[PATCH] D59603: [PR40707][PR41011][OpenCL] Allow addr space spelling without double underscore in C++ mode

jsji added a comment. Looks like this may cause some unexpected failures. See https://bugs.llvm.org/show_bug.cgi?id=41247 for more details. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D59603/new/ https://reviews.llvm.org/D59603 __

[PATCH] D59924: [PowerPC][Clang] Port MMX intrinsics and basic test cases to Power

jsji requested changes to this revision. jsji added a comment. This revision now requires changes to proceed. Since we are adding new headers that is not originated by this patch, please also update the description (and commit message) to include who actually contributed to `mmintrin.h` and othe

[PATCH] D59924: [PowerPC] [Clang] Port MMX intrinsics and basic test cases to Power

jsji requested changes to this revision. jsji added a comment. This revision now requires changes to proceed. Please make sure you run all test before updating patch! Comment at: clang/test/CodeGen/ppc-mmintrin.c:23 +// CHECK: i64 @_mm_packs_pu16(i64 [[REG1:[0-9a-zA-Z_%.]+]], i

[PATCH] D59924: [PowerPC] [Clang] Port MMX intrinsics and basic test cases to Power

jsji accepted this revision. jsji added a comment. This revision is now accepted and ready to land. LGTM. Thanks for porting! Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D59924/new/ https://reviews.llvm.org/D59924 _

[PATCH] D59924: [PowerPC] [Clang] Port MMX intrinsics and basic test cases to Power

jsji added a comment. In D59924#1475363 , @MaskRay wrote: > May I ask why the mmintrin.h emulation layer must be added to the clang > resource directory? Why can't it be provided as a standalone library like > https://github.com/intel/ARM_NEON_2_x86_SSE

[PATCH] D59924: [PowerPC] [Clang] Port MMX intrinsics and basic test cases to Power

jsji added a comment. Thanks @MaskRay for help fixing and review! Comment at: lib/Headers/ppc_wrappers/mmintrin.h:3 + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to

[PATCH] D59924: [PowerPC] [Clang] Port MMX intrinsics and basic test cases to Power

jsji added inline comments. Comment at: lib/Headers/ppc_wrappers/mmintrin.h:3 + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal jsji wrote: > Mas

[PATCH] D62121: [PowerPC] [Clang] Port SSE intrinsics to PowerPC

jsji accepted this revision. jsji added a comment. This revision is now accepted and ready to land. LGTM. Thanks for fixing. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D62121/new/ https://reviews.llvm.org/D62121 ___ cfe-commits mailing li

[Diffusion] rL358949: [PowerPC] [Clang] Port MMX intrinsics and basic test cases to Power

jsji added inline comments. /cfe/trunk/lib/Driver/ToolChains/PPCLinux.cpp:26 yes, it is currently Linux only. I think we can extend it to include other OS like AIX later if needed. https://reviews.llvm.org/rL358949 ___ cfe-commits mailing list cfe-

[PATCH] D62569: [PowerPC] [Clang] Port SSE2 intrinsics to PowerPC

jsji accepted this revision. jsji added a comment. This revision is now accepted and ready to land. LGTM. Thanks for porting. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D62569/new/ https://reviews.llvm.org/D62569 ___ cfe-commits mailing l

[PATCH] D64849: Add platform guards to PPC vector intrinsics headers.

jsji accepted this revision. jsji added a comment. This revision is now accepted and ready to land. LGTM. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D64849/new/ https://reviews.llvm.org/D64849 ___ cfe-commits maili

[PATCH] D65630: [PowerPC] Port SSE3, SSSE3 and SSE4 intrinsics to PowerPC.

jsji accepted this revision. jsji added a comment. This revision is now accepted and ready to land. LGTM. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D65630/new/ https://reviews.llvm.org/D65630 ___ cfe-commits maili

[PATCH] D63636: [PowerPC][Altivec] Fix offsets for vec_xl and vec_xst

jsji added a comment. A few questions. Comment at: lib/Headers/altivec.h:16364 signed short *__ptr) { - return *(unaligned_vec_sshort *)(__ptr + __offset); + signed char *Adjusted = (signed char *)__ptr + __offset; + ret

[PATCH] D63636: [PowerPC][Altivec] Fix offsets for vec_xl and vec_xst

jsji accepted this revision. jsji added a comment. This revision is now accepted and ready to land. LGTM. Thanks for fixing. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D63636/new/ https://reviews.llvm.org/D63636 __

[PATCH] D64024: [PowerPC][Altivec] Emit correct builtin for single precision vec_all_ne

jsji accepted this revision. jsji added a comment. This revision is now accepted and ready to land. LGTM. BTW, looks like we are missing test cases all `vector float` `vec_all*` and `vec_any*` and also non-vsx path? Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D

[PATCH] D64119: [PowerPC] Support constraint code "ww"

jsji added a comment. It is great to add `ww` for compatibility. However if we are going to add `ww`, looks like we should update `ws` as well? Comment at: clang/lib/Basic/Targets/PPC.h:211 + case 's': // VSX vector register to hold scalar double data + case 'w': //

[PATCH] D64119: [PowerPC] Support constraint code "ww"

jsji accepted this revision. jsji added a comment. This revision is now accepted and ready to land. LGTM. Thanks for investigating GCC behavior. Comment at: clang/lib/Basic/Targets/PPC.h:211 + case 's': // VSX vector register to hold scalar double data + case 'w': //

[PATCH] D62121: [PowerPC] [Clang] Port SSE intrinsics to PowerPC

jsji added subscribers: MaskRay, rjmccall, aaron.ballman, rnk, jyknight. jsji added a comment. Add more subscribers, reviews and comments are welcome and appreciated. Thanks. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D62121/new/ https://reviews.llvm.org/D62121

[PATCH] D62121: [PowerPC] [Clang] Port SSE intrinsics to PowerPC

jsji accepted this revision. jsji added a comment. This revision is now accepted and ready to land. LGTM. Thanks for porting. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D62121/new/ https://reviews.llvm.org/D62121 _

[PATCH] D62121: [PowerPC] [Clang] Port SSE intrinsics to PowerPC

jsji reopened this revision. jsji added a comment. This revision is now accepted and ready to land. This has been reverted in https://reviews.llvm.org/rL361930 . Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D62121/new/ https://reviews.llvm.org/D62121 _

[PATCH] D62121: [PowerPC] [Clang] Port SSE intrinsics to PowerPC

jsji requested changes to this revision. jsji added a comment. This revision now requires changes to proceed. TEST 'Clang :: CodeGen/ppc-mm-malloc.c' FAILED Script: -- : 'RUN: at line 9'; /home/buildslave/buildslave1a/clang-with-lto-ubuntu/buil

[PATCH] D83915: [PowerPC] Remove QPX/A2Q BGQ/BGP support

jsji added a comment. In D83915#2155650 , @hubert.reinterpretcast wrote: > Should we indicate planned removal in the Release Notes for version 11 and > actual removal in those for version 12? Good suggestion. https://reviews.llvm.org/D83968 for adding

[PATCH] D83915: [PowerPC] Remove QPX/A2Q BGQ/BGP support

jsji added a comment. In D83915#2156346 , @jsji wrote: > In D83915#2155650 , > @hubert.reinterpretcast wrote: > > > Should we indicate planned removal in the Release Notes for version 11 and > > actual removal in

[PATCH] D83915: [PowerPC] Remove QPX/A2Q BGQ/BGP support

jsji added a comment. Ping ... Any other comments? Thanks. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D83915/new/ https://reviews.llvm.org/D83915 ___ cfe-commits mailing list cfe-commits@lists.llvm.o

[PATCH] D78308: [NFC][PowerPC] Refactor ppcUserFeaturesCheck()

jsji added inline comments. Comment at: clang/lib/Basic/Targets/PPC.cpp:243 + bool Passed = CheckVSXSubfeature("+power8-vector", "-mpower8-vector"); + Passed |= CheckVSXSubfeature("+direct-move", "-mdirect-move"); + Passed |= CheckVSXSubfeature("+float128", "-mfloat128");

[PATCH] D78308: [NFC][PowerPC] Refactor ppcUserFeaturesCheck()

jsji accepted this revision. jsji added a comment. LGTM. Since there is an improvement for multi errors, we should remove `[NFC]` in title and commit messages. Thanks. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D78308/new/ https://reviews.llvm.o

[PATCH] D79014: [clangd] Move non-clang base pieces into separate support/ lib.

jsji added a comment. Herald added a subscriber: wuzish. Looks like this is causing buildbot failure when built with `-DBUILD_SHARED_LIBS=ON`. http://lab.llvm.org:8011/builders/clang-ppc64le-rhel/builds/3077/steps/build%20stage%201/logs/stdio FAILED: lib/libclangDaemon.so.11git : && /home/b

[PATCH] D79014: [clangd] Move non-clang base pieces into separate support/ lib.

jsji added a comment. Thanks all, yes, it is passing now. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D79014/new/ https://reviews.llvm.org/D79014 ___ cfe-commits mailing list cfe-commits@lists.llvm.or

[PATCH] D79653: [sanitizer] Enable whitelist/blacklist in new PM

jsji created this revision. jsji added reviewers: tuktuk, vitalybuka, morehouse, kcc. Herald added subscribers: llvm-commits, Sanitizers, cfe-commits, hiraditya. Herald added projects: clang, Sanitizers, LLVM. https://reviews.llvm.org/D63616 added `-fsanitize-coverage-whitelist` and `-fsanitize-co

[PATCH] D79653: [sanitizer] Enable whitelist/blacklist in new PM

This revision was automatically updated to reflect the committed changes. Closed by commit rGa72b9dfd45cd: [sanitizer] Enable whitelist/blacklist in new PM (authored by jsji). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D79653/new/ https://reviews.

[PATCH] D77249: [MSan] Pass command line options to MSan with new pass manager

jsji added a comment. Herald added a subscriber: wuzish. NOTE that https://reviews.llvm.org/D79445 has committed a similar fix and fixed most of the failures. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D77249/new/ https://reviews.llvm.org/D77249

[PATCH] D122377: [PowerPC][Linux] Support 16-byte lock free atomics on pwr8 and up

jsji added inline comments. Comment at: clang/test/CodeGen/PowerPC/atomic-alignment.c:7 +// RUN: -emit-llvm -o - %s | FileCheck %s --check-prefixes=PPC,PPC64-PWR8 // RUN: %clang_cc1 -verify -triple powerpc64-unknown-aix -emit-llvm -o - %s | \ // RUN: FileCheck %s --check-pr

[PATCH] D106344: [PowerPC] Implement XL compatible behavior of __compare_and_swap

jsji added a comment. Doesn't look like good enough to me, the code sequence generated is not clean enough. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D106344/new/ https://reviews.llvm.org/D106344 __

[PATCH] D106344: [PowerPC] Implement XL compatible behavior of __compare_and_swap

jsji accepted this revision as: jsji. jsji added a comment. This revision is now accepted and ready to land. > that looks like a historical issue which exists for 13yrs Hmm... OK.. Let us leave it as it is for now. Thanks for the new tests. Repository: rG LLVM Github Monorepo CHANGES SINCE

[PATCH] D105527: libclang.so: Make SONAME independent from LLVM version

jsji added inline comments. Comment at: clang/tools/libclang/CMakeLists.txt:79 + set(LLVM_EXPORTED_SYMBOL_FILE) + set(USE_VERSION_SCRIPT TRUE) +endif() I think we should check LLVM_HAVE_LINK_VERSION_SCRIPT here, there are platforms that does NOT support VERSIO

[PATCH] D106914: [libclang] Check LLVM_HAVE_LINK_VERSION_SCRIPT

jsji created this revision. jsji added reviewers: PowerPC, tstellar, MaskRay, hubert.reinterpretcast. Herald added a subscriber: mgorny. jsji requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. There are some platform that might not have versio

[PATCH] D105527: libclang.so: Make SONAME independent from LLVM version

jsji added inline comments. Comment at: clang/tools/libclang/CMakeLists.txt:79 + set(LLVM_EXPORTED_SYMBOL_FILE) + set(USE_VERSION_SCRIPT TRUE) +endif() jsji wrote: > I think we should check LLVM_HAVE_LINK_VERSION_SCRIPT here, there are > platforms that does NO

[PATCH] D106914: [libclang] Check LLVM_HAVE_LINK_VERSION_SCRIPT

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGe07f9342be71: [libclang] Check LLVM_HAVE_LINK_VERSION_SCRIPT (authored by jsji). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D106920: [AIX] Update fetch_and_add type

jsji created this revision. Herald added subscribers: steven.zhang, kbarton, nemanjai. jsji requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. It turns out that AIX kernerl are definiing int instead of unsing int for fectch_and_add. Legacy XL

[PATCH] D106920: [AIX] Update fetch_and_add type

jsji updated this revision to Diff 362197. jsji edited the summary of this revision. jsji added a comment. Update tests. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D106920/new/ https://reviews.llvm.org/D106920 Files: clang/include/clang/Basic/

[PATCH] D106920: [AIX] Update fetch_and_add type

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGedbdf8e5b5aa: [AIX] Update fetch_and_add type (authored by jsji). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews

[PATCH] D107077: [PowerPC] Fix return type of XL compat CAS

jsji accepted this revision as: jsji. jsji 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/D107077/new/ https://reviews.llvm.org/D107077 _

[PATCH] D121441: [PowerPC][NFC] Add atomic alignments and ops tests for powerpc

jsji accepted this revision as: jsji. jsji 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/D121441/new/ https://reviews.llvm.org/D121441 _

[PATCH] D123498: [clang] Adding Platform/Architecture Specific Resource Header Installation Targets

jsji added a comment. @qiongsiwu1 Please also add reviewers from other targets. Thanks. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D123498/new/ https://reviews.llvm.org/D123498 ___ cfe-commits mailing

[PATCH] D123372: [PowerPC] add XLC compat builtin __abs

jsji accepted this revision as: jsji. jsji 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/D123372/new/ https://reviews.llvm.org/D123372 _

[PATCH] D125386: [clang][ppc] Creating Seperate Install Target for PPC htm Headers

jsji accepted this revision. jsji added a comment. This revision is now accepted and ready to land. LGTM Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D125386/new/ https://reviews.llvm.org/D125386 ___ cfe

[PATCH] D66699: [PowerPC][Altivec] Fix constant argument for vec_dss

jsji created this revision. jsji added reviewers: nemanjai, hfinkel. Herald added subscribers: cfe-commits, shchenz, MaskRay, kbarton. Herald added a project: clang. jsji added a reviewer: PowerPC. Herald added a subscriber: wuzish. This is similar to vec_ct* in https://reviews.llvm.org/rL304205.

[PATCH] D66748: [PowerPC][Altivec][Clang] Check compile-time constant for vec_dst*

jsji created this revision. jsji added reviewers: nemanjai, hfinkel, echristo. Herald added subscribers: cfe-commits, shchenz, MaskRay, kbarton. Herald added a project: clang. jsji added a reviewer: PowerPC. Herald added a subscriber: wuzish. This is follow up of https://reviews.llvm.org/D66699. W

[PATCH] D66699: [PowerPC][Altivec] Fix constant argument for vec_dss

jsji added a comment. Ping.. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D66699/new/ https://reviews.llvm.org/D66699 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-

[PATCH] D66748: [PowerPC][Altivec][Clang] Check compile-time constant for vec_dst*

jsji added a comment. Ping.. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D66748/new/ https://reviews.llvm.org/D66748 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-

[PATCH] D66699: [PowerPC][Altivec] Fix constant argument for vec_dss

jsji updated this revision to Diff 218479. jsji added a comment. Add range check and test. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D66699/new/ https://reviews.llvm.org/D66699 Files: clang/include/clang/Basic/BuiltinsPPC.def clang/lib/Head

[PATCH] D66699: [PowerPC][Altivec] Fix constant argument for vec_dss

jsji marked 2 inline comments as done. jsji added inline comments. Comment at: clang/test/CodeGen/builtins-ppc-error.c:78 +void testDSS(int index) { + vec_dss(index); //expected-error {{argument to '__builtin_altivec_dss' must be a constant integer}} + wuzish w

[PATCH] D66748: [PowerPC][Altivec][Clang] Check compile-time constant for vec_dst*

jsji updated this revision to Diff 218485. jsji added a comment. Add range check and tests. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D66748/new/ https://reviews.llvm.org/D66748 Files: clang/include/clang/Basic/BuiltinsPPC.def clang/lib/Sem

[PATCH] D66699: [PowerPC][Altivec] Fix constant argument for vec_dss

This revision was automatically updated to reflect the committed changes. jsji marked an inline comment as done. Closed by commit rL370902: [PowerPC][Altivec] Fix constant argument for vec_dss (authored by jsji, committed by ). Herald added a project: LLVM. Herald added a subscriber: llvm-commits.

[PATCH] D66748: [PowerPC][Altivec][Clang] Check compile-time constant for vec_dst*

jsji updated this revision to Diff 218696. jsji added a comment. Rebase to ToT. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D66748/new/ https://reviews.llvm.org/D66748 Files: clang/include/clang/Basic/BuiltinsPPC.def clang/lib/Sema/SemaChecki

[PATCH] D66748: [PowerPC][Altivec][Clang] Check compile-time constant for vec_dst*

This revision was automatically updated to reflect the committed changes. Closed by commit rL370912: [PowerPC][Altivec][Clang] Check compile-time constant for vec_dst* (authored by jsji, committed by ). Herald added a project: LLVM. Herald added a subscriber: llvm-commits. Changed prior to commit

[PATCH] D72707: [clang][OpenCL] Fix covered switch warning

jsji created this revision. jsji added a reviewer: svenvh. Herald added subscribers: cfe-commits, Anastasia, yaxunl. Herald added a project: clang. jsji edited the summary of this revision. -Werror clang build is broken now. tools/clang/lib/Sema/OpenCLBuiltins.inc:11824:5: error: default label in

[PATCH] D72707: [clang][OpenCL] Fix covered switch warning

This revision was automatically updated to reflect the committed changes. Closed by commit rGe2b8e2113a49: [clang][OpenCL] Fix covered switch warning (authored by jsji). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D72707/new/ https://reviews.llvm.o

[PATCH] D89242: [PowerPC] [Clang] Port SSE4.1-compatible insert intrinsics

jsji accepted this revision as: jsji. jsji 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/D89242/new/ https://reviews.llvm.org/D89242 ___

[PATCH] D92329: [PowerPC][Clang] Remove QPX support

jsji created this revision. Herald added subscribers: cfe-commits, dang, shchenz, kbarton, nemanjai. Herald added a project: clang. jsji requested review of this revision. Herald added a reviewer: jdoerfert. Herald added a subscriber: sstefan1. Clean up QPX code in clang missed in https://reviews.

[PATCH] D92329: [PowerPC][Clang] Remove QPX support

jsji updated this revision to Diff 308405. jsji added a comment. Herald added a project: LLVM. Herald added a subscriber: llvm-commits. Fix llvm test as well Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D92329/new/ https://reviews.llvm.org/D92329

[PATCH] D92329: [PowerPC][Clang] Remove QPX support

This revision was automatically updated to reflect the committed changes. Closed by commit rGb49b8f096c03: [PowerPC][Clang] Remove QPX support (authored by jsji). Changed prior to commit: https://reviews.llvm.org/D92329?vs=308405&id=309908#toc Repository: rG LLVM Github Monorepo CHANGES SIN

[PATCH] D102560: [AIX][clang-repl][test] Mark unsupported pending XCOFF64 integrated-as

jsji accepted this revision. jsji 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/D102560/new/ https://reviews.llvm.org/D102560 __

[PATCH] D102647: [Driver][test] Don't assume integrated-as

jsji created this revision. jsji added reviewers: PowerPC, thakis, MaskRay, phosek, hubert.reinterpretcast. jsji requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. The tests of fdebug-compilation-dir and -ffile-compilation-dir for `-x assemble

[PATCH] D102647: [Driver][test] Don't assume integrated-as

jsji added a comment. In D102647#2764376 , @thakis wrote: > Are there any platforms left that don't default to integrated as? Which ones? > Is there a tracking bug for switching them? Thanks! Oh, I should have mentioned that in description, this is expo

[PATCH] D102647: [Driver][test] Don't assume integrated-as

This revision was automatically updated to reflect the committed changes. Closed by commit rG82b52812471c: [Driver][test] Don't assume integrated-as (authored by jsji). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D102647/new/ https://reviews.llvm.o

[PATCH] D102654: [DebugInfo][test] Check specific func name to ignore codegen differences

jsji created this revision. jsji added reviewers: PowerPC, dblaikie, shchenz. jsji requested review of this revision. Herald added a project: clang. Herald added a subscriber: cfe-commits. We use `CHECK-LABEL: define` to divide input stream into functions, this works well on most platforms. But t

[PATCH] D102654: [DebugInfo][test] Check specific func name to ignore codegen differences

jsji updated this revision to Diff 346011. jsji added a comment. Address comments. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D102654/new/ https://reviews.llvm.org/D102654 Files: clang/test/CodeGenCXX/debug-info-line.cpp Index: clang/test/Co

[PATCH] D102654: [DebugInfo][test] Check specific func name to ignore codegen differences

jsji marked an inline comment as done. jsji added inline comments. Comment at: clang/test/CodeGenCXX/debug-info-line.cpp:149 -// CHECK-LABEL: define +// CHECK-LABEL: f11 __complex double f11() { dblaikie wrote: > Probably flesh that out a bit. Maybe `define{{.

[PATCH] D102654: [DebugInfo][test] Check specific func name to ignore codegen differences

This revision was automatically updated to reflect the committed changes. jsji marked an inline comment as done. Closed by commit rG7d6449322ecb: [DebugInfo][test] Check specific func name to ignore codegen differences (authored by jsji). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST

[PATCH] D102724: Revert "[AIX] Avoid structor alias; die before bad alias codegen"

jsji added a comment. > because AIX now has an alias implementation. Can we please include the patches or commits that implement the alias? Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D102724/new/ https://reviews.llvm.org/D102724 ___

[PATCH] D102814: [AIX] Print printable byte list as quoted string

jsji created this revision. jsji added reviewers: PowerPC, hubert.reinterpretcast, DiggerLin, xingxue, shchenz. Herald added subscribers: hiraditya, nemanjai. jsji requested review of this revision. Herald added projects: clang, LLVM. Herald added subscribers: llvm-commits, cfe-commits. .byte sup

[PATCH] D102724: Revert "[AIX] Avoid structor alias; die before bad alias codegen"

jsji added a comment. In D102724#2769756 , @jasonliu wrote: > Good idea, fyi, this is the patch: https://reviews.llvm.org/D83252 Thanks. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D102724/new/ https://r

[PATCH] D102814: [AIX] Print printable byte list as quoted string

jsji updated this revision to Diff 346794. jsji added a comment. Address comments. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D102814/new/ https://reviews.llvm.org/D102814 Files: clang/test/CodeGenCXX/debug-info-byval.cpp llvm/include/llvm/M

[PATCH] D102814: [AIX] Print printable byte list as quoted string

jsji added inline comments. Comment at: llvm/lib/MC/MCAsmStreamer.cpp:1008 + for (const unsigned char C : Data) { +if (!isPrint(C)) + return false; This still need some slight change to deal with the ending 0 so that we can handle .string. Repository

[PATCH] D102814: [AIX] Print printable byte list as quoted string

jsji updated this revision to Diff 346830. jsji added a comment. Update .string handling. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D102814/new/ https://reviews.llvm.org/D102814 Files: clang/test/CodeGenCXX/debug-info-byval.cpp llvm/include

[PATCH] D102814: [AIX] Print printable byte list as quoted string

jsji added inline comments. Comment at: llvm/lib/MC/MCAsmStreamer.cpp:1008 + for (const unsigned char C : Data) { +if (!isPrint(C)) + return false; jsji wrote: > This still need some slight change to deal with the ending 0 so that we can > handle .stri

[PATCH] D102814: [AIX] Print printable byte list as quoted string

jsji updated this revision to Diff 346907. jsji added a comment. Address comments. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D102814/new/ https://reviews.llvm.org/D102814 Files: clang/test/CodeGenCXX/debug-info-byval.cpp llvm/include/llvm/M

[PATCH] D102814: [AIX] Print printable byte list as quoted string

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rGedf4d69d3888: [AIX] Print printable byte list as quoted string (authored by jsji). Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION

[PATCH] D99901: [Driver][test] Test intended target only

jsji added a comment. In D99901#2787034 , @ted wrote: > Why would we not want to support this on other targets, like Arm? Fixes for > AIX shouldn't impact other things. Feel free to update the triple to include arm , as long as the tests focus on gnu to

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji created this revision. jsji added reviewers: PowerPC, Whitney, w2yehia, MaskRay. Herald added subscribers: wenlei, hiraditya, inglorion. jsji requested review of this revision. Herald added projects: clang, LLVM. Herald added subscribers: llvm-commits, cfe-commits. On AIX, we relied on LTO to

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji updated this revision to Diff 374867. jsji added a comment. Restore the limitation of sample profiling. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110422/new/ https://reviews.llvm.org/D110422 Files: clang/lib/Driver/ToolChains/AIX.cpp

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji updated this revision to Diff 374879. jsji added a comment. Address comments. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110422/new/ https://reviews.llvm.org/D110422 Files: clang/lib/Driver/ToolChains/AIX.cpp clang/lib/Driver/ToolChain

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added inline comments. Comment at: llvm/lib/Transforms/Instrumentation/InstrProfiling.cpp:869 + // symbol, so we can not ensure the correctness of the relative CounterPtr, so + // we have to use private linkage for counter and data symbols. + if (TT.isOSBinFormatXCOFF())

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added inline comments. Comment at: llvm/lib/Transforms/Instrumentation/InstrProfiling.cpp:869 + // symbol, so we can not ensure the correctness of the relative CounterPtr, so + // we have to use private linkage for counter and data symbols. + if (TT.isOSBinFormatXCOFF())

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added a comment. > In other binary formats the first weak definition is selected and other weak > definitions are discarded. > Do you mean that AIX ld doesn't pick the first weak definition? No. I think this is exactly what is causing problem here. eg: if we have two weak symbols （weak_fun

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added a comment. Pos_Rel __profc_weak_func notation is just the normal relocation to symbol __profc_weak_func. The relocation here is to calculate the relative pointer (offset) to __profc__weak_func from __profd__weak__func. In example above, the offsets value after binding will be the sa

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added a comment. In D110422#3021535 , @MaskRay wrote: > The description is still unclear. > > Say a.o has a weak `foo, __profc_foo, __profd_foo`, b.o has a weak `foo, > __profc_foo, __profd_foo`. > The linker picks the definitions from `a.o`. In the

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added a comment. We have verified all the profile counters in SPEC , all are OK. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110422/new/ https://reviews.llvm.org/D110422 ___ cfe-commits mailing li

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added a comment. Using to `PrivateLinkage`, all the profc/profd for each weak symbol will be *local* to objects, and all kept in the csect, so we won't have problem. The downside is that we won't be able to discard the duplicated counters and profile data, but those can not be discarded ev

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added a comment. @MaskRay Do you have further comments or alternative solutions? Thanks. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110422/new/ https://reviews.llvm.org/D110422 ___ cfe-commits ma

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added a comment. In D110422#3025133 , @MaskRay wrote: > The description isn't clear why the previous `weak` symbol usage does not > work. Have you verified that it does not work? Yes, we started with the `weak` symbols for `profc`/`profd` first, an

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added a comment. OK, I may not describe the example clearly. Let me use example code with offsets and llvm internal calculations as the example, so that you might be clearer. Let us say we have 2 objects , which both have weak function foo (and some non-weak functions). [2054] m 0x1

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added a comment. > "as binder can NOT discard only part of a csect." this part caused confusion > to me... OK, yeah, csect is special to XCOFF, so this is not the same concept of sections in ELF. > I haven't verified but it is possible that PGO picks the first pair (zero > value) and hav

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji added a comment. > "as binder can NOT discard a subset of a csect."? Thanks, this does looks better! > The linker (binder) doesn't have to discard a subset of csect, but it should > ensure there are not two non-local symbols with the same name. > The latter is what I request. > Without

[PATCH] D110422: [AIX] Enable PGO without LTO

jsji updated this revision to Diff 375688. jsji marked an inline comment as done. jsji added a comment. Address comments. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110422/new/ https://reviews.llvm.org/D110422 Files: clang/lib/Driver/ToolChai

[PATCH] D110422: [AIX] Change the linkage of profiling counter/data to be private

jsji updated this revision to Diff 375709. jsji added a comment. Split the Driver changes to another patch, also fixed the duplicate code. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110422/new/ https://reviews.llvm.org/D110422 Files: clang/te

[PATCH] D110422: [AIX] Change the linkage of profiling counter/data to be private

jsji updated this revision to Diff 375716. jsji added a comment. Remove empty line. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110422/new/ https://reviews.llvm.org/D110422 Files: clang/test/Profile/cxx-templates.cpp llvm/lib/Transforms/Inst

[PATCH] D110422: [AIX] Change the linkage of profiling counter/data to be private

jsji added a comment. @MaskRay Thank you so much for your review and discussion! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D110422/new/ https://reviews.llvm.org/D110422 ___ cfe-commits mailing list cf

[PATCH] D110422: [AIX] Change the linkage of profiling counter/data to be private

This revision was landed with ongoing or failed builds. This revision was automatically updated to reflect the committed changes. Closed by commit rG25c30324e953: [AIX] Change the linkage of profiling counter/data to be private (authored by jsji). Repository: rG LLVM Github Monorepo CHANGES SI

  1   2   3   >