r298208 - [X86][AVX512][Clang][Intrinsics] Adding missing intrinsics to Clang .

Author: ibreger Date: Sun Mar 19 03:27:16 2017 New Revision: 298208 URL: http://llvm.org/viewvc/llvm-project?rev=298208=rev Log: [X86][AVX512][Clang][Intrinsics] Adding missing intrinsics to Clang . Summary: Adding missing intrinsics : _mm512_set_epi16, _mm512_set_epi8,

[PATCH] D26021: [X86][AVX512][Clang] Add support for mask_{move|store|load}_s{s/d} and int2mask/mask2int intrinsics.

igorb accepted this revision. igorb added a comment. This revision is now accepted and ready to land. LGTM Comment at: lib/Headers/avx512fintrin.h:9123 + __m128 res=__A; + res[0] = (__U&1)? __B[0]:__W[0]; + return res; . Comment at:

Re: [PATCH] D24961: [avx512] Add aliases to some missing avx512 intrinsics.

igorb accepted this revision. igorb added a comment. This revision is now accepted and ready to land. LGTM https://reviews.llvm.org/D24961 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

Re: [PATCH] D24785: [AVX512] Fix return types on __builtin_ia32_gather3XivXdi builtins

igorb accepted this revision. igorb added a comment. This revision is now accepted and ready to land. LGTM Repository: rL LLVM https://reviews.llvm.org/D24785 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

Re: [PATCH] D21836: [AVX512][BUILTIN][vpermilps][intrinsics] Fixing two incorrect IMM check.

igorb accepted this revision. igorb added a comment. This revision is now accepted and ready to land. LGTM http://reviews.llvm.org/D21836 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

r274110 - [AVX512] Zero extend cmp intrinsic return value.

Author: ibreger Date: Wed Jun 29 03:14:17 2016 New Revision: 274110 URL: http://llvm.org/viewvc/llvm-project?rev=274110=rev Log: [AVX512] Zero extend cmp intrinsic return value. Differential Revision: http://reviews.llvm.org/D21746 Modified: cfe/trunk/lib/CodeGen/CGBuiltin.cpp

[PATCH] D21746: [AVX512] Zero extend cmp intrinsic return value.

igorb created this revision. igorb added reviewers: delena, craig.topper. igorb added a subscriber: cfe-commits. igorb set the repository for this revision to rL LLVM. [AVX512] Zero extend cmp intrinsic return value. Repository: rL LLVM http://reviews.llvm.org/D21746 Files:

Re: [PATCH] D21734: [Intrinsics][AVX512][BuiltIn] adding intrinsics for vrangesd instruction set

igorb accepted this revision. igorb added a comment. This revision is now accepted and ready to land. LGTM http://reviews.llvm.org/D21734 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

Re: [PATCH] D21322: [Clang][avx512][Intrinsics] adding prefetch gather intrinsics

igorb accepted this revision. igorb added a comment. This revision is now accepted and ready to land. LGTM http://reviews.llvm.org/D21322 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

r272141 - [AVX512] Emit select instruction instead of using x86 specific instrinsics.

Author: ibreger Date: Wed Jun 8 08:59:20 2016 New Revision: 272141 URL: http://llvm.org/viewvc/llvm-project?rev=272141=rev Log: [AVX512] Emit select instruction instead of using x86 specific instrinsics. This will allow us to remove the x86 instrinics from the backend. Differential Revision:

Re: [PATCH] D21058: [Clang][AVX512][BUILTIN]Adding missing intrinsics srl and sll

igorb accepted this revision. igorb added a comment. This revision is now accepted and ready to land. LGTM http://reviews.llvm.org/D21058 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

[PATCH] D21060: [AVX512] Emit select instruction instead of using x86 specific instrinsics.

igorb created this revision. igorb added reviewers: craig.topper, delena. igorb added a subscriber: cfe-commits. igorb set the repository for this revision to rL LLVM. [AVX512] Emit select instruction instead of using x86 specific instrinsics. This will allow us to remove the x86 instrinics

Re: [PATCH] D20861: [AVX512] add missing integer to float conversion

igorb accepted this revision. igorb added a comment. This revision is now accepted and ready to land. LGTM Repository: rL LLVM http://reviews.llvm.org/D20861 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

Re: [PATCH] D20620: [Clang][AVX512][Builtin] Fix palignr intrinsics header

igorb accepted this revision. igorb added a comment. This revision is now accepted and ready to land. LGTM http://reviews.llvm.org/D20620 ___ cfe-commits mailing list cfe-commits@lists.llvm.org

Re: [PATCH] D20321: [Clang][AVX512][intrinsics] Fix vperm intrinsics.

igorb added a comment. In http://reviews.llvm.org/D20321#436494, @craig.topper wrote: > Looking at this again. This doesn't match the gcc implementation of the > builtins. Unless their header file is also wrong. Can you clarify? Thanks, You are correct. I implemented all changes in code-gen

Re: [PATCH] D20321: [Clang][AVX512][intrinsics] Fix vperm intrinsics.

igorb updated this revision to Diff 58055. igorb added a comment. Update path according to comments. Thanks for review, Craig ! Repository: rL LLVM http://reviews.llvm.org/D20321 Files: include/clang/Basic/BuiltinsX86.def lib/Headers/avx512bwintrin.h lib/Headers/avx512fintrin.h

[PATCH] D20321: [Clang][AVX512][intrinsics] Fix vperm intrinsics.

igorb created this revision. igorb added reviewers: m_zuckerman, AsafBadouh, delena. igorb added a subscriber: cfe-commits. igorb set the repository for this revision to rL LLVM. [Clang][AVX512][intrinsics] Fix vperm{w|d|q|ps|pd} intrinsics. Index is first argument to buildin function.

r247277 - AVX-512: Changed nidx parameter in extractf64/32 intrinsic from i8 to i32 according to the Intel Spec

Author: ibreger Date: Thu Sep 10 07:55:54 2015 New Revision: 247277 URL: http://llvm.org/viewvc/llvm-project?rev=247277=rev Log: AVX-512: Changed nidx parameter in extractf64/32 intrinsic from i8 to i32 according to the Intel Spec Differential Revision: http://reviews.llvm.org/D12752 Modified:

r246430 - AVX-512: Changed cnt parameter in valignq/d intrinsic from i8 to i32 according to the Intel Spec

Author: ibreger Date: Mon Aug 31 06:15:06 2015 New Revision: 246430 URL: http://llvm.org/viewvc/llvm-project?rev=246430=rev Log: AVX-512: Changed cnt parameter in valignq/d intrinsic from i8 to i32 according to the Intel Spec Differential Revision: http://reviews.llvm.org/D12274 Modified: