Due to raising the minimum binutils version to version >=2.28,
the used cfi_escape for cfi_val_offset can now be ommitted.
The commit 0fc76d8762
has already adjusted it for the 64bit part of mcount.
This patch also adjusts it for the 31bit part of mcount.
Checked with "objdump -WF" / "objdump -Wf" that the previous
cfi_escape and the new cfi_val_offset are equal.
Due to raising the minimum binutils version to version >=2.28,
the used cfi_escape for cfi_val_offset can now be ommitted.
Checked with "objdump -WF" / "objdump -Wf" that the previous
cfi_escape and the new cfi_val_offset are equal.
The glibc-hwcaps subdirectories are extended by "z17". Libraries are loaded if
the z17 facility bits are active:
- Miscellaneous-instruction-extensions facility 4
- Vector-enhancements-facility 3
- Vector-Packed-Decimal-Enhancement Facility 3
- CPU: Concurrent-Functions Facility
tst-glibc-hwcaps.c is extended in order to test z17 via new marker6.
In case of running on a z17 with a kernel not recognizing z17 yet,
AT_PLATFORM will be z900 but vector-bit in AT_HWCAP is set. This situation
is now recognized and this testcase does not fail.
A fatal glibc error is dumped if glibc was build with architecture
level set for z17, but run on an older machine (See dl-hwcap-check.h).
Note, you might get an SIGILL before this check if you don't use:
configure --with-rtld-early-cflags=-march=<older-machine>
ld.so --list-diagnostics now also dumps information about s390.cpu_features.
Independent from z17, the s390x kernel won't introduce new HWCAP-Bits if there
is no special handling needed in kernel itself. For z17, we don't have new
HWCAP flags, but have to check the facility bits retrieved by
stfle-instruction.
Instead of storing all the stfle-bits (currently four 64bit values) in the
cpu_features struct, we now only store those bits, which are needed within
glibc itself. Note that we have this list twice, one with original values and
the other one which can be filtered with GLIBC_TUNABLES=glibc.cpu.hwcaps.
Those new fields are stored in so far reserved space in cpu_features struct.
Thus processes started in between the update of glibc package and we e.g. have
a new ld.so and an old libc.so, won't crash. The glibc internal ifunc-resolvers
would not select the best optimized variant.
The users of stfle-bits are also updated:
- parsing of GLIBC_TUNABLES=glibc.cpu.hwcaps
- glibc internal ifunc-resolvers
- __libc_ifunc_impl_list
- sysconf
The current approach tracks math maximum supported errors by explicitly
setting them per function and architecture. On newer implementations or
new compiler versions, the file is updated with newer values if it
shows higher results. The idea is to track the maximum known error, to
update the manual with the obtained values.
The constant libm-test-ulps shows little value, where it is usually a
mechanical change done by the maintainer, for past releases it is
usually ignored whether the ulp change resulted from a compiler
regression, and the math tests already have a maximum ulp error that
triggers a regression.
It was shown by a recent update after the new acosf [1] implementation
that is correctly rounded, where the libm-test-ulps was indeed from a
compiler issue.
This patch removes all arch-specific libm-test-ulps, adds system generic
libm-test-ulps where applicable, and changes its semantics. The generic
files now track specific implementation constraints, like if it is
expected to be correctly rounded, or if the system-specific has
different error expectations.
Now multiple libm-test-ulps can be defined, and system-specific
overrides generic implementation. This is for the case where
arch-specific implementation might show worse precision than generic
implementation, for instance, the cbrtf on i686.
Regressions are only reported if the implementation shows larger errors
than 9 ulps (13 for IBM long double) unless it is overridden by
libm-test-ulps and the maximum error is not printed at the end of tests.
The regen-ulps rule is also removed since it does not make sense to
update the libm-test-ulps automatically.
The manual error table is also removed, Paul Zimmermann and others have
been tracking libm precision with a more comprehensive analysis for some
releases; so link to his work instead.
[1] https://sourceware.org/git/?p=glibc.git;a=commit;h=9cc9f8e11e8fb8f54f1e84d9f024917634a78201
Remove unused _dl_hwcap_string defines. As a result many dl-procinfo.h headers
can be removed. This also removes target specific _dl_procinfo implementations
which only printed HWCAP strings using dl_hwcap_string.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
The CORE-MATH implementation is correctly rounded (for any rounding mode),
although it should worse performance than current one. The current
implementation performance comes mainly from the internal usage of
the optimize expf implementation, and shows a maximum ULPs of 2 for
FE_TONEAREST and 3 for other rounding modes.
The code was adapted to glibc style and to use the definition of
math_config.h (to handle errno, overflow, and underflow).
Benchtest on x64_64 (Ryzen 9 5900X, gcc 14.2.1), aarch64 (Neoverse-N1,
gcc 13.3.1), and powerpc (POWER10, gcc 13.2.1):
Latency master patched improvement
x86_64 40.6995 49.0737 -20.58%
x86_64v2 40.5841 44.3604 -9.30%
x86_64v3 39.3879 39.7502 -0.92%
i686 112.3380 129.8570 -15.59%
aarch64 (Neoverse) 18.6914 17.0946 8.54%
power10 11.1343 9.3245 16.25%
reciprocal-throughput master patched improvement
x86_64 18.6471 24.1077 -29.28%
x86_64v2 17.7501 20.2946 -14.34%
x86_64v3 17.8262 17.1877 3.58%
i686 64.1454 86.5645 -34.95%
aarch64 (Neoverse) 9.77226 12.2314 -25.16%
power10 4.0200 5.3316 -32.63%
Signed-off-by: Alexei Sibidanov <sibid@uvic.ca>
Signed-off-by: Paul Zimmermann <Paul.Zimmermann@inria.fr>
Signed-off-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Reviewed-by: DJ Delorie <dj@redhat.com>
If an executable is static PIE and has a non-zero load address
(compare to elf/tst-pie-address-static), it segfaults as
elf_machine_load_address() returns 0x0 and elf_machine_dynamic()
returns the run-time instead of link-time address of _DYNAMIC.
Now rely on __ehdr_start and _DYNAMIC as also done on other
architectures.
Checked back to old arch-levels that this approach works fine:
- 31bit: -march=g5
- 64bit: -march=z900
Note, that there is no static-PIE support on 31bit, but this
approach cleans it also up.
Furthermore this cleanup in glibc does not change anything
regarding the first GOT-element as the s390 ABI
(https://github.com/IBM/s390x-abi) explicitely defines:
The doubleword at _GLOBAL_OFFSET_TABLE_[0] is set by the linkage
editor to hold the address of the dynamic structure, referenced
with the symbol _DYNAMIC. This allows a program, such as the dynamic
linker, to find its own dynamic structure without having yet processed
its relocation entries. This is especially important for the dynamic
linker, because it must initialize itself without relying on other
programs to relocate its memory image.
The CORE-MATH implementation is correctly rounded (for any rounding mode)
and shows better performance to the generic tanf.
The code was adapted to glibc style, to use the definition of
math_config.h, to remove errno handling, and to use a generic
128 bit routine for ABIs that do not support it natively.
Benchtest on x64_64 (Ryzen 9 5900X, gcc 14.2.1), aarch64 (neoverse1,
gcc 13.2.1), and powerpc (POWER10, gcc 13.2.1):
latency master patched improvement
x86_64 82.3961 54.8052 33.49%
x86_64v2 82.3415 54.8052 33.44%
x86_64v3 69.3661 50.4864 27.22%
i686 219.271 45.5396 79.23%
aarch64 29.2127 19.1951 34.29%
power10 19.5060 16.2760 16.56%
reciprocal-throughput master patched improvement
x86_64 28.3976 19.7334 30.51%
x86_64v2 28.4568 19.7334 30.65%
x86_64v3 21.1815 16.1811 23.61%
i686 105.016 15.1426 85.58%
aarch64 18.1573 10.7681 40.70%
power10 8.7207 8.7097 0.13%
Signed-off-by: Alexei Sibidanov <sibid@uvic.ca>
Signed-off-by: Paul Zimmermann <Paul.Zimmermann@inria.fr>
Signed-off-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Reviewed-by: DJ Delorie <dj@redhat.com>
The CORE-MATH implementation is correctly rounded (for any rounding mode)
and shows better performance compared to the generic exp2m1f.
The code was adapted to glibc style and to use the definition of
math_config.h (to handle errno, overflow, and underflow). The
only change is to handle FLT_MAX_EXP for FE_DOWNWARD or FE_TOWARDZERO.
The benchmark inputs are based on exp2f ones.
Benchtest on x64_64 (Ryzen 9 5900X, gcc 14.2.1), aarch64 (Neoverse-N1,
gcc 13.3.1), and powerpc (POWER10, gcc 13.2.1):
Latency master patched improvement
x86_64 40.6042 48.7104 -19.96%
x86_64v2 40.7506 35.9032 11.90%
x86_64v3 35.2301 31.7956 9.75%
i686 102.094 94.6657 7.28%
aarch64 18.2704 15.1387 17.14%
power10 11.9444 8.2402 31.01%
reciprocal-throughput master patched improvement
x86_64 20.8683 16.1428 22.64%
x86_64v2 19.5076 10.4474 46.44%
x86_64v3 19.2106 10.4014 45.86%
i686 56.4054 59.3004 -5.13%
aarch64 12.0781 7.3953 38.77%
power10 6.5306 5.9388 9.06%
The generic implementation calls __ieee754_exp2f and x86_64 provides
an optimized ifunc version (built with -mfma -mavx2, not correctly
rounded). This explains the performance difference for x86_64.
Same for i686, where the ABI provides an optimized __ieee754_exp2f
version built with '-msse2 -mfpmath=sse'. When built wth same
flags, the new algorithm shows a better performance:
master patched improvement
latency 102.094 91.2823 10.59%
reciprocal-throughput 56.4054 52.7984 6.39%
Signed-off-by: Alexei Sibidanov <sibid@uvic.ca>
Signed-off-by: Paul Zimmermann <Paul.Zimmermann@inria.fr>
Signed-off-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Reviewed-by: DJ Delorie <dj@redhat.com>
The CORE-MATH implementation is correctly rounded (for any rounding mode)
and shows better performance compared to the generic exp10m1f.
The code was adapted to glibc style and to use the definition of
math_config.h (to handle errno, overflow, and underflow). I mostly
fixed some small issues in corner cases (sNaN handling, -INFINITY,
a specific overflow check).
Benchtest on x64_64 (Ryzen 9 5900X, gcc 14.2.1), aarch64 (Neoverse-N1,
gcc 13.3.1), and powerpc (POWER10, gcc 13.2.1):
Latency master patched improvement
x86_64 45.4690 49.5845 -9.05%
x86_64v2 46.1604 36.2665 21.43%
x86_64v3 37.8442 31.0359 17.99%
i686 121.367 93.0079 23.37%
aarch64 21.1126 15.0165 28.87%
power10 12.7426 8.4929 33.35%
reciprocal-throughput master patched improvement
x86_64 19.6005 17.4005 11.22%
x86_64v2 19.6008 11.1977 42.87%
x86_64v3 17.5427 10.2898 41.34%
i686 59.4215 60.9675 -2.60%
aarch64 13.9814 7.9173 43.37%
power10 6.7814 6.4258 5.24%
The generic implementation calls __ieee754_exp10f which has an
optimized version, although it is not correctly rounded, which is
the main culprit of the the latency difference for x86_64 and
throughp for i686.
Signed-off-by: Alexei Sibidanov <sibid@uvic.ca>
Signed-off-by: Paul Zimmermann <Paul.Zimmermann@inria.fr>
Signed-off-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Reviewed-by: DJ Delorie <dj@redhat.com>
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>
Building the s390 specific iconv modules - utf16-utf32-z9.c, utf8-utf32-z9.c
and utf8-utf16-z9.c - with -fno-omit-frame-pointer leads to a build error
"error: %r11 cannot be used in 'asm' here" as r11 is needed as frame-pointer.
The cuXY-instructions need two even-odd register pairs. Therefore the register
pinning is used. This patch just uses a different register pair.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
The z13/vector-optimized wcsncmp implementation segfaults if n=1
and there is only one character (equal on both strings) before
the page end. Then it loads and compares one character and misses
to check n again. The following load fails.
This patch removes the extra load and compare of the first character
and just start with the loop which uses vector-load-to-block-boundary.
This code-path also checks n.
With this patch both tests are passing:
- the simplified one mentioned in the bugzilla 31934
- the full one in Florian Weimer's patch:
"manual: Document a GNU extension for strncmp/wcsncmp"
(https://patchwork.sourceware.org/project/glibc/patch/874j9eml6y.fsf@oldenburg.str.redhat.com/):
On s390x-linux-gnu (z16), the new wcsncmp test fails due to bug 31934.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
