r304127 - IRGen: Add optnone attribute on function during O0

Author: mehdi_amini Date: Mon May 29 00:38:20 2017 New Revision: 304127 URL: http://llvm.org/viewvc/llvm-project?rev=304127&view=rev Log: IRGen: Add optnone attribute on function during O0 Amongst other, this will help LTO to correctly handle/honor files compiled with O0, helping debugging failur

r311857 - Emit static constexpr member as available_externally definition

Author: mehdi_amini Date: Sun Aug 27 13:24:09 2017 New Revision: 311857 URL: http://llvm.org/viewvc/llvm-project?rev=311857&view=rev Log: Emit static constexpr member as available_externally definition By exposing the constant initializer, the optimizer can fold many of these constructs. Differe

r311858 - CMake: only try to find the Z3 package when `CLANG_ANALYZER_BUILD_Z3` is ON

Author: mehdi_amini Date: Sun Aug 27 13:24:23 2017 New Revision: 311858 URL: http://llvm.org/viewvc/llvm-project?rev=311858&view=rev Log: CMake: only try to find the Z3 package when `CLANG_ANALYZER_BUILD_Z3` is ON This avoids the spurious message: Could NOT find Z3 (missing: Z3_LIBRARIES Z3_IN

Re: r311857 - Emit static constexpr member as available_externally definition

Sorry for the inconvenience! I will follow-up in the PR. -- Mehdi 2017-08-28 12:42 GMT-07:00 Hans Wennborg : > I reverted this in r311898 as it caused Chromium builds to fail with > an assertion; see PR34348. > > On Sun, Aug 27, 2017 at 1:24 PM, Mehdi Amini via cfe-comm

r312512 - Emit static constexpr member as available_externally definition

Author: mehdi_amini Date: Mon Sep 4 20:58:35 2017 New Revision: 312512 URL: http://llvm.org/viewvc/llvm-project?rev=312512&view=rev Log: Emit static constexpr member as available_externally definition By exposing the constant initializer, the optimizer can fold many of these constructs. This is

Re: r311857 - Emit static constexpr member as available_externally definition

in the PR. > > -- > Mehdi > > > 2017-08-28 12:42 GMT-07:00 Hans Wennborg : > >> I reverted this in r311898 as it caused Chromium builds to fail with >> an assertion; see PR34348. >> >> On Sun, Aug 27, 2017 at 1:24 PM, Mehdi Amini via cfe-commits >> wrote:

[clang] 57d2d48 - Revert "[DomTree] Replace ChildrenGetter with GraphTraits over GraphDiff."

Author: Mehdi Amini Date: 2020-04-10T06:27:59Z New Revision: 57d2d48399b63c0ef1dda490fdaf28efbb80c2c0 URL: https://github.com/llvm/llvm-project/commit/57d2d48399b63c0ef1dda490fdaf28efbb80c2c0 DIFF: https://github.com/llvm/llvm-project/commit/57d2d48399b63c0ef1dda490fdaf28efbb80c2c0.diff LOG: R

[clang] 0445c64 - [DomTree] Replace ChildrenGetter with GraphTraits over GraphDiff.

Author: Alina Sbirlea Date: 2020-04-10T07:38:53Z New Revision: 0445c64998d14b81f0d3a3182011fc5eae47fa71 URL: https://github.com/llvm/llvm-project/commit/0445c64998d14b81f0d3a3182011fc5eae47fa71 DIFF: https://github.com/llvm/llvm-project/commit/0445c64998d14b81f0d3a3182011fc5eae47fa71.diff LOG:

[clang] bbeeb35 - Revert "[DomTree] Replace ChildrenGetter with GraphTraits over GraphDiff."

Author: Mehdi Amini Date: 2020-04-10T07:44:06Z New Revision: bbeeb35c1ffaf7aeb3241bdfcb69bf4708370ac1 URL: https://github.com/llvm/llvm-project/commit/bbeeb35c1ffaf7aeb3241bdfcb69bf4708370ac1 DIFF: https://github.com/llvm/llvm-project/commit/bbeeb35c1ffaf7aeb3241bdfcb69bf4708370ac1.diff LOG: R

[clang] ed03d94 - Revert "[TLI] Per-function fveclib for math library used for vectorization"

Author: Mehdi Amini Date: 2020-04-11T01:05:01Z New Revision: ed03d9485eb5b502708f8a21ce34fa6da05a9ca5 URL: https://github.com/llvm/llvm-project/commit/ed03d9485eb5b502708f8a21ce34fa6da05a9ca5 DIFF: https://github.com/llvm/llvm-project/commit/ed03d9485eb5b502708f8a21ce34fa6da05a9ca5.diff LOG: R

[clang] d31c9e5 - Change filecheck default to dump input on failure

Author: Mehdi Amini Date: 2020-06-09T18:57:46Z New Revision: d31c9e5a46ee692daf2430b52626afcea1db18ab URL: https://github.com/llvm/llvm-project/commit/d31c9e5a46ee692daf2430b52626afcea1db18ab DIFF: https://github.com/llvm/llvm-project/commit/d31c9e5a46ee692daf2430b52626afcea1db18ab.diff LOG: C

[clang] 72d20b9 - [LLVM] Change isa<> to a variadic function template

Author: Rahul Joshi Date: 2020-06-15T18:46:57Z New Revision: 72d20b9604f65e35eedb3d00098a2c9412953946 URL: https://github.com/llvm/llvm-project/commit/72d20b9604f65e35eedb3d00098a2c9412953946 DIFF: https://github.com/llvm/llvm-project/commit/72d20b9604f65e35eedb3d00098a2c9412953946.diff LOG: [

Re: [PATCH] D24806: [docs] Add ThinLTO user documentation

mehdi_amini accepted this revision. mehdi_amini added a comment. This revision is now accepted and ready to land. Great! Thanks for writing it :) Comment at: docs/CommandGuide/clang.rst:339 @@ +338,3 @@ + The default for :option:`-flto` is :option:`-flto=full`, in which the +

Re: r282089 - [docs] Add ThinLTO user documentation

It is automatically refreshed frequently I believe. If it is not there tomorrow, we should start being worried :) — Mehdi > On Sep 21, 2016, at 10:57 AM, Teresa Johnson wrote: > > Anyone know how to get these changes on http://clang.llvm.org/docs/ > ? Is that site

Re: [PATCH] D24826: [LTO] Add -flto-jobs=N to control backend parallelism

mehdi_amini added a comment. The Gold path looks fine. On OSX, we would have the clang driver relying on a LLVM cl::opt, for which I don't think there is any precedent. CC Duncan for advice. https://reviews.llvm.org/D24826 ___ cfe-commits mailing

Re: [PATCH] D24826: [LTO] Add -flto-jobs=N to control backend parallelism

mehdi_amini added a comment. Also I don't think the same option should be used for the parallel LTO codegen: it actually does not generate the same binary, which should deserve a dedicated opt-in (What if I mix ThinLTO and LTO, and I don't want // codegen?) https://reviews.llvm.org/D24826 _

Re: [PATCH] D24826: [LTO] Add -flto-jobs=N to control backend parallelism

mehdi_amini added inline comments. Comment at: include/clang/Driver/Options.td:818 @@ -815,1 +817,3 @@ + HelpText<"Controls the backend parallelism of -flto=thin (default " + "of 0 means use std::thread::hardware_concurrency)">; def fthinlto_index_EQ : Joined<["-"], "f

Re: [PATCH] D24826: [LTO] Add -flto-jobs=N to control backend parallelism

mehdi_amini accepted this revision. mehdi_amini added a comment. This revision is now accepted and ready to land. LGTM, Thanks! https://reviews.llvm.org/D24826 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mai

Re: [llvm-dev] Upgrading phabricator

One of the new “feature” is that emails are HTML only right now. Not quite nice for the archive (or for interacting by email). See for instance: http://lists.llvm.org/pipermail/cfe-commits/Week-of-Mon-20160926/172081.html (Also the funky "[Changed Subscribers] “ in the title) Another issue is

Re: [llvm-dev] Upgrading phabricator

> On Sep 29, 2016, at 11:04 PM, Eric Liu wrote: > > I've switched the default email format to be plain text only now. This option > should be per-user configurable, but somehow it is not shown in the > "Settings"; I'll try if I can make the option personalized. > > Regarding new features and

r282996 - Use StringRef in Triple API (NFC)

Author: mehdi_amini Date: Fri Sep 30 20:16:22 2016 New Revision: 282996 URL: http://llvm.org/viewvc/llvm-project?rev=282996&view=rev Log: Use StringRef in Triple API (NFC) Modified: cfe/trunk/lib/CodeGen/CGBuiltin.cpp Modified: cfe/trunk/lib/CodeGen/CGBuiltin.cpp URL: http://llvm.org/viewvc

r283004 - Use StringRef in Pass/PassManager APIs (NFC)

Author: mehdi_amini Date: Fri Sep 30 21:56:57 2016 New Revision: 283004 URL: http://llvm.org/viewvc/llvm-project?rev=283004&view=rev Log: Use StringRef in Pass/PassManager APIs (NFC) Modified: cfe/trunk/lib/CodeGen/CodeGenAction.cpp Modified: cfe/trunk/lib/CodeGen/CodeGenAction.cpp URL: htt

r283039 - Use StringRef in Registry API (NFC)

Author: mehdi_amini Date: Sat Oct 1 10:44:54 2016 New Revision: 283039 URL: http://llvm.org/viewvc/llvm-project?rev=283039&view=rev Log: Use StringRef in Registry API (NFC) Modified: cfe/trunk/lib/Tooling/CompilationDatabase.cpp Modified: cfe/trunk/lib/Tooling/CompilationDatabase.cpp URL:

r283043 - Use StringRef for MemoryBuffer identifier API (NFC)

Author: mehdi_amini Date: Sat Oct 1 11:38:28 2016 New Revision: 283043 URL: http://llvm.org/viewvc/llvm-project?rev=283043&view=rev Log: Use StringRef for MemoryBuffer identifier API (NFC) Modified: cfe/trunk/include/clang/Basic/SourceManager.h cfe/trunk/include/clang/Lex/DirectoryLookup

[PATCH] D25190: [ThinLTO doc] Fix typo

mehdi_amini accepted this revision. mehdi_amini added a comment. This revision is now accepted and ready to land. Good catch indeed! https://reviews.llvm.org/D25190 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bi

[PATCH] D24815: [clang] make reciprocal estimate codegen a function attribute

mehdi_amini added inline comments. > echristo wrote in CGCall.cpp:1735 > Would be nice to get these pulled into a single fast-math string that's set > and then used all over for sure. :) I agree with getting on a path to remove these function attributes that have an equivalent on per-instructi

[PATCH] D24815: [clang] make reciprocal estimate codegen a function attribute

mehdi_amini added inline comments. > spatel wrote in CGCall.cpp:1735 > Good point - I think we have to convert all codegen tests that have these > function-level attributes to IR FMF and make sure that the output doesn't > change. Definitely not part of this patch, but hopefully something that

[PATCH] D24815: [clang] make reciprocal estimate codegen a function attribute

mehdi_amini added inline comments. > spatel wrote in CGCall.cpp:1735 > Auto-upgrading is part of the solution. Based on how we've been doing this > with vector intrinsics that get converted to IR, it's a ~3-step process: > > 1. Prepare the backend (DAG) to handle the expected new IR patterns an

[PATCH] D24644: Pass -ffunction-sections/-fdata-sections along to gold-plugin

mehdi_amini added a comment. What about function attributes? Hey that's the trend :) You could have a subset of the functions in their own sections but not all. With LTO it means that you can disable this for a single input file. https://reviews.llvm.org/D24644 __

r283416 - Use llvm::raw_string_ostream instead of std::stringstream (NFC)

Author: mehdi_amini Date: Wed Oct 5 23:26:16 2016 New Revision: 283416 URL: http://llvm.org/viewvc/llvm-project?rev=283416&view=rev Log: Use llvm::raw_string_ostream instead of std::stringstream (NFC) As a side effect, this avoid having to call .data() on the StringRef. Modified: cfe/trunk/

r283417 - Taking StringRef in Driver.h APIs instead of raw pointers (NFC)

Author: mehdi_amini Date: Thu Oct 6 00:11:48 2016 New Revision: 283417 URL: http://llvm.org/viewvc/llvm-project?rev=283417&view=rev Log: Taking StringRef in Driver.h APIs instead of raw pointers (NFC) Modified: cfe/trunk/include/clang/Driver/Driver.h cfe/trunk/lib/Driver/Driver.cpp Modi

[clang-tools-extra] r283525 - Revert "Revert "Add a static_assert to enforce that parameters to llvm::format() are not totally unsafe""

Author: mehdi_amini Date: Fri Oct 7 03:25:42 2016 New Revision: 283525 URL: http://llvm.org/viewvc/llvm-project?rev=283525&view=rev Log: Revert "Revert "Add a static_assert to enforce that parameters to llvm::format() are not totally unsafe"" This reverts commit r283510 and reapply r283509, wit

r283527 - Use StringReg in TargetParser APIs (NFC)

Author: mehdi_amini Date: Fri Oct 7 03:37:29 2016 New Revision: 283527 URL: http://llvm.org/viewvc/llvm-project?rev=283527&view=rev Log: Use StringReg in TargetParser APIs (NFC) Modified: cfe/trunk/lib/Basic/Targets.cpp cfe/trunk/lib/Driver/Tools.cpp Modified: cfe/trunk/lib/Basic/Target

r283605 - Turn ArchName/BoundArch in Driver from raw pointer to StringRef (NFC)

Author: mehdi_amini Date: Fri Oct 7 16:27:26 2016 New Revision: 283605 URL: http://llvm.org/viewvc/llvm-project?rev=283605&view=rev Log: Turn ArchName/BoundArch in Driver from raw pointer to StringRef (NFC) Modified: cfe/trunk/include/clang/Driver/Action.h cfe/trunk/include/clang/Driver/

r283611 - Driver: use StringRef instead of raw pointer in lookupTypeForExtension() (NFC)

Author: mehdi_amini Date: Fri Oct 7 16:41:00 2016 New Revision: 283611 URL: http://llvm.org/viewvc/llvm-project?rev=283611&view=rev Log: Driver: use StringRef instead of raw pointer in lookupTypeForExtension() (NFC) Modified: cfe/trunk/include/clang/Driver/ToolChain.h cfe/trunk/include/c

r283616 - Rename variable to not use C++ reserved name (leading underscore + capital) (NFC)

Author: mehdi_amini Date: Fri Oct 7 17:03:03 2016 New Revision: 283616 URL: http://llvm.org/viewvc/llvm-project?rev=283616&view=rev Log: Rename variable to not use C++ reserved name (leading underscore + capital) (NFC) Modified: cfe/trunk/lib/Driver/Action.cpp Modified: cfe/trunk/lib/Drive

r283615 - Fix MSVC build: requires namespace in front of StringRef (NFC)

Author: mehdi_amini Date: Fri Oct 7 17:02:59 2016 New Revision: 283615 URL: http://llvm.org/viewvc/llvm-project?rev=283615&view=rev Log: Fix MSVC build: requires namespace in front of StringRef (NFC) Modified: cfe/trunk/lib/Driver/Action.cpp Modified: cfe/trunk/lib/Driver/Action.cpp URL: h

r283645 - Use StringRef in Command::printArg() instead of raw pointer (NFC)

Author: mehdi_amini Date: Fri Oct 7 20:38:43 2016 New Revision: 283645 URL: http://llvm.org/viewvc/llvm-project?rev=283645&view=rev Log: Use StringRef in Command::printArg() instead of raw pointer (NFC) Modified: cfe/trunk/include/clang/Driver/Job.h cfe/trunk/lib/Driver/Job.cpp Modified

r283671 - Turn cl::values() (for enum) from a vararg function to using C++ variadic template

Author: mehdi_amini Date: Sat Oct 8 14:41:06 2016 New Revision: 283671 URL: http://llvm.org/viewvc/llvm-project?rev=283671&view=rev Log: Turn cl::values() (for enum) from a vararg function to using C++ variadic template The core of the change is supposed to be NFC, however it also fixes what I

[clang-tools-extra] r283671 - Turn cl::values() (for enum) from a vararg function to using C++ variadic template

Author: mehdi_amini Date: Sat Oct 8 14:41:06 2016 New Revision: 283671 URL: http://llvm.org/viewvc/llvm-project?rev=283671&view=rev Log: Turn cl::values() (for enum) from a vararg function to using C++ variadic template The core of the change is supposed to be NFC, however it also fixes what I

r283776 - Use StringRef in LangOptions::isNoBuiltinFunc API (NFC)

Author: mehdi_amini Date: Mon Oct 10 11:34:07 2016 New Revision: 283776 URL: http://llvm.org/viewvc/llvm-project?rev=283776&view=rev Log: Use StringRef in LangOptions::isNoBuiltinFunc API (NFC) Modified: cfe/trunk/include/clang/Basic/LangOptions.h cfe/trunk/lib/Basic/LangOptions.cpp Modi

r283802 - Change Builtins name to be stored as StringRef instead of raw pointers (NFC)

Author: mehdi_amini Date: Mon Oct 10 16:34:29 2016 New Revision: 283802 URL: http://llvm.org/viewvc/llvm-project?rev=283802&view=rev Log: Change Builtins name to be stored as StringRef instead of raw pointers (NFC) Modified: cfe/trunk/include/clang/Basic/Builtins.h cfe/trunk/lib/CodeGen/C

r283815 - Store FileEntry::Filename as a StringRef instead of raw pointer (NFC)

Author: mehdi_amini Date: Mon Oct 10 17:52:47 2016 New Revision: 283815 URL: http://llvm.org/viewvc/llvm-project?rev=283815&view=rev Log: Store FileEntry::Filename as a StringRef instead of raw pointer (NFC) Modified: cfe/trunk/include/clang/Basic/FileManager.h cfe/trunk/lib/Basic/FileMan

[clang-tools-extra] r283819 - Fix clang-tools-extra build after r283815 (Store FileEntry::Filename as a StringRef instead of raw pointer)

Author: mehdi_amini Date: Mon Oct 10 18:24:16 2016 New Revision: 283819 URL: http://llvm.org/viewvc/llvm-project?rev=283819&view=rev Log: Fix clang-tools-extra build after r283815 (Store FileEntry::Filename as a StringRef instead of raw pointer) Modified: clang-tools-extra/trunk/clang-apply

r283856 - Turn FileManager DirectoryEntry::Name from raw pointer to StringRef (NFC)

Author: mehdi_amini Date: Tue Oct 11 02:31:29 2016 New Revision: 283856 URL: http://llvm.org/viewvc/llvm-project?rev=283856&view=rev Log: Turn FileManager DirectoryEntry::Name from raw pointer to StringRef (NFC) Modified: cfe/trunk/include/clang/Basic/FileManager.h cfe/trunk/include/clang

r283888 - Use StringRef instead of raw pointer in SourceManagerInternals LineTableInfo API (NFC)

Author: mehdi_amini Date: Tue Oct 11 10:15:32 2016 New Revision: 283888 URL: http://llvm.org/viewvc/llvm-project?rev=283888&view=rev Log: Use StringRef instead of raw pointer in SourceManagerInternals LineTableInfo API (NFC) Modified: cfe/trunk/include/clang/Basic/SourceManagerInternals.h M

Re: r283802 - Change Builtins name to be stored as StringRef instead of raw pointers (NFC)

Builtins.cpp and Targets.cpp, StringRef(const > char*) is not constexpr (because of strlen). This means you'll get a > huge generated initialization function for it. We want to reduce the > number of global initializers in LLVM, not create new ones. > > On Mon, Oct 10, 2016 at

Re: r283802 - Change Builtins name to be stored as StringRef instead of raw pointers (NFC)

gRef(const >>> char*) is not constexpr (because of strlen). This means you'll get a >>> huge generated initialization function for it. We want to reduce the >>> number of global initializers in LLVM, not create new ones. >>> >>> On Mon, Oct 10, 20

Re: r283802 - Change Builtins name to be stored as StringRef instead of raw pointers (NFC)

Oct 11, 2016 at 11:40 AM, Mehdi Amini via cfe-commits > mailto:cfe-commits@lists.llvm.org>> wrote: > This is temporary: the last patch of my series of patches adds the constexpr > ctor and remove all these static initializers. > > > On Oct 11, 2016, at 11:26 AM, Benjam

r283920 - Revert "Change Builtins name to be stored as StringRef instead of raw pointers (NFC)"

Author: mehdi_amini Date: Tue Oct 11 14:04:24 2016 New Revision: 283920 URL: http://llvm.org/viewvc/llvm-project?rev=283920&view=rev Log: Revert "Change Builtins name to be stored as StringRef instead of raw pointers (NFC)" This reverts commit r283802. It introduces temporarily static initialize

Re: r283802 - Change Builtins name to be stored as StringRef instead of raw pointers (NFC)

tin::Info in Builtins.cpp and Targets.cpp, StringRef(const >>> char*) is not constexpr (because of strlen). This means you'll get a >>> huge generated initialization function for it. We want to reduce the >>> number of global initializers in LLVM, not create new ones.

[PATCH] D24644: Pass -ffunction-sections/-fdata-sections along to gold-plugin

mehdi_amini accepted this revision. mehdi_amini added a comment. This revision is now accepted and ready to land. LGTM. https://reviews.llvm.org/D24644 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/lis

[PATCH] D25537: [ThinLTO] Update doc to include lld

mehdi_amini requested changes to this revision. mehdi_amini added a comment. This revision now requires changes to proceed. You have to add which platforms are supported. https://reviews.llvm.org/D25537 ___ cfe-commits mailing list cfe-commits@lists

Re: [PATCH] D25537: [ThinLTO] Update doc to include lld

Perfect! LGTM, thanks! — Mehdi > On Oct 12, 2016, at 5:04 PM, Davide Italiano wrote: > > davide updated this revision to Diff 74459. > davide added a comment. > > I mentioned that's ELF only. Are you OK with that or you want me to put more > detailed info? > > > https://reviews.llvm.org/D2

Re: [libcxx] r282345 - Use __attribute__((internal_linkage)) when available.

Hi Eric, This is not clear to me that this patch is correct as-is. See the last message in the thread: http://lists.llvm.org/pipermail/cfe-dev/2016-July/049985.html I think we should reach a consensus on the right course of actio

Re: [libcxx] r282345 - Use __attribute__((internal_linkage)) when available.

Thanks Eric! > On Oct 12, 2016, at 9:18 PM, Eric Fiselier wrote: > > Sure, I've reverted the change in r284101. > > I'll try and respond in detail tomorrow. > > /Eric > > On Wed, Oct 12, 2016 at 9:57 PM, Mehdi Amini > wrote: > Hi Eric, > > This is not clear to

[libcxx] [flang] [libc] [clang-tools-extra] [llvm] [clang] [compiler-rt] [mlir] Make SmallVectorImpl destructor protected (PR #71439)

@@ -95,7 +95,7 @@ class SerializeToHsacoPass std::unique_ptr> serializeISA(const std::string &isa) override; - std::unique_ptr> assembleIsa(const std::string &isa); + std::unique_ptr> assembleIsa(const std::string &isa); joker-eph wrote: @krzysz00 can y

[clang] [clang-tools-extra] [clang][NFC] Refactor `CXXNewExpr::InitializationStyle` (PR #71322)

joker-eph wrote: > I've seen those exact test failures locally, but since precommit CI was fine, > I landed this PR and kept a close eye on the bots. If you see failures locally, it's best to understand where they coming from: even if it passes on bots it just means we have a hole in our test

[clang] [clang-tools-extra] [clang][NFC] Refactor `CXXNewExpr::InitializationStyle` (PR #71322)

joker-eph wrote: > tests will fail locally but pass everywhere else (for example, people > working on slightly out-of-norm configurations will sometimes have persistent > local failures that are unrelated to changes in the patch). It never happened to me, in a way that can't be root caused an

[flang] [clang] [clang-tools-extra] [mlir] [compiler-rt] [llvm] [mlir] Verify non-negative `offset` and `size` (PR #72059)

joker-eph wrote: > > This looks like a good fix for the verifier, but we should also fix the > > canonicalization to **not** create invalid IR! > > Then I think that I'll implement the suggestions from Matthias in this PR and > leave the canonicalization for a future PR. For that future PR, co

[compiler-rt] [clang] [llvm] [mlir] [clang-tools-extra] [mlir][tensor] Document `dest` operand (PR #71726)

@@ -18,31 +18,34 @@ def Tensor_Dialect : Dialect { let description = [{ The `tensor` dialect is intended to hold core tensor creation and manipulation ops, which are not strongly associated with any particular -other dialect or domain abstraction. The primary smok

[compiler-rt] [clang] [llvm] [mlir] [clang-tools-extra] [mlir][tensor] Document `dest` operand (PR #71726)

https://github.com/joker-eph approved this pull request. LG (with one edit) https://github.com/llvm/llvm-project/pull/71726 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [mlir] [compiler-rt] [llvm] [libcxx] [clang-tools-extra] [flang] [mlir] Fix a zero stride canonicalizer crash (PR #74200)

@@ -0,0 +1,14 @@ +// RUN: mlir-opt %s -inline -split-input-file | FileCheck %s + +// CHECK-LABEL: func @inline_negative_stride +func.func @inline_negative_stride(%arg0 : memref<10xf32>) -> memref<1xf32, strided<[?], offset: 1>> { + cf.br ^bb1(%arg0 : memref<10xf32>) +^bb1(%m: me

[llvm] [clang-tools-extra] [libcxx] [compiler-rt] [mlir] [clang] [flang] [mlir] Fix a zero stride canonicalizer crash (PR #74200)

joker-eph wrote: > Is it possible to add the test case (or minimal similar example) with > `--inline` option so that we can confirm the original issue is resolved. > > #73383 It is common that the person filing the issue does not reduce the IR or the pipeline that causes the bug. In particul

[llvm] [libc] [mlir] [clang-tools-extra] [clang] [libcxx] [compiler-rt] [flang] [mlir][llvm] Fix verifier for const int and dense (PR #74340)

https://github.com/joker-eph approved this pull request. https://github.com/llvm/llvm-project/pull/74340 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[compiler-rt] [libcxx] [mlir] [clang] [llvm] [flang] [clang-tools-extra] [mlir] Fix a zero stride canonicalizer crash (PR #74200)

https://github.com/joker-eph approved this pull request. Thanks! https://github.com/llvm/llvm-project/pull/74200 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[compiler-rt] [lld] [openmp] [libcxx] [mlir] [clang] [lldb] [llvm] [flang] [libc] [clang-tools-extra] [MLIR] Enable GPU Dialect to SYCL runtime integration (PR #71430)

@@ -0,0 +1,31 @@ +//===- SPIRVToLLVMIRTranslation.cpp - Translate SPIRV to LLVM IR --===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apa

[compiler-rt] [lld] [openmp] [libcxx] [mlir] [clang] [lldb] [llvm] [flang] [libc] [clang-tools-extra] [MLIR] Enable GPU Dialect to SYCL runtime integration (PR #71430)

@@ -61,6 +63,7 @@ registerAllGPUToLLVMIRTranslations(DialectRegistry ®istry) { registerLLVMDialectTranslation(registry); registerNVVMDialectTranslation(registry); registerROCDLDialectTranslation(registry); + registerSPIRVDialectTranslation(registry); jo

[compiler-rt] [lld] [openmp] [libcxx] [mlir] [clang] [lldb] [llvm] [flang] [libc] [clang-tools-extra] [MLIR] Enable GPU Dialect to SYCL runtime integration (PR #71430)

https://github.com/joker-eph edited https://github.com/llvm/llvm-project/pull/71430 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[flang] [mlir] [llvm] [compiler-rt] [libcxx] [clang-tools-extra] [clang] [libc] [mlir][gpu] Support dynamic_shared_memory Op with vector dialect (PR #74475)

https://github.com/joker-eph approved this pull request. https://github.com/llvm/llvm-project/pull/74475 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[mlir] [llvm] [clang] [llvm] Improve implementation of StringRef::find_last_of and cie (PR #71865)

@@ -274,6 +278,23 @@ StringRef::size_type StringRef::find_first_not_of(StringRef Chars, /// Note: O(size() + Chars.size()) StringRef::size_type StringRef::find_last_of(StringRef Chars, size_t From) const { +#ifdef __SSE2__ + if (Ch

[clang-tools-extra] [compiler-rt] [llvm] [clang] [flang] [mlir] [mlir] Verify non-negative `offset` and `size` (PR #72059)

https://github.com/joker-eph approved this pull request. Nice, thanks! https://github.com/llvm/llvm-project/pull/72059 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[libc] [clang-tools-extra] [flang] [clang] [llvm] [compiler-rt] [libcxx] [lld] [mlir] [lldb] [MLIR] Enable GPU Dialect to SYCL runtime integration (PR #71430)

@@ -61,6 +63,7 @@ registerAllGPUToLLVMIRTranslations(DialectRegistry ®istry) { registerLLVMDialectTranslation(registry); registerNVVMDialectTranslation(registry); registerROCDLDialectTranslation(registry); + registerSPIRVDialectTranslation(registry); jo

[libc] [clang-tools-extra] [flang] [clang] [llvm] [compiler-rt] [libcxx] [lld] [mlir] [lldb] [MLIR] Enable GPU Dialect to SYCL runtime integration (PR #71430)

@@ -0,0 +1,31 @@ +//===- SPIRVToLLVMIRTranslation.cpp - Translate SPIRV to LLVM IR --===// +// +// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. +// See https://llvm.org/LICENSE.txt for license information. +// SPDX-License-Identifier: Apa

[libc] [clang-tools-extra] [mlir] [compiler-rt] [lld] [libcxx] [clang] [flang] [lldb] [llvm] [MLIR] Enable GPU Dialect to SYCL runtime integration (PR #71430)

@@ -61,6 +63,7 @@ registerAllGPUToLLVMIRTranslations(DialectRegistry ®istry) { registerLLVMDialectTranslation(registry); registerNVVMDialectTranslation(registry); registerROCDLDialectTranslation(registry); + registerSPIRVDialectTranslation(registry); jo

[libc] [clang-tools-extra] [mlir] [compiler-rt] [lld] [libcxx] [clang] [flang] [lldb] [llvm] [MLIR] Enable GPU Dialect to SYCL runtime integration (PR #71430)

https://github.com/joker-eph edited https://github.com/llvm/llvm-project/pull/71430 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[lldb] [compiler-rt] [llvm] [clang] [lld] [mlir] [flang] [libcxx] [mlir][async] Avoid crash when not using `func.func` (PR #72801)

https://github.com/joker-eph approved this pull request. https://github.com/llvm/llvm-project/pull/72801 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[flang] [clang] [libcxx] [lld] [lldb] [compiler-rt] [mlir] [llvm] [mlir][async] Avoid crash when not using `func.func` (PR #72801)

joker-eph wrote: FYI the build is failing (maybe re-running would be enough though) https://github.com/llvm/llvm-project/pull/72801 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [mlir] Add config for PDL (PR #69927)

joker-eph wrote: Need to iterate a bit more on framing the problem: > MLIR is a general infrastructure and as far as possible no dialect is > intended to be special or privileged Yes, and you can use MLIR without PDL, it's not a mandatory component. I believe that the minimal examples `mlir-c

[clang] 3289ecf - [𝘀𝗽𝗿] changes introduced through rebase

Author: Mehdi Amini Date: 2023-11-03T03:23:43-07:00 New Revision: 3289ecff8e8f5022cb6a40777392c98f1bcf5780 URL: https://github.com/llvm/llvm-project/commit/3289ecff8e8f5022cb6a40777392c98f1bcf5780 DIFF: https://github.com/llvm/llvm-project/commit/3289ecff8e8f5022cb6a40777392c98f1bcf5780.diff L

[libcxxabi] [flang] [libcxx] [llvm] [clang] [mlir] [clang-tools-extra] [lldb] [lld] [compiler-rt] Refactor ModuleToObject to offer more flexibility to subclass (NFC) (PR #71165)

https://github.com/joker-eph edited https://github.com/llvm/llvm-project/pull/71165 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[libcxxabi] [flang] [libcxx] [llvm] [clang] [mlir] [clang-tools-extra] [lldb] [lld] [compiler-rt] Refactor ModuleToObject to offer more flexibility to subclass (NFC) (PR #71165)

https://github.com/joker-eph closed https://github.com/llvm/llvm-project/pull/71165 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang-tools-extra] [mlir] [llvm] [compiler-rt] [flang] [mlir][vector] Fix invalid `LoadOp` indices being created (PR #76292)

@@ -866,16 +866,41 @@ struct TransferOpConversion : public VectorToSCFPattern { this->setHasBoundedRewriteRecursion(); } + static void getMaskBufferLoadIndices(OpTy xferOp, Value castedMaskBuffer, + SmallVector &loadIndices, --

[mlir] [flang] [clang-tools-extra] [llvm] [compiler-rt] [clang] [mlir][vector] Fix invalid `LoadOp` indices being created (PR #76292)

@@ -897,7 +921,8 @@ struct TransferOpConversion : public VectorToSCFPattern { } else { // It's safe to assume the mask buffer can be unpacked if the data // buffer was unpacked. -auto castedMaskType = *unpackOneDim(maskBufferType); +auto m

[flang] [clang-tools-extra] [clang] [mlir] [llvm] [compiler-rt] [mlir][vector] Fix invalid `LoadOp` indices being created (PR #76292)

@@ -866,16 +866,41 @@ struct TransferOpConversion : public VectorToSCFPattern { this->setHasBoundedRewriteRecursion(); } + static void getMaskBufferLoadIndices(OpTy xferOp, Value castedMaskBuffer, + SmallVector &loadIndices, +

[llvm] [libcxxabi] [lldb] [libcxx] [polly] [openmp] [compiler-rt] [flang] [libc] [mlir] [clang-tools-extra] [clang] [mlir][vector] Fix invalid `LoadOp` indices being created (PR #76292)

@@ -897,7 +921,8 @@ struct TransferOpConversion : public VectorToSCFPattern { } else { // It's safe to assume the mask buffer can be unpacked if the data // buffer was unpacked. -auto castedMaskType = *unpackOneDim(maskBufferType); +auto m

[lldb] [polly] [libcxx] [openmp] [libcxxabi] [mlir] [compiler-rt] [clang] [llvm] [clang-tools-extra] [libc] [flang] [mlir][vector] Fix invalid `LoadOp` indices being created (PR #76292)

https://github.com/joker-eph approved this pull request. https://github.com/llvm/llvm-project/pull/76292 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] 42f588f - Use ManagedStatic and lazy initialization of cl::opt in libSupport to make it free of global initializer

Author: Mehdi Amini Date: 2021-07-16T03:33:20Z New Revision: 42f588f39c5ce6f521e3709b8871d1fdd076292f URL: https://github.com/llvm/llvm-project/commit/42f588f39c5ce6f521e3709b8871d1fdd076292f DIFF: https://github.com/llvm/llvm-project/commit/42f588f39c5ce6f521e3709b8871d1fdd076292f.diff LOG: U

[clang-tools-extra] 16b5e9d - Revert "Use ManagedStatic and lazy initialization of cl::opt in libSupport to make it free of global initializer"

Author: Mehdi Amini Date: 2021-07-16T03:46:53Z New Revision: 16b5e9d6a269913e8da0fa037e8af32eaf304c8f URL: https://github.com/llvm/llvm-project/commit/16b5e9d6a269913e8da0fa037e8af32eaf304c8f DIFF: https://github.com/llvm/llvm-project/commit/16b5e9d6a269913e8da0fa037e8af32eaf304c8f.diff LOG: R

[clang] 16b5e9d - Revert "Use ManagedStatic and lazy initialization of cl::opt in libSupport to make it free of global initializer"

Author: Mehdi Amini Date: 2021-07-16T03:46:53Z New Revision: 16b5e9d6a269913e8da0fa037e8af32eaf304c8f URL: https://github.com/llvm/llvm-project/commit/16b5e9d6a269913e8da0fa037e8af32eaf304c8f DIFF: https://github.com/llvm/llvm-project/commit/16b5e9d6a269913e8da0fa037e8af32eaf304c8f.diff LOG: R

[clang-tools-extra] af93217 - Use ManagedStatic and lazy initialization of cl::opt in libSupport to make it free of global initializer

Author: Mehdi Amini Date: 2021-07-16T06:54:26Z New Revision: af9321739b20becf170e6bb5060b8d780e1dc8dd URL: https://github.com/llvm/llvm-project/commit/af9321739b20becf170e6bb5060b8d780e1dc8dd DIFF: https://github.com/llvm/llvm-project/commit/af9321739b20becf170e6bb5060b8d780e1dc8dd.diff LOG: U

[clang-tools-extra] 8d051d8 - Revert "Use ManagedStatic and lazy initialization of cl::opt in libSupport to make it free of global initializer"

Author: Mehdi Amini Date: 2021-07-16T07:35:13Z New Revision: 8d051d854619956de633047409149cdab1e3319a URL: https://github.com/llvm/llvm-project/commit/8d051d854619956de633047409149cdab1e3319a DIFF: https://github.com/llvm/llvm-project/commit/8d051d854619956de633047409149cdab1e3319a.diff LOG: R

[clang-tools-extra] 7637457 - Use ManagedStatic and lazy initialization of cl::opt in libSupport to make it free of global initializer

Author: Mehdi Amini Date: 2021-07-16T07:38:16Z New Revision: 76374573ce829b083b95b74937a11e9b91f8f45f URL: https://github.com/llvm/llvm-project/commit/76374573ce829b083b95b74937a11e9b91f8f45f DIFF: https://github.com/llvm/llvm-project/commit/76374573ce829b083b95b74937a11e9b91f8f45f.diff LOG: U

[clang] 26eec9e - Revert "[clang] Implement Change scope of lambda trailing-return-type"

Author: Mehdi Amini Date: 2022-04-13T19:35:13Z New Revision: 26eec9e9dbc20186f6b810fe01d3a01aac8ae6f4 URL: https://github.com/llvm/llvm-project/commit/26eec9e9dbc20186f6b810fe01d3a01aac8ae6f4 DIFF: https://github.com/llvm/llvm-project/commit/26eec9e9dbc20186f6b810fe01d3a01aac8ae6f4.diff LOG: R

[clang] [llvm] Add missing clang to the monolithic pre-merge build (PR #85354)

https://github.com/joker-eph created https://github.com/llvm/llvm-project/pull/85354 Clang has a custom separate pipeline integrated with libc++ that only runs in release mode. It means that changes which touches only clang won't run the clang tests in the configuration used by LLVM premerge a

[clang] [clang-tools-extra] [compiler-rt] [flang] [lld] [lldb] [llvm] [mlir] [openmp] [pstl] Finally formalise our defacto line-ending policy (PR #86318)

joker-eph wrote: LGTM overall, but the CI failure looks real. https://github.com/llvm/llvm-project/pull/86318 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang-tools-extra] [lld] [lldb] [llvm] [mlir] Rename llvm::ThreadPool -> llvm::DefaultThreadPool (NFC) (PR #83702)

joker-eph wrote: Actually no: the first patch landed already, so `ThreadPoolInterface` is now the base class in the codebase. I have some mixup here in that when renaming ThreadPool -> DefaultThreadPool, I used the base class ThreadPoolInterface when updating some of the uses. I will push the

[clang] [clang-tools-extra] [lld] [lldb] [llvm] [mlir] Rename llvm::ThreadPool -> llvm::DefaultThreadPool (NFC) (PR #83702)

https://github.com/joker-eph edited https://github.com/llvm/llvm-project/pull/83702 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang-tools-extra] [lld] [lldb] [llvm] [mlir] Rename llvm::ThreadPool -> llvm::DefaultThreadPool (NFC) (PR #83702)

https://github.com/joker-eph updated https://github.com/llvm/llvm-project/pull/83702 >From 1b407d9d5abc9a1cf58afaf7f32ed40446d55f52 Mon Sep 17 00:00:00 2001 From: Mehdi Amini Date: Tue, 5 Mar 2024 10:38:41 -0800 Subject: [PATCH] Rename llvm::ThreadPool -> llvm::DefaultThreadPool (NFC) The base

[clang] [clang-tools-extra] [lld] [lldb] [llvm] [mlir] Rename llvm::ThreadPool -> llvm::DefaultThreadPool (NFC) (PR #83702)

https://github.com/joker-eph updated https://github.com/llvm/llvm-project/pull/83702 >From ea79b6037497230b23caf36024a9e6883d3cac04 Mon Sep 17 00:00:00 2001 From: Mehdi Amini Date: Tue, 5 Mar 2024 10:38:41 -0800 Subject: [PATCH] Rename llvm::ThreadPool -> llvm::DefaultThreadPool (NFC) The base

  1   2   3   4   5   >