[PATCH] D86626: [OpenCL][Docs] 10.x release notes

Anastasia closed this revision. Anastasia added a comment. https://reviews.llvm.org/rGdae9fe408793def8a49f5e1d10d2a859627785e3 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D86626/new/ https://reviews.llvm.org/D86626 ___ cfe-commits mailing li

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

Anastasia added a comment. In D89909#2423750 , @bader wrote: >> It was mentioned that the changes in the type system with address spaces is >> undesirable for SYCL because you indend to parse existing C++ code as-is. >> This contradicts the intended sema

[PATCH] D92721: [PATCH] [clang] Create SPIRABIInfo to enable SPIR_FUNC calling convention

Anastasia added a comment. In D92721#2437941 , @mibintc wrote: > There are many RuntimeCalls that are created in Clang, including _read_pipe, > all the OpenMP runtime calls, etc. Shall they all have their calling > conventions set from the ABIInfo? Cur

[PATCH] D92231: [OpenCL] Implement extended subgroups fully in headers

Anastasia added inline comments. Comment at: clang/lib/Headers/opencl-c-base.h:17-23 +#define cl_khr_subgroup_extended_types +#define cl_khr_subgroup_non_uniform_vote +#define cl_khr_subgroup_ballot +#define cl_khr_subgroup_non_uniform_arithmetic +#define cl_khr_subgroup_shuffle

[PATCH] D92231: [OpenCL] Implement extended subgroups fully in headers

Anastasia updated this revision to Diff 310227. Anastasia added a comment. - Set all macros to value 1 CHANGES SINCE LAST ACTION https://reviews.llvm.org/D92231/new/ https://reviews.llvm.org/D92231 Files: clang/include/clang/Basic/OpenCLExtensions.def clang/lib/Headers/opencl-c-base.h

[PATCH] D92231: [OpenCL] Implement extended subgroups fully in headers

Anastasia added inline comments. Comment at: clang/lib/Headers/opencl-c-base.h:17-23 +#define cl_khr_subgroup_extended_types +#define cl_khr_subgroup_non_uniform_vote +#define cl_khr_subgroup_ballot +#define cl_khr_subgroup_non_uniform_arithmetic +#define cl_khr_subgroup_shuffle

[PATCH] D92231: [OpenCL] Implement extended subgroups fully in headers

This revision was automatically updated to reflect the committed changes. Closed by commit rGa84599f177a6: [OpenCL] Implement extended subgroups fully in headers. (authored by Anastasia). Herald added a project: clang. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://rev

[PATCH] D92277: [OpenCL] Refactor of targets OpenCL option settings

Anastasia added a subscriber: cfe-commits. Anastasia added a comment. This looks much cleaner than the current flow! Thanks! We should just figure out a better place for defining the macros (see detailed comment in Targets.cpp). I am adding @cfe-commits since it's an important change to be vis

[PATCH] D92721: [PATCH] [clang] Create SPIRABIInfo to enable SPIR_FUNC calling convention

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/D92721/new/ https://reviews.llvm.org/D92721 __

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

Anastasia added a comment. In D89909#2442573 , @bader wrote: > In D89909#2439837 , @Anastasia wrote: > >> In D89909#2423750 , @bader wrote: >> It was mentioned that the

[PATCH] D90766: [OpenCL] Support vec_step in C++ for OpenCL mode

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Good spot! CHANGES SINCE LAST ACTION https://reviews.llvm.org/D90766/new/ https://reviews.llvm.org/D90766 ___ cfe-commits mailing li

[PATCH] D89869: [OpenCL] Define OpenCL feature macros for all versions

Anastasia added a comment. I guess targets like SPIR will be supporting all features by default? At this point, I would prefer that we only add the features that affect the parsing directly i.e. used in the clang source code. FYI I think we need to explain the common design of features and exte

[PATCH] D89869: [OpenCL] Define OpenCL feature macros for all versions

Anastasia added a comment. PS I think it's better to move both test files to SemaOpenCL folder as they are not related to common parsing but OpenCL specifically. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D89869/new/ https://reviews.llvm.org/D89869 __

[PATCH] D89869: [OpenCL] Define OpenCL feature macros for all versions

Anastasia added a comment. In D89869#2378235 , @azabaznov wrote: >> I guess targets like SPIR will be supporting all features by default? > > It sounds confusing for me: can you please elaborate about why does SPIR-V > target should support all features/e

[PATCH] D89869: [OpenCL] Define OpenCL feature macros for all versions

Anastasia added inline comments. Comment at: clang/include/clang/Basic/OpenCLExtensions.def:110 +OPENCLFEAT_INTERNAL(__opencl_c_generic_address_space, 200, ~0U) +OPENCLFEAT_INTERNAL(__opencl_c_work_group_collective_functions, 200, ~0U) +OPENCLFEAT_INTERNAL(__opencl_c_atomic_order

[PATCH] D90928: [OpenCL] Add assertions to extension lookup

Anastasia added a comment. Ok, it would still segfault but perhaps it is acceptable as this is an internal frontend only option for now. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D90928/new/ https://reviews.llvm.org/D90928

[PATCH] D90928: [OpenCL] Add assertions to extension lookup

Anastasia added a comment. In D90928#2382111 , @erik2020 wrote: > In D90928#2379322 , @Anastasia wrote: > >> Ok, it would still segfault but perhaps it is acceptable as this is an >> internal frontend only option f

[PATCH] D91237: [OpenCL] Make Clang recognize -cl-std=1.0 as a value argument

Anastasia added a comment. In D91237#2388168 , @Elvina wrote: > Failure on test "linux > HWAddressSanitizer-x86_64.TestCases::sizes.cpp" > looks bogus. I had the same issue with another PR > https://reviews.llvm.org/D89972 and other people also faced it

[PATCH] D91237: [OpenCL] Make Clang recognize -cl-std=1.0 as a value argument

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks for fixing this. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91237/new/ https://reviews.llvm.org/D91237 __

[PATCH] D89869: [OpenCL] Define OpenCL feature macros for all versions

Anastasia added a comment. In D89869#2388499 , @azabaznov wrote: > Addressed all concerns except replacing //__opencl_c_int64 //definition into > header. The reason for this as follows: this macro should be predefined > regardless if clang includes defau

[PATCH] D91429: [OpenCL] Stop opencl-c-base.h leaking extension enabling

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/D91429/new/ https://reviews.llvm.org/D91429 __

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

Anastasia added a comment. In D89909#2388622 , @bader wrote: > In D89909#2353931 , @Anastasia wrote: > >> In the RFC it has been discussed to either use target address spaces or >> perhaps to introduce a new attrib

[PATCH] D89869: [OpenCL] Define OpenCL feature macros for all versions

Anastasia added a comment. In D89869#2392032 , @Anastasia wrote: > In D89869#2388499 , @azabaznov wrote: > >> Addressed all concerns except replacing //__opencl_c_int64 //definition into >> header. The reason for t

[PATCH] D91531: [RFC][OpenCL] Provide mechanisms for defining extension macros

Anastasia updated this revision to Diff 305721. Anastasia added a comment. - Added full diffs. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91531/new/ https://reviews.llvm.org/D91531 Files: clang/include/clang/Basic/OpenCLExtensions.def clang/include/clang/Basic/TargetInfo.h cla

[PATCH] D91534: [RFC][OpenCL] Add new diagnostic for extension pragma with no effect

Anastasia updated this revision to Diff 305723. Anastasia added a comment. Added full diff. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91534/new/ https://reviews.llvm.org/D91534 Files: clang/docs/DiagnosticsReference.rst clang/include/clang/Basic/DiagnosticParseKinds.td clang/

[PATCH] D91538: [RFC][OpenCL] Make Tablegen header work with extensions that are not added in clang

Anastasia updated this revision to Diff 305724. Anastasia added a comment. Added full diffs. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D91538/new/ https://reviews.llvm.org/D91538 Files: clang/include/clang/Basic/OpenCLExtensions.def clang/lib/Basic/Targets/AMDGPU.h clang/lib/H

[PATCH] D89869: [OpenCL] Define OpenCL feature macros for all versions

Anastasia added inline comments. Comment at: clang/include/clang/Basic/OpenCLExtensions.def:100 +// OpenCL features +OPENCLFEAT_INTERNAL(__opencl_c_pipes, 200, ~0U) +OPENCLFEAT_INTERNAL(__opencl_c_generic_address_space, 200, ~0U) azabaznov wrote: > Anastasia wrot

[PATCH] D89869: [OpenCL] Define OpenCL feature macros for all versions

Anastasia added inline comments. Comment at: clang/include/clang/Basic/OpenCLOptions.h:51 + // check specific version of feature) + void supportFeatureForPreOCL30(StringRef Feat, bool On = true) { +assert(CLVer < 300 && "Can'be called for OpenCL C higher 3.0");

[PATCH] D90928: [OpenCL] Check for extension string extension lookup

Anastasia added a comment. LGTM for the fix! Do you think we could improve testing? I presume there is something that triggers a failure without your change... Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D90928/new/ https://reviews.llvm.org/D9092

[PATCH] D88300: [OpenCL] Initial patch for OpenCL C 3.0 support

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks! Even though I don't imagine the discussion on the RFC (http://lists.llvm.org/pipermail/cfe-dev/2020-September/066883.html) will affect this functionality I think it's good

[PATCH] D88300: [OpenCL] Initial patch for OpenCL C 3.0 support

Anastasia added a comment. I think it should be ok to proceed with this now. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D88300/new/ https://reviews.llvm.org/D88300 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.o

[PATCH] D88300: [OpenCL] Initial patch for OpenCL C 3.0 support

This revision was automatically updated to reflect the committed changes. Closed by commit rG71d3b7ec7b62: [OpenCL] Add new compilation mode for OpenCL 3.0. (authored by Anastasia). Herald added a project: clang. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.l

[PATCH] D89143: [OpenCL][Docs] Improved description of the supported language functionality.

Anastasia created this revision. Anastasia added reviewers: azabaznov, svenvh. Herald added subscribers: ebevhan, yaxunl. Anastasia requested review of this revision. As suggested on RFC: http://lists.llvm.org/pipermail/cfe-dev/2020-October/066932.html https://reviews.llvm.org/D89143 Files:

[PATCH] D89143: [OpenCL][Docs] Improved description of the supported language functionality.

Anastasia added inline comments. Comment at: clang/docs/UsersManual.rst:2718 +Clang only supports the full OpenCL profile, and not `the embedded profile +`_. + I am surprised t

[PATCH] D62574: Add support for target-configurable address spaces.

Anastasia added a comment. In D62574#2249817 , @ebevhan wrote: > In D62574#2242471 , @Anastasia wrote: > >> The only thing is that it would be good to test the new target setting logic >> somehow... do you have any

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia added a comment. Yes, this is a group of extensions that doesn't seem to change anything in the kernel language. So if the macro is available I don't understand how it can be used to do anything different in the kernel code because it just doesn't modify anything in the kernel code.

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia added a comment. In D89372#2330217 , @yaxunl wrote: > With this change, clang basically will have no knowledge about the removed > extensions, i.e., it will not know which extension is supported in which > version of OpenCL and have no way to e

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia added inline comments. Comment at: clang/include/clang/Basic/OpenCLExtensions.def:74 OPENCLEXT_INTERNAL(cl_khr_mipmap_image_writes, 200, ~0U) -OPENCLEXT_INTERNAL(cl_khr_srgb_image_writes, 200, ~0U) OPENCLEXT_INTERNAL(cl_khr_subgroups, 200, ~0U) azabaz

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia added a comment. In D89372#2330822 , @yaxunl wrote: > In D89372#2330362 , @Anastasia wrote: > >> In D89372#2330217 , @yaxunl wrote: >> >>> With this change, clang b

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia added inline comments. Comment at: clang/include/clang/Basic/OpenCLExtensions.def:16 // // If the extensions are to be enumerated without the supported OpenCL version, // define OPENCLEXT(ext) where ext is the name of the extension. yaxunl wrote: > C

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia added a comment. In D89372#2332853 , @jvesely wrote: > `cl_khr_byte_addressable_stores` changes language semantics. without it, > pointer dereferences of types smaller than 32 bits are illegal. Ok, does it mean that we are missing to diagnose t

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia added a comment. In D89372#2333937 , @jvesely wrote: > In D89372#2332997 , @Anastasia wrote: > >> In D89372#2332853 , @jvesely wrote: >> >>> `cl_khr_byte_addressabl

[PATCH] D89520: Don't permit array bound constant folding in OpenCL.

Anastasia added a comment. > Permitting non-standards-driven "do the best you can" constant-folding > of array bounds is permitted solely as a GNU compatibility feature. We > should not be doing it in any language mode that is attempting to be > conforming. > > From https://reviews.llvm.org/D20090

[PATCH] D90928: [OpenCL] Check for extension string extension lookup

Anastasia added a comment. In D90928#2408249 , @erik2020 wrote: > In D90928#2405796 , @Anastasia wrote: > >> Do you think we could improve testing? I presume there is something that >> triggers a failure without yo

[PATCH] D91531: [RFC][OpenCL] Provide mechanisms for defining extension macros

Anastasia added a comment. In D91531#2406390 , @azabaznov wrote: > Yes, in general this approach looks good to me conceptually. I have two > suggestions: > > 1. As we discussed, the term //core functionality// should be revisited here. > There's no clear

[PATCH] D91531: [RFC][OpenCL] Provide mechanisms for defining extension macros

Anastasia added a comment. In D91531#2411725 , @azabaznov wrote: >> Perhaps if you give me an example it would help to understand > > I was meant that this can potentially be used to undefine macros inside clang > directly. In this case there will no need

[PATCH] D90928: [OpenCL] Check for extension string extension lookup

Anastasia added inline comments. Comment at: clang/include/clang/Basic/OpenCLOptions.h:47 bool isSupported(llvm::StringRef Ext, const LangOptions &LO) const { +auto E = OptMap.find(Ext); +if (E == OptMap.end()) { erik2020 wrote: > Anastasia wrote: > >

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

Anastasia added a comment. Btw how about making some checks simpler. We could always define feature macros `__opencl_c_atomic_scope_device`, `__opencl_c_generic_address_space` for OpenCL 2.0 or C++ for OpenCL . Then everywhere we would only need to check feature macros instead of language vers

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

Anastasia added a subscriber: azabaznov. Anastasia added a comment. I assume your change depends on some other changes that add the feature macro definitions. So I think we should link the other changes in at some point. CCing to @azabaznov here who is working on the features too. Repository:

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

Anastasia added a comment. In D92004#2412953 , @airlied wrote: > this file has never been clang-format clean, happy to make changes if > reviewer decides they are needed. This has not been considered a conventional clang source so to say, so the coding

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

Anastasia added a subscriber: svenvh. Anastasia added a comment. > It will need some decent review and maybe some testing by other CL users to > make sure I've haven't messed up when used with a fully featured CL 3.0 stack. Clang functionality is expected to be tested in llvm-project and with Op

[PATCH] D91531: [RFC][OpenCL] Provide mechanisms for defining extension macros

Anastasia added a comment. In D91531#2412532 , @azabaznov wrote: >> So if I understand it well you are suggesting to perform a check for every >> parsed macro definition that would identify whether the macro has a name >> matching what is provided in -cl

[PATCH] D92033: [OpenCL] Move kernel arg type tests into one file

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/D92033/new/ https://reviews.llvm.org/D92033 ___

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

Anastasia added a comment. In D92004#2414692 , @airlied wrote: > In D92004#2413560 , @Anastasia wrote: > >> Btw how about making some checks simpler. We could always define feature >> macros `__opencl_c_atomic_scop

[PATCH] D92091: [OpenCL] Allow pointer-to-pointer kernel args beyond CL 1.2

Anastasia added inline comments. Comment at: clang/lib/Sema/SemaDecl.cpp:8638 // array, this recursive call only happens once. return getOpenCLKernelParameterType(S, QualType(UnderlyingTy, 0)); } Btw I am surprised that we recurse to check the underl

[PATCH] D90928: [OpenCL] Check for extension string extension lookup

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. Great! LGTM! Thanks! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D90928/new/ https://reviews.llvm.org/D90928 ___

[PATCH] D91531: [RFC][OpenCL] Provide mechanisms for defining extension macros

Anastasia added a comment. In D91531#2418237 , @mantognini wrote: > When reading the documentation [1] for -cl-ext (which I've never used so > far), I've noticed nothing is said about non-standard configurations (such as > disabling cl_khr_depth_images w

[PATCH] D89869: [OpenCL] Define OpenCL feature macros for all versions

Anastasia added a comment. After a recent discussion with Khronos on https://github.com/KhronosGroup/OpenCL-Docs/issues/500 and more thinking I start to realize that there is absolutely nothing new in OpenCL 3.0 features as it isn't a new concept at all. They use to be explained as optional cor

[PATCH] D89869: [OpenCL] Define OpenCL feature macros for all versions

Anastasia added inline comments. Comment at: clang/include/clang/Basic/OpenCLOptions.h:37 + // OpenCL Version + unsigned CLVer = 120; + bool IsOpenCLCPlusPlus = false; Anastasia wrote: > Anastasia wrote: > > azabaznov wrote: > > > Anastasia wrote: > > > > Ok,

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

Anastasia added a comment. In D89909#2403069 , @bader wrote: >> Did anyone conclude there that the language address spaces should be added >> for SYCL? I can't see any of this. In fact I don't even think there was any >> conclusion on the RFC. You should

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

Anastasia added a comment. Thanks! This looks much cleaner. I am still unclear what should we do with testing though. Comment at: clang/lib/Headers/opencl-c-base.h:284 memory_scope_work_group = __OPENCL_MEMORY_SCOPE_WORK_GROUP, +#if defined(__OPENCL_CPP_VERSION__) || (__OPE

[PATCH] D92231: [OpenCL] Implement extended subgroups fully in headers

Anastasia created this revision. Anastasia added reviewers: azabaznov, AlexeySotkin, mantognini, svenvh, PiotrFusik. Herald added subscribers: ebevhan, yaxunl. Anastasia requested review of this revision. Extended subgroups extensions were added in https://reviews.llvm.org/D79781, however, in th

[PATCH] D92244: [OpenCL] Prevent adding vendor extensions for all targets

Anastasia created this revision. Anastasia added reviewers: svenvh, AlexeySotkin, azabaznov. Herald added subscribers: ebevhan, yaxunl. Anastasia requested review of this revision. Some extensions were defined in the headers without guarding by the target. This commit fixes this and also removed

[PATCH] D92091: [OpenCL] Allow pointer-to-pointer kernel args beyond CL 1.2

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks! The testing can be improved before committing! Comment at: clang/lib/Sema/SemaDecl.cpp:8638 // array, this recursive call only happens once. re

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

Anastasia added a comment. In D92004#2420121 , @svenvh wrote: >> I am still unclear what should we do with testing though. > > We probably don't want to hold up this patch until we have more thorough > testing in place, since we don't even have any concre

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

Anastasia added a comment. In D92004#2423441 , @svenvh wrote: >>> We probably don't want to hold up this patch until we have more thorough >>> testing in place, since we don't even have any concrete plans for such >>> testing at the moment. >> >> Well ac

[PATCH] D92231: [OpenCL] Implement extended subgroups fully in headers

Anastasia added a comment. In D92231#2425305 , @PiotrFusik wrote: > The specification for these extensions was edited by Ben Ashbaugh @Intel. > I've asked him about this change. Sure. Thanks! Happy to hold off committing for a few days to see if Ben has

[PATCH] D92406: [OpenCL] Add some more kernel argument tests

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks Comment at: clang/test/SemaOpenCL/invalid-kernel-parameters.cl:40 +void no_addrspace_param(global int x) { } // expected-error{{parameter may not be qua

[PATCH] D92231: [OpenCL] Implement extended subgroups fully in headers

Anastasia added a comment. In D92231#2425323 , @Anastasia wrote: > In D92231#2425305 , @PiotrFusik > wrote: > >> The specification for these extensions was edited by Ben Ashbaugh @Intel. >> I've asked him about th

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia added a comment. In D89372#2334728 , @jvesely wrote: > In D89372#2334225 , @Anastasia wrote: > >>> >> >> Ok, so the pragma is supposed to control the pointer dereferencing which >> seems like a valid cas

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia added a comment. In D89372#2335627 , @jvesely wrote: > In D89372#2335027 , @Anastasia wrote: > >> In D89372#2334728 , @jvesely wrote: >> >>> In D89372#2334225

[PATCH] D89520: Don't permit array bound constant folding in OpenCL.

Anastasia added a comment. In D89520#2335628 , @rsmith wrote: > In D89520#2334477 , @Anastasia wrote: > >>> I couldn't find any spec justification for accepting the kinds of cases >>> that D20090

[PATCH] D89520: Don't permit array bound constant folding in OpenCL.

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/D89520/new/ https://reviews.llvm.org/D89520 ___

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia added a comment. In D89372#2341638 , @mantognini wrote: > I don't want to stop the wider discussion, that being said I think I've > addressed the comment regarding the content of this PR. Let me know if the > latest version is fine or needs fur

[PATCH] D89372: [OpenCL] Remove unused extensions

Anastasia accepted this revision. Anastasia added a comment. LGTM! Thanks! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D89372/new/ https://reviews.llvm.org/D89372 ___ cfe-commits mailing list cfe-commit

[PATCH] D82756: Port some floating point options to new option marshalling infrastructure

Anastasia added inline comments. Comment at: clang/test/CodeGen/fp-function-attrs.cpp:2 +// RUN: %clang_cc1 -triple x86_64-linux-gnu -ffast-math -ffinite-math-only -menable-unsafe-fp-math \ +// RUN: -menable-no-infs -menable-no-nans -fno-signed-zeros -freciprocal-math \ +// R

[PATCH] D82756: Port some floating point options to new option marshalling infrastructure

Anastasia resigned from this revision. Anastasia added a comment. Just to clarify aside from the concern I have raised regarding internal testing I am not in any strong opposition of this feature. So if the community decides that it is more important to have this feature than to keep the tests s

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

Anastasia added a comment. In the RFC it has been discussed to either use target address spaces or perhaps to introduce a new attribute to reflect a semantic needed for SYCL, but it seems to me in this change you are building on top of the existing language address space attribute with new entr

[PATCH] D60193: [OpenCL] Added addrspace_cast operator

Anastasia added inline comments. Comment at: clang/include/clang-c/Index.h:2057 + */ + CXCursor_CXXAddrspaceCastExpr = 129, + hans wrote: > akyrtzi wrote: > > Hi Anastasia, apologies for not catching this earlier, but libclang is > > intended to keep a stable

[PATCH] D101519: [C++4OpenCL] Fix reinterpret_cast of vectors

Anastasia added inline comments. Comment at: clang/lib/Sema/SemaCast.cpp:2333 + if (DestType->isExtVectorType() || SrcType->isExtVectorType()) { +// FIXME: Allow for reinterpret cast between 3 and 4 element vectors +if (Self.areVectorTypesSameSize(SrcType, De

[PATCH] D102248: [C++4OpenCL] Fix initialization of __constant constructors without arguments

Anastasia added inline comments. Comment at: clang/lib/Sema/SemaDecl.cpp:12643 + if (CXXRecordDecl *RD = Var->getType()->getAsCXXRecordDecl()) { +for (auto *ctor : RD->ctors()) { + if (ctor->isConstexpr() && ctor->getNumParams() == 0 && ctor

[PATCH] D100976: [OpenCL] Simplify use of C11 atomic types

Anastasia updated this revision to Diff 344781. Anastasia added a comment. Added a comment in the test. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D100976/new/ https://reviews.llvm.org/D100976 Files: clang/lib/Sema/Sema.cpp clang/test/Parser/opencl-atomics-cl20.cl Index: clang/

[PATCH] D100976: [OpenCL] Simplify use of C11 atomic types

Anastasia marked an inline comment as done. Anastasia added inline comments. Comment at: clang/test/Parser/opencl-atomics-cl20.cl:12 void atomic_types_test() { -// OpenCL v2.0 s6.13.11.6 defines supported atomic types. + // OpenCL v2.0 s6.13.11.6 defines supported atomic types.

[PATCH] D101843: [OpenCL] Add clang extension for bitfields

Anastasia updated this revision to Diff 344784. Anastasia added a comment. Added suggestions from Sven. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D101843/new/ https://reviews.llvm.org/D101843 Files: clang/docs/LanguageExtensions.rst clang/include/clang/Basic/OpenCLExtensions.def

[PATCH] D101843: [OpenCL] Add clang extension for bitfields

Anastasia added inline comments. Comment at: clang/lib/Sema/SemaDecl.cpp:16798 } // OpenCL v1.2 s6.9.c: bitfields are not supported. +if (BitWidth && !getOpenCLOptions().isAvailableOption( svenvh wrote: > Slightly modified to avoid repeating the ex

[PATCH] D102248: [C++4OpenCL] Fix initialization of __constant constructors without arguments

Anastasia added inline comments. Comment at: clang/test/SemaOpenCLCXX/addrspace-constructors.clcpp:36 + Z() __local = default; + Z() __global = default; + constexpr Z() __constant : z(0) {} olestrohm wrote: > Anastasia wrote: > > We seem to be missing the cove

[PATCH] D100985: [OpenCL] Remove pragma requirement for functions from Arm dot extension

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG58d18dde5cca: [OpenCL] Remove pragma requirement from Arm dot extension. (authored by Anastasia). Herald added a subscriber: ldrumm. Herald added a p

[PATCH] D102248: [C++4OpenCL] Fix initialization of __constant constructors without arguments

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks! Comment at: clang/test/CodeGenOpenCLCXX/addrspace-constructors.clcpp:22 + +// CHECK: @cx1 = dso_local addrspace(2) constant %struct.X zeroinitializer, al

[PATCH] D101519: [C++4OpenCL] Fix reinterpret_cast of vectors

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks! Just remember to reflect your final change in the commit message description. :) CHANGES SINCE LAST ACTION https://reviews.llvm.org/D101519/new/ https://reviews.llvm.or

[PATCH] D100976: [OpenCL] Simplify use of C11 atomic types

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG769cc335e6e6: [OpenCL] Simplify use of C11 atomic types. (authored by Anastasia). Herald added a subscriber: ldrumm. Herald added a project: clang.

[PATCH] D96524: [OpenCL] Add support of OpenCL C 3.0 __opencl_c_fp64

Anastasia added inline comments. Comment at: clang/lib/Sema/Sema.cpp:306 +// Set conditionally to provide correct diagnostics for 'double' type +llvm::StringRef FP64Feature( +getLangOpts().OpenCLVersion >= 300 ? "__opencl_c_fp64" : "cl_khr_fp64"); ---

[PATCH] D100396: [SYCL] Enable `opencl_global_[host,device]` attributes for SYCL

Anastasia added inline comments. Comment at: clang/docs/SYCLSupport.rst:96 - global_space, constant_space + * - ``__attribute__((opencl_global_device))`` + - global_space I think we should extend conversion rules too. Repository: rG LLVM Github Mo

[PATCH] D101043: [OpenCL] Drop extension pragma handling for extension types/declarations

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG3549466ac05e: [OpenCL] Drop pragma handling for extension types/decls. (authored by Anastasia). Herald added a subscriber: ldrumm. Herald added a pro

[PATCH] D102248: [C++4OpenCL] Fix initialization of __constant constructors without arguments

Anastasia added a comment. LGTM! Thanks! CHANGES SINCE LAST ACTION https://reviews.llvm.org/D102248/new/ https://reviews.llvm.org/D102248 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-co

[PATCH] D100396: [SYCL] Enable `opencl_global_[host,device]` attributes for SYCL

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/D100396/new/ https://reviews.llvm.org/D100396

[PATCH] D96524: [OpenCL] Add support of OpenCL C 3.0 __opencl_c_fp64

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Thanks! Comment at: clang/lib/Sema/Sema.cpp:304 addImplicitTypedef("event_t", Context.OCLEventTy); + if (getLangOpts().OpenCLCPlusPlus || getLangOpts().

[PATCH] D102689: [C++4OpenCL] Allow address space conversion in reinterpret_cast

Anastasia added a comment. Just to move the discussion from https://reviews.llvm.org/D101519 here. First of all I think this change should not be OpenCL specific i.e. there is not reason to do something different from C++ in general. Embedded C doesn't regulate the conversions between non-poin

[PATCH] D102850: [C++4OpenCL] Fix overloading resolution of addrspace constructors

Anastasia added inline comments. Comment at: clang/lib/Sema/SemaOverload.cpp:9870 + if (S.getLangOpts().OpenCL) { +if (const auto *CD1 = dyn_cast_or_null(Cand1.Function)) { I think we should remove the OpenCL check since it is not OpenCL specific rule an

[PATCH] D102853: [OpenCL] Align definition of __IMAGE_SUPPORT__ feature macro with OpenCL version and __opencl_c_images feature macro definition

Anastasia accepted this revision. Anastasia added a comment. This revision is now accepted and ready to land. LGTM! Although I think it would be better if we set `__IMAGE_SUPPORT__` conditionally also for earlier OpenCL versions. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D102850: [C++4OpenCL] Fix overloading resolution of addrspace constructors

Anastasia added inline comments. Comment at: clang/lib/Sema/SemaOverload.cpp:9870 + if (S.getLangOpts().OpenCL) { +if (const auto *CD1 = dyn_cast_or_null(Cand1.Function)) { olestrohm wrote: > Anastasia wrote: > > I think we should remove the OpenCL check

<    4   5   6   7   8   9   10   11   12   13   >