks take less than 10 seconds on older, slower cores and
more than 0.5 seconds on fast cores.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
1. Subnormals: 128 inputs.
2. Normal numbers with large exponent difference (|x/y| > 2^8):
1024 inputs between FLT_MIN and FLT_MAX;
3. Close exponents (ey >= -103 and |x/y| < 2^8): 1024 inputs with
exponents between -10 and 10.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Add three different dataset, from random floating point numbers:
1. Subnormals: 128 inputs.
2. Normal numbers with large exponent difference (|x/y| > 2^52):
1024 inputs between DBL_MIN and DBL_MAX;
3. Close exponents (ey >= -907 and |x/y| < 2^52): 1024 inputs with
exponents between -10 and 10.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
This allows other targets to use the same inputs for their own libmvec
microbenchmarks without having to duplicate them in their own
subdirectory.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Remove simple_strcspn/strpbrk/strsep which are significantly slower than the
generic implementations. Also remove oldstrsep and oldstrtok since they are
practically identical to the generic implementation. Adjust iteration count
to reduce benchmark time.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Remove memchr_strnlen since it is now the same as generic_strnlen. Adjust
iteration count to reduce benchmark time. Keep memchr_strlen since the
generic strlen does not use memchr.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Almost all uses of rawmemchr find the end of a string. Since most targets use
a generic implementation, replacing it with strchr is better since that is
optimized by compilers into strlen (s) + s. Also fix the generic rawmemchr
implementation to use a cast to unsigned char in the if statement.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
It shows both throughput (total bytes obtained in the test duration)
and latecy for both arc4random and arc4random_buf with different
sizes.
Checked on x86_64-linux-gnu, aarch64-linux, and powerpc64le-linux-gnu.
Add a second iteration for memrchr to set `pos` starting from the end
of the buffer.
Previously `pos` was only set relative to the beginning of the
buffer. This isn't really useful for memrchr because the beginning
of the search space is (buf + len).
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
1. Use json_ctx for output to help standardize format across all
benchtests.
2. Add some additional tests to strstr and memchr expanding alignments
and adding more small values.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
1. Output results in json format so its easier to parse
2. Increase max alignment to `getpagesize () - 1` to make it possible
to test page cross cases.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Unroll slightly and enforce good instruction scheduling. This improves
performance on out-of-order machines. The unrolling allows for
pipelined multiplies.
As well, as an optional sysdep, reorder the operations and prevent
reassosiation for better scheduling and higher ILP. This commit
only adds the barrier for x86, although it should be either no
change or a win for any architecture.
Unrolling further started to induce slowdowns for sizes [0, 4]
but can help the loop so if larger sizes are the target further
unrolling can be beneficial.
Results for _dl_new_hash
Benchmarked on Tigerlake: 11th Gen Intel(R) Core(TM) i7-1165G7 @ 2.80GHz
Time as Geometric Mean of N=30 runs
Geometric of all benchmark New / Old: 0.674
type, length, New Time, Old Time, New Time / Old Time
fixed, 0, 2.865, 2.72, 1.053
fixed, 1, 3.567, 2.489, 1.433
fixed, 2, 2.577, 3.649, 0.706
fixed, 3, 3.644, 5.983, 0.609
fixed, 4, 4.211, 6.833, 0.616
fixed, 5, 4.741, 9.372, 0.506
fixed, 6, 5.415, 9.561, 0.566
fixed, 7, 6.649, 10.789, 0.616
fixed, 8, 8.081, 11.808, 0.684
fixed, 9, 8.427, 12.935, 0.651
fixed, 10, 8.673, 14.134, 0.614
fixed, 11, 10.69, 15.408, 0.694
fixed, 12, 10.789, 16.982, 0.635
fixed, 13, 12.169, 18.411, 0.661
fixed, 14, 12.659, 19.914, 0.636
fixed, 15, 13.526, 21.541, 0.628
fixed, 16, 14.211, 23.088, 0.616
fixed, 32, 29.412, 52.722, 0.558
fixed, 64, 65.41, 142.351, 0.459
fixed, 128, 138.505, 295.625, 0.469
fixed, 256, 291.707, 601.983, 0.485
random, 2, 12.698, 12.849, 0.988
random, 4, 16.065, 15.857, 1.013
random, 8, 19.564, 21.105, 0.927
random, 16, 23.919, 26.823, 0.892
random, 32, 31.987, 39.591, 0.808
random, 64, 49.282, 71.487, 0.689
random, 128, 82.23, 145.364, 0.566
random, 256, 152.209, 298.434, 0.51
Co-authored-by: Alexander Monakov <amonakov@ispras.ru>
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Add a simple benchmark that measures wcrtomb performance with various
locales with 1-4 byte characters.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Improve libmvec benchmark integration so that in future other
architectures may be able to run their libmvec benchmarks as well. This
now allows libmvec benchmarks to be run with `make BENCHSET=bench-math`.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
The libmvec benchmarks print a message indicating that a certain CPU
feature is unsupported and exit prematurelyi, which breaks the JSON in
bench.out.
Handle this more elegantly in the bench makefile target by adding
support for an UNSUPPORTED exit status (77) so that bench.out continues
to have output for valid tests.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Benchmark for testing pthread mutex locks performance with different
threads and critical sections.
The test configuration consists of 3 parts:
1. thread number
2. critical-section length
3. non-critical-section length
Thread number starts from 1 and increased by 2x until num of CPU cores
(nprocs). An additional over-saturation case (1.25 * nprocs) is also
included.
Critical-section is represented by a loop of shared do_filler(),
length can be determined by the loop iters.
Non-critical-section is similiar to the critical-section, except it's
based on non-shared do_filler().
Currently, adaptive pthread_mutex lock is tested.
1. Use json-lib for printing results.
2. Expose all parameters (before pos, seek_char, and max_char where
not printed).
3. Add benchmarks that test multiple occurence of seek_char in the
string.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
