Noah Goldstein [Mon, 1 Nov 2021 05:49:52 +0000 (00:49 -0500)]
x86: Double size of ERMS rep_movsb_threshold in dl-cacheinfo.h
No bug.
This patch doubles the rep_movsb_threshold when using ERMS. Based on
benchmarks the vector copy loop, especially now that it handles 4k
aliasing, is better for these medium ranged.
On Skylake with ERMS:
Size, Align1, Align2, dst>src,(rep movsb) / (vec copy)
4096, 0, 0, 0, 0.975
4096, 0, 0, 1, 0.953
4096, 12, 0, 0, 0.969
4096, 12, 0, 1, 0.872
4096, 44, 0, 0, 0.979
4096, 44, 0, 1, 0.83
4096, 0, 12, 0, 1.006
4096, 0, 12, 1, 0.989
4096, 0, 44, 0, 0.739
4096, 0, 44, 1, 0.942
4096, 12, 12, 0, 1.009
4096, 12, 12, 1, 0.973
4096, 44, 44, 0, 0.791
4096, 44, 44, 1, 0.961
4096, 2048, 0, 0, 0.978
4096, 2048, 0, 1, 0.951
4096, 2060, 0, 0, 0.986
4096, 2060, 0, 1, 0.963
4096, 2048, 12, 0, 0.971
4096, 2048, 12, 1, 0.941
4096, 2060, 12, 0, 0.977
4096, 2060, 12, 1, 0.949
8192, 0, 0, 0, 0.85
8192, 0, 0, 1, 0.845
8192, 13, 0, 0, 0.937
8192, 13, 0, 1, 0.939
8192, 45, 0, 0, 0.932
8192, 45, 0, 1, 0.927
8192, 0, 13, 0, 0.621
8192, 0, 13, 1, 0.62
8192, 0, 45, 0, 0.53
8192, 0, 45, 1, 0.516
8192, 13, 13, 0, 0.664
8192, 13, 13, 1, 0.659
8192, 45, 45, 0, 0.593
8192, 45, 45, 1, 0.575
8192, 2048, 0, 0, 0.854
8192, 2048, 0, 1, 0.834
8192, 2061, 0, 0, 0.863
8192, 2061, 0, 1, 0.857
8192, 2048, 13, 0, 0.63
8192, 2048, 13, 1, 0.629
8192, 2061, 13, 0, 0.627
8192, 2061, 13, 1, 0.62
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Noah Goldstein [Mon, 1 Nov 2021 05:49:51 +0000 (00:49 -0500)]
x86: Optimize memmove-vec-unaligned-erms.S
No bug.
The optimizations are as follows:
1) Always align entry to 64 bytes. This makes behavior more
predictable and makes other frontend optimizations easier.
2) Make the L(more_8x_vec) cases 4k aliasing aware. This can have
significant benefits in the case that:
0 < (dst - src) < [256, 512]
3) Align before `rep movsb`. For ERMS this is roughly a [0, 30%]
improvement and for FSRM [-10%, 25%].
In addition to these primary changes there is general cleanup
throughout to optimize the aligning routines and control flow logic.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Noah Goldstein [Mon, 1 Nov 2021 05:49:50 +0000 (00:49 -0500)]
benchtests: Add partial overlap case in bench-memmove-walk.c
This commit adds a new partial overlap benchmark. This is generally
the most interesting performance case for memmove and was missing.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Noah Goldstein [Mon, 1 Nov 2021 05:49:49 +0000 (00:49 -0500)]
benchtests: Add additional cases to bench-memcpy.c and bench-memmove.c
This commit adds more benchmarks for the common memcpy/memmove
benchmarks. The most signifcant cases are the half page offsets. The
current versions leaves dst and src near page aligned which leads to
false 4k aliasing on x86_64. This can add noise due to false
dependencies from one run to the next. As well, this seems like more
of an edge case that common case so it shouldn't be the only thing
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Noah Goldstein [Mon, 1 Nov 2021 05:49:48 +0000 (00:49 -0500)]
string: Make tests birdirectional test-memcpy.c
This commit updates the memcpy tests to test both dst > src and dst <
src. This is because there is logic in the code based on the
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
H.J. Lu [Sat, 6 Nov 2021 13:16:30 +0000 (06:16 -0700)]
Remove the last trace of generate-md5 [BZ #28554]
generate-md5 was removed by
commit
d73f5331ce5370ca5a879229e3842f5de98689cd
Author: Roland McGrath <roland@gnu.org>
Date: Fri May 2 02:20:45 2003 +0000
2003-05-01 Roland McGrath <roland@redhat.com>
Remove its last trace. This fixes BZ #28554.
Sunil K Pandey [Fri, 5 Nov 2021 23:13:12 +0000 (16:13 -0700)]
Revert "benchtests: Add acosf function to bench-math"
This reverts commit
79d0fc65395716c1d95931064c7bf37852203c66.
H.J. Lu [Wed, 9 Jun 2021 12:29:16 +0000 (05:29 -0700)]
Configure GCC with --enable-initfini-array [BZ #27945]
Starting from GCC 12, the .init_array and .fini_array sections are enabled
unconditionally by
commit
13a39886940331149173b25d6ebde0850668d8b9
Author: H.J. Lu <hjl.tools@gmail.com>
Date: Tue Jun 8 16:09:24 2021 -0700
Always enable DT_INIT_ARRAY/DT_FINI_ARRAY on Linux
configure GCC with --enable-initfini-array to enable them when using GCC
release branches.
Fixes BZ #27945.
Florian Weimer [Fri, 5 Nov 2021 16:01:24 +0000 (17:01 +0100)]
elf: Earlier missing dynamic segment check in _dl_map_object_from_fd
Separated debuginfo files have PT_DYNAMIC with p_filesz == 0. We
need to check for that before the _dl_map_segments call because
that could attempt to write to mappings that extend beyond the end
of the file, resulting in SIGBUS.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Nikita Popov [Tue, 2 Nov 2021 08:21:42 +0000 (13:21 +0500)]
gconv: Do not emit spurious NUL character in ISO-2022-JP-3 (bug 28524)
Bugfix 27256 has introduced another issue:
In conversion from ISO-2022-JP-3 encoding, it is possible
to force iconv to emit extra NUL character on internal state reset.
To do this, it is sufficient to feed iconv with escape sequence
which switches active character set.
The simplified check 'data->__statep->__count != ASCII_set'
introduced by the aforementioned bugfix picks that case and
behaves as if '\0' character has been queued thus emitting it.
To eliminate this issue, these steps are taken:
* Restore original condition
'(data->__statep->__count & ~7) != ASCII_set'.
It is necessary since bits 0-2 may contain
number of buffered input characters.
* Check that queued character is not NUL.
Similar step is taken for main conversion loop.
Bundled test case follows following logic:
* Try to convert ISO-2022-JP-3 escape sequence
switching active character set
* Reset internal state by providing NULL as input buffer
* Ensure that nothing has been converted.
Signed-off-by: Nikita Popov <npv1310@gmail.com>
Paul A. Clarke [Tue, 19 Oct 2021 14:49:12 +0000 (09:49 -0500)]
[powerpc] Tighten contraints for asm constant parameters
There are a few places where only known numeric values are acceptable for
`asm` parameters, yet the constraint "i" is used. "i" can include
"symbolic constants whose values will be known only at assembly time or
later."
Use "n" instead of "i" where known numeric values are required.
Suggested-by: Segher Boessenkool <segher@kernel.crashing.org>
Reviewed-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
Adhemerval Zanella [Wed, 3 Nov 2021 12:19:30 +0000 (09:19 -0300)]
elf: Do not run DSO sorting if tunables is not enabled
Since the argorithm selection requires tunables.
Checked on x86_64-linux-gnu with --enable-tunables=no.
Adhemerval Zanella [Tue, 2 Nov 2021 19:51:39 +0000 (16:51 -0300)]
riscv: Build with -mno-relax if linker does not support R_RISCV_ALIGN
It allows build both glibc and tests with lld (Since lld does not
support R_RISCV_ALIGN linker relaxation).
Checked with a build for riscv32-linux-gnu-rv32imafdc-ilp32d and
riscv64-linux-gnu-rv64imafdc-lp64d.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Reviewed-by: Fangrui Song <maskray@google.com>
Fangrui Song [Wed, 3 Nov 2021 03:59:52 +0000 (20:59 -0700)]
x86-64: Replace movzx with movzbl
Clang cannot assemble movzx in the AT&T dialect mode.
../sysdeps/x86_64/strcmp.S:2232:16: error: invalid operand for instruction
movzx (%rsi), %ecx
^~~~
Change movzx to movzbl, which follows the AT&T dialect and is used
elsewhere in the file.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Fangrui Song [Tue, 2 Nov 2021 17:07:59 +0000 (10:07 -0700)]
regex: Unnest nested functions in regcomp.c
This refactor moves four functions out of a nested scope and converts
them into static always_inline functions. collseqwc, table_size,
symb_table, extra are now initialized to zero because they are passed as
function arguments.
On x86-64, .text is 16 byte larger likely due to the 4 stores.
This is nothing compared to the amount of work that regcomp has to do
looking up the collation weights, or other functions.
If the non-buildable `sysdeps/generic/dl-machine.h` doesn't count,
this patch removes the last `auto inline` usage from glibc.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Joseph Myers [Tue, 2 Nov 2021 16:54:59 +0000 (16:54 +0000)]
Use Linux 5.15 in build-many-glibcs.py
This patch makes build-many-glibcs.py use Linux 5.15.
Tested with build-many-glibcs.py (host-libraries, compilers and glibcs
builds).
Adhemerval Zanella [Fri, 29 Oct 2021 13:29:54 +0000 (10:29 -0300)]
elf: Assume disjointed .rela.dyn and .rela.plt for loader
The patch removes the the ELF_DURING_STARTUP optimization and assume
both .rel.dyn and .rel.plt might not be subsequent. This allows some
code simplification since relocation will be handled independently
where it is done on bootstrap.
At least on x86_64_64, I can not measure any performance implications.
Running 10000 time the command
LD_DEBUG=statistics ./elf/ld.so ./libc.so
And filtering the "total startup time in dynamic loader" result,
the geometric mean is:
patched master
Ryzen 7 5900x 24140 24952
i7-4510U 45957 45982
(The results do show some variation, I did not make any statistical
analysis).
It also allows build arm with lld, since it inserts ".ARM.exidx"
between ".rel.dyn" and ".rel.plt" for the loader.
Checked on x86_64-linux-gnu and arm-linux-gnueabihf.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Florian Weimer [Tue, 2 Nov 2021 09:26:23 +0000 (10:26 +0100)]
i386: Explain why __HAVE_64B_ATOMICS has to be 0
Adhemerval Zanella [Mon, 5 Apr 2021 17:08:42 +0000 (14:08 -0300)]
benchtests: Add hypotf
Based on random input arguments. About 85% tuples have exponents
of the two arguments close together (+-1 range).
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Adhemerval Zanella [Mon, 8 Mar 2021 18:21:56 +0000 (15:21 -0300)]
benchtests: Make hypot input random
Instead of inputs based on the algorithm implementation details.
About 85% tuples have exponents of the two arguments close
together (+-1 range).
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Adhemerval Zanella [Thu, 28 Oct 2021 11:15:10 +0000 (08:15 -0300)]
arm: Use have-mtls-dialect-gnu2 to check for ARM TLS descriptors support
The lld linker does not support TLSDESC for arm. The have-arm-tls-desc
is a leftover of
56583289b1 to support NaCL.
Reviewed-by: Fangrui Song <maskray@google.com>
Adhemerval Zanella [Thu, 28 Oct 2021 10:55:36 +0000 (07:55 -0300)]
arm: Use internal symbol for _dl_argv on _dl_start_user
The lld does not support R_ARM_GOTOFF32 to preemptible symbol (_dl_argv
has default visibility). Use the internal alias instead (one option
would to use HIDDEN_JUMPTARGET, bu the macro is not defined for
!__ASSEMBLER__ and I made this patch arm-specific to avoid require to
check extensivelly on other architecture it this might break something).
Checked on arm-linux-gnueabihf.
Reviewed-by: Fangrui Song <maskray@google.com>
H.J. Lu [Fri, 29 Oct 2021 19:56:53 +0000 (12:56 -0700)]
x86-64: Remove Prefer_AVX2_STRCMP
Remove Prefer_AVX2_STRCMP to enable EVEX strcmp. When comparing 2 32-byte
strings, EVEX strcmp has been improved to require 1 load, 1 VPTESTM, 1
VPCMP, 1 KMOVD and 1 INCL instead of 2 loads, 3 VPCMPs, 2 KORDs, 1 KMOVD
and 1 TESTL while AVX2 strcmp requires 1 load, 2 VPCMPEQs, 1 VPMINU, 1
VPMOVMSKB and 1 TESTL. EVEX strcmp is now faster than AVX2 strcmp by up
to 40% on Tiger Lake and Ice Lake.
H.J. Lu [Fri, 29 Oct 2021 19:40:20 +0000 (12:40 -0700)]
x86-64: Improve EVEX strcmp with masked load
In strcmp-evex.S, to compare 2 32-byte strings, replace
VMOVU (%rdi, %rdx), %YMM0
VMOVU (%rsi, %rdx), %YMM1
/* Each bit in K0 represents a mismatch in YMM0 and YMM1. */
VPCMP $4, %YMM0, %YMM1, %k0
VPCMP $0, %YMMZERO, %YMM0, %k1
VPCMP $0, %YMMZERO, %YMM1, %k2
/* Each bit in K1 represents a NULL in YMM0 or YMM1. */
kord %k1, %k2, %k1
/* Each bit in K1 represents a NULL or a mismatch. */
kord %k0, %k1, %k1
kmovd %k1, %ecx
testl %ecx, %ecx
jne L(last_vector)
with
VMOVU (%rdi, %rdx), %YMM0
VPTESTM %YMM0, %YMM0, %k2
/* Each bit cleared in K1 represents a mismatch or a null CHAR
in YMM0 and 32 bytes at (%rsi, %rdx). */
VPCMP $0, (%rsi, %rdx), %YMM0, %k1{%k2}
kmovd %k1, %ecx
incl %ecx
jne L(last_vector)
It makes EVEX strcmp faster than AVX2 strcmp by up to 40% on Tiger Lake
and Ice Lake.
Co-Authored-By: Noah Goldstein <goldstein.w.n@gmail.com>
Sunil K Pandey [Wed, 27 Oct 2021 20:50:16 +0000 (13:50 -0700)]
benchtests: Add acosf function to bench-math
Add acosf function to bench-math and copy acosf-inputs to benchtests.
Motivation for this patch is to prepare for upcoming libmvec new
functions. Float and double version of libmvec functions stays
together.
acosf-inputs file generated from acos-inputs file using following
scaling formula:
f = d * (FLT_MAX/DBL_MAX)
Where d is input(double) and f is output(float). If scaled float value
is duplicate in new input file, nextafterf() function used to find next
float value, ensuring no duplicates.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Wilco Dijkstra [Fri, 29 Oct 2021 14:41:18 +0000 (15:41 +0100)]
benchtests: Improve bench-memcpy-random
Improve the random memcpy benchmark. Double the number of tests and increase
the size of the memory region to test between 32KB and 1024KB. This improves
accuracy on modern cores. Clean up formatting of the frequency array.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Joseph Myers [Fri, 29 Oct 2021 14:40:45 +0000 (14:40 +0000)]
Disable -Waggressive-loop-optimizations warnings in tst-dynarray.c
My build-many-glibcs.py bot shows -Waggressive-loop-optimizations
errors building the glibc testsuite for 32-bit architectures with GCC
mainline, which seem to have appeared between GCC commits
4abc0c196b10251dc80d0743ba9e8ab3e56c61ed and
d8edfadfc7a9795b65177a50ce44fd348858e844:
In function 'dynarray_long_noscratch_resize',
inlined from 'test_long_overflow' at tst-dynarray.c:489:5,
inlined from 'do_test' at tst-dynarray.c:571:3:
../malloc/dynarray-skeleton.c:391:36: error: iteration
1073741823 invokes undefined behavior [-Werror=aggressive-loop-optimizations]
391 | DYNARRAY_ELEMENT_INIT (&list->u.dynarray_header.array[i]);
tst-dynarray.c:39:37: note: in definition of macro 'DYNARRAY_ELEMENT_INIT'
39 | #define DYNARRAY_ELEMENT_INIT(e) (*(e) = 23)
| ^
In file included from tst-dynarray.c:42:
../malloc/dynarray-skeleton.c:389:37: note: within this loop
389 | for (size_t i = old_size; i < size; ++i)
| ~~^~~~~~
In function 'dynarray_long_resize',
inlined from 'test_long_overflow' at tst-dynarray.c:479:5,
inlined from 'do_test' at tst-dynarray.c:571:3:
../malloc/dynarray-skeleton.c:391:36: error: iteration
1073741823 invokes undefined behavior [-Werror=aggressive-loop-optimizations]
391 | DYNARRAY_ELEMENT_INIT (&list->u.dynarray_header.array[i]);
tst-dynarray.c:27:37: note: in definition of macro 'DYNARRAY_ELEMENT_INIT'
27 | #define DYNARRAY_ELEMENT_INIT(e) (*(e) = 17)
| ^
In file included from tst-dynarray.c:28:
../malloc/dynarray-skeleton.c:389:37: note: within this loop
389 | for (size_t i = old_size; i < size; ++i)
| ~~^~~~~~
I don't know what GCC change made these errors appear, or why they
only appear for 32-bit architectures. However, the warnings appear to
be both true (that iteration would indeed involve undefined behavior
if executed) and useless in this particular case (that iteration is
never executed, because the allocation size overflows and so the
allocation fails - but the check for allocation size overflow is in a
separate source file and so can't be seen by the compiler when
compiling this test). So use the DIAG_* macros to disable
-Waggressive-loop-optimizations around the calls in question to
dynarray_long_resize and dynarray_long_noscratch_resize in this test.
Tested with build-many-glibcs.py (GCC mainline) for arm-linux-gnueabi,
where it restores a clean testsuite build.
Stafford Horne [Thu, 28 Oct 2021 21:27:17 +0000 (06:27 +0900)]
Fix compiler issue with mmap_internal
Compiling mmap_internal fails to compile when we use -1 for MMAP2_PAGE_UNIT
on 32 bit architectures. The error is as follows:
../sysdeps/unix/sysv/linux/mmap_internal.h:30:8: error: unknown type
name 'uint64_t'
|
30 | static uint64_t page_unit;
|
| ^~~~~~~~
Fix by adding including stdint.h.
Adhemerval Zanella [Tue, 26 Oct 2021 13:58:50 +0000 (10:58 -0300)]
Check if linker also support -mtls-dialect=gnu2
Since some linkers (for instance lld for i386) does not support it
for all architectures.
Checked on i686-linux-gnu.
Reviewed-by: Fangrui Song <maskray@google.com>
Adhemerval Zanella [Tue, 26 Oct 2021 12:02:57 +0000 (09:02 -0300)]
Fix LIBC_PROG_BINUTILS for -fuse-ld=lld
GCC does not print the correct linker when -fuse-ld=lld is used with
the -print-prog-name=ld:
$ gcc -v 2>&1 | tail -n 1
gcc version 11.2.0 (Ubuntu 11.2.0-7ubuntu2)
$ gcc
ld
This is different than for gold:
$ gcc -fuse-ld=gold -print-prog-name=ld
ld.gold
Using ld.lld as the static linker name prints the expected result.
This is only required when -fuse-ld=lld is used, if lld is used as
the 'ld' programs (through a symlink) LIBC_PROG_BINUTILS works
as expected.
Checked on x86_64-linux-gnu.
Reviewed-by: Fangrui Song <maskray@google.com>
Adhemerval Zanella [Tue, 26 Oct 2021 11:30:18 +0000 (08:30 -0300)]
elf: Disable ifuncmain{1,5,5pic,5pie} when using LLD
These tests takes the address of a protected symbol (foo_protected)
and lld does not support copy relocations on protected data symbols.
Checked on x86_64-linux-gnu.
Reviewed-by: Fangrui Song <maskray@google.com>
Siddhesh Poyarekar [Fri, 29 Oct 2021 09:23:55 +0000 (14:53 +0530)]
Handle NULL input to malloc_usable_size [BZ #28506]
Hoist the NULL check for malloc_usable_size into its entry points in
malloc-debug and malloc and assume non-NULL in all callees. This fixes
BZ #28506
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Reviewed-by: Richard W.M. Jones <rjones@redhat.com>
Noah Goldstein [Thu, 28 Oct 2021 19:41:03 +0000 (14:41 -0500)]
x86_64: Add memcmpeq.S to fix disable-multi-arch build
The following commit:
commit
cf4fd28ea453d1a9cec93939bc88b58ccef5437a
Author: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Tue Oct 26 19:43:18 2021 -0500
Broke --disable-multi-arch build for x86_64 because x86_64/memcmpeq.S
was not defined outside of multiarch and the alias for __memcmpeq in
x86_64/memcmp.S was removed.
This commit fixes that issue by adding x86_64/memcmpeq.S.
make xcheck passes on x86_64 with and without --disable-multi-arch
Stafford Horne [Wed, 13 Oct 2021 12:43:50 +0000 (21:43 +0900)]
login: Add back libutil as an empty library
There are several packages like sysvinit and buildroot that expect
-lutil to work. Rather than impacting them with having to change
the linker flags provide an empty libutil.a.
Fangrui Song [Thu, 28 Oct 2021 18:39:49 +0000 (11:39 -0700)]
riscv: Fix incorrect jal with HIDDEN_JUMPTARGET
A non-local STV_DEFAULT defined symbol is by default preemptible in a
shared object. j/jal cannot target a preemptible symbol. On other
architectures, such a jump instruction either causes PLT [BZ #18822], or
if short-ranged, sometimes rejected by the linker (but not by GNU ld's
riscv port [ld PR/28509]).
Use HIDDEN_JUMPTARGET to target a non-preemptible symbol instead.
With this patch, ld.so and libc.so can be linked with LLD if source
files are compiled/assembled with -mno-relax/-Wa,-mno-relax.
Acked-by: Palmer Dabbelt <palmer@dabbelt.com>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Noah Goldstein [Thu, 30 Sep 2021 20:41:00 +0000 (15:41 -0500)]
x86_64: Add evex optimized __memcmpeq in memcmpeq-evex.S
No bug. This commit adds new optimized __memcmpeq implementation for
evex.
The primary optimizations are:
1) skipping the logic to find the difference of the first mismatched
byte.
2) not updating src/dst addresses as the non-equals logic does not
need to be reused by different areas.
Noah Goldstein [Thu, 30 Sep 2021 20:37:09 +0000 (15:37 -0500)]
x86_64: Add avx2 optimized __memcmpeq in memcmpeq-avx2.S
No bug. This commit adds new optimized __memcmpeq implementation for
avx2.
The primary optimizations are:
1) skipping the logic to find the difference of the first mismatched
byte.
2) not updating src/dst addresses as the non-equals logic does not
need to be reused by different areas.
Noah Goldstein [Thu, 30 Sep 2021 20:32:50 +0000 (15:32 -0500)]
x86_64: Add sse2 optimized __memcmpeq in memcmp-sse2.S
No bug. This commit does not modify any of the memcmp
implementation. It just adds __memcmpeq ifdefs to skip obvious cases
where computing the proper 1/-1 required by memcmp is not needed.
Noah Goldstein [Wed, 27 Oct 2021 00:43:18 +0000 (19:43 -0500)]
x86_64: Add support for __memcmpeq using sse2, avx2, and evex
No bug. This commit adds support for __memcmpeq to be implemented
seperately from memcmp. Support is added for versions optimized with
sse2, avx2, and evex.
Noah Goldstein [Sat, 23 Oct 2021 05:36:30 +0000 (01:36 -0400)]
Benchtests: Add benchtests for __memcmpeq
No bug. This commit adds __memcmpeq benchmarks. The benchmarks just
use the existing ones in memcmp. This will be useful for testing
implementations of __memcmpeq that do not just alias memcmp.
Noah Goldstein [Wed, 27 Oct 2021 00:42:59 +0000 (19:42 -0500)]
String: Add __memcmpeq as build target
No bug. This commit just adds __memcmpeq as a build target so that
implementations for __memcmpeq that are not just aliases to memcmp can
be supported.
Noah Goldstein [Fri, 22 Oct 2021 16:04:55 +0000 (11:04 -0500)]
NEWS: Add item for __memcmpeq
Noah Goldstein [Thu, 21 Oct 2021 20:07:18 +0000 (15:07 -0500)]
String: Add tests for __memcmpeq
No bug.
This commit adds tests for the new function __memcmpeq. The new tests
use the existing tests in 'test-memcmp.c' but relax the result
requirement to only check for zero or non-zero returns.
All string tests include test-memcmpeq are passing.
Noah Goldstein [Thu, 21 Oct 2021 20:53:58 +0000 (15:53 -0500)]
String: Add hidden defs for __memcmpeq() to enable internal usage
No bug.
This commit adds hidden defs for all declarations of __memcmpeq. This
enables usage of __memcmpeq without the PLT for usage internal to
GLIBC.
Noah Goldstein [Thu, 21 Oct 2021 20:54:57 +0000 (15:54 -0500)]
String: Add support for __memcmpeq() ABI on all targets
No bug.
This commit adds support for __memcmpeq() as a new ABI for all
targets. In this commit __memcmpeq() is implemented only as an alias
to the corresponding targets memcmp() implementation. __memcmpeq() is
added as a new symbol starting with GLIBC_2.35 and defined in string.h
with comments explaining its behavior. Basic tests that it is callable
and works where added in string/tester.c
As discussed in the proposal "Add new ABI '__memcmpeq()' to libc"
__memcmpeq() is essentially a reserved namespace for bcmp(). The means
is shares the same specifications as memcmp() except the return value
for non-equal byte sequences is any non-zero value. This is less
strict than memcmp()'s return value specification and can be better
optimized when a boolean return is all that is needed.
__memcmpeq() is meant to only be called by compilers if they can prove
that the return value of a memcmp() call is only used for its boolean
value.
All tests in string/tester.c passed. As well build succeeds on
x86_64-linux-gnu target.
Fangrui Song [Mon, 25 Oct 2021 20:17:44 +0000 (13:17 -0700)]
configure: Don't check LD -v --help for LIBC_LINKER_FEATURE
When LIBC_LINKER_FEATURE is used to check a linker option with the equal
sign, it will likely fail because the LD -v --help output may look like
`-z lam-report=[none|warning|error]` while the needle is something like
`-z lam-report=warning`.
The LD -v --help filter doesn't save much time, so just remove it.
H.J. Lu [Fri, 15 Oct 2021 18:29:24 +0000 (11:29 -0700)]
elf: Make global.out depend on reldepmod4.so [BZ #28457]
The global test is linked with globalmod1.so which dlopens reldepmod4.so.
Make global.out depend on reldepmod4.so. This fixes BZ #28457.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Noah Goldstein [Sat, 23 Oct 2021 05:26:47 +0000 (01:26 -0400)]
x86: Replace sse2 instructions with avx in memcmp-evex-movbe.S
This commit replaces two usages of SSE2 'movups' with AVX 'vmovdqu'.
it could potentially be dangerous to use SSE2 if this function is ever
called without using 'vzeroupper' beforehand. While compilers appear
to use 'vzeroupper' before function calls if AVX2 has been used, using
SSE2 here is more brittle. Since it is not absolutely necessary it
should be avoided.
It costs 2-extra bytes but the extra bytes should only eat into
alignment padding.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
H.J. Lu [Fri, 22 Oct 2021 23:13:11 +0000 (16:13 -0700)]
bench-math: Sort and put each bench per line
Sort and put each math bench per line to prepare for new math benches.
Reviewed-by: Noah Goldstein <goldstein.w.n@gmail.com>
Sunil K Pandey [Wed, 13 Oct 2021 20:37:24 +0000 (13:37 -0700)]
x86_64: Add missing libmvec ABI tests
Add vector ABI tests for cos, exp, log, pow and sin functions.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Adhemerval Zanella [Thu, 21 Oct 2021 20:26:32 +0000 (17:26 -0300)]
elf: Fix
e6fd79f379 build with --enable-tunables=no
The _dl_sort_maps_init() is not defined when tunables is not enabled.
Checked on x86_64-linux-gnu.
Chung-Lin Tang [Thu, 21 Oct 2021 13:41:22 +0000 (21:41 +0800)]
elf: Fix slow DSO sorting behavior in dynamic loader (BZ #17645)
This second patch contains the actual implementation of a new sorting algorithm
for shared objects in the dynamic loader, which solves the slow behavior that
the current "old" algorithm falls into when the DSO set contains circular
dependencies.
The new algorithm implemented here is simply depth-first search (DFS) to obtain
the Reverse-Post Order (RPO) sequence, a topological sort. A new l_visited:1
bitfield is added to struct link_map to more elegantly facilitate such a search.
The DFS algorithm is applied to the input maps[nmap-1] backwards towards
maps[0]. This has the effect of a more "shallow" recursion depth in general
since the input is in BFS. Also, when combined with the natural order of
processing l_initfini[] at each node, this creates a resulting output sorting
closer to the intuitive "left-to-right" order in most cases.
Another notable implementation adjustment related to this _dl_sort_maps change
is the removing of two char arrays 'used' and 'done' in _dl_close_worker to
represent two per-map attributes. This has been changed to simply use two new
bit-fields l_map_used:1, l_map_done:1 added to struct link_map. This also allows
discarding the clunky 'used' array sorting that _dl_sort_maps had to sometimes
do along the way.
Tunable support for switching between different sorting algorithms at runtime is
also added. A new tunable 'glibc.rtld.dynamic_sort' with current valid values 1
(old algorithm) and 2 (new DFS algorithm) has been added. At time of commit
of this patch, the default setting is 1 (old algorithm).
Signed-off-by: Chung-Lin Tang <cltang@codesourcery.com>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Chung-Lin Tang [Thu, 21 Oct 2021 13:41:21 +0000 (21:41 +0800)]
elf: Testing infrastructure for ld.so DSO sorting (BZ #17645)
This is the first of a 2-part patch set that fixes slow DSO sorting behavior in
the dynamic loader, as reported in BZ #17645. In order to facilitate such a
large modification to the dynamic loader, this first patch implements a testing
framework for validating shared object sorting behavior, to enable comparison
between old/new sorting algorithms, and any later enhancements.
This testing infrastructure consists of a Python script
scripts/dso-ordering-test.py' which takes in a description language, consisting
of strings that describe a set of link dependency relations between DSOs, and
generates testcase programs and Makefile fragments to automatically test the
described situation, for example:
a->b->c->d # four objects linked one after another
a->[bc]->d;b->c # a depends on b and c, which both depend on d,
# b depends on c (b,c linked to object a in fixed order)
a->b->c;{+a;%a;-a} # a, b, c serially dependent, main program uses
# dlopen/dlsym/dlclose on object a
a->b->c;{}!->[abc] # a, b, c serially dependent; multiple tests generated
# to test all permutations of a, b, c ordering linked
# to main program
(Above is just a short description of what the script can do, more
documentation is in the script comments.)
Two files containing several new tests, elf/dso-sort-tests-[12].def are added,
including test scenarios for BZ #15311 and Redhat issue #1162810 [1].
Due to the nature of dynamic loader tests, where the sorting behavior and test
output occurs before/after main(), generating testcases to use
support/test-driver.c does not suffice to control meaningful timeout for ld.so.
Therefore a new utility program 'support/test-run-command', based on
test-driver.c/support_test_main.c has been added. This does the same testcase
control, but for a program specified through a command-line rather than at the
source code level. This utility is used to run the dynamic loader testcases
generated by dso-ordering-test.py.
[1] https://bugzilla.redhat.com/show_bug.cgi?id=1162810
Signed-off-by: Chung-Lin Tang <cltang@codesourcery.com>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Stafford Horne [Tue, 5 Oct 2021 01:56:10 +0000 (10:56 +0900)]
iconv: Use TIMEOUTFACTOR for iconv test timeout
Currently the timeout for each iconv test is hard coded to 3 seconds.
On my OpenRISC test platform this is too slow and the test fails with a
HANG error.
This change uses the available TIMEOUTFACTOR to compute the timeout.
The default value is still 3.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Adhemerval Zanella [Mon, 4 Jan 2021 18:46:52 +0000 (15:46 -0300)]
posix: Remove alloca usage for internal fnmatch implementation
This patch replaces the internal fnmatch pattern list generation
to use a dynamic array.
Checked on x86_64-linux-gnu.
Jonathan Wakely [Wed, 6 Oct 2021 19:05:48 +0000 (20:05 +0100)]
Add alloc_align attribute to memalign et al
GCC 4.9.0 added the alloc_align attribute to say that a function
argument specifies the alignment of the returned pointer. Clang supports
the attribute too. Using the attribute can allow a compiler to generate
better code if it knows the returned pointer has a minimum alignment.
See https://gcc.gnu.org/PR60092 for more details.
GCC implicitly knows the semantics of aligned_alloc and posix_memalign,
but not the obsolete memalign. As a result, GCC generates worse code
when memalign is used, compared to aligned_alloc. Clang knows about
aligned_alloc and memalign, but not posix_memalign.
This change adds a new __attribute_alloc_align__ macro to <sys/cdefs.h>
and then uses it on memalign (where it helps GCC) and aligned_alloc
(where GCC and Clang already know the semantics, but it doesn't hurt)
and xposix_memalign. It can't be used on posix_memalign because that
doesn't return a pointer (the allocated pointer is returned via a void**
parameter instead).
Unlike the alloc_size attribute, alloc_align only allows a single
argument. That means the new __attribute_alloc_align__ macro doesn't
really need to be used with double parentheses to protect a comma
between its arguments. For consistency with __attribute_alloc_size__
this patch defines it the same way, so that double parentheses are
required.
Signed-off-by: Jonathan Wakely <jwakely@redhat.com>
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Tested-by: Carlos O'Donell <carlos@redhat.com>
Fangrui Song [Wed, 20 Oct 2021 21:22:43 +0000 (14:22 -0700)]
linux: Fix a possibly non-constant expression in _Static_assert
According to C11 6.6p6, `const int` as an operand may not make up a
constant expression. GCC -O0 errors:
../sysdeps/unix/sysv/linux/opendir.c:107:19: error: static_assert expression is not an integral constant expression
_Static_assert (allocation_size >= sizeof (struct dirent64),
-O2 -Wpedantic has a similar warning.
See https://gcc.gnu.org/PR102502 for GCC's inconsistency.
Use enum which is guaranteed to be a constant expression.
This also makes the file compilable with Clang.
Fixes:
4b962c9e859de23b461d61f860dbd3f21311e83a ("linux: Simplify opendir buffer allocation")
H.J. Lu [Tue, 12 Oct 2021 01:07:09 +0000 (18:07 -0700)]
x86-64: Add sysdeps/x86_64/fpu/Makeconfig
1. Add sysdeps/x86_64/fpu/Makeconfig to auto-generate libmvec.mk, which
contains libmvec ABI test dependencies and CFLAGS, in the build directory.
2. Include libmvec.mk for libmvec ABI test dependencies and CFLAGS.
Tested on SSE4, AVX, AVX2 and AVX512 machines.
Reviewed-by: Noah Goldstein <goldstein.w.n@gmail.com>
omain GEISSLER [Wed, 20 Oct 2021 15:01:40 +0000 (12:01 -0300)]
stdlib: Fix tst-canon-bz26341 when the glibc build current working directory is itself using symlinks.
Adhemerval Zanella [Fri, 12 Feb 2021 16:20:27 +0000 (19:20 +0300)]
powerpc: Remove backtrace implementation
The powerpc optimization to provide a fast stacktrace requires some
ad-hoc code to handle Linux signal frames and the change is fragile
once the kernel decides to slight change its execution sequence [1].
The generic implementation work as-is and it should be future proof
since the kernel provides the expected CFI directives in vDSO shared
page.
Checked on powerpc-linux-gnu, powerpc64le-linux-gnu, and
powerpc64-linux-gnu.
[1] https://sourceware.org/pipermail/libc-alpha/2021-January/122027.html
Joseph Myers [Wed, 20 Oct 2021 13:38:50 +0000 (13:38 +0000)]
Correct access attribute on memfrob (bug 28475)
As noted in bug 28475, the access attribute on memfrob in <string.h>
is incorrect: the function both reads and writes the memory pointed to
by its argument, so it needs to use __read_write__, not
__write_only__. This incorrect attribute results in a build failure
for accessing uninitialized memory for s390x-linux-gnu-O3 with
build-many-glibcs.py using GCC mainline.
Correct the attribute. Fixing this shows up that some calls to
memfrob in elf/ tests are reading uninitialized memory; I'm not
entirely sure of the purpose of those calls, but guess they are about
ensuring that the stack space is indeed allocated at that point in the
function, and so it matters that they are calling a function whose
semantics are unknown to the compiler. Thus, change the first memfrob
call in those tests to use explicit_bzero instead, as suggested by
Florian in
<https://sourceware.org/pipermail/libc-alpha/2021-October/132119.html>,
to avoid the use of uninitialized memory.
Tested for x86_64, and with build-many-glibcs.py (GCC mainline) for
s390x-linux-gnu-O3.
Siddhesh Poyarekar [Wed, 20 Oct 2021 12:43:05 +0000 (18:13 +0530)]
debug: Add tests for _FORTIFY_SOURCE=3
Add some testing coverage for _FORTIFY_SOURCE=3.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Siddhesh Poyarekar [Wed, 20 Oct 2021 12:42:41 +0000 (18:12 +0530)]
Make sure that the fortified function conditionals are constant
In _FORTIFY_SOURCE=3, the size expression may be non-constant,
resulting in branches in the inline functions remaining intact and
causing a tiny overhead. Clang (and in future, gcc) make sure that
the -1 case is always safe, i.e. any comparison of the generated
expression with (size_t)-1 is always false so that bit is taken care
of. The rest is avoidable since we want the _chk variant whenever we
have a size expression and it's not -1.
Rework the conditionals in a uniform way to clearly indicate two
conditions at compile time:
- Either the size is unknown (-1) or we know at compile time that the
operation length is less than the object size. We can call the
original function in this case. It could be that either the length,
object size or both are non-constant, but the compiler, through
range analysis, is able to fold the *comparison* to a constant.
- The size and length are known and the compiler can see at compile
time that operation length > object size. This is valid grounds for
a warning at compile time, followed by emitting the _chk variant.
For everything else, emit the _chk variant.
This simplifies most of the fortified function implementations and at
the same time, ensures that only one call from _chk or the regular
function is emitted.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Siddhesh Poyarekar [Tue, 12 Oct 2021 06:59:13 +0000 (12:29 +0530)]
Don't add access size hints to fortifiable functions
In the context of a function definition, the size hints imply that the
size of an object pointed to by one parameter is another parameter.
This doesn't make sense for the fortified versions of the functions
since that's the bit it's trying to validate.
This is harmless with __builtin_object_size since it has fairly simple
semantics when it comes to objects passed as function parameters.
With __builtin_dynamic_object_size we could (as my patchset for gcc[1]
already does) use the access attribute to determine the object size in
the general case but it misleads the fortified functions.
Basically the problem occurs when access attributes are present on
regular functions that have inline fortified definitions to generate
_chk variants; the attributes get inherited by these definitions,
causing problems when analyzing them. For example with poll(fds, nfds,
timeout), nfds is hinted using the __attr_access as being the size of
fds.
Now, when analyzing the inline function definition in bits/poll2.h, the
compiler sees that nfds is the size of fds and tries to use that
information in the function body. In _FORTIFY_SOURCE=3 case, where the
object size could be a non-constant expression, this information results
in the conclusion that nfds is the size of fds, which defeats the
purpose of the implementation because we're trying to check here if nfds
does indeed represent the size of fds. Hence for this case, it is best
to not have the access attribute.
With the attributes gone, the expression evaluation should get delayed
until the function is actually inlined into its destinations.
Disable the access attribute for fortified function inline functions
when building at _FORTIFY_SOURCE=3 to make this work better. The
access attributes remain for the _chk variants since they can be used
by the compiler to warn when the caller is passing invalid arguments.
[1] https://gcc.gnu.org/pipermail/gcc-patches/2021-October/581125.html
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Fangrui Song [Tue, 19 Oct 2021 16:58:16 +0000 (09:58 -0700)]
glibcextract.py: Place un-assemblable @@@ in a comment
Unlike GCC, Clang parses asm statements and verifies they are valid
instructions/directives. Place the magic @@@ into a comment to avoid
a parse error.
Fangrui Song [Tue, 19 Oct 2021 16:36:43 +0000 (09:36 -0700)]
nss: Unnest nested function add_key
This makes makedb.c compilable with Clang which does not support nested
functions.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
H.J. Lu [Thu, 23 Sep 2021 16:06:49 +0000 (09:06 -0700)]
ld.so: Initialize bootstrap_map.l_ld_readonly [BZ #28340]
1. Define DL_RO_DYN_SECTION to initalize bootstrap_map.l_ld_readonly
before calling elf_get_dynamic_info to get dynamic info in bootstrap_map,
2. Define a single
static inline bool
dl_relocate_ld (const struct link_map *l)
{
/* Don't relocate dynamic section if it is readonly */
return !(l->l_ld_readonly || DL_RO_DYN_SECTION);
}
This updates BZ #28340 fix.
Stafford Horne [Thu, 14 Oct 2021 21:17:41 +0000 (06:17 +0900)]
timex: Use 64-bit fields on 32-bit TIMESIZE=64 systems (BZ #28469)
This was found when testing the OpenRISC port I am working on. These
two tests fail with SIGSEGV:
FAIL: misc/tst-ntp_gettime
FAIL: misc/tst-ntp_gettimex
This was found to be due to the kernel overwriting the stack space
allocated by the timex structure. The reason for the overwrite being
that the kernel timex has 64-bit fields and user space code only
allocates enough stack space for timex with 32-bit fields.
On 32-bit systems with TIMESIZE=64 __USE_TIME_BITS64 is not defined.
This causes the timex structure to use 32-bit fields with type
__syscall_slong_t.
This patch adjusts the ifdef condition to allow 32-bit systems with
TIMESIZE=64 to use the 64-bit long long timex definition.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Stafford Horne [Thu, 14 Oct 2021 21:55:31 +0000 (06:55 +0900)]
manual: Update _TIME_BITS to clarify it's user defined
The current language reads "This macro determines...", changing to
"Define this macro...". This is consistent with other feature macro
documentation language.
When I first read the previous language it seems to indicate that the
macro is already defined. By changing the language to "Define this
macro..." it's clear that its the user's responsibility to define it.
Stafford Horne [Sat, 16 Oct 2021 22:41:54 +0000 (07:41 +0900)]
nptl: Fix tst-cancel7 and tst-cancelx7 pidfile race
The check for waiting for the pidfile to be created looks wrong. At the
point when ACCESS is run the pid file will always be created and
accessible as it is created during DO_PREPARE. This means that thread
cancellation may be performed before the pid is written to the pidfile.
This was found to be flaky when testing on my OpenRISC platform.
Fix this by using the semaphore to wait for pidfile pid write
completion.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Adhemerval Zanella [Fri, 15 Oct 2021 17:35:31 +0000 (14:35 -0300)]
elf: Fix elf_get_dynamic_info() for bootstrap
THe
d6d89608ac8c broke powerpc for --enable-bind-now because it turned
out that different than patch assumption rtld elf_get_dynamic_info()
does require to handle RTLD_BOOTSTRAP to avoid DT_FLAGS and
DT_RUNPATH (more specially the GLRO usage which is not reallocate
yet).
This patch fixes by passing two arguments to elf_get_dynamic_info()
to inform that by rtld (bootstrap) or static pie initialization
(static_pie_bootstrap). I think using explicit argument is way more
clear and burried C preprocessor, and compiler should remove the
dead code.
I checked on x86_64 and i686 with default options, --enable-bind-now,
and --enable-bind-now and --enable--static-pie. I also check on
aarch64, armhf, powerpc64, and powerpc with default and
--enable-bind-now.
Samuel Thibault [Sun, 17 Oct 2021 23:39:02 +0000 (01:39 +0200)]
hurd if_index: Explicitly use AF_INET for if index discovery
5bf07e1b3a74 ("Linux: Simplify __opensock and fix race condition [BZ #28353]")
made __opensock try NETLINK then UNIX then INET. On the Hurd, only INET
knows about network interfaces, so better actually specify that in
if_index.
Samuel Thibault [Sun, 17 Oct 2021 22:43:23 +0000 (00:43 +0200)]
hurd: Fix intr-msg parameter/stack kludge
INTR_MSG_TRAP was tinkering with esp to make it point to
_hurd_intr_rpc_mach_msg's parameters, and notably use (&msg)[-1] which is
meaningless in C.
Instead, just push the parameters on the stack, which also avoids leaving
local variables of _hurd_intr_rpc_mach_msg below esp. We now also
properly express that OPTION and TIMEOUT may be updated during the trap
call.
H.J. Lu [Wed, 13 Oct 2021 20:46:52 +0000 (13:46 -0700)]
x86-64: Add test-vector-abi.h/test-vector-abi-sincos.h
Add templates for vector ABI test and use them for vector sincos/sincosf
ABI tests.
Adhemerval Zanella [Wed, 13 Oct 2021 12:49:34 +0000 (09:49 -0300)]
elf: Fix dynamic-link.h usage on rtld.c
The
4af6982e4c fix does not fully handle RTLD_BOOTSTRAP usage on
rtld.c due two issues:
1. RTLD_BOOTSTRAP is also used on dl-machine.h on various
architectures and it changes the semantics of various machine
relocation functions.
2. The elf_get_dynamic_info() change was done sideways, previously
to
490e6c62aa get-dynamic-info.h was included by the first
dynamic-link.h include *without* RTLD_BOOTSTRAP being defined.
It means that the code within elf_get_dynamic_info() that uses
RTLD_BOOTSTRAP is in fact unused.
To fix 1. this patch now includes dynamic-link.h only once with
RTLD_BOOTSTRAP defined. The ELF_DYNAMIC_RELOCATE call will now have
the relocation fnctions with the expected semantics for the loader.
And to fix 2. part of
4af6982e4c is reverted (the check argument
elf_get_dynamic_info() is not required) and the RTLD_BOOTSTRAP
pieces are removed.
To reorganize the includes the static TLS definition is moved to
its own header to avoid a circular dependency (it is defined on
dynamic-link.h and dl-machine.h requires it at same time other
dynamic-link.h definition requires dl-machine.h defitions).
Also ELF_MACHINE_NO_REL, ELF_MACHINE_NO_RELA, and ELF_MACHINE_PLT_REL
are moved to its own header. Only ancient ABIs need special values
(arm, i386, and mips), so a generic one is used as default.
The powerpc Elf64_FuncDesc is also moved to its own header, since
csu code required its definition (which would require either include
elf/ folder or add a full path with elf/).
Checked on x86_64, i686, aarch64, armhf, powerpc64, powerpc32,
and powerpc64le.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Noah Goldstein [Mon, 20 Sep 2021 21:20:15 +0000 (16:20 -0500)]
x86: Optimize memset-vec-unaligned-erms.S
No bug.
Optimization are
1. change control flow for L(more_2x_vec) to fall through to loop and
jump for L(less_4x_vec) and L(less_8x_vec). This uses less code
size and saves jumps for length > 4x VEC_SIZE.
2. For EVEX/AVX512 move L(less_vec) closer to entry.
3. Avoid complex address mode for length > 2x VEC_SIZE
4. Slightly better aligning code for the loop from the perspective of
code size and uops.
5. Align targets so they make full use of their fetch block and if
possible cache line.
6. Try and reduce total number of icache lines that will need to be
pulled in for a given length.
7. Include "local" version of stosb target. For AVX2/EVEX/AVX512
jumping to the stosb target in the sse2 code section will almost
certainly be to a new page. The new version does increase code size
marginally by duplicating the target but should get better iTLB
behavior as a result.
test-memset, test-wmemset, and test-bzero are all passing.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Noah Goldstein [Tue, 21 Sep 2021 23:45:03 +0000 (18:45 -0500)]
x86: Optimize memcmp-evex-movbe.S for frontend behavior and size
No bug.
The frontend optimizations are to:
1. Reorganize logically connected basic blocks so they are either in
the same cache line or adjacent cache lines.
2. Avoid cases when basic blocks unnecissarily cross cache lines.
3. Try and 32 byte align any basic blocks possible without sacrificing
code size. Smaller / Less hot basic blocks are used for this.
Overall code size shrunk by 168 bytes. This should make up for any
extra costs due to aligning to 64 bytes.
In general performance before deviated a great deal dependending on
whether entry alignment % 64 was 0, 16, 32, or 48. These changes
essentially make it so that the current implementation is at least
equal to the best alignment of the original for any arguments.
The only additional optimization is in the page cross case. Branch on
equals case was removed from the size == [4, 7] case. As well the [4,
7] and [2, 3] case where swapped as [4, 7] is likely a more hot
argument size.
test-memcmp and test-wmemcmp are both passing.
Stafford Horne [Tue, 5 Oct 2021 23:31:12 +0000 (08:31 +0900)]
libio: Update tst-wfile-sync to not depend on stdin
The test expects stdin to be a file which is not the case when running
tests over ssh where stdin is piped in.
The test fails with:
error: xlseek.c:27: lseek64 (0, 0, 1): Illegal seek
Update the test to create a temporary file and use that to perform the
test.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Stafford Horne [Tue, 5 Oct 2021 23:30:38 +0000 (08:30 +0900)]
elf: Update audit tests to not depend on stdout
The tst-audit14, tst-audit15 and tst-audit16 tests all have audit
modules that write to stdout; the test reads from stdout to confirm
what was written. This assumes the stdout is a file which is not the
case when run over ssh.
This patch updates the tests to use a post run cmp command to compare
the output against and .exp file. This is similar to how many other
tests work and it fixes the stdout limitation. Also, this means the
test code can be greatly simplified.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Adhemerval Zanella [Mon, 11 Oct 2021 19:01:49 +0000 (16:01 -0300)]
elf: Fix elf_get_dynamic_info definition
Before to
490e6c62aa31a8a ('elf: Avoid nested functions in the loader
[BZ #27220]'), elf_get_dynamic_info() was defined twice on rtld.c: on
the first dynamic-link.h include and later within _dl_start(). The
former definition did not define DONT_USE_BOOTSTRAP_MAP and it is used
on setup_vdso() (since it is a global definition), while the former does
define DONT_USE_BOOTSTRAP_MAP and it is used on loader self-relocation.
With the commit change, the function is now included and defined once
instead of defined as a nested function. So rtld.c defines without
defining RTLD_BOOTSTRAP and it brokes at least powerpc32.
This patch fixes by moving the get-dynamic-info.h include out of
dynamic-link.h, which then the caller can corirectly set the expected
semantic by defining STATIC_PIE_BOOTSTRAP, RTLD_BOOTSTRAP, and/or
RESOLVE_MAP.
It also required to enable some asserts only for the loader bootstrap
to avoid issues when called from setup_vdso().
As a side note, this is another issues with nested functions: it is
not clear from pre-processed output (-E -dD) how the function will
be build and its semantic (since nested function will be local and
extra C defines may change it).
I checked on x86_64-linux-gnu (w/o --enable-static-pie),
i686-linux-gnu, powerpc64-linux-gnu, powerpc-linux-gnu-power4,
aarch64-linux-gnu, arm-linux-gnu, sparc64-linux-gnu, and
s390x-linux-gnu.
Reviewed-by: Fangrui Song <maskray@google.com>
Joseph Myers [Tue, 12 Oct 2021 13:48:39 +0000 (13:48 +0000)]
Add TEST_COMPARE_STRING_WIDE to support/check.h
I'd like to be able to test narrow and wide string interfaces, with
the narrow string tests using TEST_COMPARE_STRING and the wide string
tests using something analogous (possibly generated using macros from
a common test template for both the narrow and wide string tests where
appropriate).
Add such a TEST_COMPARE_STRING_WIDE, along with functions
support_quote_blob_wide and support_test_compare_string_wide that it
builds on. Those functions are built using macros from common
templates shared by the narrow and wide string implementations, though
I didn't do that for the tests of test functions. In
support_quote_blob_wide, I chose to use the \x{} delimited escape
sequence syntax proposed for C2X in N2785, rather than e.g. trying to
generate the end of a string and the start of a new string when
ambiguity would result from undelimited \x (when the next character
after such an escape sequence is valid hex) or forcing an escape
sequence to be used for the next character in the case of such
ambiguity.
Tested for x86_64.
Joseph Myers [Mon, 11 Oct 2021 21:47:32 +0000 (21:47 +0000)]
Fix nios2 localplt failure
Building for nios2-linux-gnu has recently started showing a localplt
test failure, arising from a reference to __floatunsidf from
getloadavg after commit
b5c8a3aa82f66f49b731ca5204104cee48bccfa5
("Linux: implement getloadavg(3) using sysinfo(2)") (this is an
architecture with soft-fp in libc). Add this as a permitted local PLT
reference in localplt.data.
Tested with build-many-glibcs.py for nios2-linux-gnu.
Fangrui Song [Mon, 11 Oct 2021 18:14:02 +0000 (11:14 -0700)]
elf: Remove Intel MPX support (lazy PLT, ld.so profile, and LD_AUDIT)
Intel MPX failed to gain wide adoption and has been deprecated for a
while. GCC 9.1 removed Intel MPX support. Linux kernel removed MPX in
2019.
This patch removes the support code from the dynamic loader.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Martin Sebor [Mon, 11 Oct 2021 15:36:57 +0000 (09:36 -0600)]
resolv: Avoid GCC 12 false positive warning [BZ #28439].
Replace a call to sprintf with an equivalent pair of stpcpy/strcpy calls
to avoid a GCC 12 -Wformat-overflow false positive due to recent optimizer
improvements.
Noah Goldstein [Fri, 24 Sep 2021 23:09:47 +0000 (18:09 -0500)]
benchtests: Add medium cases and increase iters in bench-memset.c
No bug.
This commit adds new medium size cases for lengths in [512, 1024). As
well it increase the iters to INNER_LOOP_ITERS_LARGE for more reliable
results.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Noah Goldstein [Tue, 21 Sep 2021 23:31:49 +0000 (18:31 -0500)]
x86: Modify ENTRY in sysdep.h so that p2align can be specified
No bug.
This change adds a new macro ENTRY_P2ALIGN which takes a second
argument, log2 of the desired function alignment.
The old ENTRY(name) macro is just ENTRY_P2ALIGN(name, 4) so this
doesn't affect any existing functionality.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Cristian Rodríguez [Fri, 6 Aug 2021 19:17:49 +0000 (15:17 -0400)]
resolv: make res_randomid use random_bits()
It is at least "more random" than 0xffff & __getpid ();
Signed-off-by: Cristian Rodríguez <crrodriguez@opensuse.org>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Cristian Rodríguez [Fri, 6 Aug 2021 19:17:48 +0000 (15:17 -0400)]
Linux: implement getloadavg(3) using sysinfo(2)
Signed-off-by: Cristian Rodríguez <crrodriguez@opensuse.org>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
DJ Delorie [Tue, 5 Oct 2021 18:52:05 +0000 (14:52 -0400)]
Remove unreliable parts of rt/tst-cpuclock2
This is a follow-up to the tst-cpuclock1.c change here:
9a29f1a2ae3d4bb253ee368e0d71db0ca9494120
This test, like tst-cpuclock1, may fail on heavily loaded VM
servers (and has occasionally failed on the 32bit trybot),
so tests that rely on "wall time" have been removed.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Fangrui Song [Thu, 7 Oct 2021 18:55:02 +0000 (11:55 -0700)]
elf: Avoid nested functions in the loader [BZ #27220]
dynamic-link.h is included more than once in some elf/ files (rtld.c,
dl-conflict.c, dl-reloc.c, dl-reloc-static-pie.c) and uses GCC nested
functions. This harms readability and the nested functions usage
is the biggest obstacle prevents Clang build (Clang doesn't support GCC
nested functions).
The key idea for unnesting is to add extra parameters (struct link_map
*and struct r_scope_elm *[]) to RESOLVE_MAP,
ELF_MACHINE_BEFORE_RTLD_RELOC, ELF_DYNAMIC_RELOCATE, elf_machine_rel[a],
elf_machine_lazy_rel, and elf_machine_runtime_setup. (This is inspired
by Stan Shebs' ppc64/x86-64 implementation in the
google/grte/v5-2.27/master which uses mixed extra parameters and static
variables.)
Future simplification:
* If mips elf_machine_runtime_setup no longer needs RESOLVE_GOTSYM,
elf_machine_runtime_setup can drop the `scope` parameter.
* If TLSDESC no longer need to be in elf_machine_lazy_rel,
elf_machine_lazy_rel can drop the `scope` parameter.
Tested on aarch64, i386, x86-64, powerpc64le, powerpc64, powerpc32,
sparc64, sparcv9, s390x, s390, hppa, ia64, armhf, alpha, and mips64.
In addition, tested build-many-glibcs.py with {arc,csky,microblaze,nios2}-linux-gnu
and riscv64-linux-gnu-rv64imafdc-lp64d.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
H.J. Lu [Fri, 18 Jun 2021 23:50:39 +0000 (16:50 -0700)]
Add run-time check for indirect external access
When performing symbol lookup for references in executable without
indirect external access:
1. Disallow copy relocations in executable against protected data symbols
in a shared object with indirect external access.
2. Disallow non-zero symbol values of undefined function symbols in
executable, which are used as the function pointer, against protected
function symbols in a shared object with indirect external access.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
H.J. Lu [Fri, 18 Jun 2021 18:53:55 +0000 (11:53 -0700)]
Initial support for GNU_PROPERTY_1_NEEDED
1. Add GNU_PROPERTY_1_NEEDED:
#define GNU_PROPERTY_1_NEEDED GNU_PROPERTY_UINT32_OR_LO
to indicate the needed properties by the object file.
2. Add GNU_PROPERTY_1_NEEDED_INDIRECT_EXTERN_ACCESS:
#define GNU_PROPERTY_1_NEEDED_INDIRECT_EXTERN_ACCESS (1U << 0)
to indicate that the object file requires canonical function pointers and
cannot be used with copy relocation.
3. Scan GNU_PROPERTY_1_NEEDED property and store it in l_1_needed.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Adhemerval Zanella [Wed, 25 Aug 2021 14:17:06 +0000 (11:17 -0300)]
io: Fix ftw internal realloc buffer (BZ #28126)
The
106ff08526d3ca did not take in consideration the buffer might be
reallocated if the total path is larger than PATH_MAX. The realloc
uses 'dirbuf', where 'dirstreams' is the allocated buffer.
Checked on x86_64-linux-gnu.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
Paul Eggert [Tue, 14 Sep 2021 05:49:45 +0000 (22:49 -0700)]
Fix subscript error with odd TZif file [BZ #28338]
* time/tzfile.c (__tzfile_compute): Fix unlikely off-by-one bug
that accessed before start of an array when an oddball-but-valid
TZif file was queried with an unusual time_t value.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Stefan Liebler [Tue, 5 Oct 2021 14:14:10 +0000 (16:14 +0200)]
S390: Add PCI_MIO and SIE HWCAPs
Both new HWCAPs were introduced in these kernel commits:
-
7e8403ecaf884f307b627f3c371475913dd29292
"s390: add HWCAP_S390_PCI_MIO to ELF hwcaps"
-
7e82523f2583e9813e4109df3656707162541297
"s390/hwcaps: make sie capability regular hwcap"
Also note that the kernel commit
511ad531afd4090625def4d9aba1f5227bd44b8e
"s390/hwcaps: shorten HWCAP defines" has shortened the prefix of the macros
from "HWCAP_S390_" to "HWCAP_". For compatibility reasons, we do not
change the prefix in public glibc header file.
Siddhesh Poyarekar [Wed, 6 Oct 2021 16:18:35 +0000 (21:48 +0530)]
support: Also return fd when it is 0
The fd validity check in open_dev_null checks if fd > 0, which would
lead to a leaked fd if it is == 0.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
H.J. Lu [Mon, 2 Aug 2021 20:52:36 +0000 (13:52 -0700)]
ld.so: Don't fill the DT_DEBUG entry in ld.so [BZ #28129]
Linker creates the DT_DEBUG entry only in executables. Don't fill the
non-existent DT_DEBUG entry in ld.so with the run-time address of the
r_debug structure. This fixes BZ #28129.
Stefan Liebler [Wed, 6 Oct 2021 14:34:40 +0000 (16:34 +0200)]
S390: update libm test ulps
Update after
commit
6bbf7298323bf31bc43494b2201465a449778e10.
Fixed inaccuracy of j0f (BZ #28185)
See also e.g.
commit
c75b106145c30e6c7bcf87f384a5c68ce56406e9
aarch64: update libm test ulps
Adhemerval Zanella [Wed, 6 Oct 2021 13:49:28 +0000 (16:49 +0300)]
powerpc: update libm test ulps
Update after commit
6bbf7298323bf31bc43494b2201465a449778e10
(Fixed inaccuracy of j0f (BZ #28185)).
Adhemerval Zanella [Wed, 6 Oct 2021 13:47:52 +0000 (16:47 +0300)]
math: Also xfail the new j0f tests for ibm128-libgcc
From commit
6bbf7298323bf31b.
Checked on powerpc64-linux-gnu.