[PATCH] D64592: [OpenMP] Fix unified memory implementation for multiple compilation units

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/CodeGen/CGOpenMPRuntime.cpp:2610 + auto *GV = cast(Ptr); + GV->setLinkage(llvm::GlobalValue::WeakAnyLinkage); + gtbercea wrote: > ABataev wrote: > > Better to fix

[PATCH] D64926: [OpenMP] Fx link clause

gtbercea abandoned this revision. gtbercea added a comment. Not needed as a separate patch. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D64926/new/ https://reviews.llvm.org/D64926 ___ cfe-commits mailing list cfe-co

[PATCH] D64592: [OpenMP] Fix target link implementation

gtbercea updated this revision to Diff 210645. gtbercea added a comment. - Address comments. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D64592/new/ https://reviews.llvm.org/D64592 Files: lib/CodeGen/CGOpenMPRuntime.cpp test/OpenMP/declare_target_codegen.cp

[PATCH] D64592: [OpenMP] Fix declare target link implementation

gtbercea updated this revision to Diff 210650. gtbercea added a comment. - Fix tests. Address comments. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D64592/new/ https://reviews.llvm.org/D64592 Files: lib/CodeGen/CGOpenMPRuntime.cpp test/OpenMP/declare_target

[PATCH] D64592: [OpenMP] Fix declare target link implementation

gtbercea updated this revision to Diff 210652. gtbercea added a comment. - Merge conditionals. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D64592/new/ https://reviews.llvm.org/D64592 Files: lib/CodeGen/CGOpenMPRuntime.cpp test/OpenMP/declare_target_codegen.

[PATCH] D64592: [OpenMP] Fix declare target link implementation

gtbercea updated this revision to Diff 210673. gtbercea added a comment. - Add static variable. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D64592/new/ https://reviews.llvm.org/D64592 Files: lib/CodeGen/CGOpenMPRuntime.cpp test/OpenMP/declare_target_codegen

[PATCH] D64592: [OpenMP] Fix declare target link implementation

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: test/OpenMP/declare_target_link_codegen.cpp:22-23 +// HOST-DAG: @c_decl_tgt_ref_ptr = weak global i32* @c +// HOST-DAG: @_{{.*}}d = internal global i32 2 +// HOST-DAG: @_{{.*}}d_{{.*}}_decl_tgt

[PATCH] D64592: [OpenMP] Fix declare target link implementation

gtbercea updated this revision to Diff 210854. gtbercea added a comment. - Address comments. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D64592/new/ https://reviews.llvm.org/D64592 Files: lib/CodeGen/CGOpenMPRuntime.cpp test/OpenMP/declare_target_codegen.cp

[PATCH] D64592: [OpenMP] Fix declare target link implementation

gtbercea updated this revision to Diff 210856. gtbercea added a comment. - Address comments. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D64592/new/ https://reviews.llvm.org/D64592 Files: lib/CodeGen/CGOpenMPRuntime.cpp test/OpenMP/declare_target_codegen.cp

[PATCH] D65341: [OpenMP] Add support for close map modifier in Clang

gtbercea created this revision. gtbercea added reviewers: ABataev, caomhin. Herald added subscribers: cfe-commits, guansong. Herald added a reviewer: jdoerfert. Herald added a project: clang. This patch adds support for the close map modifier in Clang. This ensures that the new map type is marked

[PATCH] D65341: [OpenMP] Add support for close map modifier in Clang

gtbercea marked 2 inline comments as done. gtbercea added inline comments. Comment at: lib/CodeGen/CGOpenMPRuntime.cpp:7695 Flags &= ~(OMP_MAP_TO | OMP_MAP_FROM | OMP_MAP_ALWAYS | - OMP_MAP_DELETE); + OMP_MAP_DELETE |

[PATCH] D65341: [OpenMP] Add support for close map modifier in Clang

gtbercea updated this revision to Diff 212011. gtbercea added a comment. - Improve test. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65341/new/ https://reviews.llvm.org/D65341 Files: lib/CodeGen/CGOpenMPRuntime.cpp test/OpenMP/target_data_codegen.cpp Ind

[PATCH] D65341: [OpenMP] Add support for close map modifier in Clang

gtbercea added a comment. In D65341#1604440 , @ABataev wrote: > In D65341#1604411 , @Hahnfeld wrote: > > > There's already D55892 with a better set > > of tests, including `target

[PATCH] D65341: [OpenMP] Add support for close map modifier in Clang

gtbercea added a comment. In D65341#1604562 , @kkwli0 wrote: > In D65341#1604561 , @gtbercea wrote: > > > In D65341#1604440 , @ABataev wrote: > > > > > In D65341#1604411

[PATCH] D65341: [OpenMP] Add support for close map modifier in Clang

gtbercea updated this revision to Diff 212821. gtbercea added a comment. - Improve test. - Add tests from previous patch. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65341/new/ https://reviews.llvm.org/D65341 Files: lib/CodeGen/CGOpenMPRuntime.cpp test/Ope

[PATCH] D65341: [OpenMP] Add support for close map modifier in Clang

gtbercea added a comment. @Hahnfeld I have merged the tests from the previous patch as per Alexey's suggestion - with minor changes to make them pass. Let me know if this now addresses your previous comments. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65341/n

[PATCH] D65341: [OpenMP] Add support for close map modifier in Clang

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/CodeGen/CGOpenMPRuntime.cpp:7695 Flags &= ~(OMP_MAP_TO | OMP_MAP_FROM | OMP_MAP_ALWAYS | - OMP_MAP_DELETE); + OMP_MAP_DELETE |

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea created this revision. gtbercea added reviewers: jdoerfert, ABataev, hfinkel, caomhin. Herald added subscribers: cfe-commits, guansong, mgorny. Herald added a project: clang. This patches fixes an issue in which the __clang_cuda_cmath.h header is being included even when cmath or math.h

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea updated this revision to Diff 198929. gtbercea added a comment. - Remove define. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61765/new/ https://reviews.llvm.org/D61765 Files: lib/Driver/ToolChains/Clang.cpp lib/Headers/CMakeLists.txt lib/Headers

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea updated this revision to Diff 199041. gtbercea added a comment. - Fix cstdlib issue. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61765/new/ https://reviews.llvm.org/D61765 Files: lib/Driver/ToolChains/Clang.cpp lib/Headers/CMakeLists.txt lib/Hea

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea updated this revision to Diff 199043. gtbercea added a comment. - Move back functions. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61765/new/ https://reviews.llvm.org/D61765 Files: lib/Driver/ToolChains/Clang.cpp lib/Headers/CMakeLists.txt lib/H

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea marked 2 inline comments as done. gtbercea added inline comments. Comment at: lib/Headers/openmp_wrappers/__clang_openmp_math_declares.h:17 + #include + #include +#endif jdoerfert wrote: > Why do we need the stdlib includes again? They are both prone

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea updated this revision to Diff 199062. gtbercea added a comment. - Clean test header. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61765/new/ https://reviews.llvm.org/D61765 Files: lib/Driver/ToolChains/Clang.cpp lib/Headers/CMakeLists.txt lib/Hea

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea updated this revision to Diff 199064. gtbercea added a comment. - Mock cstdlib header's abs functions. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61765/new/ https://reviews.llvm.org/D61765 Files: lib/Driver/ToolChains/Clang.cpp lib/Headers/CMakeL

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea updated this revision to Diff 199073. gtbercea added a comment. - Add tests. Exclude additional abs definition. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61765/new/ https://reviews.llvm.org/D61765 Files: lib/Driver/ToolChains/Clang.cpp lib/Heade

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea updated this revision to Diff 199075. gtbercea added a comment. - Simplify conditions. Add guards. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61765/new/ https://reviews.llvm.org/D61765 Files: lib/Driver/ToolChains/Clang.cpp lib/Headers/CMakeLists

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/Headers/__clang_cuda_math_forward_declares.h:30-38 +#ifndef _OPENMP +__DEVICE__ long abs(long); +__DEVICE__ long long abs(long long); +#else +#ifndef __cplusplus __DEVICE__ long abs(long);

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/Headers/__clang_cuda_math_forward_declares.h:30-38 +#ifndef _OPENMP +__DEVICE__ long abs(long); +__DEVICE__ long long abs(long long); +#else +#ifndef __cplusplus __DEVICE__ long abs(long);

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea updated this revision to Diff 199080. gtbercea added a comment. - Error if not in OpenMP. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61765/new/ https://reviews.llvm.org/D61765 Files: lib/Driver/ToolChains/Clang.cpp lib/Headers/CMakeLists.txt li

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/Headers/__clang_cuda_math_forward_declares.h:30-38 +#ifndef _OPENMP +__DEVICE__ long abs(long); +__DEVICE__ long long abs(long long); +#else +#ifndef __cplusplus __DEVICE__ long abs(long);

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea added a comment. In D61765#1500233 , @tra wrote: > In D61765#1499957 , @jdoerfert wrote: > > > Two small changes and then it is fine with me. @tra ? > > > LGTM in general. I would still like to confirm that

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea updated this revision to Diff 199304. gtbercea added a comment. - Exclude clock functions. Reverse inclusion order. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61765/new/ https://reviews.llvm.org/D61765 Files: lib/Driver/ToolChains/Clang.cpp lib/H

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea added a comment. In D61765#1500446 , @tra wrote: > In D61765#1500309 , @gtbercea wrote: > > > As soon as libc++ the limits header included in > > > > __clang_cuda_cmath.h:15 > > ``` is not found: > >

[PATCH] D61765: [OpenMP][Clang][BugFix] Split declares and math functions inclusion.

gtbercea added a comment. In D61765#1500446 , @tra wrote: > In D61765#1500309 , @gtbercea wrote: > > > As soon as libc++ the limits header included in > > > > __clang_cuda_cmath.h:15 > > ``` is not found: > >

[PATCH] D61949: [OpenMP][bugfix] Fix issues with C++ 17 compilation when handling math functions

gtbercea created this revision. gtbercea added reviewers: ABataev, tra, jdoerfert, hfinkel, caomhin. Herald added subscribers: cfe-commits, guansong. Herald added a project: clang. In OpenMP device offloading we must ensure that unde C++ 17, the inclusion of cstdlib will works correctly. Reposi

[PATCH] D61949: [OpenMP][bugfix] Fix issues with C++ 17 compilation when handling math functions

gtbercea updated this revision to Diff 199651. gtbercea added a comment. - Refactor. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61949/new/ https://reviews.llvm.org/D61949 Files: lib/Headers/__clang_cuda_cmath.h lib/Headers/__clang_cuda_device_functions.h

[PATCH] D61949: [OpenMP][bugfix] Fix issues with C++ 17 compilation when handling math functions

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/Headers/__clang_cuda_device_functions.h:1477 #endif // CUDA_VERSION >= 9020 +#if __cplusplus >= 201703L +__DEVICE__ int abs(int __a) noexcept { return __nv_abs(__a); } tra

[PATCH] D61949: [OpenMP][bugfix] Fix issues with C++ 17 compilation when handling math functions

gtbercea updated this revision to Diff 199662. gtbercea marked an inline comment as done. gtbercea added a comment. - Remove included headers. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D61949/new/ https://reviews.llvm.org/D61949 Files: lib/Headers/__clang_c

[PATCH] D60907: [OpenMP] Add math functions support in OpenMP offloading

gtbercea abandoned this revision. gtbercea marked an inline comment as done. gtbercea added a comment. Replaced by: D61399 Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D60907/new/ https://reviews.llvm.org/D60907 __

[PATCH] D61959: [OpenMP][Bugfix] Move double and float versions of abs under c++ macro

gtbercea created this revision. gtbercea added reviewers: tra, jdoerfert, hfinkel, ABataev, caomhin. Herald added subscribers: cfe-commits, guansong. Herald added a project: clang. This is a fix for the reported bug: 41861 abs functions need to be mo

[PATCH] D47849: [OpenMP][Clang][NVPTX] Enable math functions called in an OpenMP NVPTX target device region to be resolved as device-native function calls

gtbercea abandoned this revision. gtbercea added a comment. Replaced by: D61399 Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D47849/new/ https://reviews.llvm.org/D47849 ___ cfe-commi

[PATCH] D62046: [OpenMP][bugfix] Add missing math functions variants for log and abs.

gtbercea created this revision. gtbercea added reviewers: jdoerfert, hfinkel, tra, caomhin. Herald added subscribers: cfe-commits, guansong. Herald added a project: clang. When including the random header in C++, some of the math functions it relies on are not present in the CUDA headers. We incl

[PATCH] D62046: [OpenMP][bugfix] Add missing math functions variants for log and abs.

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/Headers/__clang_cuda_cmath.h:55-56 +#if defined(_OPENMP) && defined(__cplusplus) +__DEVICE__ const float abs(const float __x) { return ::fabsf((float)__x); } +__DEVICE__ const double abs(co

[PATCH] D62046: [OpenMP][bugfix] Add missing math functions variants for log and abs.

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/Headers/__clang_cuda_device_functions.h:1604 +#if defined(_OPENMP) && defined(__cplusplus) +__DEVICE__ double log(long double __a) { return __nv_log((double)__a); } +#endif

[PATCH] D60568: [OpenMP] Add support for registering requires directives with the runtime

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/CodeGen/CGOpenMPRuntime.cpp:10364 + if (const auto *FD = dyn_cast(D)) { +if (OMPDeclareTargetDeclAttr::isDeclareTargetDeclaration(FD)) { + HasEmittedDeclareTargetRegion = true; --

[PATCH] D62046: [OpenMP][bugfix] Add missing math functions variants for log and abs.

gtbercea updated this revision to Diff 200048. gtbercea added a comment. - update patch Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D62046/new/ https://reviews.llvm.org/D62046 Files: lib/Headers/__clang_cuda_cmath.h lib/Headers/__clang_cuda_math_forward_dec

[PATCH] D62046: [OpenMP][bugfix] Add missing math functions variants for log and abs.

gtbercea added a comment. @tra I eliminated the long double definition for log and only left the declaration. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D62046/new/ https://reviews.llvm.org/D62046 ___ cfe-commits

[PATCH] D60568: [OpenMP] Add support for registering requires directives with the runtime

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/CodeGen/CGOpenMPRuntime.cpp:10364 + if (const auto *FD = dyn_cast(D)) { +if (OMPDeclareTargetDeclAttr::isDeclareTargetDeclaration(FD)) { + HasEmittedDeclareTargetRegion = true; --

[PATCH] D60568: [OpenMP] Add support for registering requires directives with the runtime

gtbercea updated this revision to Diff 200088. gtbercea added a comment. - Fix conditions. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D60568/new/ https://reviews.llvm.org/D60568 Files: lib/CodeGen/CGOpenMPRuntime.cpp lib/CodeGen/CGOpenMPRuntime.h lib/Cod

[PATCH] D60568: [OpenMP] Add support for registering requires directives with the runtime

gtbercea updated this revision to Diff 200089. gtbercea marked 3 inline comments as done. gtbercea added a comment. - Fix format. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D60568/new/ https://reviews.llvm.org/D60568 Files: lib/CodeGen/CGOpenMPRuntime.cpp

[PATCH] D60568: [OpenMP] Add support for registering requires directives with the runtime

gtbercea updated this revision to Diff 200091. gtbercea added a comment. - Fix braces. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D60568/new/ https://reviews.llvm.org/D60568 Files: lib/CodeGen/CGOpenMPRuntime.cpp lib/CodeGen/CGOpenMPRuntime.h lib/CodeGen

[PATCH] D60568: [OpenMP] Add support for registering requires directives with the runtime

gtbercea added a comment. ping Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D60568/new/ https://reviews.llvm.org/D60568 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/l

[PATCH] D60568: [OpenMP] Add support for registering requires directives with the runtime

gtbercea added a comment. In D60568#1508909 , @ABataev wrote: > Looks good, in general. Just one more question: will __tgt_register_requires > be emitted if only -fopenmp is used? If -fopenmp-targets is not provided, > this function should not be called,

[PATCH] D60568: [OpenMP] Add support for registering requires directives with the runtime

gtbercea updated this revision to Diff 200355. gtbercea added a comment. - Avoid requires registration when no targets are specified. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D60568/new/ https://reviews.llvm.org/D60568 Files: include/clang/Basic/LangOption

[PATCH] D60568: [OpenMP] Add support for registering requires directives with the runtime

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/Frontend/CompilerInvocation.cpp:2840 Opts.OpenMP && Args.hasArg(options::OPT_fopenmp_is_device); + Opts.OpenMPHasTargets = + Opts.OpenMP && Args.hasArg(options::OPT_fopenmp_tar

[PATCH] D60568: [OpenMP] Add support for registering requires directives with the runtime

gtbercea updated this revision to Diff 200504. gtbercea added a comment. - Remove new option. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D60568/new/ https://reviews.llvm.org/D60568 Files: lib/CodeGen/CGOpenMPRuntime.cpp lib/CodeGen/CGOpenMPRuntime.h lib/

[PATCH] D60883: [OpenMP] Avoid emitting maps for target link variables when unified memory is used

gtbercea updated this revision to Diff 200782. gtbercea added a comment. - Fix function call. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D60883/new/ https://reviews.llvm.org/D60883 Files: lib/CodeGen/CGOpenMPRuntime.cpp lib/CodeGen/CGOpenMPRuntime.h Inde

[PATCH] D60883: [OpenMP] Avoid emitting maps for target link variables when unified memory is used

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: lib/CodeGen/CGOpenMPRuntime.cpp:7956 +// If using unified memory, no need to do the mappings. +if (CGF.CGM.getOpenMPRuntime().hasUnifiedAddressingSupport()) + return; -

[PATCH] D60883: [OpenMP] Avoid emitting maps for target link variables when unified memory is used

gtbercea updated this revision to Diff 200818. gtbercea added a comment. - Include sema check. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D60883/new/ https://reviews.llvm.org/D60883 Files: lib/CodeGen/CGOpenMPRuntime.cpp lib/CodeGen/CGOpenMPRuntime.h lib

[PATCH] D62407: [OpenMP] Add test for requires and unified shared memory clause with declare target link

gtbercea created this revision. gtbercea added reviewers: ABataev, caomhin. Herald added subscribers: cfe-commits, jdoerfert, guansong. Herald added a project: clang. This patch adds a test for requires with unified share memory clause when a declare target link is present. This test needs to go

[PATCH] D62407: [OpenMP] Add test for requires and unified shared memory clause with declare target link

gtbercea added a comment. In D62407#1516298 , @ABataev wrote: > Just one question - why it is not the part of the whole patch for the unified > memory support in clang? This test will need to change in the unified memory case. That's why its being intr

[PATCH] D60883: [OpenMP] Avoid emitting maps for target link variables when unified memory is used

gtbercea updated this revision to Diff 201307. gtbercea added a comment. - Add test. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D60883/new/ https://reviews.llvm.org/D60883 Files: lib/CodeGen/CGOpenMPRuntime.cpp lib/CodeGen/CGOpenMPRuntime.h lib/Sema/Sema

[PATCH] D60883: [OpenMP] Avoid emitting maps for target link variables when unified memory is used

gtbercea added a comment. In D60883#1516693 , @ABataev wrote: > Just one question after looking at the test: how we're going to link the > device variable to its host copy? I think the address should be passed to the > runtime, no? The runtime will tak

[PATCH] D55928: [OpenMP] Add flag for preventing the extension to 64 bits for the collapse loop counter

gtbercea created this revision. gtbercea added reviewers: ABataev, caomhin. Herald added subscribers: cfe-commits, guansong. Introduce a compiler flag for cases when the user knows that the collapsed loop counter can be safely represented using at most 32 bits. This will prevent the emission of

[PATCH] D55928: [OpenMP] Add flag for preventing the extension to 64 bits for the collapse loop counter

gtbercea updated this revision to Diff 179076. gtbercea marked 4 inline comments as done. gtbercea edited the summary of this revision. gtbercea added a comment. - Address comments. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D55928/new/ https://reviews.llvm.org

[PATCH] D55928: [OpenMP] Add flag for preventing the extension to 64 bits for the collapse loop counter

gtbercea added inline comments. Comment at: include/clang/Basic/LangOptions.def:210 LANGOPT(OpenMPCUDABlocksPerSM , 32, 0, "Number of blocks per SM for CUDA devices.") +LANGOPT(OpenMPMax32BitCollapseWidth , 1, 0, "Use at most 32 bits to represent the collapsed loop nest coun

[PATCH] D55928: [OpenMP] Add flag for preventing the extension to 64 bits for the collapse loop counter

gtbercea updated this revision to Diff 179078. gtbercea added a comment. - Address comments. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D55928/new/ https://reviews.llvm.org/D55928 Files: docs/OpenMPSupport.rst include/clang/Basic/LangOptions.def include/

[PATCH] D55928: [OpenMP] Add flag for preventing the extension to 64 bits for the collapse loop counter

gtbercea added inline comments. Comment at: docs/OpenMPSupport.rst:119 +time. To prevent this conservative choice and use at most 32 bits, +compile your program with the `-fopenmp-max-32bit-collapse-width`. + kkwli0 wrote: > ABataev wrote: > > -fopenmp-optimistic

[PATCH] D56413: [OpenMP] Avoid remainder operations for loop index values on a collapsed loop nest.

gtbercea created this revision. gtbercea added reviewers: ABataev, caomhin. Herald added subscribers: cfe-commits, arphaman, guansong. Change the strategy for computing loop index variables after collapsing a loop nest via the collapse clause by replacing the expensive remainder operation with m

[PATCH] D56413: [OpenMP] Avoid remainder operations for loop index values on a collapsed loop nest.

gtbercea updated this revision to Diff 180709. gtbercea added a comment. Invert accumulation direction. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D56413/new/ https://reviews.llvm.org/D56413 Files: docs/OpenMPSupport.rst include/clang/Basic/LangOptions.d

[PATCH] D56413: [OpenMP] Avoid remainder operations for loop index values on a collapsed loop nest.

gtbercea updated this revision to Diff 180711. gtbercea added a comment. Fix update. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D56413/new/ https://reviews.llvm.org/D56413 Files: lib/Sema/SemaOpenMP.cpp test/OpenMP/for_codegen.cpp test/OpenMP/for_simd_co

[PATCH] D56413: [OpenMP] Avoid remainder operations for loop index values on a collapsed loop nest.

gtbercea updated this revision to Diff 180713. gtbercea added a comment. Remove redundant initalization. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D56413/new/ https://reviews.llvm.org/D56413 Files: lib/Sema/SemaOpenMP.cpp test/OpenMP/for_codegen.cpp tes

[PATCH] D56413: [OpenMP] Avoid remainder operations for loop index values on a collapsed loop nest.

gtbercea updated this revision to Diff 180722. gtbercea added a comment. Fix. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D56413/new/ https://reviews.llvm.org/D56413 Files: lib/Sema/SemaOpenMP.cpp test/OpenMP/for_codegen.cpp test/OpenMP/for_simd_codegen

[PATCH] D55928: [OpenMP] Add flag for preventing the extension to 64 bits for the collapse loop counter

gtbercea marked an inline comment as done. gtbercea added inline comments. Comment at: docs/OpenMPSupport.rst:120 +compile your program with the `-fopenmp-optimistic-collapse`. + + hfinkel wrote: > Can you please clarify here what happens when the loop induction

[PATCH] D49188: [OpenMP] Initialize data sharing for SPMD case

gtbercea created this revision. gtbercea added reviewers: ABataev, carlo.bertolli, caomhin. Herald added subscribers: cfe-commits, guansong, jholewinski. In the SPMD case, we need to initialize the data sharing and globalization infrastructure. This covers the case when an SPMD region calls a fun

[PATCH] D49188: [OpenMP] Initialize data sharing stack for SPMD case

gtbercea updated this revision to Diff 155002. gtbercea added a comment. Fix test. Repository: rC Clang https://reviews.llvm.org/D49188 Files: lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp test/OpenMP/nvptx_data_sharing.cpp Index: test/OpenMP/nvptx_data_sharing.cpp =

[PATCH] D49188: [OpenMP] Initialize data sharing stack for SPMD case

gtbercea updated this revision to Diff 155037. gtbercea added a comment. Add test for spmd stack init function. Repository: rC Clang https://reviews.llvm.org/D49188 Files: lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp test/OpenMP/nvptx_data_sharing.cpp test/OpenMP/nvptx_data_sharing_spmd.cpp

[PATCH] D49188: [OpenMP] Initialize data sharing stack for SPMD case

gtbercea marked 2 inline comments as done. gtbercea added inline comments. Comment at: lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp:84-85 OMPRTL_NVPTX__kmpc_data_sharing_init_stack, + /// Call to void __kmpc_data_sharing_init_stack_spmd(); + OMPRTL_NVPTX__kmpc_data_sharing_init_stac

[PATCH] D49188: [OpenMP] Initialize data sharing stack for SPMD case

gtbercea updated this revision to Diff 155391. gtbercea added a comment. Fix tests. Repository: rC Clang https://reviews.llvm.org/D49188 Files: lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp test/OpenMP/nvptx_data_sharing.cpp test/OpenMP/nvptx_target_parallel_codegen.cpp test/OpenMP/nvptx_targ

[PATCH] D38978: [OpenMP] Enable the lowering of implicitly shared variables in OpenMP GPU-offloaded target regions to the GPU shared memory

gtbercea updated this revision to Diff 124243. gtbercea added a comment. Add regression tests and allow for shared memory lowering to be disabled at function level. Repository: rL LLVM https://reviews.llvm.org/D38978 Files: include/llvm/CodeGen/TargetPassConfig.h lib/CodeGen/TargetPassC

[PATCH] D40451: [OpenMP] Add function attribute for triggering shared memory lowering in the LLVM backend

gtbercea created this revision. Herald added a subscriber: jholewinski. Since OpenMP and CUDA share the same toolchain we need to disable: - the lowering of variables to shared memory in the LLVM NVPTX backend - the emission of the shared depot - the emission of shared stack pointers when compil

[PATCH] D40451: [OpenMP] Add function attribute for triggering shared memory lowering in the LLVM backend

gtbercea updated this revision to Diff 124246. Repository: rL LLVM https://reviews.llvm.org/D40451 Files: lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp test/OpenMP/nvptx_data_sharing.cpp Index: test/OpenMP/nvptx_data_sharing.cpp ===

[PATCH] D41123: [OpenMP] Add function attribute for triggering data sharing.

gtbercea created this revision. gtbercea added reviewers: hfinkel, Hahnfeld, ABataev, carlo.bertolli, caomhin. Herald added a subscriber: jholewinski. The backend should only emit data sharing code for the cases where it is needed. A new function attribute is used by Clang to enable data sharing o

[PATCH] D41123: [OpenMP] Add function attribute for triggering data sharing.

gtbercea updated this revision to Diff 126594. gtbercea added a comment. Fix test. Repository: rL LLVM https://reviews.llvm.org/D41123 Files: lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp test/OpenMP/nvptx_data_sharing.cpp Index: test/OpenMP/nvptx_data_sharing.cpp ==

[PATCH] D133694: [Clang][OpenMP] Fix use_device_addr

doru1004 created this revision. Herald added subscribers: guansong, yaxunl. Herald added a project: All. doru1004 requested review of this revision. Herald added a reviewer: jdoerfert. Herald added subscribers: cfe-commits, sstefan1. Herald added a project: clang. With the current implementation,

[PATCH] D133694: [Clang][OpenMP] Fix use_device_addr

doru1004 updated this revision to Diff 459452. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D133694/new/ https://reviews.llvm.org/D133694 Files: clang/lib/CodeGen/CGOpenMPRuntime.cpp clang/lib/CodeGen/CGStmtOpenMP.cpp clang/test/OpenMP/target_data_use_device_addr_codegen_ptr.cpp

[PATCH] D133694: [Clang][OpenMP] Fix use_device_addr

doru1004 updated this revision to Diff 459475. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D133694/new/ https://reviews.llvm.org/D133694 Files: clang/lib/CodeGen/CGOpenMPRuntime.cpp clang/lib/CodeGen/CGStmtOpenMP.cpp clang/test/OpenMP/target_data_use_device_addr_codegen_ptr.cpp

[PATCH] D133694: [Clang][OpenMP] Fix use_device_addr

doru1004 added inline comments. Comment at: clang/test/OpenMP/target_data_use_device_addr_codegen_ptr.cpp:14 +{ +#pragma omp target data use_device_addr(x) +{ ye-luo wrote: > In my understanding of the spec. > `map(tofrom:x[0:256])` only maps

[PATCH] D133694: [Clang][OpenMP] Fix use_device_addr

doru1004 added inline comments. Comment at: clang/test/OpenMP/target_data_use_device_addr_codegen_ptr.cpp:14 +{ +#pragma omp target data use_device_addr(x) +{ ye-luo wrote: > doru1004 wrote: > > ye-luo wrote: > > > In my understanding of the s

[PATCH] D133694: [Clang][OpenMP] Fix use_device_addr

doru1004 added inline comments. Comment at: clang/test/OpenMP/target_data_use_device_addr_codegen_ptr.cpp:14 +{ +#pragma omp target data use_device_addr(x) +{ ye-luo wrote: > doru1004 wrote: > > ye-luo wrote: > > > doru1004 wrote: > > > > ye-l

[PATCH] D133694: [Clang][OpenMP] Fix use_device_addr

doru1004 added inline comments. Comment at: clang/test/OpenMP/target_data_use_device_addr_codegen_ptr.cpp:14 +{ +#pragma omp target data use_device_addr(x) +{ ye-luo wrote: > doru1004 wrote: > > ye-luo wrote: > > > doru1004 wrote: > > > > ye-l

[PATCH] D34784: [OpenMP] Add flag for specifying the target device architecture for OpenMP device offloading

gtbercea updated this revision to Diff 109903. gtbercea added a comment. New way to handle OpenMP target flags. https://reviews.llvm.org/D34784 Files: include/clang/Basic/DiagnosticDriverKinds.td include/clang/Driver/Options.td include/clang/Driver/ToolChain.h lib/Driver/Compilation.c

[PATCH] D34784: [OpenMP] Add flag for specifying the target device architecture for OpenMP device offloading

gtbercea updated this revision to Diff 109904. gtbercea added a comment. Don't exclude flags when host matches offload toolchain. https://reviews.llvm.org/D34784 Files: include/clang/Basic/DiagnosticDriverKinds.td include/clang/Driver/Options.td include/clang/Driver/ToolChain.h lib/Dr

[PATCH] D34784: [OpenMP] Add flag for specifying the target device architecture for OpenMP device offloading

gtbercea updated this revision to Diff 109938. gtbercea added a comment. Address comments. https://reviews.llvm.org/D34784 Files: include/clang/Basic/DiagnosticDriverKinds.td include/clang/Driver/Options.td include/clang/Driver/ToolChain.h lib/Driver/Compilation.cpp lib/Driver/ToolCha

[PATCH] D34784: [OpenMP] Add flag for specifying the target device architecture for OpenMP device offloading

gtbercea added inline comments. Comment at: lib/Driver/ToolChain.cpp:808 + continue; + } else if (XOpenMPTargetNoTriple) +// Passing device args: -Xopenmp-target -opt=val. hfinkel wrote: > Please include {} around this else-if code, even th

[PATCH] D34784: [OpenMP] Add flag for specifying the target device architecture for OpenMP device offloading

gtbercea updated this revision to Diff 109939. gtbercea added a comment. Fix -march special casing. https://reviews.llvm.org/D34784 Files: include/clang/Basic/DiagnosticDriverKinds.td include/clang/Driver/Options.td include/clang/Driver/ToolChain.h lib/Driver/Compilation.cpp lib/Drive

[PATCH] D34784: [OpenMP] Add flag for specifying the target device architecture for OpenMP device offloading

gtbercea updated this revision to Diff 110007. gtbercea added a comment. Fix test comments. https://reviews.llvm.org/D34784 Files: include/clang/Basic/DiagnosticDriverKinds.td include/clang/Driver/Options.td include/clang/Driver/ToolChain.h lib/Driver/Compilation.cpp lib/Driver/ToolCh

[PATCH] D29654: [OpenMP] Integrate OpenMP target region cubin into host binary

gtbercea updated this revision to Diff 110056. gtbercea added a comment. Add -no-canonical-prefixes to tests. https://reviews.llvm.org/D29654 Files: lib/Driver/Driver.cpp lib/Driver/ToolChains/Clang.cpp lib/Driver/ToolChains/CommonArgs.cpp lib/Driver/ToolChains/CommonArgs.h lib/Driver

[PATCH] D29654: [OpenMP] Integrate OpenMP target region cubin into host binary

gtbercea added a comment. In https://reviews.llvm.org/D29654#835045, @arphaman wrote: > Hi @gtbercea, > I couldn't reply to the email as cfe-commits didn't even register this > commit somehow, so I'm replying here. > > Unfortunately I had to revert this commit (r310291), + two others for a clea

[PATCH] D29654: [OpenMP] Integrate OpenMP target region cubin into host binary

gtbercea added a comment. In https://reviews.llvm.org/D29654#835256, @arphaman wrote: > Great, thanks! I think that you can just revert my revert with the fix > applied in one commit Hi Alex, I just commited the changes again. Let me know if it still fails for you. I think the issue was actu

<    1   2   3   4   5   6   7   >