| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
|
|
| |
geometric_mean(N=40) of all benchmarks EVEX / SSE42: .621
All string/memory tests pass.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 84e7c46df4086873eae28a1fb87d2cf5388b1e16)
|
|
|
|
|
|
|
|
|
| |
geometric_mean(N=40) of all benchmarks AVX2 / SSE42: .702
All string/memory tests pass.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit bbf81222343fed5cd704001a2ae0d86c71544151)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Slightly faster method of doing TOLOWER that saves an
instruction.
Also replace the hard coded 5-byte no with .p2align 4. On builds with
CET enabled this misaligned entry to strcasecmp.
geometric_mean(N=40) of all benchmarks New / Original: .920
All string/memory tests pass.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit d154758e618ec9324f5d339c46db0aa27e8b1226)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Slightly faster method of doing TOLOWER that saves an
instruction.
Also replace the hard coded 5-byte no with .p2align 4. On builds with
CET enabled this misaligned entry to strcasecmp.
geometric_mean(N=40) of all benchmarks New / Original: .894
All string/memory tests pass.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 670b54bc585ea4a94f3b2e9272ba44aa6b730b73)
|
|
|
|
|
|
|
|
|
|
|
| |
The generic implementation is faster.
geometric_mean(N=20) of all benchmarks New / Original: .710
All string/memory tests pass.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 9c8a6ad620b49a27120ecdd7049c26bf05900397)
|
|
|
|
|
|
|
|
|
| |
The generic implementation is faster (see strcspn commit).
All string/memory tests pass.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 653358535280a599382cb6c77538a187dac6a87f)
|
|
|
|
|
|
|
|
|
|
|
| |
The generic implementation is faster.
geometric_mean(N=20) of all benchmarks New / Original: .678
All string/memory tests pass.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit fe28e7d9d9535ebab4081d195c553b4fbf39d9ae)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Use _mm_cmpeq_epi8 and _mm_movemask_epi8 to get strlen instead of
_mm_cmpistri. Also change offset to unsigned to avoid unnecessary
sign extensions.
geometric_mean(N=20) of all benchmarks that dont fallback on
sse2; New / Original: .901
All string/memory tests pass.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 412d10343168b05b8cf6c3683457cf9711d28046)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Use _mm_cmpeq_epi8 and _mm_movemask_epi8 to get strlen instead of
_mm_cmpistri. Also change offset to unsigned to avoid unnecessary
sign extensions.
geometric_mean(N=20) of all benchmarks that dont fallback on
sse2/strlen; New / Original: .928
All string/memory tests pass.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 30d627d477d7255345a4b713cf352ac32d644d61)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Small code cleanup for size: -81 bytes.
Add comment justifying using a branch to do NULL/non-null return.
All string/memory tests pass and no regressions in benchtests.
geometric_mean(N=20) of all benchmarks New / Original: .985
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit ec285ea90415458225623ddc0492ae3f705af043)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Small code cleanup for size: -53 bytes.
Add comment justifying using a branch to do NULL/non-null return.
All string/memory tests pass and no regressions in benchtests.
geometric_mean(N=20) of all benchmarks Original / New: 1.00
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit a6fbf4d51e9ba8063c4f8331564892ead9c67344)
|
|
|
|
|
|
|
| |
The symbols is not present in current POSIX specification and compiler
already generates memmove call.
(cherry picked from commit bf92893a14ebc161b08b28acc24fa06ae6be19cb)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
commit 3d9f171bfb5325bd5f427e9fc386453358c6e840
Author: H.J. Lu <hjl.tools@gmail.com>
Date: Mon Feb 7 05:55:15 2022 -0800
x86-64: Optimize bzero
added the optimized bzero. Remove bzero weak alias in SS2 memset to
avoid undefined __bzero in memset-sse2-unaligned-erms.
(cherry picked from commit 0fb8800029d230b3711bf722b2a47db92d0e273f)
|
|
|
|
| |
(cherry picked from commit c328d0152d4b14cca58407ec68143894c8863004)
|
|
|
|
| |
(cherry picked from commit 1283948f236f209b7d3f44b69a42b96806fa6da0)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Overflow case for __wcsncmp_avx2_rtm should be __wcscmp_avx2_rtm not
__wcscmp_avx2.
commit ddf0992cf57a93200e0c782e2a94d0733a5a0b87
Author: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Sun Jan 9 16:02:21 2022 -0600
x86: Fix __wcsncmp_avx2 in strcmp-avx2.S [BZ# 28755]
Set the wrong fallback function for `__wcsncmp_avx2_rtm`. It was set
to fallback on to `__wcscmp_avx2` instead of `__wcscmp_avx2_rtm` which
can cause spurious aborts.
This change will need to be backported.
All string/memory tests pass.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 9fef7039a7d04947bc89296ee0d187bc8d89b772)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Logic can read before the start of `s1` / `s2` if both `s1` and `s2`
are near the start of a page. To avoid having the result contimated by
these comparisons the `strcmp` variants would mask off these
comparisons. This was missing in the `strncmp` variants causing
the bug. This commit adds the masking to `strncmp` so that out of
range comparisons don't affect the result.
test-strcmp, test-strncmp, test-wcscmp, and test-wcsncmp all pass as
well a full xcheck on x86_64 linux.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit e108c02a5e23c8c88ce66d8705d4a24bb6b9a8bf)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
commit 3d9f171bfb5325bd5f427e9fc386453358c6e840
Author: H.J. Lu <hjl.tools@gmail.com>
Date: Mon Feb 7 05:55:15 2022 -0800
x86-64: Optimize bzero
Remove setting the .text section for the code. This commit
adds that back.
(cherry picked from commit 7912236f4a597deb092650ca79f33504ddb4af28)
|
|
|
|
|
|
|
|
|
|
|
| |
memset with zero as the value to set is by far the majority value (99%+
for Python3 and GCC).
bzero can be slightly more optimized for this case by using a zero-idiom
xor for broadcasting the set value to a register (vector or GPR).
Co-developed-by: Noah Goldstein <goldstein.w.n@gmail.com>
(cherry picked from commit 3d9f171bfb5325bd5f427e9fc386453358c6e840)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
commit b62ace2740a106222e124cc86956448fa07abf4d
Author: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Sun Feb 6 00:54:18 2022 -0600
x86: Improve vec generation in memset-vec-unaligned-erms.S
Revert usage of 'pshufb' in broadcast logic as it is an SSSE3
instruction and memset.S is restricted to only SSE2 instructions.
(cherry picked from commit 1b0c60f95bbe2eded80b2bb5be75c0e45b11cde1)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
No bug.
Split vec generation into multiple steps. This allows the
broadcast in AVX2 to use 'xmm' registers for the L(less_vec)
case. This saves an expensive lane-cross instruction and removes
the need for 'vzeroupper'.
For SSE2 replace 2x 'punpck' instructions with zero-idiom 'pxor' for
byte broadcast.
Results for memset-avx2 small (geomean of N = 20 benchset runs).
size, New Time, Old Time, New / Old
0, 4.100, 3.831, 0.934
1, 5.074, 4.399, 0.867
2, 4.433, 4.411, 0.995
4, 4.487, 4.415, 0.984
8, 4.454, 4.396, 0.987
16, 4.502, 4.443, 0.987
All relevant string/wcsmbs tests are passing.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit b62ace2740a106222e124cc86956448fa07abf4d)
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change "movl %edx, %rdx" to "movl %edx, %edx" in:
commit 8418eb3ff4b781d31c4ed5dc6c0bd7356bc45db9
Author: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Mon Jan 10 15:35:39 2022 -0600
x86: Optimize strcmp-evex.S
(cherry picked from commit 0e0199a9e02ebe42e2b36958964d63f03573c382)
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change "movl %edx, %rdx" to "movl %edx, %edx" in:
commit b77b06e0e296f1a2276c27a67e1d44f2cfa38d45
Author: Noah Goldstein <goldstein.w.n@gmail.com>
Date: Mon Jan 10 15:35:38 2022 -0600
x86: Optimize strcmp-avx2.S
(cherry picked from commit c15efd011cea3d8f0494269eb539583215a1feed)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Optimization are primarily to the loop logic and how the page cross
logic interacts with the loop.
The page cross logic is at times more expensive for short strings near
the end of a page but not crossing the page. This is done to retest
the page cross conditions with a non-faulty check and to improve the
logic for entering the loop afterwards. This is only particular cases,
however, and is general made up for by more than 10x improvements on
the transition from the page cross -> loop case.
The non-page cross cases as well are nearly universally improved.
test-strcmp, test-strncmp, test-wcscmp, and test-wcsncmp all pass.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
(cherry picked from commit 8418eb3ff4b781d31c4ed5dc6c0bd7356bc45db9)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Optimization are primarily to the loop logic and how the page cross
logic interacts with the loop.
The page cross logic is at times more expensive for short strings near
the end of a page but not crossing the page. This is done to retest
the page cross conditions with a non-faulty check and to improve the
logic for entering the loop afterwards. This is only particular cases,
however, and is general made up for by more than 10x improvements on
the transition from the page cross -> loop case.
The non-page cross cases are improved most for smaller sizes [0, 128]
and go about even for (128, 4096]. The loop page cross logic is
improved so some more significant speedup is seen there as well.
test-strcmp, test-strncmp, test-wcscmp, and test-wcsncmp all pass.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
(cherry picked from commit b77b06e0e296f1a2276c27a67e1d44f2cfa38d45)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
No bug.
Optimizations are twofold.
1) Replace page cross and 0/1 checks with masked load instructions in
L(less_vec). In applications this reduces branch-misses in the
hot [0, 32] case.
2) Change controlflow so that L(less_vec) case gets the fall through.
Change 2) helps copies in the [0, 32] size range but comes at the cost
of copies in the [33, 64] size range. From profiles of GCC and
Python3, 94%+ and 99%+ of calls are in the [0, 32] range so this
appears to the the right tradeoff.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit abddd61de090ae84e380aff68a98bd94ef704667)
|
|
|
|
|
|
|
|
| |
Don't set Prefer_No_AVX512 on processors with AVX512 and AVX-VNNI since
they won't lower CPU frequency when ZMM load and store instructions are
used.
(cherry picked from commit ceeffe968c01b1202e482f4855cb6baf5c6cb713)
|
|
|
|
|
|
|
|
|
| |
Must use notl %edi here as lower bits are for CHAR comparisons
potentially out of range thus can be 0 without indicating mismatch.
This fixes BZ #28646.
Co-Authored-By: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 4df1fa6ddc8925a75f3da644d5da3bb16eb33f02)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
No bug.
This implementation refactors memcmp-sse4.S primarily with minimizing
code size in mind. It does this by removing the lookup table logic and
removing the unrolled check from (256, 512] bytes.
memcmp-sse4 code size reduction : -3487 bytes
wmemcmp-sse4 code size reduction: -1472 bytes
The current memcmp-sse4.S implementation has a large code size
cost. This has serious adverse affects on the ICache / ITLB. While
in micro-benchmarks the implementations appears fast, traces of
real-world code have shown that the speed in micro benchmarks does not
translate when the ICache/ITLB are not primed, and that the cost
of the code size has measurable negative affects on overall
application performance.
See https://research.google/pubs/pub48320/ for more details.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 2f9062d7171850451e6044ef78d91ff8c017b9c0)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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>
(cherry picked from commit 475b63702ef38b69558fc3d31a0b66776a70f1d3)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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>
(cherry picked from commit a6b7502ec0c2da89a7437f43171f160d713e39c6)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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>
(cherry picked from commit 6720d36b6623c5e48c070d86acf61198b33e144e)
|
|
|
|
|
|
|
|
|
|
|
| |
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.
(cherry picked from commit 14dbbf46a007ae5df36646b51ad0c9e5f5259f30)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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>
(cherry picked from commit c46e9afb2df5fc9e39ff4d13777e4b4c26e04e55)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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>
(cherry picked from commit bad852b61b79503fcb3c5fc379c70f768df3e1fb)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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>
(cherry picked from commit e59ced238482fd71f3e493717f14f6507346741e)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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.
(cherry picked from commit 1bd8b8d58fc9967cc073d2c13bfb6befefca2faa)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
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>
(cherry picked from commit fc5bd179ef3a953dff8d1655bd530d0e230ffe71)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Optimize loads of all bits set into ZMM register in AVX512 SVML codes
by replacing
vpbroadcastq .L_2il0floatpacket.16(%rip), %zmmX
and
vmovups .L_2il0floatpacket.13(%rip), %zmmX
with
vpternlogd $0xff, %zmmX, %zmmX, %zmmX
This fixes BZ #28252.
(cherry picked from commit 78c9ec9000f873abe7a15a91b87080a2e4308260)
|
|
|
|
|
|
|
|
|
|
|
| |
Fixes [BZ# 28755] for wcsncmp by redirecting length >= 2^56 to
__wcscmp_evex. For x86_64 this covers the entire address range so any
length larger could not possibly be used to bound `s1` or `s2`.
test-strcmp, test-strncmp, test-wcscmp, and test-wcsncmp all pass.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
(cherry picked from commit 7e08db3359c86c94918feb33a1182cd0ff3bb10b)
|
|
|
|
|
|
|
|
| |
Use testl, instead of andl, to check __x86_string_control to avoid
updating __x86_string_control.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
(cherry picked from commit 3c8b9879cab6d41787bc5b14c1748f62fd6d0e5f)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
commit 3ec5d83d2a237d39e7fd6ef7a0bc8ac4c171a4a5
Author: H.J. Lu <hjl.tools@gmail.com>
Date: Sat Jan 25 14:19:40 2020 -0800
x86-64: Avoid rep movsb with short distance [BZ #27130]
introduced some regressions on Intel processors without Fast Short REP
MOV (FSRM). Add Avoid_Short_Distance_REP_MOVSB to avoid rep movsb with
short distance only on Intel processors with FSRM. bench-memmove-large
on Skylake server shows that cycles of __memmove_evex_unaligned_erms
improves for the following data size:
before after Improvement
length=4127, align1=3, align2=0: 479.38 349.25 27%
length=4223, align1=9, align2=5: 405.62 333.25 18%
length=8223, align1=3, align2=0: 786.12 496.38 37%
length=8319, align1=9, align2=5: 727.50 501.38 31%
length=16415, align1=3, align2=0: 1436.88 840.00 41%
length=16511, align1=9, align2=5: 1375.50 836.38 39%
length=32799, align1=3, align2=0: 2890.00 1860.12 36%
length=32895, align1=9, align2=5: 2891.38 1931.88 33%
(cherry picked from commit 91cc803d27bda34919717b496b53cf279e44a922)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
No bug. The way wcsnlen will check if near the end of maxlen
is the following macro:
mov %r11, %rsi; \
subq %rax, %rsi; \
andq $-64, %rax; \
testq $-64, %rsi; \
je L(strnlen_ret)
Which words independently of s + maxlen overflowing. So the
second overflow check is unnecissary for correctness and
just extra overhead in the common no overflow case.
test-strlen.c, test-wcslen.c, test-strnlen.c and test-wcsnlen.c are
all passing
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 08cbcd4dbc686bb38ec3093aff2f919fbff5ec17)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch changes the condition for copy 4x VEC so that if length is
exactly equal to 4 * VEC_SIZE it will use the 4x VEC case instead of
8x VEC case.
Results For Skylake memcpy-avx2-erms
size, al1 , al2 , Cur T , New T , Win , New / Cur
128 , 0 , 0 , 9.137 , 6.873 , New , 75.22
128 , 7 , 0 , 12.933 , 7.732 , New , 59.79
128 , 0 , 7 , 11.852 , 6.76 , New , 57.04
128 , 7 , 7 , 12.587 , 6.808 , New , 54.09
Results For Icelake memcpy-evex-erms
size, al1 , al2 , Cur T , New T , Win , New / Cur
128 , 0 , 0 , 9.963 , 5.416 , New , 54.36
128 , 7 , 0 , 16.467 , 8.061 , New , 48.95
128 , 0 , 7 , 14.388 , 7.644 , New , 53.13
128 , 7 , 7 , 14.546 , 7.642 , New , 52.54
Results For Tigerlake memcpy-evex-erms
size, al1 , al2 , Cur T , New T , Win , New / Cur
128 , 0 , 0 , 8.979 , 4.95 , New , 55.13
128 , 7 , 0 , 14.245 , 7.122 , New , 50.0
128 , 0 , 7 , 12.668 , 6.675 , New , 52.69
128 , 7 , 7 , 13.042 , 6.802 , New , 52.15
Results For Skylake memmove-avx2-erms
size, al1 , al2 , Cur T , New T , Win , New / Cur
128 , 0 , 32 , 6.181 , 5.691 , New , 92.07
128 , 32 , 0 , 6.165 , 5.752 , New , 93.3
128 , 0 , 7 , 13.923 , 9.37 , New , 67.3
128 , 7 , 0 , 12.049 , 10.182 , New , 84.5
Results For Icelake memmove-evex-erms
size, al1 , al2 , Cur T , New T , Win , New / Cur
128 , 0 , 32 , 5.479 , 4.889 , New , 89.23
128 , 32 , 0 , 5.127 , 4.911 , New , 95.79
128 , 0 , 7 , 18.885 , 13.547 , New , 71.73
128 , 7 , 0 , 15.565 , 14.436 , New , 92.75
Results For Tigerlake memmove-evex-erms
size, al1 , al2 , Cur T , New T , Win , New / Cur
128 , 0 , 32 , 5.275 , 4.815 , New , 91.28
128 , 32 , 0 , 5.376 , 4.565 , New , 84.91
128 , 0 , 7 , 19.426 , 14.273 , New , 73.47
128 , 7 , 0 , 15.924 , 14.951 , New , 93.89
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
(cherry picked from commit 1b992204f68af851e905c16016756fd4421e1934)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
No bug. This commit makes a few small improvements to
memset-vec-unaligned-erms.S. The changes are 1) only aligning to 64
instead of 128. Either alignment will perform equally well in a loop
and 128 just increases the odds of having to do an extra iteration
which can be significant overhead for small values. 2) Align some
targets and the loop. 3) Remove an ALU from the alignment process. 4)
Reorder the last 4x VEC so that they are stored after the loop. 5)
Move the condition for leq 8x VEC to before the alignment
process. test-memset and test-wmemset are both passing.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 6abf27980a947f9b6e514d6b33b83059d39566ae)
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
No bug. This commit optimizes memcmp-evex.S. The optimizations include
adding a new vec compare path for small sizes, reorganizing the entry
control flow, removing some unnecissary ALU instructions from the main
loop, and most importantly replacing the heavy use of vpcmp + kand
logic with vpxor + vptern. test-memcmp and test-wmemcmp are both
passing.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 4ad473e97acdc5f6d811755b67c09f2128a644ce)
|
|
|
|
|
|
|
|
|
|
|
| |
No bug. This commit optimizes memcmp-avx2.S. The optimizations include
adding a new vec compare path for small sizes, reorganizing the entry
control flow, and removing some unnecissary ALU instructions from the
main loop. test-memcmp and test-wmemcmp are both passing.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 16d12015c57701b08d7bbed6ec536641bcafb428)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
No bug.
This commit adds a new implementation for EVEX memchr that is not safe
for RTM because it uses vzeroupper. The benefit is that by using
ymm0-ymm15 it can use vpcmpeq and vpternlogd in the 4x loop which is
faster than the RTM safe version which cannot use vpcmpeq because
there is no EVEX encoding for the instruction. All parts of the
implementation aside from the 4x loop are the same for the two
versions and the optimization is only relevant for large sizes.
Tigerlake:
size , algn , Pos , Cur T , New T , Win , Dif
512 , 6 , 192 , 9.2 , 9.04 , no-RTM , 0.16
512 , 7 , 224 , 9.19 , 8.98 , no-RTM , 0.21
2048 , 0 , 256 , 10.74 , 10.54 , no-RTM , 0.2
2048 , 0 , 512 , 14.81 , 14.87 , RTM , 0.06
2048 , 0 , 1024 , 22.97 , 22.57 , no-RTM , 0.4
2048 , 0 , 2048 , 37.49 , 34.51 , no-RTM , 2.98 <--
Icelake:
size , algn , Pos , Cur T , New T , Win , Dif
512 , 6 , 192 , 7.6 , 7.3 , no-RTM , 0.3
512 , 7 , 224 , 7.63 , 7.27 , no-RTM , 0.36
2048 , 0 , 256 , 8.48 , 8.38 , no-RTM , 0.1
2048 , 0 , 512 , 11.57 , 11.42 , no-RTM , 0.15
2048 , 0 , 1024 , 17.92 , 17.38 , no-RTM , 0.54
2048 , 0 , 2048 , 30.37 , 27.34 , no-RTM , 3.03 <--
test-memchr, test-wmemchr, and test-rawmemchr are all passing.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit 104c7b1967c3e78435c6f7eab5e225a7eddf9c6e)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The glibc memcpy benchmark on Intel Core i7-1065G7 (Ice Lake) showed
that REP MOVSB became faster after 2112 bytes:
Vector Move REP MOVSB
length=2112, align1=0, align2=0: 24.20 24.40
length=2112, align1=1, align2=0: 26.07 23.13
length=2112, align1=0, align2=1: 27.18 28.13
length=2112, align1=1, align2=1: 26.23 25.16
length=2176, align1=0, align2=0: 23.18 22.52
length=2176, align1=2, align2=0: 25.45 22.52
length=2176, align1=0, align2=2: 27.14 27.82
length=2176, align1=2, align2=2: 22.73 25.56
length=2240, align1=0, align2=0: 24.62 24.25
length=2240, align1=3, align2=0: 29.77 27.15
length=2240, align1=0, align2=3: 35.55 29.93
length=2240, align1=3, align2=3: 34.49 25.15
length=2304, align1=0, align2=0: 34.75 26.64
length=2304, align1=4, align2=0: 32.09 22.63
length=2304, align1=0, align2=4: 28.43 31.24
Use REP MOVSB for data size > 2112 bytes in memcpy on processors with
fast short REP MOVSB (FSRM).
* sysdeps/x86/dl-cacheinfo.h (dl_init_cacheinfo): Set
rep_movsb_threshold to 2112 on processors with fast short REP
MOVSB (FSRM).
(cherry picked from commit cf2c57526ba4b57e6863ad4db8a868e2678adce8)
|
|
|
|
|
|
|
|
|
|
|
| |
No bug. This commit optimizes strchr-evex.S. The optimizations are
mostly small things such as save an ALU in the alignment process,
saving a few instructions in the loop return. The one significant
change is saving 2 instructions in the 4x loop. test-strchr,
test-strchrnul, test-wcschr, and test-wcschrnul are all passing.
Signed-off-by: Noah Goldstein <goldstein.w.n@gmail.com>
(cherry picked from commit 7f3e7c262cab4e2401e4331a6ef29c428de02044)
|