[PATCH] D69979: clang: Guess at some platform FTZ/DAZ default settings

arsenm created this revision. arsenm added reviewers: spatel, craig.topper, RKSimon, hfinkel, probinson. Herald added a subscriber: wdng. arsenm added a parent revision: D69978: Separately track input and output denormal mode. This is to avoid performance regressions when the default attribute be

[PATCH] D69878: Consoldiate internal denormal flushing controls

arsenm updated this revision to Diff 228348. arsenm added a comment. Fix name in documentation CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69878/new/ https://reviews.llvm.org/D69878 Files: clang/include/clang/Basic/CodeGenOptions.def clang/include/clang/Basic/CodeGenOptions.h c

[PATCH] D69598: Work on cleaning up denormal mode handling

arsenm marked an inline comment as done. arsenm added a comment. In D69598#1739655 , @andrew.w.kaylor wrote: > I'm unclear as to the expectations surrounding this option. I suppose this is > somewhat beyond the scope of the current changes, but I'm confu

[PATCH] D69979: clang: Guess at some platform FTZ/DAZ default settings

arsenm added a comment. In D69979#1738723 , @spatel wrote: > Also, I may have missed some discussions. Does this patch series replace the > proposal to add instruction-level FMF for denorms? > http://lists.llvm.org/pipermail/llvm-dev/2019-September/13518

[PATCH] D69979: clang: Guess at some platform FTZ/DAZ default settings

arsenm added a comment. In D69979#1738099 , @craig.topper wrote: > I checked Redhat 7.4 that's on the server I'm using for work. And I had a > coworker check his Ubuntu 18.04 system with this program. And both systems > printed 1f80 as the value of MXCS

[PATCH] D70366: Add new 'flatten' LLVM attribute to fix clang's 'flatten' function attribute

arsenm added inline comments. Comment at: llvm/docs/LangRef.rst:1428 can prove that the call/invoke cannot call a convergent function. +``flatten`` +This attribute is similar to ``alwaysinline``, but applies recursively to It's not obvious to me what the

[PATCH] D69979: clang: Guess at some platform FTZ/DAZ default settings

arsenm added a comment. In D69979#1746043 , @spatel wrote: > In D69979#1740294 , @arsenm wrote: > > > In D69979#1738099 , @craig.topper > > wrote: > > > > > I checked Redhat

[PATCH] D69978: Separately track input and output denormal mode

arsenm marked an inline comment as done. arsenm added inline comments. Comment at: llvm/docs/LangRef.rst:1822 ``"denorm-fp-mode"`` + This indicates the subnormal handling that may be assumed for the andrew.w.kaylor wrote: > I don't like the definition of thi

[PATCH] D70424: clang/AMDGPU: Fix default for frame-pointer attribute

arsenm created this revision. arsenm added reviewers: yaxunl, scott.linder, t-tye. Herald added subscribers: tpr, dstuttard, wdng, kzhuravl. Enabling optimization should allow frame pointer elimination. https://reviews.llvm.org/D70424 Files: clang/lib/Driver/ToolChains/Clang.cpp clang/test/

[PATCH] D69598: Work on cleaning up denormal mode handling

arsenm added a comment. In D69598#1742740 , @andrew.w.kaylor wrote: > Thanks. I understand your direction for denormal handling now, and I'm OK > with this patch apart from the remaining references to subnormal that Sanjay > mentioned. > > In D69598#173

[PATCH] D69598: Work on cleaning up denormal mode handling

arsenm closed this revision. arsenm added a comment. 7fe9435dc88050ee78eb1d4adec87610dce468f7 This does now need to be merged with the FPEnv.h header CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69598/new/ https://

[PATCH] D69878: Consoldiate internal denormal flushing controls

arsenm marked an inline comment as done. arsenm added inline comments. Comment at: clang/lib/CodeGen/CGCall.cpp:1775 -if (getLangOpts().OpenCL) - FuncAttrs.addAttribute("denorms-are-zero", - llvm::toStringRef(CodeGenOpts.FlushDenorm)); ---

[PATCH] D69493: Add -fconvergent-functions flag

arsenm closed this revision. arsenm added a comment. e531750c6cf9ab6ca987ffbfe100b1d766269eb5 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69493/new/ https://reviews.llvm.org/D69493 __

[PATCH] D59321: WIP: AMDGPU: Teach toolchain to link rocm device libs

arsenm marked an inline comment as done. arsenm added inline comments. Comment at: lib/Driver/ToolChains/AMDGPU.h:25 +/// TODO: Generalize to handle libclc. +class RocmInstallationDetector { +private: arsenm wrote: > yaxunl wrote: > > I don't think we should dete

[PATCH] D59321: WIP: AMDGPU: Teach toolchain to link rocm device libs

arsenm updated this revision to Diff 230197. arsenm added a comment. Rebase CHANGES SINCE LAST ACTION https://reviews.llvm.org/D59321/new/ https://reviews.llvm.org/D59321 Files: clang/include/clang/Basic/DiagnosticDriverKinds.td clang/include/clang/Driver/Options.td clang/lib/Driver/Dr

[PATCH] D70424: clang/AMDGPU: Fix default for frame-pointer attribute

arsenm added a comment. Ping CHANGES SINCE LAST ACTION https://reviews.llvm.org/D70424/new/ https://reviews.llvm.org/D70424 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D70424: clang/AMDGPU: Fix default for frame-pointer attribute

arsenm added a comment. In D70424#1761298 , @arsenm wrote: > In D70424#1758902 , @yaxunl wrote: > > > LGTM. But I am wondering how it affects -g. Do we need to keep frame > > pointer when -g is specified? Should we

[PATCH] D70424: clang/AMDGPU: Fix default for frame-pointer attribute

arsenm added a comment. In D70424#1758902 , @yaxunl wrote: > LGTM. But I am wondering how it affects -g. Do we need to keep frame pointer > when -g is specified? Should we add a test for -O3 -g? It’s independent for every other target. This matches the

[PATCH] D69878: Consoldiate internal denormal flushing controls

arsenm marked an inline comment as done. arsenm added inline comments. Comment at: clang/lib/CodeGen/CGCall.cpp:1775 -if (getLangOpts().OpenCL) - FuncAttrs.addAttribute("denorms-are-zero", - llvm::toStringRef(CodeGenOpts.FlushDenorm)); ---

[PATCH] D69979: clang: Guess at some platform FTZ/DAZ default settings

arsenm updated this revision to Diff 231726. arsenm added a comment. DAZ/FTZ seem to be set in crtfastmath.o, so try to reproduce the logic for linking that CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69979/new/ https://reviews.llvm.org/D69979 Files: clang/include/clang/Driver/Too

[PATCH] D69979: clang: Guess at some platform FTZ/DAZ default settings

arsenm updated this revision to Diff 232088. arsenm added a comment. Rename functions CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69979/new/ https://reviews.llvm.org/D69979 Files: clang/include/clang/Driver/ToolChain.h clang/lib/Driver/ToolChain.cpp clang/lib/Driver/ToolChains/

[PATCH] D70424: clang/AMDGPU: Fix default for frame-pointer attribute

arsenm accepted this revision. arsenm added a comment. This revision is now accepted and ready to land. 2cc11941a2e88236e0b4842229454ae6d85142cd CHANGES SINCE LAST ACTION https://reviews.llvm.org/D70424/new/ https://review

[PATCH] D69878: Consoldiate internal denormal flushing controls

arsenm updated this revision to Diff 233059. arsenm added a comment. Reword langref, fix name in langref CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69878/new/ https://reviews.llvm.org/D69878 Files: clang/include/clang/Basic/CodeGenOptions.def clang/include/clang/Basic/CodeGenOpt

[PATCH] D69979: clang: Guess at some platform FTZ/DAZ default settings

arsenm added a comment. ping CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69979/new/ https://reviews.llvm.org/D69979 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D76472: AMDGPU: Emit llvm.fshr for __builtin_amdgcn_alignbit

arsenm created this revision. arsenm added reviewers: yaxunl, rampitec. Herald added subscribers: kerbowa, t-tye, tpr, dstuttard, nhaehnle, wdng, jvesely, kzhuravl. These are equivalent. The generic rotate builtins do not directly map to the fshr intrinsic. https://reviews.llvm.org/D76472 File

[PATCH] D69978: Separately track input and output denormal mode

arsenm closed this revision. arsenm added a comment. a3c814d23497bc71b8ed53c35f773366aff02922 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69978/new/ https://reviews.llvm.org/D69978 __

[PATCH] D69979: clang: Guess at some platform FTZ/DAZ default settings

arsenm updated this revision to Diff 243553. arsenm added a comment. Rebase and fix check prefix name CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69979/new/ https://reviews.llvm.org/D69979 Files: clang/include/clang/Driver/ToolChain.h clang/lib/Driver/ToolChain.cpp clang/lib/Dr

[PATCH] D69979: clang: Guess at some platform FTZ/DAZ default settings

arsenm closed this revision. arsenm added a comment. fa7cd549d604bfd8f9dce5d649a19720cbc39cca CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69979/new/ https://reviews.llvm.org/D69979 __

[PATCH] D74500: clang: Treat ieee mode as the default for denormal-fp-math

arsenm created this revision. arsenm added reviewers: spatel, cameron.mcinally, andrew.w.kaylor. Herald added subscribers: llvm-commits, kerbowa, dexonsmith, nhaehnle, wdng, jvesely. Herald added a project: LLVM. The IR hasn't switched the default yet, so explicitly add the ieee attributes. htt

[PATCH] D69878: Consoldiate internal denormal flushing controls

arsenm updated this revision to Diff 238670. arsenm added a comment. Forgot clang parts CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69878/new/ https://reviews.llvm.org/D69878 Files: clang/include/clang/Basic/CodeGenOptions.def clang/include/clang/Basic/CodeGenOptions.h clang/in

[PATCH] D69878: Consoldiate internal denormal flushing controls

arsenm marked an inline comment as done. arsenm added inline comments. Comment at: clang/lib/Driver/ToolChains/Clang.cpp:2757 // subsequent options conflict then emit warning diagnostic. + // TODO: How should this interact with DenormalFP32Math? if (HonorINFs &

[PATCH] D69878: Consoldiate internal denormal flushing controls

arsenm closed this revision. arsenm added a comment. a4451d88ee456304c26d552749aea6a7f5154bde CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69878/new/ https://reviews.llvm.org/D69878 __

[PATCH] D69978: Separately track input and output denormal mode

arsenm updated this revision to Diff 239196. arsenm added a comment. Herald added a subscriber: kerbowa. Rebase CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69978/new/ https://reviews.llvm.org/D69978 Files: clang/include/clang/Basic/CodeGenOptions.h clang/include/clang/Driver/Tool

[PATCH] D69978: Separately track input and output denormal mode

arsenm marked 3 inline comments as done. arsenm added inline comments. Comment at: llvm/docs/LangRef.rst:1829 + operations. The second indicates the handling of denormal inputs to + floating point instructions. + andrew.w.kaylor wrote: > Based on the changes

[PATCH] D69978: Separately track input and output denormal mode

arsenm updated this revision to Diff 241714. arsenm marked an inline comment as done. arsenm added a comment. Tweak langref CHANGES SINCE LAST ACTION https://reviews.llvm.org/D69978/new/ https://reviews.llvm.org/D69978 Files: clang/include/clang/Basic/CodeGenOptions.h clang/include/clang

[PATCH] D44533: [AMDGPU] Fix codegen for inline assembly

arsenm accepted this revision. arsenm added a comment. This revision is now accepted and ready to land. LGTM https://reviews.llvm.org/D44533 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-c

[PATCH] D39784: OpenCL: Assume inline asm is convergent

arsenm created this revision. Herald added subscribers: yaxunl, wdng. Already done for CUDA. https://reviews.llvm.org/D39784 Files: lib/CodeGen/CGStmt.cpp test/CodeGenOpenCL/convergent.cl Index: test/CodeGenOpenCL/convergent.cl =

[PATCH] D39784: OpenCL: Assume inline asm is convergent

arsenm closed this revision. arsenm added a comment. r318098 https://reviews.llvm.org/D39784 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D40062: Update tests for llvm.invariant.group.barrier becoming mangled

arsenm accepted this revision. arsenm added a comment. This revision is now accepted and ready to land. LGTM https://reviews.llvm.org/D40062 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-c

[PATCH] D145150: clang: Emit nofpclass(nan inf) for -ffinite-math-only

arsenm updated this revision to Diff 505253. arsenm marked an inline comment as done. arsenm added a comment. Drop unnecessary part CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145150/new/ https://reviews.llvm.org/D145150 Files: clang/lib/CodeGen/CGCall.cpp clang/test/CodeGen/comp

[PATCH] D142907: LangRef: Add "dynamic" option to "denormal-fp-math"

arsenm updated this revision to Diff 505274. arsenm marked an inline comment as done. arsenm added a comment. Update doxygen comment CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142907/new/ https://reviews.llvm.org/D142907 Files: clang/lib/CodeGen/CGCall.cpp clang/lib/CodeGen/Code

[PATCH] D145150: clang: Emit nofpclass(nan inf) for -ffinite-math-only

arsenm closed this revision. arsenm added a comment. dd8181055494d8f973dfaaba282bb1ae6eab4d40 Comment at: clang/lib/CodeGen/CGCall.cpp:3052 +AI->addAttrs( +llvm::AttrBuilder(getLLVMContex

[PATCH] D145151: clang: Handle MatrixType in hasFloatingRepresentation

arsenm closed this revision. arsenm added a comment. ffe12e765cb9ba77dc30ccec89fb163234b98541 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145151/new/ https://reviews.llvm.org/D145151 __

[PATCH] D142934: clang: Use ptrmask for pointer alignment

arsenm closed this revision. arsenm added a comment. 8e009348e8a2e9c4577538eba6ca5c6cb286776f CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142934/new/ https://reviews.llvm.org/D142934 __

[PATCH] D142907: LangRef: Add "dynamic" option to "denormal-fp-math"

arsenm added a comment. ping CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142907/new/ https://reviews.llvm.org/D142907 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D144590: [Clang][OpenMP] Fix shared memory allocation on AMDGPU

arsenm added a comment. In D144590#4220648 , @ZwFink wrote: > I wanted to follow up on this. Is anything additional needed from me? This needs a test (I would hope there is already one that needs updating), and also should switch to using PoisonValue in

[PATCH] D140992: clang: Add __builtin_elementwise_fma

arsenm closed this revision. arsenm added a comment. 8709bcacfb3a06847b47bb6b47e8556db43f3a43 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D140992/new/ https://reviews.llvm.org/D140992 __

[PATCH] D144802: clang: Add __builtin_elementwise_round

arsenm created this revision. arsenm added reviewers: aaron.ballman, RKSimon, fhahn, junaire, bob80905, python3kgae, erichkeane. Herald added a subscriber: StephenFan. Herald added a project: All. arsenm requested review of this revision. Herald added a subscriber: wdng. This would only differ fr

[PATCH] D145150: clang: Emit nofpclass(nan inf) for -ffinite-math-only

arsenm created this revision. arsenm added reviewers: efriedma, nikic, jcranmer-intel, kpn, andrew.w.kaylor, cameron.mcinally, sepavloff. Herald added subscribers: StephenFan, jdoerfert, hiraditya. Herald added a project: All. arsenm requested review of this revision. Herald added subscribers: llv

[PATCH] D145151: clang: Handle MatrixType in hasFloatingRepresentation

arsenm created this revision. arsenm added a reviewer: fhahn. Herald added a subscriber: StephenFan. Herald added a project: All. arsenm requested review of this revision. Herald added a subscriber: wdng. Allows applying nofpclass to matrix arguments. https://reviews.llvm.org/D145151 Files: c

[PATCH] D145150: clang: Emit nofpclass(nan inf) for -ffinite-math-only

arsenm updated this revision to Diff 502015. arsenm added a comment. Drop todo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145150/new/ https://reviews.llvm.org/D145150 Files: clang/lib/CodeGen/CGCall.cpp clang/test/CodeGen/complex-math.c clang/test/CodeGen/fp-function-attrs.cpp

[PATCH] D142907: LangRef: Add "dynamic" option to "denormal-fp-math"

arsenm added a comment. ping CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142907/new/ https://reviews.llvm.org/D142907 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D145343: [AMDGPU] Emit predefined macro `__AMDGCN_CUMODE_OPTION`

arsenm added a comment. I think exposing whether or not the flag was used is weird/broken, as is including _OPTION in the name. Should just define to whether it's enabled or not CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145343/new/ https://reviews.llvm.org/D145343

[PATCH] D145343: [AMDGPU] Emit predefined macro `__AMDGCN_CUMODE_OPTION`

arsenm requested changes to this revision. arsenm added a comment. This revision now requires changes to proceed. We can’t let specific flag usage leak into the semantics. Cu mode is on or off. If someone really cares about supporting older compilers they could always define their own macro CH

[PATCH] D145941: [Clang] Always use -zdefs when linking AMDGPU images

arsenm added a comment. Release note change should be here Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D145941/new/ https://reviews.llvm.org/D145941 ___ cfe-commits mailing list cfe-commits@lists.llvm.o

[PATCH] D146023: [AMDGPU] Remove Code Object V2

arsenm added a comment. Is clover still relying on the cov2 support? Comment at: clang/include/clang/Basic/TargetOptions.h:85 COV_None, -COV_2 = 200, COV_3 = 300, I wouldn't remove the enum field, just add a comment that emission is unsupported o

[PATCH] D145767: [Verifier][NFC] Refactor check for associated metadata to allow multiple operands on AIX

arsenm added a comment. Missing verifier test in llvm/test/Verifier (e.g., the ones I added in 87f2e9448e82bbed4ac59bb61bea03256aa5f4de ) Comment at: clang/test/CodeGen/PowerPC/aix-init-ref-null.cpp:2 +// RU

[PATCH] D153226: OpenMP: Don't include stdbool.h in builtin headers

arsenm created this revision. arsenm added reviewers: ronlieb, jdoerfert, doru1004. Herald added subscribers: sunshaoce, guansong, yaxunl. Herald added a project: All. arsenm requested review of this revision. Herald added subscribers: jplehr, sstefan1, wdng. Pre-C99 didn't include bool, and C99 a

[PATCH] D144802: clang: Add __builtin_elementwise_round

arsenm updated this revision to Diff 532481. arsenm marked 2 inline comments as done. arsenm added a comment. Fix description and add release notes CHANGES SINCE LAST ACTION https://reviews.llvm.org/D144802/new/ https://reviews.llvm.org/D144802 Files: clang/docs/LanguageExtensions.rst cl

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm created this revision. arsenm added reviewers: aaron.ballman, RKSimon, fhahn, junaire, bob80905, python3kgae, erichkeane, sepavloff, kpn, andrew.w.kaylor. Herald added a subscriber: StephenFan. Herald added a project: All. arsenm requested review of this revision. Herald added a subscriber:

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm updated this revision to Diff 532590. arsenm marked 3 inline comments as done. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D153233/new/ https://reviews.llvm.org/D153233 Files: clang/docs/LanguageExtensions.rst clang/docs/ReleaseNotes.rst clang/include/clang/Basic/Builtins.d

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm added inline comments. Comment at: clang/test/CodeGen/strictfp-elementwise-bulitins.cpp:13-14 +// CHECK-NEXT:[[ADD:%.*]] = tail call <2 x float> @llvm.experimental.constrained.fadd.v2f32(<2 x float> [[TMP0]], <2 x float> [[TMP1]], metadata !"round.dynamic", metadata

[PATCH] D85471: Make clang HIP headers compatible with C++98

arsenm added a comment. Herald added a project: All. I don't understand why this is necessary. What does cmake have to do with it? Is HIP just not defaulting to the correct default C++ standard? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D85471/n

[PATCH] D85471: Make clang HIP headers compatible with C++98

arsenm added a comment. Can you either 1. Add test coverage 2. Delete this entirely Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D85471/new/ https://reviews.llvm.org/D85471 ___ cfe-commits mailing list

[PATCH] D138397: HIP: Directly call ldexp builtins

arsenm added a comment. ping CHANGES SINCE LAST ACTION https://reviews.llvm.org/D138397/new/ https://reviews.llvm.org/D138397 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D138504: clang/HIP: Remove __llvm_amdgcn_* wrapper hacks

arsenm closed this revision. arsenm added a comment. f407a7399575a6821940973c54754d42e72dd9ce CHANGES SINCE LAST ACTION https://reviews.llvm.org/D138504/new/ https://reviews.llvm.org/D138504 ___ cfe-commits mailing list cfe-commits@lists.llvm.org h

[PATCH] D144802: clang: Add __builtin_elementwise_round

arsenm added inline comments. Comment at: clang/docs/LanguageExtensions.rst:647 + T __builtin_elementwise_round(T x) round x to the nearest integer value in floating point format, floating point types + rounding halfway cases to

[PATCH] D144802: clang: Add __builtin_elementwise_round

arsenm closed this revision. arsenm added a comment. 2a488b4443a9c0aa0f368aed901676508ced202f CHANGES SINCE LAST ACTION https://reviews.llvm.org/D144802/new/ https://reviews.llvm.org/D144802 __

[PATCH] D147732: [AMDGPU] Add type mangling for {read, write, readfirst, perm}lane intrinsics

arsenm added a comment. I think this may not hard break mesa. I believe mesa bypasses the intrinsic creation API, and just declares the string name of the intrinsic. The type name mangling suffix is technically irrelevant, and as long as you use a consistent type with a consistent suffix things

[PATCH] D147732: [AMDGPU] Add type mangling for {read, write, readfirst, perm}lane intrinsics

arsenm added inline comments. Comment at: llvm/lib/Target/AMDGPU/AMDGPULateCodeGenPrepare.cpp:209 +bool is16Bit = +(EltType->isIntegerTy() && EltType->getIntegerBitWidth() == 16) || +(EltType->isHalfTy()); isIntegerTy(16). Also, just check the

[PATCH] D147732: [AMDGPU] Add type mangling for {read, write, readfirst, perm}lane intrinsics

arsenm added inline comments. Comment at: llvm/test/CodeGen/AMDGPU/llvm.amdgcn.readlane.ll:5 +declare i16 @llvm.amdgcn.readlane.i16(i16, i32) #0 +declare half @llvm.amdgcn.readlane.f16(half, i32) #0 +declare float @llvm.amdgcn.readlane.f32(float, i32) #0 arsenm w

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm updated this revision to Diff 533007. arsenm marked an inline comment as done. arsenm added a comment. Keep fixing documentation CHANGES SINCE LAST ACTION https://reviews.llvm.org/D153233/new/ https://reviews.llvm.org/D153233 Files: clang/docs/LanguageExtensions.rst clang/docs/Rel

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm added inline comments. Comment at: clang/docs/LanguageExtensions.rst:656 + T __builtin_elementwise_nearbyint(T x) round x to the nearest integer value in floating point format, floating point types + rounding accordin

[PATCH] D138397: HIP: Directly call ldexp builtins

arsenm closed this revision. arsenm added a comment. c448e1dabeeb066c41d893fc8e7a3d8bde2056c8 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D138397/new/ https://reviews.llvm.org/D138397 __

[PATCH] D142823: Intrinsics: Allow tablegen to mark parameters with dereferenceable

arsenm closed this revision. arsenm added a comment. 1717c18664d5880f78db85eb0075a2c1379df2d9 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D142823/new/ https://reviews.llvm.org/D142823 __

[PATCH] D149716: clang: Use new frexp intrinsic for builtins and add f16 version

arsenm added a comment. ping CHANGES SINCE LAST ACTION https://reviews.llvm.org/D149716/new/ https://reviews.llvm.org/D149716 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D152858: OpenMP: Use generated checks and pragma declare target

arsenm added a comment. ping CHANGES SINCE LAST ACTION https://reviews.llvm.org/D152858/new/ https://reviews.llvm.org/D152858 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D152857: OpenMP: Don't use target regions in library function test

arsenm added a comment. ping CHANGES SINCE LAST ACTION https://reviews.llvm.org/D152857/new/ https://reviews.llvm.org/D152857 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm added inline comments. Comment at: clang/lib/Sema/SemaChecking.cpp:18583 << Callee << CallerTCB; } } arsenm wrote: > bob80905 wrote: > > I don't believe you intended to remove all this code in your latest update, > > did you? > I didn't

[PATCH] D139629: clang: Stop emitting "strictfp"

arsenm added a comment. ping CHANGES SINCE LAST ACTION https://reviews.llvm.org/D139629/new/ https://reviews.llvm.org/D139629 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D112932: Use llvm.is_fpclass to implement FP classification functions

arsenm accepted this revision. arsenm added a comment. This revision is now accepted and ready to land. LGTM Comment at: clang/test/CodeGen/isfpclass.c:2 +// RUN: %clang_cc1 -triple x86_64-linux-gnu -S -O1 -emit-llvm %s -o - | FileCheck %s + +inline __attribute__((always_inlin

[PATCH] D112932: Use llvm.is_fpclass to implement FP classification functions

arsenm added a comment. Description now needs updating since this diff no longer adds all the split out pieces Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D112932/new/ https://reviews.llvm.org/D112932

[PATCH] D140294: clang: Replace implementation of __builtin_isnormal

arsenm abandoned this revision. arsenm added a comment. Herald added a subscriber: StephenFan. Obsoleted by D112932 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D140294/new/ https://reviews.llvm.org/D140294

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm added inline comments. Comment at: clang/test/CodeGen/strictfp-elementwise-bulitins.cpp:190 +// CHECK-NEXT: entry: +// CHECK-NEXT:[[ELT_TRUNC:%.*]] = tail call <4 x float> @llvm.canonicalize.v4f32(<4 x float> [[A]]) #[[ATTR4]] +// CHECK-NEXT:ret <4 x float> [[ELT

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm updated this revision to Diff 533779. arsenm added a comment. Rebase on fix for wrong debug name CHANGES SINCE LAST ACTION https://reviews.llvm.org/D153233/new/ https://reviews.llvm.org/D153233 Files: clang/docs/LanguageExtensions.rst clang/docs/ReleaseNotes.rst clang/include/cl

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm updated this revision to Diff 533789. arsenm added a comment. SemaCXX test CHANGES SINCE LAST ACTION https://reviews.llvm.org/D153233/new/ https://reviews.llvm.org/D153233 Files: clang/docs/LanguageExtensions.rst clang/docs/ReleaseNotes.rst clang/include/clang/Basic/Builtins.def

[PATCH] D149716: clang: Use new frexp intrinsic for builtins and add f16 version

arsenm added a comment. In D149716#4445411 , @yaxunl wrote: > should we have tests for llvm.frexp.* for LLVM codegen? Those are added in the parent that adds the intrinsic, this is just the clang side CHANGES SINCE LAST ACTION https://reviews.llvm.o

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm added a comment. In D153233#4443000 , @bob80905 wrote: > Should there be tests added for usage of scalable vector types for RISC-V / > AArch64? > I typically have added such tests in the past, as shown here: > https://reviews.llvm.org/D135011 Te

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm updated this revision to Diff 534111. arsenm added a comment. Add fixme to test CHANGES SINCE LAST ACTION https://reviews.llvm.org/D153233/new/ https://reviews.llvm.org/D153233 Files: clang/docs/LanguageExtensions.rst clang/docs/ReleaseNotes.rst clang/include/clang/Basic/Builtin

[PATCH] D153233: clang: Add __builtin_elementwise_rint and nearbyint

arsenm closed this revision. arsenm added a comment. 9d84f8dc948b3188fdec9a1a080eb6d845c2082d CHANGES SINCE LAST ACTION https://reviews.llvm.org/D153233/new/ https://reviews.llvm.org/D153233 __

[PATCH] D151587: [clang][ConstantEmitter] have tryEmitPrivate[ForVarInit] try ConstExprEmitter fast-path first

arsenm added inline comments. Comment at: clang/test/CodeGenOpenCL/amdgpu-nullptr.cl:60 -// CHECK: @fold_priv ={{.*}} local_unnamed_addr addrspace(1) global ptr addrspace(5) addrspacecast (ptr null to ptr addrspace(5)), align 4 +// CHECK: @fold_priv ={{.*}} local_unnamed_addr

[PATCH] D152829: clang: Add start of header test for __clang_hip_libdevice_declares

arsenm added a comment. ping CHANGES SINCE LAST ACTION https://reviews.llvm.org/D152829/new/ https://reviews.llvm.org/D152829 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D152858: OpenMP: Use generated checks and pragma declare target

arsenm closed this revision. arsenm added a comment. 53d28b2a71063b974ec058ced614d1c95d5584fe CHANGES SINCE LAST ACTION https://reviews.llvm.org/D152858/new/ https://reviews.llvm.org/D152858 __

[PATCH] D153226: OpenMP: Don't include stdbool.h in builtin headers

arsenm closed this revision. arsenm added a comment. dee1f5b32c3a6a5694c3bb2fbf68d162447a5970 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D153226/new/ https://reviews.llvm.org/D153226 __

[PATCH] D152850: OpenMP: Add missing test coverage for nothrow new/delete

arsenm closed this revision. arsenm added a comment. 51bb9b12f87ea315f750490c39cd993507202861 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D152850/new/ https://reviews.llvm.org/D152850 __

[PATCH] D152851: OpenMP: Fix nothrow new/delete for amdgpu

arsenm closed this revision. arsenm added a comment. 0e17174d40b2d25a755d2ce9acf7c905f254746d CHANGES SINCE LAST ACTION https://reviews.llvm.org/D152851/new/ https://reviews.llvm.org/D152851 __

[PATCH] D153883: [Clang][OpenMP] Enable use of __kmpc_alloc_shared for VLAs defined in AMD GPU offloaded regions

arsenm added inline comments. Comment at: clang/lib/CodeGen/CGDecl.cpp:1603 +// deallocation call of __kmpc_free_shared() is emitted later. +if (getLangOpts().OpenMP && getTarget().getTriple().isAMDGCN()) { + // Emit call to __kmpc_alloc_shared() instead of the alloc

[PATCH] D153883: [Clang][OpenMP] Enable use of __kmpc_alloc_shared for VLAs defined in AMD GPU offloaded regions

arsenm added inline comments. Comment at: clang/lib/CodeGen/CGDecl.cpp:1603 +// deallocation call of __kmpc_free_shared() is emitted later. +if (getLangOpts().OpenMP && getTarget().getTriple().isAMDGCN()) { + // Emit call to __kmpc_alloc_shared() instead of the alloc

[PATCH] D153883: [Clang][OpenMP] Enable use of __kmpc_alloc_shared for VLAs defined in AMD GPU offloaded regions

arsenm added a comment. In D153883#4456342 , @tianshilei1992 wrote: > I think it's better to just limit it to AMDGPU for now. > BTW, it might be worth to check if heap-to-stack will push it back to stack. If you're really going to go for backend workaro

[PATCH] D150860: [OpenMP] Change clang emitTargetDataCalls to use OMPIRBuilder

arsenm added inline comments. Comment at: clang/lib/CodeGen/CGOpenMPRuntime.cpp:10395-10399 +if (CGF.ConstantFoldsToSimpleInteger(IfCond, CondConstant)) { + IfCondVal = CGF.Builder.getInt1(CondConstant); +} else { + IfCondVal = CGF.EvaluateExprAsBool(IfCond); +

<    5   6   7   8   9   10   11   12   13   >