[gcc(refs/users/aoliva/heads/testme)] [testsuite] [x86] strlenopt-80 needs -msse2 on ia32

2025-05-16 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:27737095d9d2eff025363f031bcacbc1ad73f340 commit 27737095d9d2eff025363f031bcacbc1ad73f340 Author: Alexandre Oliva Date: Fri May 16 07:52:41 2025 -0300 [testsuite] [x86] strlenopt-80 needs -msse2 on ia32 The string length optimizations at 8-byte blocks requires

[gcc(refs/users/aoliva/heads/testme)] [testsuite] [x86] memcpy-6 needs -msse2

2025-05-16 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:42834053a4ab2ee0adb5aeb0fcc4aab8b4757419 commit 42834053a4ab2ee0adb5aeb0fcc4aab8b4757419 Author: Alexandre Oliva Date: Fri May 16 07:52:24 2025 -0300 [testsuite] [x86] memcpy-6 needs -msse2 The 8-byte memory operations will only be inlined on ia32 with -m

[gcc(refs/users/aoliva/heads/testme)] [testsuite] require sysconf for vect-early-break_109-pr113588.c

2025-05-16 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:72d5ed7a53c8a8dab17494c8593358eab823c3fd commit 72d5ed7a53c8a8dab17494c8593358eab823c3fd Author: Alexandre Oliva Date: Thu May 15 05:50:57 2025 -0300 [testsuite] require sysconf for vect-early-break_109-pr113588.c The test calls sysconf but it doesn't require

[gcc(refs/users/aoliva/heads/testme)] [testsuite] [x86] no-callee-saved-16.c needs -fomit-frame-pointer

2025-05-16 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:cbd499db356a42e2cd9fd4f4635acd8d08a3f074 commit cbd499db356a42e2cd9fd4f4635acd8d08a3f074 Author: Alexandre Oliva Date: Fri May 16 07:53:30 2025 -0300 [testsuite] [x86] no-callee-saved-16.c needs -fomit-frame-pointer If the toolchain is built with --enable-fra

[gcc r16-789] vxworks: libgcc: include string.h for memset

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:99a65bfe59208c33a74718ef5fc05e255a76393a commit r16-789-g99a65bfe59208c33a74718ef5fc05e255a76393a Author: Alexandre Oliva Date: Wed May 21 06:19:46 2025 -0300 vxworks: libgcc: include string.h for memset gthr-vxworks-thread.c calls memset in __ghtread_cond_si

[gcc r16-790] [testsuite] [vxworks] netinet includes atomic, reqs c++11

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:659fe2a28e8cbaf4672d4db8ef3f13c6efed9c0c commit r16-790-g659fe2a28e8cbaf4672d4db8ef3f13c6efed9c0c Author: Alexandre Oliva Date: Wed May 21 06:19:57 2025 -0300 [testsuite] [vxworks] netinet includes atomic, reqs c++11 On vxworks, the included netinet/in.h head

[gcc r16-794] [testsuite] [aarch64] match alt cache clear names in sme nonlocal_goto tests

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:4b75decf5297838ef9ddfb842da0117adbe3f975 commit r16-794-g4b75decf5297838ef9ddfb842da0117adbe3f975 Author: Alexandre Oliva Date: Wed May 21 06:20:22 2025 -0300 [testsuite] [aarch64] match alt cache clear names in sme nonlocal_goto tests vxworks calls cacheText

[gcc r16-795] [testsuite] [x86] double copysign requires -msse2

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:02788cde86264559ca9cb3323c73c72fd0211c5d commit r16-795-g02788cde86264559ca9cb3323c73c72fd0211c5d Author: Alexandre Oliva Date: Wed May 21 06:20:29 2025 -0300 [testsuite] [x86] double copysign requires -msse2 SSE_FLOAT_MODE_P only holds for DFmode with SSE2,

[gcc r16-791] [testsuite] [analyzer] [vxworks] define __STDC_WANT_LIB_EXT1__ to 1

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:f3c5e0a2091ddd5cae4d7381a847aac5f546f04c commit r16-791-gf3c5e0a2091ddd5cae4d7381a847aac5f546f04c Author: Alexandre Oliva Date: Wed May 21 06:20:03 2025 -0300 [testsuite] [analyzer] [vxworks] define __STDC_WANT_LIB_EXT1__ to 1 vxworks' headers use #if instead

[gcc r16-793] [testsuite] [aarch64] use uint64_t in rwsr tests

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:d41028df8b39087ef71de84a5daf68cb305b9f7f commit r16-793-gd41028df8b39087ef71de84a5daf68cb305b9f7f Author: Alexandre Oliva Date: Wed May 21 06:20:17 2025 -0300 [testsuite] [aarch64] use uint64_t in rwsr tests stdint.h defines uint64_t instead of __uint64_t, so

[gcc r16-797] [testsuite] [x86] strlenopt-80 needs -msse2 on ia32

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:fe9be609d7a10f2cd706aecd772e3e3427868daf commit r16-797-gfe9be609d7a10f2cd706aecd772e3e3427868daf Author: Alexandre Oliva Date: Wed May 21 06:20:37 2025 -0300 [testsuite] [x86] strlenopt-80 needs -msse2 on ia32 The string length optimizations at 8-byte blocks

[gcc r16-800] [testsuite] [x86] no-callee-saved-16.c needs -fomit-frame-pointer

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:012a857d1eb5b45baee8752e3b5a434fa25c52e2 commit r16-800-g012a857d1eb5b45baee8752e3b5a434fa25c52e2 Author: Alexandre Oliva Date: Wed May 21 06:20:54 2025 -0300 [testsuite] [x86] no-callee-saved-16.c needs -fomit-frame-pointer If the toolchain is built with --e

[gcc r16-802] [testsuite] [x86] pr31985.c needs -fomit-frame-pointer to match movl count

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:6621311c309fcc68ecdd395bceb9ad7994bed41f commit r16-802-g6621311c309fcc68ecdd395bceb9ad7994bed41f Author: Alexandre Oliva Date: Wed May 21 06:21:04 2025 -0300 [testsuite] [x86] pr31985.c needs -fomit-frame-pointer to match movl count On an --enable-frame-poin

[gcc r16-801] [testsuite] [x86] pr108938-3.c needs -msse2 for bswap in foo2 with -m32

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:3392849511c9b9eb1d912a547f2441cece766e3b commit r16-801-g3392849511c9b9eb1d912a547f2441cece766e3b Author: Alexandre Oliva Date: Wed May 21 06:20:59 2025 -0300 [testsuite] [x86] pr108938-3.c needs -msse2 for bswap in foo2 with -m32 Without SSE2, we don't combi

[gcc r16-803] [testsuite] [x86] vect-simd-clone-1[678]e.c adjust

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:e82a9f6536ba8fcf50a7172650a86519c50aeabd commit r16-803-ge82a9f6536ba8fcf50a7172650a86519c50aeabd Author: Alexandre Oliva Date: Wed May 21 06:21:08 2025 -0300 [testsuite] [x86] vect-simd-clone-1[678]e.c adjust Since r13-6296, we haven't got 4 simdclone calls

[gcc r16-796] [testsuite] [x86] memcpy-6 needs -msse2

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:f3a758e9a2b2aa40bda68a18157996167d772e4d commit r16-796-gf3a758e9a2b2aa40bda68a18157996167d772e4d Author: Alexandre Oliva Date: Wed May 21 06:20:33 2025 -0300 [testsuite] [x86] memcpy-6 needs -msse2 The 8-byte memory operations will only be inlined on ia32 wi

[gcc r16-798] [testsuite] [x86] forwprop-41 needs -msse

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:8bb72b737c38adb08bfefabc43cb4f25f7d9e95d commit r16-798-g8bb72b737c38adb08bfefabc43cb4f25f7d9e95d Author: Alexandre Oliva Date: Wed May 21 06:20:42 2025 -0300 [testsuite] [x86] forwprop-41 needs -msse The vector operations are only turned into BIT_INSERT_EXPR

[gcc r16-799] [testsuite] add missing require vect_early_break_hw for vect-tsvc

2025-05-21 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:57cc4f8bf31f8c25fd338b18b5e709d77cc2a0ad commit r16-799-g57cc4f8bf31f8c25fd338b18b5e709d77cc2a0ad Author: Alexandre Oliva Date: Wed May 21 06:20:48 2025 -0300 [testsuite] add missing require vect_early_break_hw for vect-tsvc Some tsvc tests add vect_early_bre

[gcc(refs/users/aoliva/heads/testme)] [lra] force reg update after spilling to memory [PR120424]

2025-05-24 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:82fd9054ef3f5d0381591e9c62f031e5e4a9e1d9 commit 82fd9054ef3f5d0381591e9c62f031e5e4a9e1d9 Author: Alexandre Oliva Date: Sat May 24 19:48:23 2025 -0300 [lra] force reg update after spilling to memory [PR120424] In the added C++ testcase, a stack slot at a negat

[gcc/aoliva/heads/testme] [lra] force reg update after spilling to memory [PR120424]

2025-05-24 Thread Alexandre Oliva via Gcc-cvs
The branch 'aoliva/heads/testme' was updated to point to: 82fd9054ef3f... [lra] force reg update after spilling to memory [PR120424] It previously pointed to: 1445b105ff49... [lra] force reg update after spilling to memory Diff: !!! WARNING: THE FOLLOWING COMMITS ARE NO LONGER ACCESSIBLE (LO

[gcc/aoliva/heads/testme] (180 commits) [lra] force reg update after spilling to memory

2025-05-24 Thread Alexandre Oliva via Gcc-cvs
The branch 'aoliva/heads/testme' was updated to point to: 1445b105ff49... [lra] force reg update after spilling to memory It previously pointed to: 64388b09fdcf... [testsuite] [x86] vect-simd-clone-1[678]e.c adjust Diff: !!! WARNING: THE FOLLOWING COMMITS ARE NO LONGER ACCESSIBLE (LOST): ---

[gcc/aoliva/heads/testbase] (179 commits) [vxworks] build partial libatomic

2025-05-24 Thread Alexandre Oliva via Gcc-cvs
The branch 'aoliva/heads/testbase' was updated to point to: 10360c1b0d45... [vxworks] build partial libatomic It previously pointed to: f99017c3125f... OpenMP/Fortran: Fix allocatable-component mapping of derive Diff: Summary of changes (added commits): ---

[gcc(refs/users/aoliva/heads/testme)] [lra] force reg update after spilling to memory

2025-05-24 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:1445b105ff49c01c048cc204c205d39017da7874 commit 1445b105ff49c01c048cc204c205d39017da7874 Author: Alexandre Oliva Date: Sat May 24 20:15:02 2025 -0300 [lra] force reg update after spilling to memory Diff: --- gcc/lra-eliminations.cc | 41 +

[gcc r16-826] iesFrom: Alexandre Oliva

2025-05-22 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:a4f45de95eed90e3d166bb706bb883e3a6cda0bd commit r16-826-ga4f45de95eed90e3d166bb706bb883e3a6cda0bd Author: Alexandre Oliva Date: Thu May 22 15:06:24 2025 -0300 iesFrom: Alexandre Oliva [aarch64] [vxworks] mark x18 as fixed, adjust tests VxWorks uses

[gcc r16-827] [vxworks] build partial libatomic

2025-05-22 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:10360c1b0d45ae129df616a9e9b1db5f2a2eaef8 commit r16-827-g10360c1b0d45ae129df616a9e9b1db5f2a2eaef8 Author: Alexandre Oliva Date: Thu May 22 15:15:31 2025 -0300 [vxworks] build partial libatomic Since vxworks' libc contains much of libatomic, in not-very-granul

[gcc r16-1065] [lra] force reg update after spilling to memory [PR120424]

2025-06-02 Thread Alexandre Oliva via Gcc-cvs
https://gcc.gnu.org/g:2edb50a310896da72710dba7724a3b4a541a1cbb commit r16-1065-g2edb50a310896da72710dba7724a3b4a541a1cbb Author: Alexandre Oliva Date: Mon Jun 2 20:21:45 2025 -0300 [lra] force reg update after spilling to memory [PR120424] In the added C++ testcase, a stack slot a

<    5   6   7   8   9   10