[PATCH] D159250: [X86][RFC] Add new option `-m[no-]evex512` to disable ZMM and 64-bit mask instructions for AVX512 features

RKSimon accepted this revision. RKSimon added a comment. LGTM with a couple of minors Comment at: clang/lib/CodeGen/Targets/X86.cpp:1493 const llvm::StringMap &CalleeMap, QualType Ty, StringRef Feature,

[PATCH] D159250: [X86][RFC] Add new option `-m[no-]evex512` to disable ZMM and 64-bit mask instructions for AVX512 features

RKSimon added a comment. Would it be possible to add function multiversioning tests to ensure the evex512 attribute would work with it? Comment at: clang/lib/CodeGen/Targets/X86.cpp:1517 + bool Caller256 = CallerMap.lookup("avx512f") && !CallerMap.lookup("evex512"); + bool C

[PATCH] D158046: [X86] Support -march=gracemont

RKSimon accepted this revision. RKSimon added a comment. This revision is now accepted and ready to land. LGTM - cheers Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D158046/new/ https://reviews.llvm.org/D158046

[PATCH] D158046: [X86] Support -march=gracemont

RKSimon added inline comments. Comment at: clang/lib/Basic/Targets/X86.cpp:551 case CK_Lunarlake: + case CK_Gracemont: case CK_Sierraforest: FreddyYe wrote: > RKSimon wrote: > > Why not handle this above (below tremont) as the next in the *mont series? > G

[PATCH] D158046: [X86] Support -march=gracemont

RKSimon added inline comments. Comment at: clang/lib/Basic/Targets/X86.cpp:551 case CK_Lunarlake: + case CK_Gracemont: case CK_Sierraforest: Why not handle this above (below tremont) as the next in the *mont series? Repository: rG LLVM Github Monorepo

[PATCH] D157420: [clang] Enable constexpr on LZCNT/POPCNT MS extension intrinsics

This revision was automatically updated to reflect the committed changes. Closed by commit rG00158ae236dd: [clang] Enable constexpr on LZCNT/POPCNT MS extension intrinsics (authored by alexguirre, committed by RKSimon). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://re

[PATCH] D157420: [clang] Enable constexpr on LZCNT/POPCNT MS extension intrinsics

RKSimon accepted this revision. RKSimon added a comment. This revision is now accepted and ready to land. LGTM - cheers @alexguirre Please post your full name / email and I'll commit this for you Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D15742

[PATCH] D157297: [clang] Fixes compile error like error: expected unqualified-id for ::_tzcnt_u32(mask);

RKSimon added a comment. Would we be better off creating proper function definitions - we already have similar duplicates for `_bextr_u64` (Intel) vs `__bextr_u64` (AMD) names that do this Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D157297/new/

[PATCH] D157420: [clang] Enable constexpr on LZCNT/POPCNT MS extension intrinsics

RKSimon added reviewers: erichkeane, aaron.ballman. RKSimon added a comment. Please can you add them to the constexpr lists in LanguageExtensions.rst ? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D157420/new/ https://reviews.llvm.org/D157420

[PATCH] D157078: [include-cleaner] Handle files with unnamed buffers

RKSimon added a comment. In D157078#4562788 , @tylanphear wrote: > Seeing the link error downstream. I think clangSerialization needs to be > added to the link libraries. Should be fixed by 36daf3532d91bb

[PATCH] D154130: [lit][clang] Avoid realpath on Windows due to MAX_PATH limitations

RKSimon added inline comments. Comment at: llvm/utils/lit/lit/discovery.py:60 +cfgpath = util.abs_path_preserve_drive(cfgpath) +target = config_map.get(cfgpath) if target: RKSimon wrote: > RKSimon wrote: > > Found the problem

[PATCH] D154130: [lit][clang] Avoid realpath on Windows due to MAX_PATH limitations

RKSimon added inline comments. Comment at: llvm/utils/lit/lit/discovery.py:60 +cfgpath = util.abs_path_preserve_drive(cfgpath) +target = config_map.get(cfgpath) if target: RKSimon wrote: > Found the problem - you have moved th

[PATCH] D154130: [lit][clang] Avoid realpath on Windows due to MAX_PATH limitations

RKSimon added inline comments. Comment at: llvm/utils/lit/lit/discovery.py:60 +cfgpath = util.abs_path_preserve_drive(cfgpath) +target = config_map.get(cfgpath) if target: Found the problem - you have moved the os.path.normpat

[PATCH] D154130: [lit][clang] Avoid realpath on Windows due to MAX_PATH limitations

RKSimon added a comment. @MrTrillian This is failing for me with: C:\LLVM\ninja>ninja check-llvm-codegen-x86 [0/1/0/1] Running lit suite C:/LLVM/llvm-project/llvm/test/CodeGen/X86llvm-lit.py: C:\LLVM\llvm-project\llvm\utils\lit\lit\TestingConfig.py:151: fatal: unable to parse config file '

[PATCH] D156239: [X86] Support -march=arrowlake, arrowlake-s, lunarlake

RKSimon added inline comments. Comment at: llvm/lib/TargetParser/X86TargetParser.cpp:430 + { {"arrowlake-s"}, CK_ArrowlakeS, FEATURE_AVX2, FeaturesArrowlakeS, '\0', false }, + { {"arrowlake_s"}, CK_Lunarlake, FEATURE_AVX2, FeaturesArrowlakeS, 'p', true }, + // Lunarlake micr

[PATCH] D156239: [X86] Support -march=arrowlake, arrowlake-s, lunarlake

RKSimon added inline comments. Comment at: llvm/test/CodeGen/X86/cpus-intel.ll:37 +; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=arrowlake-s 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty +; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unk

[PATCH] D156239: [X86] Support -march=arrowlake, arrowlake-s, lunarlake

RKSimon added inline comments. Comment at: llvm/test/CodeGen/X86/cpus-intel.ll:37 +; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=arrowlake-s 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty +; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unk

[PATCH] D156239: [X86] Support -march=arrowlake, arrowlake-s, lunarlake

RKSimon added inline comments. Comment at: clang/test/Preprocessor/predefined-arch-macros.c:2570 // CHECK_SRF_M64: #define __AVXNECONVERT__ 1 +// CHECK_ARLS_M64: #define __AVXVNNIINT16__ 1 // CHECK_SRF_M64: #define __AVXVNNIINT8__ 1 Do the CHECK_SRF_M64-NOT tri

[PATCH] D155798: [X86] Support -march=graniterapids-d and update -march=graniterapids

RKSimon accepted this revision. RKSimon added a comment. LGTM - cheers Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D155798/new/ https://reviews.llvm.org/D155798 ___ cfe-commits mailing list cfe-commits@

[PATCH] D155798: [X86] Support -march=graniterapids-d and update -march=graniterapids

RKSimon added inline comments. Comment at: llvm/lib/Target/X86/X86.td:1082 FeaturePREFETCHI, - FeatureSHA512, FeatureAMXCOMPLEX];

[PATCH] D155798: [X86] Support -march=graniterapids-d and update -march=graniterapids

RKSimon added inline comments. Comment at: clang/test/Preprocessor/predefined-arch-macros.c:1925 // CHECK_GNR_M64: #define __AMX_BF16__ 1 -// CHECK_GNR_M64: #define __AMX_COMPLEX__ 1 +// CHECK_GNR_M64-NOT: #define __AMX_COMPLEX__ 1 +// CHECK_GNRD_M64: #define __AMX_COMPLEX__ 1 -

[PATCH] D155798: [X86] Support -march=graniterapids-d and update -march=graniterapids

RKSimon added inline comments. Comment at: clang/test/Preprocessor/predefined-arch-macros.c:1922 +// RUN: --target=x86_64 \ +// RUN: | FileCheck -match-full-lines %s -check-prefix=CHECK_GNRD_M64 // CHECK_GNR_M64: #define __AES__ 1 Maybe create a common CHE

[PATCH] D155147: [X86] Add SM3 instructions.

RKSimon added a comment. @pengfei Are you happy with the intrinsics doxygen descriptions? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D155147/new/ https://reviews.llvm.org/D155147 ___ cfe-commits mailin

[PATCH] D155148: [X86] Add SM4 instructions.

RKSimon added a comment. @pengfei Are you happy with the intrinsics doxygen descriptions? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D155148/new/ https://reviews.llvm.org/D155148 ___ cfe-commits mailin

[PATCH] D155148: [X86] Add SM4 instructions.

RKSimon added inline comments. Comment at: clang/lib/Headers/sm4intrin.h:26 +/// +/// This intrinsic corresponds to the \c VPDPBSSD instruction. +/// VPDPBSSD ? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D155148/

[PATCH] D155146: [X86] Add SHA512 instructions.

RKSimon accepted this revision. RKSimon added a comment. LGTM - but I'd prefer more complete 32-bit vs 64-bit test coverage (similar to the SM3/SM4 patches) if its possible. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D155146/new/ https://reviews

[PATCH] D155147: [X86] Add SM3 instructions.

RKSimon added inline comments. Comment at: clang/lib/Headers/sm3intrin.h:31 +/// +/// This intrinsic corresponds to the \c VPDPBSSD instruction. +/// VPDPBSSD ? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D155147/

[PATCH] D155145: [X86] Add AVX-VNNI-INT16 instructions.

RKSimon added inline comments. Comment at: clang/lib/Headers/avxvnniint16intrin.h:26 + +static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_dpwsud_epi32(__m128i __A, + __m128i __B, doxygen descriptio

[PATCH] D155146: Add SHA512 instructions.

RKSimon added inline comments. Comment at: clang/lib/Headers/sha512intrin.h:21 + +static __inline__ __m256i __DEFAULT_FN_ATTRS256 +_mm256_sha512msg1_epi64(__m256i __A, __m128i __B) { doxygen descriptions? Comment at: clang/test/CodeGen/X86/sha5

[PATCH] D155148: Add SM4 instructions.

RKSimon added inline comments. Comment at: clang/lib/Headers/sm4intrin.h:17 + +#define _mm_sm4key4_epi32(A, B) \ + (__m128i) __builtin_ia32_vsm4key4128((__v4su)A, (__v4su)B) doxygen descriptions?

[PATCH] D155147: Add SM3 instructions.

RKSimon added inline comments. Comment at: clang/lib/Headers/sm3intrin.h:21 + +static __inline__ __m128i __DEFAULT_FN_ATTRS128 _mm_sm3msg1_epi32(__m128i __A, + __m128i __B, Doxygen descriptions? =

[PATCH] D154493: [X86] Support some Intel CPUs for cpu_specific/dispatch feature

RKSimon accepted this revision. RKSimon added a comment. This revision is now accepted and ready to land. LGTM Comment at: llvm/lib/TargetParser/X86TargetParser.cpp:333 { {"pentium"}, CK_Pentium, ~0U, FeatureX87 | FeatureCMPXCHG8B, 'B', false }, { {"pentium-mmx"}, CK_Pent

[PATCH] D154493: [X86] Support some Intel CPUs for cpu_specific/dispatch feature

RKSimon added inline comments. Comment at: llvm/lib/TargetParser/X86TargetParser.cpp:333 { {"pentium"}, CK_Pentium, ~0U, FeatureX87 | FeatureCMPXCHG8B, 'B', false }, { {"pentium-mmx"}, CK_PentiumMMX, ~0U, FeaturesPentiumMMX, '\0', false }, { {"pentium_mmx"}, CK_PentiumMM

[PATCH] D151696: [X86] Remove CPU_SPECIFIC* MACROs and add getCPUDispatchMangling

RKSimon accepted this revision. RKSimon added a comment. LGTM - cheers Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D151696/new/ https://reviews.llvm.org/D151696 ___ cfe-commits mailing list cfe-commits@

[PATCH] D151696: [x86] Remove CPU_SPECIFIC* MACROs and add getCPUDispatchMangling

RKSimon added inline comments. Comment at: clang/test/CodeGen/attr-cpuspecific-cpus.c:40 ATTR(cpu_specific(knm)) void CPU(void){} +ATTR(cpu_specific(cascadelake)) void CPU(void){} +ATTR(cpu_specific(cooperlake)) void CPU(void){} FreddyYe wrote: > In this patch,

[PATCH] D151696: [x86] Remove CPU_SPECIFIC* MACROs and add getCPUDispatchMangling

RKSimon added inline comments. Comment at: llvm/lib/TargetParser/X86TargetParser.cpp:378 + { {"core_3rd_gen_avx"}, CK_IvyBridge, FEATURE_AVX, FeaturesIvyBridge, 'S', true }, + { {"core-avx-i"}, CK_IvyBridge, FEATURE_AVX, FeaturesIvyBridge, '\0', false }, // Haswell microar

[PATCH] D154209: [X86] Add missing features for ivybridge, sandybridge and knl in X86TargetParser.def.

RKSimon accepted this revision. RKSimon added a comment. This revision is now accepted and ready to land. LGTM - but it would be good to fix silvermont / westmere cases as well with suitable test coverage Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.or

[PATCH] D153006: [clang][dataflow] Perform deep copies in copy and move operations.

RKSimon added a comment. @mboehme https://lab.llvm.org/buildbot/#/builders/124 is still broken - please can you revert the patch series? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D153006/new/ https://reviews.llvm.org/D153006 __

[PATCH] D154209: [X86] Add missing features for ivybridge, sandybridge and knl in X86TargetParser.def.

RKSimon added inline comments. Comment at: llvm/include/llvm/TargetParser/X86TargetParser.def:262 CPU_SPECIFIC("core_i7_sse4_2", "nehalem", 'P', "+cmov,+mmx,+sse,+sse2,+sse3,+ssse3,+sse4.1,+sse4.2,+popcnt") CPU_SPECIFIC("core_aes_pclmulqdq", "westmere", 'Q', "+cmov,+mmx,+sse,

[PATCH] D151696: [x86] Remove CPU_SPECIFIC* MACROs and add getCPUDispatchMangling

RKSimon added inline comments. Comment at: clang/test/CodeGen/attr-cpuspecific.c:47 // LINUX: %[[FEAT_INIT:.+]] = load i32, ptr getelementptr inbounds ({ i32, i32, i32, [1 x i32] }, ptr @__cpu_model, i32 0, i32 3, i32 0), align 4 -// LINUX: %[[FEAT_JOIN:.+]] = and i32 %[[FEAT_I

[PATCH] D151696: [x86] Remove CPU_SPECIFIC* MACROs and add getCPUDispatchMangling

RKSimon added inline comments. Comment at: clang/test/CodeGen/attr-cpuspecific.c:47 // LINUX: %[[FEAT_INIT:.+]] = load i32, ptr getelementptr inbounds ({ i32, i32, i32, [1 x i32] }, ptr @__cpu_model, i32 0, i32 3, i32 0), align 4 -// LINUX: %[[FEAT_JOIN:.+]] = and i32 %[[FEAT_I

[PATCH] D151696: [x86] Remove CPU_SPECIFIC* MACROs and add getCPUDispatchMangling

RKSimon added a comment. In D151696#4458751 , @FreddyYe wrote: > In D151696#4458443 , @pengfei wrote: > >> It looks to me the failed unit tests might be related to this patch, please >> take a look. > > This is d

[PATCH] D151696: [x86] Remove CPU_SPECIFIC* MACROs and add getCPUDispatchMangling

RKSimon added a comment. Please can you cleanup the summary, as it isn't very easy to understand at the moment. Possibly split into a series of bullet points describing the changes? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D151696/new/ https:/

[PATCH] D151696: [x86] Remove CPU_SPECIFIC* MACROs and add getManglingForCPU

RKSimon added inline comments. Comment at: llvm/lib/TargetParser/X86TargetParser.cpp:110 FeatureBitset Features; + char Mangling; + bool OnlyForCPUDispatchSpecific; Do we have a documented list of the mangling values anywhere? The values below look too much

[PATCH] D151696: [x86] Remove CPU_SPECIFIC* MACROs and add getManglingForCPU

RKSimon added inline comments. Comment at: llvm/test/CodeGen/X86/cpus-intel.ll:8 ; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown -mcpu=pentium 2>&1 | FileCheck %s --check-prefix=CHECK-NO-ERROR --allow-empty -; RUN: llc < %s -o /dev/null -mtriple=i686-unknown-unknown

[PATCH] D151696: [x86] Remove CPU_SPECIFIC* MACROs and add getManglingForCPU

RKSimon added inline comments. Comment at: llvm/lib/TargetParser/X86TargetParser.cpp:15 #include "llvm/ADT/StringSwitch.h" +#include "llvm/ADT/StringExtras.h" #include (clang-format) - include order Comment at: llvm/lib/TargetParser/X86Targe

[PATCH] D150114: [Headers][doc] Add "add/sub/mul" intrinsic descriptions to avx2intrin.h

RKSimon added a comment. @pengfei Any objections? CHANGES SINCE LAST ACTION https://reviews.llvm.org/D150114/new/ https://reviews.llvm.org/D150114 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listin

[PATCH] D143287: [Clang][X86] Change X86 cast intrinsics to use __builtin_nondeterministic_value

RKSimon accepted this revision. RKSimon added a comment. LGTM - just simplify the shuffle masks (even if it break 80-col). Please keep an eye out for any regressions, I'm not certain we've shaken out every possible issue. Comment at: clang/lib/Headers/avx512fintrin.h:401 + r

[PATCH] D147901: [NFC][CLANG][API] Fix coverity remarks about large copies by values

RKSimon added inline comments. Comment at: clang/include/clang/ExtractAPI/API.h:138 APIRecord(RecordKind Kind, StringRef USR, StringRef Name, -PresumedLoc Location, AvailabilitySet Availabilities, +PresumedLoc Location, const AvailabilitySet &Availabili

[PATCH] D147481: [M68k] Add basic Clang supports for M68881/2

RKSimon added inline comments. Comment at: clang/test/Driver/m68k-macros.cpp:2 // Check macro definitions // RUN: %clang -target m68k-unknown-linux -m68000 -dM -E %s | FileCheck --check-prefix=CHECK-MX %s +// RUN: %clang -target m68k-unknown-linux -m68000 -mhard-float -dM -E %

[PATCH] D145715: Remove -lower-global-dtors-via-cxa-atexit flag

RKSimon added a comment. In D145715#4185070 , @thetruestblue wrote: > This looks good to me. I'm not sure the policy on what should be added to the > release notes. But this was a temporary flag only added to slowly deprecate > this, and was noted when

[PATCH] D145715: Remove -lower-global-dtors-via-cxa-atexit flag

RKSimon added a comment. It might be worth adding something to the release notes explaining that removal of the flag Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145715/new/ https://reviews.llvm.org/D145715 __

[PATCH] D144903: [X86] Drop single use check for freeze(undef) in LowerAVXCONCAT_VECTORS

RKSimon added a comment. In D144903#4169623 , @ManuelJBrito wrote: > It seems the build failure was caused by a known crash > https://github.com/llvm/llvm-project/issues/55263. I tried to find some > workaround but unsuccessfully. @pengfei Do you kno

[PATCH] D144903: [X86] Drop single use check for freeze(undef) in LowerAVXCONCAT_VECTORS

RKSimon added a comment. @ManuelJBrito Any luck with getting this committed? Your first attempt was reverted but was it just because of the bad Differential Revision tag? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D144903/new/ https://reviews.ll

[PATCH] D144903: [X86] Drop single use check for freeze(undef) in LowerAVXCONCAT_VECTORS

RKSimon accepted this revision. RKSimon added a comment. This revision is now accepted and ready to land. LGTM Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D144903/new/ https://reviews.llvm.org/D144903 _

[PATCH] D144903: [X86] Drop single use check for freeze(undef) in LowerAVXCONCAT_VECTORS

RKSimon added inline comments. Comment at: clang/test/CodeGen/X86/avx-cast-builtins.c:1 -// RUN: %clang_cc1 %s -O3 -flax-vector-conversions=none -ffreestanding %s -triple=x86_64-unknown-unknown -target-feature +avx -target-feature +avx512f -target-feature +avx512fp16 -S -o - |

[PATCH] D143287: [Clang][X86] Change X86 cast intrinsics to use __builtin_nondeterministic_value

RKSimon added a comment. In D143287#4150186 , @ManuelJBrito wrote: > There are some performance regressions with casts from 128 to 512. The > backend inserts vinsertf instructions. So that has to be fixed. > In D130339

[PATCH] D143287: [Clang][x86] Change x86 cast intrinsics to use __builtin_nondeterministic_value

RKSimon added inline comments. Comment at: clang/test/CodeGen/X86/avx-builtins.c:146 // CHECK-LABEL: test_mm256_castpd128_pd256 - // CHECK: shufflevector <2 x double> %{{.*}}, <2 x double> %{{.*}}, <4 x i32> + // CHECK: shufflevector <2 x double> %{{.*}}, <2 x double> %{{.

[PATCH] D143318: [Support] Move ItaniumManglingCanonicalizer and SymbolRemappingReader from Support to ProfileData

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG6c8fe9658236: [Support] Move ItaniumManglingCanonicalizer and SymbolRemappingReader from… (authored by RKSimon). Changed prior to commit: https://

[PATCH] D143318: [Support] Move ItaniumManglingCanonicalizer and SymbolRemappingReader from Support to ProfileData

RKSimon updated this revision to Diff 494925. RKSimon added a comment. fix the git diff lost renames Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D143318/new/ https://reviews.llvm.org/D143318 Files: clang/docs/tools/clang-formatted-files.txt l

[PATCH] D143318: [Support] Move ItaniumManglingCanonicalizer and SymbolRemappingReader from Support to ProfileData

RKSimon updated this revision to Diff 494917. RKSimon added a comment. Add bazel build fix for llvm-cxxmap Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D143318/new/ https://reviews.llvm.org/D143318 Files: clang/docs/tools/clang-formatted-files.t

[PATCH] D143318: [Support] Move ItaniumManglingCanonicalizer and SymbolRemappingReader from Support to ProfileData

RKSimon added a comment. In D143318#4104410 , @phosek wrote: > Another alternative would be to move these classes to the Demangle library. I'm happy to move this to LLVMDemangle instead if everyone agrees. Repository: rG LLVM Github Monorepo CHANGES

[PATCH] D143318: [Support] Move ItaniumManglingCanonicalizer and SymbolRemappingReader from Support to ProfileData

RKSimon added a comment. In D143318#4104408 , @MaskRay wrote: > `ItaniumManglingCanonicalizer.cpp` doesn't have many includes. I think it is > fine to remain in llvm/lib/Support ... The frontend cost might not be high - but ItaniumManglingCanonicalizer.

[PATCH] D143318: [Support] Move ItaniumManglingCanonicalizer and SymbolRemappingReader from Support to ProfileData

RKSimon created this revision. RKSimon added reviewers: rsmith, chapuni, MaskRay, kristina, erik.pilkington. Herald added subscribers: wenlei, hiraditya. Herald added a project: All. RKSimon requested review of this revision. Herald added projects: clang, LLVM. Herald added a subscriber: cfe-commit

[PATCH] D143287: [Clang][x86] Change x86 cast intrinsics to use __builtin_nondeterministic_value

RKSimon added a comment. What do we gain from using __builtin_nondeterministic_value instead of just setzero? https://godbolt.org/z/zrb6858Mr Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D143287/new/ https://reviews.llvm.org/D143287 _

[PATCH] D142477: [X86] Ensure the _mm_test_all_ones macro does not reuse argument (PR60006)

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGc9b28233599a: [X86] Ensure the _mm_test_all_ones macro does not reuse argument (PR60006) (authored by RKSimon). Repository: rG LLVM Github Monorep

[PATCH] D142477: [X86] Ensure the _mm_test_all_ones macro does not reuse argument (PR60006)

RKSimon created this revision. RKSimon added reviewers: pengfei, craig.topper. Herald added a project: All. RKSimon requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. The macro `_mm_test_all_ones(V)` was defined as `_mm_testc_si128((V), _mm_c

[PATCH] D78028: move shebangs from python2 to python3

RKSimon added a comment. @LocutusOfBorg - a grep of llvm-project *.py files is not finding many matches for python2 - I'm assuming this was handled (e.g. update_cc_test_checks.py was updated in D129590 ). There might still be a few missed cases (or references t

[PATCH] D141798: Drop the ZeroBehavior parameter from countLeadingZeros and the like (NFC)

RKSimon added a comment. @kazu Thanks for dealing with this! I'd like to build on this and create llvm variants of the C++20 countl_zero/countr_zero/countl_one/countr_one template functions similar to what I did for popcount in D132407 (and have MathExtras.h

[PATCH] D140992: clang: Add __builtin_elementwise_fma

RKSimon added inline comments. Comment at: clang/docs/LanguageExtensions.rst:634 the most negative integer remains the most negative integer + T __builtin_elementwise_fma(T x, T y, T z) fused multiply add.

[PATCH] D140547: Perform access checking to private members in simple requirement.

RKSimon added inline comments. Comment at: clang/docs/ReleaseNotes.rst:739 +- Correctly handle access-checks in requires expression. Fixes `GH53364 `_, + `GH53334 `_. C++2b Fe

[PATCH] D140547: Perform access checking to private members in simple requirement.

RKSimon added inline comments. Comment at: clang/docs/ReleaseNotes.rst:739 +- Correctly handle access-checks in requires expression. Fixes `GH53364 `_, + `GH53334 `_. C++2b Fe

[PATCH] D139073: [X86] AMD Zen 4 Initial enablement

RKSimon accepted this revision. RKSimon added a comment. cheers Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D139073/new/ https://reviews.llvm.org/D139073 ___ cfe-commits mailing list cfe-commits@lists.l

[PATCH] D139073: [X86] AMD Zen 4 Initial enablement

RKSimon accepted this revision. RKSimon added a comment. This revision is now accepted and ready to land. Thanks @GGanesh - LGTM with one minor - the orphan ZNVER4 checks from the llvm-mca tests need removing Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.ll

[PATCH] D139073: [X86] AMD Zen 4 Initial enablement

RKSimon added a comment. @GGanesh reverse-ping Comment at: llvm/lib/Target/X86/X86.td:1619 ProcessorFeatures.ZN3Tuning>; +def : ProcModel<"znver4", Znver3Model, ProcessorFeatures.ZN4Features, +ProcessorFeatures.ZN4Tuning>; RKSim

[PATCH] D139507: [Intrinsic] Add get.rounding as alias to flt.rounds and rename related DAG nodes

RKSimon added a comment. In D139507#3978449 , @sepavloff wrote: > Thank you for working on this! > > Is there any reason why we should keep the old intrinsic? I'm not clear on the general policy, but for x86 we tend to provide bc of old intrinsics/sigta

[PATCH] D139073: [X86] AMD Zen 4 Initial enablement

RKSimon added a comment. Thank you @gganesh I thought you'd forgotten about us :) Please add znver4 test coverage to: llvm/test/CodeGen/X86/cpus-amd.ll llvm/test/CodeGen/X86/slow-unaligned-mem.ll Comment at: llvm/lib/Target/X86/X86.td:1619 ProcessorFeatures.Z

[PATCH] D136919: [X86][RFC] Change mangle name of __bf16 from u6__bf16 to DF16b

RKSimon added reviewers: stuij, tra. RKSimon added a comment. Do you think this patch needs to be expanded to handle ARM/AArch64/NVPTX - all of which override getBFloat16Mangling in similar ways? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D136919

[PATCH] D136919: [X86][RFC] Change mangle name of __bf16 from u6__bf16 to DF16b

RKSimon added a comment. What are the rules on this? Do we just handle this as an ABI breaking change and document it in the release notes - or do we need to provide any auto-upgrade path (with a warning?)? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm

[PATCH] D137153: [WIP][X86] Support -march=sierraforest, grandridge, graniterapids.

RKSimon added inline comments. Comment at: llvm/lib/Target/X86/X86.td:1528 +ProcessorFeatures.TRMTuning>; +def : ProcModel<"grandridge", SLMModel, ProcessorFeatures.GRRFeatures, +ProcessorFeatures.TRMTuning>; SLMModel explicitly do

[PATCH] D137054: [X86][Driver] Remove stale FIXME. NFC

RKSimon accepted this revision. RKSimon added a comment. This revision is now accepted and ready to land. LGTM Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D137054/new/ https://reviews.llvm.org/D137054 _

[PATCH] D135930: [X86] Add AVX-NE-CONVERT instructions.

RKSimon added inline comments. Comment at: clang/lib/Headers/avx512vlbf16intrin.h:164 +#define _mm_cvtneps_pbh(A) \ + ((__m128bh)__builtin_ia32_vcvtneps2bf16128((__v4sf)(A))) Is there no way for __attribute__ to allow different attribute permutations? Also, c

[PATCH] D135951: [X86][1/2] SUPPORT RAO-INT

RKSimon accepted this revision. RKSimon added a comment. This revision is now accepted and ready to land. LGTM Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D135951/new/ https://reviews.llvm.org/D135951 _

[PATCH] D135930: [X86] Add AVX-NE-CONVERT instructions.

RKSimon added a comment. Possibly rename the x86-64-* test files to *-64 (and *-32 equivalent) so that the 32/64 bit files are closer together for tracking (and to help avoid bitrot). Comment at: clang/lib/Headers/immintrin.h:257 +/* FIXME: Change these When _Float16 type is

[PATCH] D135938: [X86] Add AVX-VNNI-INT8 instructions.

RKSimon added inline comments. Comment at: clang/include/clang/Basic/BuiltinsX86.def:2098 +TARGET_BUILTIN(__builtin_ia32_vpdpbssd128, "V4iV4iV4iV4i", "ncV:128:", "avxvnniint8") +TARGET_BUILTIN(__builtin_ia32_vpdpbssd256, "V8iV8iV8iV8i", "ncV:256:", "avxvnniint8")

[PATCH] D123064: [Clang][C++23] P2071 Named universal character escapes

RKSimon added inline comments. Comment at: llvm/utils/UnicodeData/UnicodeNameMappingGenerator.cpp:46 + auto SecondSemiPos = Line.find(';', FirstSemiPos + 1); + if (FirstSemiPos == std::string::npos) +continue; @cor3ntin Should this be SecondSemi

[PATCH] D120395: [X86] Prohibit arithmetic operations on type `__bfloat16`

RKSimon added a comment. Herald added a subscriber: StephenFan. @pengfei Do we still need this? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D120395/new/ https://reviews.llvm.org/D120395 ___ cfe-commits

[PATCH] D135937: [X86] Support -march=raptorlake, meteorlake

RKSimon added inline comments. Comment at: clang/test/Driver/x86-march.c:120 +// RUN: | FileCheck %s -check-prefix=meteorlake +// meteorlake: "-target-cpu" "meteorlake" // Move these after alderlake instead of the old atom cores? Repository: rG LLVM Github

[PATCH] D135938: [X86] Add AVX-VNNI-INT8 instructions.

RKSimon added a comment. In D135938#3867170 , @FreddyYe wrote: > I get your point of "close to each other" and updated. And I merged the > Disasm tests, while I didn't merge the MC tests because it is not so > convenient to do. See latest updated. > > D

[PATCH] D132329: [X86][RFC] Using `__bf16` for AVX512_BF16 intrinsics

RKSimon accepted this revision. RKSimon added a comment. LGTM - cheers Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D132329/new/ https://reviews.llvm.org/D132329 ___ cfe-commits mailing list cfe-commits

[PATCH] D132329: [X86][RFC] Using `__bf16` for AVX512_BF16 intrinsics

RKSimon added a comment. Add a short description to clang ReleaseNotes about the new bf/bh types? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D132329/new/ https://reviews.llvm.org/D132329 ___ cfe-commit

[PATCH] D135930: [X86] Add AVX-NE-CONVERT instructions.

RKSimon added a comment. merge att/intel test coverage files and rename the 32/64 bit files so that they are close together in the file lists Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D135930/new/ https://reviews.llvm.org/D135930 _

[PATCH] D135937: [X86] Support -march=meteorlake

RKSimon added a comment. Please can you update the summary now that raptorlake + meteorlake are in the same patch? Comment at: clang/test/Preprocessor/predefined-arch-macros.c:2233 +// RUN: %clang -march=raptorlake -m32 -E -dM %s -o - 2>&1 \ +// RUN: -target i386-unknown

[PATCH] D135938: [X86] Add AVX-VNNI-INT8 instructions.

RKSimon added a comment. Can you fix the MC + disasm test file names - drop att/intel and ensure you test both syntaxes for 32 and 64 bits. Ideally the 32/64 bit names should be close to each other in a file list (e.g. avx-vnni-int8-32.s + avx-vnni-int8-64.s ?) Repository: rG LLVM Github Mo

[PATCH] D135933: [X86] Add CMPCCXADD instructions.

RKSimon added inline comments. Comment at: llvm/test/MC/X86/x86-64-cmpccxadd-att.s:1 +// RUN: llvm-mc -triple x86_64 --show-encoding %s | FileCheck %s + Drop the -att.s and add intel test coverage? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTIO

[PATCH] D135930: [X86] Add AVX-NE-CONVERT instructions.

RKSimon added inline comments. Comment at: llvm/test/MC/X86/avx-ne-convert-att.s:1 +// RUN: llvm-mc -triple i686-unknown-unknown --show-encoding %s | FileCheck %s + merge the att + intel test files and use --check-prefixes to test both Repository: rG LLVM Gi

[PATCH] D135938: [X86] Add AVX-VNNI-INT8 instructions.

RKSimon added inline comments. Comment at: clang/docs/ReleaseNotes.rst:553 - Fix 32-bit ``__fastcall`` and ``__vectorcall`` ABI mismatch with MSVC. +- Support ISA of ``AVX-VNNI-INT8``. Please add a bullet list of the added intrinsics Comment

[PATCH] D135951: [X86] SUPPORT RAO-INT

RKSimon added inline comments. Comment at: llvm/lib/Target/X86/X86ISelLowering.h:801 +RXOR, +RAND, + very pedantic, but are these likely to get confused with ROR / RAND instructions? Would it be better to use a RAO_ prefix? Comment at:

[PATCH] D135966: [X86] Use unsigned int for return type of __get_cpuid_max.

RKSimon accepted this revision. RKSimon added a comment. This revision is now accepted and ready to land. LGTM Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D135966/new/ https://reviews.llvm.org/D135966 _

[PATCH] D135937: [X86] Support -march=meteorlake

RKSimon added a comment. Please add this to llvm-project\llvm\test\CodeGen\X86\cpus-intel.ll Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D135937/new/ https://reviews.llvm.org/D135937 ___ cfe-commits mai

  1   2   3   4   5   6   7   >