[PATCH] D152206: [Basic] Support 64-bit x86 target for UEFI

kosarev added a comment. @Prabhuk This reproduces with my setup as well. Did you try building with `-DBUILD_SHARED_LIBS=ON`? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D152206/new/ https://reviews.llvm.org/D152206

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev added a comment. Committing this revealed some more failures, fixed in rG8894c05b0d351f998b1836542ba791247394bd12 . Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D125604: [FileCheck] Catch missspelled directives.

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGad1d60c3befd: [FileCheck] Catch missspelled directives. (authored by kosarev). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev updated this revision to Diff 431979. kosarev added a comment. Rebased on top of fixes for non-trivial cases. Going to submit this tomorrow, if no objections. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D125604/new/

[PATCH] D126191: [Clang][CodeGen] Fix the cmse-clear-return.c test.

This revision was automatically updated to reflect the committed changes. Closed by commit rGbe4eaf10eef7: [Clang][CodeGen] Fix the cmse-clear-return.c test. (authored by kosarev). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D126191/new/

[PATCH] D126191: [Clang][CodeGen] Fix the cmse-clear-return.c test.

kosarev created this revision. Herald added a project: All. kosarev requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. Caught with D125604 . Repository: rG LLVM Github Monorepo

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev updated this revision to Diff 430392. kosarev added a comment. Herald added a project: Flang. Added Flang fixes and rebased. Thanks Simon for the quick turnaround! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D125604/new/

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev added subscribers: chill, momchil.velikov, az, alex-t, keith.walker.arm, plotfi, aivchenk, zer0, awanpand, levivic, zlei. kosarev added a comment. Tagging more people in hope to add visibility. Comment at: clang/test/CodeGen/cmse-clear-return.c:235 //

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev updated this revision to Diff 430322. kosarev added a comment. Herald added a reviewer: bollu. Add Polly fixes. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D125604/new/ https://reviews.llvm.org/D125604 Files:

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev added inline comments. Comment at: llvm/lib/FileCheck/FileCheck.cpp:1774-1781 +static std::pair +FindCheckType(const FileCheckRequest , StringRef Buffer, StringRef Prefix) { + bool Misspelled = false; + auto Res = FindCheckType(Req, Buffer, Prefix, Misspelled); + if

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev added inline comments. Comment at: llvm/lib/FileCheck/FileCheck.cpp:1774-1781 +static std::pair +FindCheckType(const FileCheckRequest , StringRef Buffer, StringRef Prefix) { + bool Misspelled = false; + auto Res = FindCheckType(Req, Buffer, Prefix, Misspelled); + if

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev updated this revision to Diff 429997. kosarev added a comment. Herald added subscribers: ThomasRaoux, sdasgup3, wenzhicui, wrengr, Chia-hungDuan, dcaballe, cota, mravishankar, teijeong, rdzhabarov, tatianashp, msifontes, jurahul, Kayjukh, grosul1, Joonsoo, stephenneuendorffer, liufengdb,

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev added inline comments. Comment at: llvm/lib/FileCheck/FileCheck.cpp:1774-1781 +static std::pair +FindCheckType(const FileCheckRequest , StringRef Buffer, StringRef Prefix) { + bool Misspelled = false; + auto Res = FindCheckType(Req, Buffer, Prefix, Misspelled); + if

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev added a comment. In D125604#3514983 , @jhenderson wrote: > Are the TODO cases where the test fails if changing them? Yes, this and where the fix is not perfectly obvious. Comment at:

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev added a comment. A work in progress. Contains changes that we likely want to see addressed separately. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D125604/new/ https://reviews.llvm.org/D125604

[PATCH] D125604: [FileCheck] Catch missspelled directives.

kosarev created this revision. Herald added subscribers: mattd, gchakrabarti, pmatos, asb, asavonic, wenlei, kerbowa, pengfei, Jim, asbirlea, thopre, rupprecht, george.burgess.iv, kbarton, hiraditya, jgravelle-google, sbc100, jvesely, nemanjai, dylanmckay, dschuff, jholewinski. Herald added a

[PATCH] D49075: [NEON] Define fp16 vld and vst intrinsics conditionally

This revision was automatically updated to reflect the committed changes. Closed by commit rC340140: [NEON] Define fp16 vld and vst intrinsics conditionally (authored by kosarev, committed by ). Repository: rC Clang https://reviews.llvm.org/D49075 Files: include/clang/Basic/arm_neon.td

[PATCH] D49075: [NEON] Define fp16 vld and vst intrinsics conditionally

This revision was automatically updated to reflect the committed changes. Closed by commit rL340140: [NEON] Define fp16 vld and vst intrinsics conditionally (authored by kosarev, committed by ). Herald added a subscriber: llvm-commits. Changed prior to commit:

[PATCH] D49075: [NEON] Define fp16 vld and vst intrinsics conditionally

kosarev added a comment. Ping. https://reviews.llvm.org/D49075 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D49075: [NEON] Define fp16 vld and vst intrinsics conditionally

kosarev updated this revision to Diff 157960. kosarev added a comment. Test that the affected intrinsics are not defined in no-fp16 mode. https://reviews.llvm.org/D49075 Files: include/clang/Basic/arm_neon.td test/CodeGen/arm-neon-vld.c test/CodeGen/arm-neon-vst.c

[PATCH] D49375: [NEON] Define half-precision vmaxnm intrinsics only when available

This revision was automatically updated to reflect the committed changes. Closed by commit rC337704: [NEON] Define half-precision vmaxnm intrinsics only when available (authored by kosarev, committed by ). Changed prior to commit: https://reviews.llvm.org/D49375?vs=155675=156802#toc

[PATCH] D49376: [NEON] Define half-precision vrnd intrinsics only when available

This revision was automatically updated to reflect the committed changes. Closed by commit rC337699: [NEON] Define half-precision vrnd intrinsics only when available (authored by kosarev, committed by ). Repository: rC Clang https://reviews.llvm.org/D49376 Files:

[PATCH] D49376: [NEON] Define half-precision vrnd intrinsics only when available

kosarev added inline comments. Comment at: include/clang/Basic/arm_neon.td:1419 // Vector rounding - def FRINTZH : SInst<"vrnd", "dd", "hQh">; - def FRINTNH : SInst<"vrndn", "dd", "hQh">; - def FRINTAH : SInst<"vrnda", "dd", "hQh">; - def FRINTPH :

[PATCH] D49375: [NEON] Define half-precision vmaxnm intrinsics only when available

kosarev added a comment. Thanks for reviewing! Comment at: include/clang/Basic/arm_neon.td:1466 def VMINH : SInst<"vmin", "ddd", "hQh">; - def FMAXNMH : SInst<"vmaxnm", "ddd", "hQh">; - def FMINNMH : SInst<"vminnm", "ddd", "hQh">; + let ArchGuard =

[PATCH] D48829: [NEON] Fix support for vrndi_f32(), vrndiq_f32() and vrndns_f32() intrinsics

This revision was automatically updated to reflect the committed changes. Closed by commit rC337690: [NEON] Fix support for vrndi_f32(), vrndiq_f32() and vrndns_f32() intrinsics (authored by kosarev, committed by ). Repository: rC Clang https://reviews.llvm.org/D48829 Files:

[PATCH] D49075: [NEON] Define fp16 vld and vst intrinsics conditionally

kosarev added a comment. Ping. https://reviews.llvm.org/D49075 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D48829: [NEON] Fix support for vrndi_f32(), vrndiq_f32() and vrndns_f32() intrinsics

kosarev added a comment. Ping. https://reviews.llvm.org/D48829 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D49375: [NEON] Define half-precision vmaxnm intrinsics only when available

kosarev added a comment. Ping. https://reviews.llvm.org/D49375 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D49376: [NEON] Define half-precision vrnd intrinsics only when available

kosarev added a comment. Ping. https://reviews.llvm.org/D49376 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D49075: [NEON] Define fp16 vld and vst intrinsics conditionally

kosarev added a comment. Ping. https://reviews.llvm.org/D49075 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D48829: [NEON] Fix support for vrndi_f32(), vrndiq_f32() and vrndns_f32() intrinsics

kosarev added a comment. Ping. https://reviews.llvm.org/D48829 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D49376: [NEON] Define half-precision vrnd intrinsics only when available

kosarev created this revision. kosarev added reviewers: SjoerdMeijer, jgreenhalgh, rengolin. kosarev added a project: clang. Herald added a reviewer: javed.absar. https://reviews.llvm.org/D49376 Files: include/clang/Basic/arm_neon.td test/Sema/arm-no-fp16.c Index: test/Sema/arm-no-fp16.c

[PATCH] D49375: [NEON] Define half-precision vmaxnm intrinsics only when available

kosarev created this revision. kosarev added reviewers: SjoerdMeijer, jgreenhalgh, rengolin. kosarev added a project: clang. Herald added a reviewer: javed.absar. https://reviews.llvm.org/D49375 Files: include/clang/Basic/arm_neon.td test/Sema/arm-no-fp16.c Index: test/Sema/arm-no-fp16.c

[PATCH] D48829: [NEON] Fix support for vrndi_f32(), vrndiq_f32() and vrndns_f32() intrinsics

kosarev added a comment. Ping. https://reviews.llvm.org/D48829 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D49075: [NEON] Define fp16 vld and vst intrinsics conditionally

kosarev created this revision. kosarev added reviewers: SjoerdMeijer, jgreenhalgh, rengolin. kosarev added a project: clang. Herald added a reviewer: javed.absar. This patch fixes definitions of vld and vst NEON intrinsics so that we only define them if half-precision arithmetic is supported on

[PATCH] D48829: [NEON] Fix support for vrndi_f32(), vrndiq_f32() and vrndns_f32() intrinsics

kosarev created this revision. kosarev added reviewers: SjoerdMeijer, jgreenhalgh, rengolin. kosarev added a project: clang. Herald added a reviewer: javed.absar. Herald added a subscriber: kristof.beyls. This patch adds support for vrndi_f32() and vrndiq_f32() intrinsics in AArch32 mode and for

[PATCH] D48440: [NEON] Support vldNq intrinsics in AArch32 (Clang part)

kosarev added a comment. Yep, this patch removes `vld_dup.c` as it duplicates part of what we have in `arm-neon-vld.c`. Clicking 'Show File Contents' below the file displays its contents. Thanks for reviewing! https://reviews.llvm.org/D48440 ___

[PATCH] D48440: [NEON] Support vldNq intrinsics in AArch32 (Clang part)

kosarev added a comment. Ping. https://reviews.llvm.org/D48440 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D47446: [NEON] Support VST1xN intrinsics in AArch32 mode (Clang part)

kosarev added a comment. Ping. https://reviews.llvm.org/D47446 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D47121: [NEON] Support VLD1xN intrinsics in AArch32 mode (Clang part)

This revision was automatically updated to reflect the committed changes. Closed by commit rL333819: [NEON] Support VLD1xN intrinsics in AArch32 mode (Clang part) (authored by kosarev, committed by ). Herald added a subscriber: llvm-commits. Changed prior to commit:

[PATCH] D47121: [NEON] Support VLD1xN intrinsics in AArch32 mode (Clang part)

kosarev added a comment. Thanks for reviewing. Comment at: lib/CodeGen/CGBuiltin.cpp:7865 } // FIXME: Sharing loads & stores with 32-bit is complicated by the absence // of an Align parameter here. SjoerdMeijer wrote: > How about this FIXME? Is it

[PATCH] D47121: [NEON] Support VLD1xN intrinsics in AArch32 mode (Clang part)

kosarev added a comment. Ping. https://reviews.llvm.org/D47121 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D45668: [NEON] Define vget_high_f16() and vget_low_f16() intrinsics in AArch64 mode only

kosarev added a comment. Thanks Sjoerd and James. Just added a comment referring to this revision in https://reviews.llvm.org/rL330420. Repository: rL LLVM https://reviews.llvm.org/D45668 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D45670: [NEON] Define vfma_n_f32() and vfmaq_n_f32() intrinsics in AArch32 mode

This revision was automatically updated to reflect the committed changes. Closed by commit rL330336: [NEON] Define vfma_n_f32() and vfmaq_n_f32() intrinsics in AArch32 mode (authored by kosarev, committed by ). Herald added a subscriber: llvm-commits. Changed prior to commit:

[PATCH] D45668: [NEON] Define vget_high_f16() and vget_low_f16() intrinsics in AArch64 mode only

kosarev added a comment. Sure, will do. Should we treat these intrinsics as ARMv8 or ARMv7/v8? Also, would you mind if I commit a comment under this differential revision explaining the situation? Repository: rL LLVM https://reviews.llvm.org/D45668

[PATCH] D45668: [NEON] Define vget_high_f16() and vget_low_f16() intrinsics in AArch64 mode only

This revision was automatically updated to reflect the committed changes. Closed by commit rL330195: [NEON] Define vget_high_f16() and vget_low_f16() intrinsics in AArch64 mode only (authored by kosarev, committed by ). Herald added a subscriber: llvm-commits. Changed prior to commit:

[PATCH] D45670: [NEON} Define vfma_n_f32() and vfmaq_n_f32() intrinsics in AArch32 mode

kosarev updated this revision to Diff 142784. kosarev added a comment. Removed checks for unused bitcasts. Thanks for catching! https://reviews.llvm.org/D45670 Files: include/clang/Basic/arm_neon.td test/CodeGen/arm-neon-fma.c Index: test/CodeGen/arm-neon-fma.c

[PATCH] D45668: [NEON] Define vget_high_f16() and vget_low_f16() intrinsics in AArch64 mode only

kosarev added a comment. The NEON Intrinsics Reference (http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ihi0073a/index.html) reads like they are AArch64-only. https://reviews.llvm.org/D45668 ___ cfe-commits mailing list

[PATCH] D45669: [NEON] Fix the architecture condition for the crypto intrinsics

This revision was automatically updated to reflect the committed changes. Closed by commit rL330187: [NEON] Fix the architecture condition for the crypto intrinsics (authored by kosarev, committed by ). Herald added a subscriber: llvm-commits. Changed prior to commit:

[PATCH] D45669: [NEON] Fix the architecture condition for the crypto intrinsics

This revision was automatically updated to reflect the committed changes. Closed by commit rC330187: [NEON] Fix the architecture condition for the crypto intrinsics (authored by kosarev, committed by ). Repository: rL LLVM https://reviews.llvm.org/D45669 Files:

[PATCH] D45670: [NEON} Define vfma_n_f32() and vfmaq_n_f32() intrinsics in AArch32 mode

kosarev created this revision. kosarev added reviewers: t.p.northover, rengolin, SjoerdMeijer. kosarev added a project: clang. Herald added subscribers: kristof.beyls, javed.absar. Currently we only support them in AArch64 mode. The AArch64 test cases are in

[PATCH] D45669: [NEON] Fix the architecture condition for the crypto intrinsics

kosarev created this revision. kosarev added reviewers: t.p.northover, rengolin, SjoerdMeijer. kosarev added a project: clang. Herald added subscribers: kristof.beyls, javed.absar. This is rather a cosmetic change as on ARMv7 targets we do not define __ARM_FEATURE_CRYPTO, even if it was

[PATCH] D45668: [NEON] Define vget_high_f16() and vget_low_f16() intrinsics in AArch64 mode only

kosarev created this revision. kosarev added reviewers: t.p.northover, rengolin, SjoerdMeijer. kosarev added a project: clang. Herald added subscribers: kristof.beyls, javed.absar. These are AArch64-specific intrinsics. The patch removes AArch32-mode test cases and maintains AArch64 ones in

[PATCH] D45515: [NEON] Support vrndns_f32 intrinsic

This revision was automatically updated to reflect the committed changes. Closed by commit rC330012: [NEON] Support vrndns_f32 intrinsic (authored by kosarev, committed by ). Repository: rC Clang https://reviews.llvm.org/D45515 Files: include/clang/Basic/arm_neon.td

[PATCH] D45483: [NEON] Support vfma_n and vfms_n intrinsics

This revision was automatically updated to reflect the committed changes. Closed by commit rL329814: [NEON] Support vfma_n and vfms_n intrinsics (authored by kosarev, committed by ). Herald added a subscriber: llvm-commits. Changed prior to commit:

[PATCH] D45483: [NEON] Support vfma_n and vfms_n intrinsics

This revision was automatically updated to reflect the committed changes. Closed by commit rC329814: [NEON] Support vfma_n and vfms_n intrinsics (authored by kosarev, committed by ). Repository: rL LLVM https://reviews.llvm.org/D45483 Files: include/clang/Basic/arm_neon.td

[PATCH] D45515: [NEON] Support vrndns_f32 intrinsic

kosarev created this revision. kosarev added reviewers: t.p.northover, rengolin, SjoerdMeijer. kosarev added a project: clang. Herald added a subscriber: javed.absar. https://reviews.llvm.org/D45515 Files: include/clang/Basic/arm_neon.td lib/CodeGen/CGBuiltin.cpp

[PATCH] D45483: [NEON] Support vfma_n and vfms_n intrinsics

kosarev created this revision. kosarev added reviewers: az, sbaranga, t.p.northover. kosarev added a project: clang. Herald added subscribers: javed.absar, rengolin. https://reviews.llvm.org/D45483 Files: include/clang/Basic/arm_neon.td test/CodeGen/aarch64-neon-2velem.c Index:

[PATCH] D44616: Update existed CodeGen TBAA tests

kosarev added a comment. Looks fine to me. Repository: rC Clang https://reviews.llvm.org/D44616 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D42366: [CodeGen] Fix generation of TBAA tags for may-alias accesses

kosarev added a comment. I think zero would serve better as the unknown-size value. People who are not aware of TBAA internals would guess that since zero-sized accesses make no sense, they are likely to have some special meaning. Similarly, for code that is supposed to process the size fields

[PATCH] D42366: [CodeGen] Fix generation of TBAA tags for may-alias accesses

This revision was automatically updated to reflect the committed changes. Closed by commit rC325575: [CodeGen] Fix generation of TBAA tags for may-alias accesses (authored by kosarev, committed by ). Repository: rC Clang https://reviews.llvm.org/D42366 Files: lib/CodeGen/CodeGenModule.cpp

[PATCH] D42366: [CodeGen] Fix generation of TBAA tags for may-alias accesses

kosarev updated this revision to Diff 134877. kosarev edited the summary of this revision. kosarev added a comment. Do not change the size field of may-alias TBAA descriptors for now. https://reviews.llvm.org/D42366 Files: lib/CodeGen/CodeGenModule.cpp lib/CodeGen/CodeGenTBAA.cpp

[PATCH] D43181: [CodeGen] Initialize large arrays by copying from a global

This revision was automatically updated to reflect the committed changes. Closed by commit rL325478: [CodeGen] Initialize large arrays by copying from a global (authored by kosarev, committed by ). Changed prior to commit: https://reviews.llvm.org/D43181?vs=134590=134865#toc Repository: rL

[PATCH] D42366: [CodeGen] Fix generation of TBAA tags for may-alias accesses

kosarev added a comment. John, the patch hangs for about a month now and Hal still seems to be extremely busy. Can we commit this with the constant to be whatever you think is most correct and the fix it as needed and if needed later? At this very moment we don't analyze the size fields of

[PATCH] D43181: [CodeGen] Initialize large arrays by copying from a global

kosarev updated this revision to Diff 134590. kosarev added a comment. Updated to consider the total size of the explicit initializers instead of their number. The threshold value is adjusted respectively. https://reviews.llvm.org/D43181 Files: lib/CodeGen/CGExprAgg.cpp

[PATCH] D43187: [AST] Refine the condition for element-dependent array fillers

This revision was automatically updated to reflect the committed changes. Closed by commit rL325120: [AST] Refine the condition for element-dependent array fillers (authored by kosarev, committed by ). Changed prior to commit: https://reviews.llvm.org/D43187?vs=133846=134200#toc Repository:

[PATCH] D43181: [CodeGen] Initialize large arrays by copying from a global

kosarev updated this revision to Diff 134042. kosarev added a comment. Improved as suggested to cover all trivially-copyable types. https://reviews.llvm.org/D43181 Files: lib/CodeGen/CGExprAgg.cpp test/CodeGen/init.c Index: test/CodeGen/init.c

[PATCH] D43187: [AST] Refine the condition for element-dependent array fillers

kosarev added a comment. John, maybe you can review this or suggest some other reviewers? Thanks. Repository: rL LLVM https://reviews.llvm.org/D43187 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D43181: [CodeGen] Initialize large arrays by copying from a global

kosarev added inline comments. Comment at: lib/CodeGen/CGExprAgg.cpp:421 + // with explicit initializers should be large enough. + if (NumInitElements > 8 && elementType->isBuiltinType()) { +CodeGen::CodeGenModule = CGF.CGM; rjmccall wrote: > Is there a

[PATCH] D43187: [AST] Refine the condition for element-dependent array fillers

kosarev created this revision. kosarev added a reviewer: rsmith. kosarev added a project: clang. Herald added a subscriber: llvm-commits. This patch fixes clang to not consider braced initializers for aggregate elements of arrays to be potentially dependent on the indices of the initialized

[PATCH] D43181: [CodeGen] Initialize large arrays by copying from a global

kosarev created this revision. kosarev added a reviewer: rjmccall. kosarev added a project: clang. Herald added a subscriber: llvm-commits. Currently, clang compiles explicit initializers for array elements into series of store instructions. For large arrays of built-in types this results in

[PATCH] D42366: [CodeGen] Fix generation of TBAA tags for may-alias accesses

kosarev added a comment. Hal, a friendly ping. Repository: rL LLVM https://reviews.llvm.org/D42366 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D42366: [CodeGen] Fix generation of TBAA tags for may-alias accesses

kosarev added a comment. Hal, will you please take a look? Repository: rL LLVM https://reviews.llvm.org/D42366 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D42366: [CodeGen] Fix generation of TBAA tags for may-alias accesses

kosarev added inline comments. Comment at: lib/CodeGen/CodeGenTBAA.h:67 /* BaseType= */ nullptr, /* AccessType= */ nullptr, - /* Offset= */ 0, /* Size= */ 0); + /* Offset= */ 0, /* Size= */ UINT64_MAX);

[PATCH] D41539: [CodeGen] Decorate aggregate accesses with TBAA tags

This revision was automatically updated to reflect the committed changes. Closed by commit rC323421: [CodeGen] Decorate aggregate accesses with TBAA tags (authored by kosarev, committed by ). Changed prior to commit: https://reviews.llvm.org/D41539?vs=131257=131434#toc Repository: rC Clang

[PATCH] D41539: [CodeGen] Decorate aggregate accesses with TBAA tags

This revision was automatically updated to reflect the committed changes. Closed by commit rL323421: [CodeGen] Decorate aggregate accesses with TBAA tags (authored by kosarev, committed by ). Herald added a subscriber: llvm-commits. Changed prior to commit:

[PATCH] D41539: [CodeGen] Decorate aggregate accesses with TBAA tags

kosarev updated this revision to Diff 131257. kosarev added a comment. The copying functions changed to take LValues. It seems Address-taking versions are not very useful so we don't bother with overloading. https://reviews.llvm.org/D41539 Files: lib/CodeGen/CGAtomic.cpp

[PATCH] D41539: [CodeGen] Decorate aggregate accesses with TBAA tags

kosarev updated this revision to Diff 130599. kosarev added a comment. - Supported propagation of TBAA info for aggregate transfers. - Added tests on copying of `may_alias` aggregates. https://reviews.llvm.org/D41539 Files: lib/CodeGen/CGAtomic.cpp lib/CodeGen/CGCall.cpp

[PATCH] D41539: [CodeGen] Decorate aggregate accesses with TBAA tags

kosarev added a comment. OK, I'm reading your response so that this patch may significantly increase the number of cases where we propagate TBAA tags from memory-transfer intrinsic calls, which means potentially more cases where ignoring may_alias would lead to problems. If so, then I tend to

[PATCH] D41539: [CodeGen] Decorate aggregate accesses with TBAA tags

kosarev added a comment. Sure, but since it is not a trivial change we could first replace 'tbaa.struct' with 'tbaa' and then decide how to attach multiple TBAA tags to an instruction. I didn't think enough about it, but one way to do that is to have different kinds of TBAA tags for read and

[PATCH] D41562: [CodeGen] Generate TBAA info on passing arguments and returning values

kosarev created this revision. kosarev added reviewers: rjmccall, hfinkel. kosarev added a project: clang. We only do this for the new format as the old format does not allow to represent accesses to aggregates. This patch significantly improves TBAA coverage on -O1 builds. Repository: rL

[PATCH] D41547: [CodeGen] Fix TBAA info for accesses to members of base classes

kosarev created this revision. kosarev added reviewers: rjmccall, hfinkel. kosarev added a project: clang. Resolves: Bug 35724 - regression (r315984): fatal error: error in backend: Broken function found (Did not see access type in access path!) https://bugs.llvm.org/show_bug.cgi?id=35724

[PATCH] D41539: [CodeGen] Decorate aggregate accesses with TBAA tags

kosarev updated this revision to Diff 128017. kosarev added a comment. Reworked to not add another test file. https://reviews.llvm.org/D41539 Files: lib/CodeGen/CGExprAgg.cpp test/CodeGen/tbaa-struct.cpp Index: test/CodeGen/tbaa-struct.cpp

[PATCH] D41399: [CodeGen] Represent array members in new-format TBAA type descriptors

This revision was automatically updated to reflect the committed changes. Closed by commit rL321352: [CodeGen] Represent array members in new-format TBAA type descriptors (authored by kosarev, committed by ). Changed prior to commit: https://reviews.llvm.org/D41399?vs=127742=127994#toc

[PATCH] D41394: [CodeGen] Support generation of TBAA info in the new format

This revision was automatically updated to reflect the committed changes. Closed by commit rL321351: [CodeGen] Support generation of TBAA info in the new format (authored by kosarev, committed by ). Changed prior to commit: https://reviews.llvm.org/D41394?vs=127838=127993#toc Repository: rL

[PATCH] D41394: [CodeGen] Support generation of TBAA info in the new format

kosarev updated this revision to Diff 127838. kosarev added a comment. Updated as suggested. https://reviews.llvm.org/D41394 Files: lib/CodeGen/CodeGenTBAA.cpp test/CodeGen/tbaa.cpp Index: test/CodeGen/tbaa.cpp === ---

[PATCH] D41452: [CodeGen] Fix access sizes in new-format TBAA tags

This revision was automatically updated to reflect the committed changes. Closed by commit rL321250: [CodeGen] Fix access sizes in new-format TBAA tags (authored by kosarev, committed by ). Changed prior to commit: https://reviews.llvm.org/D41452?vs=127724=127834#toc Repository: rL LLVM

[PATCH] D41399: [CodeGen] Represent array members in new-format TBAA type descriptors

kosarev updated this revision to Diff 127742. kosarev added a comment. - Fixed the access size. - Added the suggested tests. https://reviews.llvm.org/D41399 Files: lib/CodeGen/CodeGenTBAA.cpp test/CodeGen/tbaa-array.cpp Index: test/CodeGen/tbaa-array.cpp

[PATCH] D41399: [CodeGen] Represent array members in new-format TBAA type descriptors

kosarev added inline comments. Comment at: test/CodeGen/tbaa-array.cpp:24 +// CHECK-DAG: [[TAG_A_i]] = !{[[TYPE_A:!.*]], [[TYPE_int:!.*]], i64 0, i64 4} +// CHECK-DAG: [[TAG_C_i]] = !{[[TYPE_C:!.*]], [[TYPE_int:!.*]], i64 0, i64 16} +// CHECK-DAG: [[TYPE_A]] =

[PATCH] D41399: [CodeGen] Represent array members in new-format TBAA type descriptors

kosarev added inline comments. Comment at: test/CodeGen/tbaa-array.cpp:24 +// CHECK-DAG: [[TAG_A_i]] = !{[[TYPE_A:!.*]], [[TYPE_int:!.*]], i64 0, i64 4} +// CHECK-DAG: [[TAG_C_i]] = !{[[TYPE_C:!.*]], [[TYPE_int:!.*]], i64 0, i64 16} +// CHECK-DAG: [[TYPE_A]] =

[PATCH] D41452: [CodeGen] Fix access sizes in new-format TBAA tags

kosarev created this revision. kosarev added reviewers: rjmccall, hfinkel. kosarev added a project: clang. The new format requires to specify both the type of the access and its size. This patch fixes setting access sizes for TBAA tags that denote accesses to structure members. This fix affects

[PATCH] D41394: [CodeGen] Support generation of TBAA info in the new format

kosarev updated this revision to Diff 127722. kosarev added a comment. Added tests. https://reviews.llvm.org/D41394 Files: lib/CodeGen/CodeGenTBAA.cpp test/CodeGen/tbaa.cpp Index: test/CodeGen/tbaa.cpp === ---

[PATCH] D41394: [CodeGen] Support generation of TBAA info in the new format

kosarev added a comment. There are more tests that we will need to rewrite during migration to the new format than I think we could handle with a single patch. I was going to address them with separate patches. Or, I can rewrite some most basic of them as part of this patch. Repository: rL

[PATCH] D41399: [CodeGen] Represent array members in new-format TBAA type descriptors

kosarev created this revision. kosarev added reviewers: rjmccall, hfinkel. kosarev added a project: clang. Now that in the new TBAA format we allow access types to be of any object types, including aggregate ones, it becomes critical to specify types of all sub-objects such aggregates comprise

[PATCH] D41394: [CodeGen] Support generation of TBAA info in the new format

kosarev created this revision. kosarev added reviewers: rjmccall, hfinkel. kosarev added a project: clang. Now that the MDBuilder helpers generating TBAA type and access descriptors in the new format are in place, we can teach clang to use them when requested. Repository: rL LLVM

[PATCH] D39455: [CodeGen] Add initial support for union members in TBAA

kosarev added a comment. Responded here: https://bugs.llvm.org/show_bug.cgi?id=26603 Repository: rL LLVM https://reviews.llvm.org/D39455 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D39455: [CodeGen] Add initial support for union members in TBAA

This revision was automatically updated to reflect the committed changes. Closed by commit rL319413: [CodeGen] Add initial support for union members in TBAA (authored by kosarev). Changed prior to commit: https://reviews.llvm.org/D39455?vs=124726=124883#toc Repository: rL LLVM

[PATCH] D39455: [CodeGen] Add initial support for union members in TBAA

kosarev updated this revision to Diff 124726. kosarev added a comment. - Added tests with nested unions. - Rebased. https://reviews.llvm.org/D39455 Files: lib/CodeGen/CGExpr.cpp lib/CodeGen/CodeGenModule.h lib/CodeGen/CodeGenTBAA.cpp lib/CodeGen/CodeGenTBAA.h

[PATCH] D40176: [CodeGen] Collect information about sizes of accesses and access types for TBAA

This revision was automatically updated to reflect the committed changes. Closed by commit rL319012: [CodeGen] Collect information about sizes of accesses and access types for TBAA (authored by kosarev). Changed prior to commit: https://reviews.llvm.org/D40176?vs=124084=124332#toc Repository:

[PATCH] D40176: [CodeGen] Collect information about sizes of accesses and access types for TBAA

kosarev updated this revision to Diff 124084. kosarev added a comment. Rebased. https://reviews.llvm.org/D40176 Files: lib/CodeGen/CGClass.cpp lib/CodeGen/CodeGenModule.cpp lib/CodeGen/CodeGenModule.h lib/CodeGen/CodeGenTBAA.cpp lib/CodeGen/CodeGenTBAA.h Index:

  1   2   >