[PATCH] D97052: [OpenCL] Prevent adding extension pragma by default

Anastasia updated this revision to Diff 327400. Anastasia added a comment. Reuploaded with changes from the 1st review round. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97052/new/ https://reviews.llvm.org/D97052 Files: clang/include/clang/Basic/OpenCLExtensions.def clang/include

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added a comment. In D97058#2587033 , @azabaznov wrote: >> FYI, I would even be ok if we remove the need for enabling non-core > > AFAIU this can be done not for every extension to maintain backward > compatibility. This should not result in any

[PATCH] D97052: [OpenCL] Prevent adding extension pragma by default

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG25ad188bfcdb: [OpenCL] Prevent adding extension pragma by default. (authored by Anastasia). Herald added a project: clang. Repository: rG LLVM Git

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added a comment. In D97058#2600160 , @azabaznov wrote: > Check 'isEnabled' is now private: it is used only for non-core or > non-optional core features; > creation of implicit type definitions is guarder with extension support > check; minor re

[PATCH] D97869: [OpenCL][Draft] Add OpenCL builtin test generator

Anastasia added a comment. If possible to add a fragment of what is generated it would be great! CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97869/new/ https://reviews.llvm.org/D97869 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D96771: [OpenCL] Add distinct file extension for C++ for OpenCL

Anastasia updated this revision to Diff 327915. Anastasia added a comment. Herald added subscribers: kbarton, nemanjai. - Renamed extensions to `cppcl` as it was more popular in RFC - Minimized number of changes in driver setup - Changed file extension in all C++ for OpenCL tests CHANGES SINCE L

[PATCH] D96771: [OpenCL] Add distinct file extension for C++ for OpenCL

Anastasia added a comment. In D96771#2574638 , @awarzynski wrote: > In D96771#2571855 , @Anastasia wrote: > >> This is only the initial patch and for the moment the primary goal is to >> remove the need for the fla

[PATCH] D97869: [OpenCL][Draft] Add OpenCL builtin test generator

Anastasia added a comment. > `` > > float test6092_cos(float arg1) { > return cos(arg1); > } > double test6093_cos(double arg1) { > return cos(arg1); > } > half test6094_cos(half arg1) { > return cos(arg1); > } > float2 test6095_cos(float2 arg1) { > return cos(arg1);

[PATCH] D89909: [SYCL] Implement SYCL address space attributes handling

Anastasia added a comment. In D89909#2600859 , @aaron.ballman wrote: > Just a few minor nits from me, but I'm mostly wondering: where are we at with > this and are there still substantive changes required? (I looked through the > comments, but there's a

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added inline comments. Comment at: clang/lib/Sema/Sema.cpp:364 for (auto &I : Atomic64BitTypes) setOpenCLExtensionForType(I, "cl_khr_int64_base_atomics cl_khr_int64_extended_atomics"); azabaznov wrote: > Anastasia wrote: > >

[PATCH] D97930: [OpenCL] Fix builtins that require multiple extensions

Anastasia added inline comments. Comment at: clang/test/SemaOpenCL/fdeclare-opencl-builtins.cl:21 +#if __OPENCL_C_VERSION__ <= CL_VERSION_1_2 +#pragma OPENCL EXTENSION cl_khr_3d_image_writes : enable +#endif You don't seem to be using `write_imagef` for OpenCL ve

[PATCH] D97869: [OpenCL][Draft] Add OpenCL builtin test generator

Anastasia added a comment. In D97869#2603249 , @azabaznov wrote: > That's awesome! > > I'm thinking of how can you track correctness of generated built-ins > according to spec.  Perhaps you can compare the number of distinct > declarations for each  buil

[PATCH] D97869: [OpenCL][Draft] Add OpenCL builtin test generator

Anastasia added a comment. In D97869#2603068 , @svenvh wrote: > In D97869#2602943 , @Anastasia wrote: > >> I was just thinking if we could combine the calls into one function to >> minimize the number of lines to p

[PATCH] D98076: [OpenCL][Docs] Release 12.0 notes

Anastasia created this revision. Anastasia added reviewers: svenvh, azabaznov. Herald added subscribers: ebevhan, yaxunl. Anastasia requested review of this revision. The notes cover functionality in clang added by commits from 16 July to 27th Jan. https://reviews.llvm.org/D98076 Files: clan

[PATCH] D97930: [OpenCL] Fix builtins that require multiple extensions

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks. Comment at: clang/test/SemaOpenCL/fdeclare-opencl-builtins.cl:188 + // expected-error@-2{{no matching function for call to 'write_imagef'}} + // expecte

[PATCH] D98039: [OpenCL] Set calling convention for -fdeclare-opencl-builtins

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM. Thanks! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D98039/new/ https://reviews.llvm.org/D98039 __

[PATCH] D97930: [OpenCL] Fix builtins that require multiple extensions

Anastasia added a comment. LGTM! CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97930/new/ https://reviews.llvm.org/D97930 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D97072: [OpenCL][Docs] Add guidelines for adding new extensions and features

Anastasia updated this revision to Diff 329602. Anastasia added a comment. Added sentence to elaborate the meaning of "useful". CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97072/new/ https://reviews.llvm.org/D97072 Files: clang/docs/OpenCLSupport.rst clang/lib/Driver/Types.cpp I

[PATCH] D97072: [OpenCL][Docs] Add guidelines for adding new extensions and features

Anastasia updated this revision to Diff 329605. Anastasia added a comment. Fixed patch reupload issue. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D97072/new/ https://reviews.llvm.org/D97072 Files: clang/docs/OpenCLSupport.rst Index: clang/docs/OpenCLSupport.rst ==

[PATCH] D98076: [OpenCL][Docs] Release 12.0 notes

Anastasia updated this revision to Diff 329617. Anastasia added a comment. Added corrections from Sven. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D98076/new/ https://reviews.llvm.org/D98076 Files: clang/docs/ReleaseNotes.rst Index: clang/docs/ReleaseNotes.rst ===

[PATCH] D98076: [OpenCL][Docs] Release 12.0 notes

Anastasia marked 3 inline comments as done. Anastasia added inline comments. Comment at: clang/docs/ReleaseNotes.rst:214 +- Added ``global_device`` and ``global_host`` address spaces for USM + allocations. + svenvh wrote: > Perhaps one more point to mention: > `

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added a comment. In D97058#2615926 , @azabaznov wrote: > Corrected some mistakes, added a test for diagnosing undeclared identifiers > when an extension is unsupported. Generally leaving the change as it is as > completely removing pragma may b

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Although some further improvements seem to be necessary they can be done separately. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D9

[PATCH] D97869: [OpenCL][Draft] Add OpenCL builtin test generator

Anastasia added a comment. Thanks for the comprehensive summary. I don't have many other suggestions in mind but I would like to add a few thoughts. We could always consider writing the tests manually from scratch too. But it doesn't feel like a good cost/benefit trade-off. Regarding 2 and 4 I

[PATCH] D77923: OpenCL: Fix some missing predefined macros

Anastasia added a comment. Herald added a subscriber: dexonsmith. FYI there has been a related spec change that makes this functionality optional in the offline compilation: https://github.com/KhronosGroup/OpenCL-Docs/pull/522/files CHANGES SINCE LAST ACTION https://reviews.llvm.org/D77923/n

[PATCH] D98076: [OpenCL][Docs] Release 12.0 notes

Anastasia added inline comments. Comment at: clang/docs/ReleaseNotes.rst:214 +- Added ``global_device`` and ``global_host`` address spaces for USM + allocations. + svenvh wrote: > Anastasia wrote: > > svenvh wrote: > > > Perhaps one more point to mention: > > >

[PATCH] D98076: [OpenCL][Docs] Release 12.0 notes

Anastasia added inline comments. Comment at: clang/docs/ReleaseNotes.rst:214 +- Added ``global_device`` and ``global_host`` address spaces for USM + allocations. + svenvh wrote: > Anastasia wrote: > > svenvh wrote: > > > Anastasia wrote: > > > > svenvh wrote: >

[PATCH] D98076: [OpenCL][Docs] Release 12.0 notes

Anastasia updated this revision to Diff 329944. Anastasia added a comment. - Added nested pointer item - Elaborated list of removed extensions with no kernel language changes. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D98076/new/ https://reviews.llvm.org/D98076 Files: clang/docs/R

[PATCH] D98076: [OpenCL][Docs] Release 12.0 notes

Anastasia added a comment. Sorry for the last-minute change, I have decided to add this too: +- Removed extensions without kernel language changes: + ``cl_khr_select_fprounding_mode``, ``cl_khr_gl_sharing``, ``cl_khr_icd``, + ``cl_khr_gl_event``, ``cl_khr_d3d10_sharing``, ``cl_khr_context

[PATCH] D98076: [OpenCL][Docs] Release 12.0 notes

Anastasia added a comment. @tstellar is it ok if I commit this to the release branch or do you prefer to do it yourself? CHANGES SINCE LAST ACTION https://reviews.llvm.org/D98076/new/ https://reviews.llvm.org/D98076 ___ cfe-commits mailing list cf

[PATCH] D97072: [OpenCL][Docs] Add guidelines for adding new extensions and features

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGbafcb4c6841a: [OpenCL][Docs] Add guidelines for new extensions and features. (authored by Anastasia). Herald added a project: clang. Changed prior t

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added a comment. In D97058#2622681 , @svenvh wrote: > In D97058#2622669 , @thakis wrote: > >> Does it repro if you add `-target arm64-apple-macosx` as arg to c-index-test >> on the RUN line of that test?

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added a comment. > (The `fatal error: 'opencl-c-base.h' file not found` bit is printed when the > test passes too, and is unrelated.) I can't reproduce this locally neither with nor without this commit. I think there is something wrong with the way this test runs, this might explai

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added a comment. In D97058#2622717 , @thakis wrote: > In case it's useful, here's the output of the RUN line on an m1 mac with this > patch locally reverted (where the test passes): > http://pastie.org/p/2F3Y0xUMtH5RP9TVRzG4LI Thanks, this see

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added a comment. In D97058#2622750 , @Anastasia wrote: > In D97058#2622717 , @thakis wrote: > >> In case it's useful, here's the output of the RUN line on an m1 mac with >> this patch locally reverted (wh

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added a comment. I can't find why and how `opencl-c-base.h` is used for the test. But it feels it is adding the random noise to the testing as it is not expected to be used. I don't feel that this commit is affecting the header though, so it is something that might have already been b

[PATCH] D98539: [OpenCL] Set target as spir for c-index-test for OpenCL

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! I can't see why we would need to tests anything target-specific here anyway, so I think the change is reasonable. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added a comment. In D97058#2622883 , @azabaznov wrote: > Yes, I was able to reproduce the failure with `-target arm64-apple-macosx` > flag, I provided the fix to set explicit target: > https://reviews.llvm.org/D98539. I have approved it. Thank

[PATCH] D98539: [OpenCL] Set target as spir for c-index-test for OpenCL

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGeed88e91f331: [OpenCL] Use spir target for CIndex tests for OpenCL. (authored by Anastasia). Repository: rG LLVM Github Monorepo CHANGES SINCE LA

[PATCH] D98539: [OpenCL] Set target as spir for c-index-test for OpenCL

Anastasia added a comment. In D98539#2623038 , @thakis wrote: > Can we get this landed asap please? Done! Sorry for the noise. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D98539/new/ https://reviews.llvm.

[PATCH] D89909: [SYCL] Implement SYCL address space attributes handling

Anastasia added a comment. In D89909#2617194 , @bader wrote: > In D89909#2606180 , @Anastasia wrote: > >> In D89909#2600859 , @aaron.ballman >> wrote: >> >>> Just a few mino

[PATCH] D97058: [OpenCL] Refactor diagnostic for OpenCL extension/feature

Anastasia added a comment. In D97058#2623160 , @thakis wrote: > I'm going to revert this in 30 minutes. The tree has been red for 12h due to > this then. I have committed the fix mentioned above. Apology for the delay. Repository: rG LLVM Github Mono

[PATCH] D89909: [SYCL] Implement SYCL address space attributes handling

Anastasia added a comment. In D89909#2623250 , @aaron.ballman wrote: > In D89909#2623211 , @Anastasia wrote: > >> In D89909#2617194 , @bader wrote: >> >>> @Anastasia, do you

[PATCH] D89909: [SYCL] Implement SYCL address space attributes handling

Anastasia added a comment. > If what you're suggesting is that Clang have a SYCL-specific page that serves > a similar purpose to https://clang.llvm.org/docs/OpenCLSupport.html, then I > agree we should have that (and it should be linked to from > https://clang.llvm.org/docs/index.html the same

[PATCH] D96771: [OpenCL] Add distinct file extension for C++ for OpenCL

Anastasia updated this revision to Diff 330698. Anastasia added a comment. Updated Driver test to check -x option. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D96771/new/ https://reviews.llvm.org/D96771 Files: clang/include/clang/Basic/LangStandard.h clang/include/clang/Driver/Typ

[PATCH] D96771: [OpenCL] Add distinct file extension for C++ for OpenCL

Anastasia added a comment. I had a second thought about the extension name and I realized that the reason why I initially wanted to use `clcpp` is that it aligns better with `clc++` which is used in `-cl-std`. Even though from the RFC the preference was towards `cppcl` it felt like there was no

[PATCH] D97869: [OpenCL][Draft] Add OpenCL builtin test generator

Anastasia added a comment. In D97869#2617063 , @svenvh wrote: > In D97869#2616772 , @Anastasia wrote: > >> Regarding 2 and 4 I think we should drive towards deprecation of >> `opencl-c.h` as it is a maintenance ove

[PATCH] D92004: [OpenCL] add CL 3.0 optional feature support to opencl-c.h

Anastasia added a comment. FYI the testing of headers functionality is now being discussed in https://reviews.llvm.org/D97869 Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D92004/new/ https://reviews.llvm.org/D92004 ___

[PATCH] D108621: [HIPSPV] Add CUDA->SPIR-V address space mapping

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108621/new/ https://reviews.llvm.org/D108621 _

[PATCH] D110618: [HIPSPV][2/4] Add HIPSPV tool chain

Anastasia added a comment. Could you please clarify the interface to SPIRV-LLVM-Translator tool, specifically: - Does clang lookup the path to the translator or assume any default path? - Is there any diagnostic provided if the translator not installed/found? - How does clang synchronize with th

[PATCH] D112230: [OpenCL] Add support of __opencl_c_device_enqueue feature macro.

Anastasia added inline comments. Comment at: clang/lib/AST/ExprConstant.cpp:9554 + +const auto &OpenCLFeaturesMap = +Info.Ctx.getTargetInfo().getSupportedOpenCLOpts(); azabaznov wrote: > azabaznov wrote: > > Anastasia wrote: > > > What test case cover

[PATCH] D112410: [SPIR-V] Add a toolchain for SPIR-V in clang

Anastasia updated this revision to Diff 391037. Anastasia retitled this revision from "[SPIR-V] Add a tool chain for SPIR-V (incomplete)" to "[SPIR-V] Add a toolchain for SPIR-V in clang". Anastasia edited the summary of this revision. Anastasia added reviewers: svenvh, azabaznov. Anastasia added

[PATCH] D112410: [SPIR-V] Add a toolchain for SPIR-V in clang

Anastasia updated this revision to Diff 391064. Anastasia added a comment. - Exported full diff. - Added forgotten test cases. - Fixed typo in docs. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D112410/new/ https://reviews.llvm.org/D112410 Files: clang/docs/UsersManual.rst clang/in

[PATCH] D110618: [HIPSPV][2/4] Add HIPSPV tool chain

Anastasia added a comment. In D110618#3152939 , @linjamaki wrote: > In D110618#3148501 , @Anastasia > wrote: > >> Could you please clarify the interface to SPIRV-LLVM-Translator tool, >> specifically: >> >> - Do

[PATCH] D112110: [OpenCL] queue_t and ndrange_t can't be defined in program scope.

Anastasia added inline comments. Herald added a subscriber: Naghasan. Comment at: clang/test/SemaOpenCL/invalid-types.cl:10 + +typedef image2d_t test; +global test qtt; // expected-error {{the '__global ndrange_t' type cannot be used to declare a program scope variable}} ---

[PATCH] D108621: [HIPSPV] Add CUDA->SPIR-V address space mapping

This revision was automatically updated to reflect the committed changes. Closed by commit rGf4d3cb4ca833: [HIPSPV] Add CUDA->SPIR-V address space mapping (authored by Anastasia). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D108621/new/ https://rev

[PATCH] D110155: [OpenCL] Allow optional __generic in __remove_address_space utility

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110155/new/ https://reviews.llvm.org/D110155 ___ cfe-commits mailing list

[PATCH] D134445: [PR57881][OpenCL] Fix incorrect diagnostics with templated types in kernel arguments

Anastasia updated this revision to Diff 470458. Anastasia added a comment. Addressed review comments CHANGES SINCE LAST ACTION https://reviews.llvm.org/D134445/new/ https://reviews.llvm.org/D134445 Files: clang/lib/Sema/SemaDecl.cpp clang/test/SemaOpenCLCXX/invalid-kernel.clcpp Index:

[PATCH] D132473: [Docs][OpenCL][SPIR-V] Release 15 notes for Clang

Anastasia created this revision. Anastasia added a reviewer: svenvh. Herald added subscribers: ebevhan, yaxunl. Herald added a project: All. Anastasia requested review of this revision. Added list of functional changes to release notes. https://reviews.llvm.org/D132473 Files: clang/docs/Relea

[PATCH] D132473: [Docs][OpenCL][SPIR-V] Release 15 notes for Clang

Anastasia added a comment. @svenvh as you made quite a lot of changes in the headers feel free to expand the description if you feel we should document some of those in more detail. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D132473/new/ https://reviews.llvm.org/D132473

[PATCH] D132473: [Docs][OpenCL][SPIR-V] Release 15 notes for Clang

Anastasia updated this revision to Diff 454838. Anastasia added a comment. Minor formatting changes CHANGES SINCE LAST ACTION https://reviews.llvm.org/D132473/new/ https://reviews.llvm.org/D132473 Files: clang/docs/ReleaseNotes.rst Index: clang/docs/ReleaseNotes.rst =

[PATCH] D132473: [Docs][OpenCL][SPIR-V] Release 15 notes for Clang

Anastasia updated this revision to Diff 455139. Anastasia added a comment. - Added clarification about the opaque pointers in SPIR-V. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D132473/new/ https://reviews.llvm.org/D132473 Files: clang/docs/ReleaseNotes.rst Index: clang/docs/Rele

[PATCH] D132473: [Docs][OpenCL][SPIR-V] Release 15 notes for Clang

Anastasia added inline comments. Comment at: clang/docs/ReleaseNotes.rst:676 + LLVM backend when generating SPIR-V binary. + Floating Point Support in Clang svenvh wrote: > Should we say anything about opaque pointers? Something like: > > ``` > - Although LL

[PATCH] D134445: [PR57881][OpenCL] Fix incorrect diagnostics with templated types in kernel arguments

Anastasia created this revision. Anastasia added reviewers: svenvh, olestrohm, rjmccall. Herald added subscribers: Naghasan, ebevhan, yaxunl. Herald added a project: All. Anastasia requested review of this revision. It seems relying on `isStandardLayoutType` helpers is fragile when kernel argumen

[PATCH] D134445: [PR57881][OpenCL] Fix incorrect diagnostics with templated types in kernel arguments

Anastasia added inline comments. Comment at: clang/test/SemaOpenCLCXX/invalid-kernel.clcpp:110 + +// FIXME: Use of templates doesn't work due to lazy instantiation of reference types. +//template I think here should be something like - `the use of templated typ

[PATCH] D141008: [Clang][SPIR-V] Emit target extension types for OpenCL types on SPIR-V.

Anastasia added a comment. Can you link relevant LLVM reviews or documentation and perhaps add some more details into the description? It's a bit suboptimal though to emit this form of types as SPIR-V specific. Ideally Clang should worry as less as possible about target specifics. So it woul

[PATCH] D141008: [Clang][SPIR-V] Emit target extension types for OpenCL types on SPIR-V.

Anastasia added inline comments. Comment at: clang/lib/CodeGen/CGExprScalar.cpp:2260 -return CGF.CGM.getNullPointer(cast(ConvertType(DestTy)), - DestTy); +// The type may be a target extension type instead of a pointer type +// (e.g., Op

[PATCH] D141297: [OpenCL] Allow undefining header-only features

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM, thanks! Btw I wonder if in the future we could add some error or warning in case someone uses the same approach for frontend specific features, i.e. #ifdef __undef___opencl_c_ge

[PATCH] D141449: clang/OpenCL: Fix not setting convergent on block invoke kernels

Anastasia added inline comments. Comment at: clang/lib/CodeGen/TargetInfo.cpp:12426 // FIXME: Apply default attributes F->addFnAttr(llvm::Attribute::NoUnwind); If `ConstructDefaultFnAttrList` was used it would get convergent correctly... Should we be usi

[PATCH] D134445: [PR57881][OpenCL] Fix incorrect diagnostics with templated types in kernel arguments

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG790cbaafc9e2: [OpenCL] Fix diagnostics with templates in kernel args. (authored by Anastasia). Herald added a subscriber: ldrumm. Herald added a proj

[PATCH] D134445: [PR57881][OpenCL] Fix incorrect diagnostics with templated types in kernel arguments

Anastasia added a comment. In D134445#3920188 , @rjmccall wrote: > You really can't ask whether a class template pattern is standard layout; > it's not meaningful. Well the templates have to be instantiated with concrete types as kernels can't be calle

[PATCH] D134445: [PR57881][OpenCL] Fix incorrect diagnostics with templated types in kernel arguments

Anastasia added a comment. In D134445#3920448 , @rjmccall wrote: > In D134445#3920257 , @Anastasia > wrote: > >> In D134445#3920188 , @rjmccall >> wrote: >> >>> You real

[PATCH] D142948: [OpenCL] Disable vector to scalar types coercion for OpenCL

Anastasia added a comment. It doesn't seem to make sense to have a `LangOpt` for this, if anything a new `CodeGenOpt` feels like a better fit. Then you should probably extend existing classify functions to prevent coercion instead of adding yet another function otherwise you will need to handle

[PATCH] D143348: [Clang][Doc][OpenCL] Release 16 notes

Anastasia created this revision. Anastasia added a reviewer: svenvh. Herald added subscribers: Naghasan, ebevhan, yaxunl. Herald added a project: All. Anastasia requested review of this revision. Documented major OpenCL features in release 16. https://reviews.llvm.org/D143348 Files: clang/doc

[PATCH] D141297: [OpenCL] Allow undefining header-only features

Anastasia added a comment. In D141297#4047148 , @svenvh wrote: > In D141297#4043122 , @Anastasia > wrote: > >> Btw I wonder if in the future we could add some error or warning in case >> someone uses the same ap

[PATCH] D141700: AMDGPU: Move enqueued block handling into clang

Anastasia added inline comments. Comment at: clang/lib/CodeGen/TargetInfo.cpp:12440 +/// AMDHSAKernelDescriptor.h) +static llvm::StructType *getKernelDescriptorType(llvm::LLVMContext &C) { + llvm::Type *Int8 = llvm::IntegerType::getInt8Ty(C); Is this AMDGPU targ

[PATCH] D141620: clang/OpenCL: Apply default attributes to enqueued blocks

Anastasia added inline comments. Comment at: clang/test/CodeGenOpenCL/cl20-device-side-enqueue-attributes.cl:26 +// SPIR32-SAME: (ptr addrspace(1) align 4 [[A:%.*]], ptr addrspace(1) align 4 [[B:%.*]], i32 [[I:%.*]]) #[[ATTR0:[0-9]+]] !kernel_arg_addr_space !2 !kernel_arg_acces

[PATCH] D127579: [clang][WIP] add option to keep types of ptr args for non-kernel functions in metadata

Anastasia added a comment. The patch seems very straight forward. I wonder though if we should always enable this for SPIR-V w/o adding any flag until the untyped pointers are added to SPIR-V so it will become a target setting controlled instead of a flag? Also is there anything that we don't

[PATCH] D127579: [clang][WIP] add option to keep types of ptr args for non-kernel functions in metadata

Anastasia added a comment. In D127579#3585537 , @nikic wrote: > In D127579#3585516 , @beanz wrote: > >> @nikic the most important thing you need to know about SPIR-V is that it is >> a virtual ISA based on LLVM I

[PATCH] D127579: [clang][WIP] add option to keep types of ptr args for non-kernel functions in metadata

Anastasia added a comment. In D127579#3586092 , @nikic wrote: > @Anastasia Thanks, that does sound like a legitimate reason to include the > information. I want to double check though, does linking the modules actually > fail if the functions have signa

[PATCH] D127579: [clang][WIP] add option to keep types of ptr args for non-kernel functions in metadata

Anastasia added a comment. In D127579#3588651 , @bader wrote: >> The way I understand a bitcast instruction in SPIR-V (`OpBitcast` in >> https://www.khronos.org/registry/SPIR-V/specs/unified1/SPIRV.html#_conversion_instructions) >> is that it can only a

[PATCH] D127579: [clang][WIP] add option to keep types of ptr args for non-kernel functions in metadata

Anastasia added a comment. In D127579#3590632 , @aeubanks wrote: > If I'm understanding correctly, previously the LLVM pointee type was used to > represent a frontend OpenCL type. An alternative to marking everything with > `elementtype` or adding metad

[PATCH] D127579: [clang][WIP] add option to keep types of ptr args for non-kernel functions in metadata

Anastasia added a comment. In D127579#3585537 , @nikic wrote: > In D127579#3585516 , @beanz wrote: > >> > > Clang is only permitted to encode pointer type information if that pointer > type has some kind of dire

[PATCH] D127961: [OpenCL] Reduce emitting candidate notes for builtins

Anastasia added inline comments. Comment at: clang/lib/Sema/SemaOverload.cpp:11224 + // so do not generate such notes. + if (S.getLangOpts().OpenCL && Fn->isImplicit() && + Cand->FailureKind != ovl_fail_bad_conversion) It would have been nice to print each

[PATCH] D128012: [HLSL] Add ExternalSemaSource & vector alias

Anastasia added a comment. I wonder if accepting type alias with identifier `vector` while parsing would be simpler to implement? Then you could just add those type aliases into the internal header. I assume user code is not allowed to redefine those i.e. it would result in undefined behavior o

[PATCH] D128012: [HLSL] Add ExternalSemaSource & vector alias

Anastasia added a comment. In D128012#3591828 , @beanz wrote: > For just the type alias, adding it during parsing would probably work. Where > things get more complicated is we have a whole mess of other data types that > we’ll need to add too eventuall

[PATCH] D127579: [clang][WIP] add option to keep types of ptr args for non-kernel functions in metadata

Anastasia added a comment. Btw do we have any issue tracking this failure, I am having a feeling that the translation of special types might be related to https://github.com/llvm/llvm-project/issues/55121? I am not aware at which point libclc is expected to be linked for SPIR-V but if linking h

[PATCH] D127579: [clang][WIP] add option to keep types of ptr args for non-kernel functions in metadata

Anastasia added a comment. In D127579#3595461 , @nikic wrote: > In D127579#3588626 , @Anastasia > wrote: > >> In D127579#3586092 , @nikic wrote: >> >>> @Anastasia Thanks,

[PATCH] D128012: [HLSL] Add ExternalSemaSource & vector alias

Anastasia added a comment. In D128012#3592126 , @beanz wrote: > In D128012#3592098 , @Anastasia > wrote: > >> aha, are you having a lot of these types then? Ok that sounds similar >> problem to OpenCL builtin fu

[PATCH] D127961: [OpenCL] Reduce emitting candidate notes for builtins

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D127961/new/ https://reviews.llvm.org/D127961 _

[PATCH] D127579: [clang][WIP] add option to keep types of ptr args for non-kernel functions in metadata

Anastasia added a comment. In D127579#3598308 , @nikic wrote: > In D127579#3598306 , @Anastasia > wrote: > >> In D127579#3595461 , @nikic wrote: >> >>> In D127579#3588626

[PATCH] D128434: [OpenCL] Remove half scalar vload/vstore builtins

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D128434/new/ https://reviews.llvm.org/D128434 _

[PATCH] D128436: [OpenCL] Remove fast_ half geometric builtins

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D128436/new/ https://reviews.llvm.org/D128436 _

[PATCH] D125679: [Clang] Added options for integrated backend only used for SPIR-V for now

This revision was automatically updated to reflect the committed changes. Closed by commit rG730dc4e9bce8: [Clang] Added options for integrated backend. (authored by Anastasia). Herald added a project: clang. Changed prior to commit: https://reviews.llvm.org/D125679?vs=429676&id=431938#toc Rep

[PATCH] D125052: [HLSL] Enable vector types for hlsl.

Anastasia added inline comments. Comment at: clang/lib/Driver/ToolChains/Clang.cpp:3490 + if (!Args.hasArg(options::OPT_dxc_no_stdinc)) +CmdArgs.push_back("-finclude-default-header"); } Sharing command line options is great, let's just update the help messa

[PATCH] D124753: [HLSL] Set main as default entry.

Anastasia added inline comments. Comment at: clang/test/CodeGenHLSL/entry_default.hlsl:1 +// RUN: %clang --driver-mode=dxc -Tcs_6_1 -fcgl -Fo - %s | FileCheck %s + Would it make sense to test the opposite i.e. if `-E` is passed in the command line `main` is not

[PATCH] D124752: [HLSL] clang codeGen for HLSLShaderAttr.

Anastasia added inline comments. Comment at: clang/lib/CodeGen/CGHLSLRuntime.cpp:54 + +void clang::CodeGen::CGHLSLRuntime::setHLSLFnuctionAttributes( +llvm::Function *F, const FunctionDecl *FD) { I don't think you need `clang::CodeGen::`? C

[PATCH] D124752: [HLSL] clang codeGen for HLSLShaderAttr.

Anastasia added inline comments. Comment at: clang/test/CodeGenHLSL/entry.hlsl:1 +// RUN: %clang --driver-mode=dxc -Tcs_6_1 -Efoo -fcgl -Fo - %s | FileCheck %s + Is this test here accidental? Comment at: clang/test/CodeGenHLSL/shader_type_attr.

[PATCH] D124752: [HLSL] clang codeGen for HLSLShaderAttr.

Anastasia added inline comments. Comment at: clang/lib/CodeGen/CGCall.cpp:2606 + if (getLangOpts().HLSL && TargetDecl) { +if (const FunctionDecl *FD = dyn_cast_or_null(TargetDecl)) + getHLSLRuntime().addHLSLFunctionAttributes(FuncAttrs, RetAttrs, FD); I

[PATCH] D124753: [HLSL] Set main as default entry.

Anastasia added a comment. From the current change it seems to me that what you need to be testing is a just that the frontend options are being passed correctly? This should then be a driver test with `-###` checking for the options to be set for the frontend invocation... Repository: rG L

<    7   8   9   10   11   12   13   14   15   16   >