[clang] 175e423 - AMDGPU: Make HIPToolChain a subclass of AMDGPUToolChain

Author: Matt Arsenault Date: 2020-03-31T18:22:46-04:00 New Revision: 175e42303bb2a4253c6512b1ae05b32b0004 URL: https://github.com/llvm/llvm-project/commit/175e42303bb2a4253c6512b1ae05b32b0004 DIFF: https://github.com/llvm/llvm-project/commit/175e42303bb2a4253c6512b1ae05b32b0004.diff

[clang] 6593360 - AMDGPU: Fix consistently backwards logic for default denormal mode

Author: Matt Arsenault Date: 2020-04-01T12:36:22-04:00 New Revision: 6593360ee760c0c98cf7fb05be70aa4ea455a42a URL: https://github.com/llvm/llvm-project/commit/6593360ee760c0c98cf7fb05be70aa4ea455a42a DIFF: https://github.com/llvm/llvm-project/commit/6593360ee760c0c98cf7fb05be70aa4ea455a42a.diff

[clang] 4ea3650 - HIP: Link correct denormal mode library

Author: Matt Arsenault Date: 2020-04-01T12:36:22-04:00 New Revision: 4ea3650c212ae471657d3a253cd424ce9d1316ac URL: https://github.com/llvm/llvm-project/commit/4ea3650c212ae471657d3a253cd424ce9d1316ac DIFF: https://github.com/llvm/llvm-project/commit/4ea3650c212ae471657d3a253cd424ce9d1316ac.diff

[clang] ce2258c - clang/AMDGPU: Stop setting old denormal subtarget features

Author: Matt Arsenault Date: 2020-04-02T17:17:12-04:00 New Revision: ce2258c1cd5dc9cf20040d1b1e540d80250c1435 URL: https://github.com/llvm/llvm-project/commit/ce2258c1cd5dc9cf20040d1b1e540d80250c1435 DIFF: https://github.com/llvm/llvm-project/commit/ce2258c1cd5dc9cf20040d1b1e540d80250c1435.diff

[clang] c64ca93 - clang: Treat ieee mode as the default for denormal-fp-math

Author: Matt Arsenault Date: 2020-03-04T23:34:02-05:00 New Revision: c64ca93053af235bac0ca4dcdcd21c8882478310 URL: https://github.com/llvm/llvm-project/commit/c64ca93053af235bac0ca4dcdcd21c8882478310 DIFF: https://github.com/llvm/llvm-project/commit/c64ca93053af235bac0ca4dcdcd21c8882478310.diff

[clang] c6da9ec - clang: Fix assert on void pointer arithmetic with address_space

Author: Matt Arsenault Date: 2019-10-31T20:07:23-07:00 New Revision: c6da9ec0e90ea8798ecae583bb8d26bdf6b9b79f URL: https://github.com/llvm/llvm-project/commit/c6da9ec0e90ea8798ecae583bb8d26bdf6b9b79f DIFF: https://github.com/llvm/llvm-project/commit/c6da9ec0e90ea8798ecae583bb8d26bdf6b9b79f.diff

[clang] 7fe9435 - Work on cleaning up denormal mode handling

Author: Matt Arsenault Date: 2019-11-19T22:01:14+05:30 New Revision: 7fe9435dc88050ee78eb1d4adec87610dce468f7 URL: https://github.com/llvm/llvm-project/commit/7fe9435dc88050ee78eb1d4adec87610dce468f7 DIFF: https://github.com/llvm/llvm-project/commit/7fe9435dc88050ee78eb1d4adec87610dce468f7.diff

[clang] e531750 - clang: Add -fconvergent-functions flag

Author: Matt Arsenault Date: 2019-11-19T23:20:15+05:30 New Revision: e531750c6cf9ab6ca987ffbfe100b1d766269eb5 URL: https://github.com/llvm/llvm-project/commit/e531750c6cf9ab6ca987ffbfe100b1d766269eb5 DIFF: https://github.com/llvm/llvm-project/commit/e531750c6cf9ab6ca987ffbfe100b1d766269eb5.diff

[clang] 2cc1194 - clang/AMDGPU: Fix default for frame-pointer attribute

Author: Matt Arsenault Date: 2019-12-07T00:09:10+05:30 New Revision: 2cc11941a2e88236e0b4842229454ae6d85142cd URL: https://github.com/llvm/llvm-project/commit/2cc11941a2e88236e0b4842229454ae6d85142cd DIFF: https://github.com/llvm/llvm-project/commit/2cc11941a2e88236e0b4842229454ae6d85142cd.diff

[clang] 00b2a9d - Reapply "clang: Treat ieee mode as the default for denormal-fp-math"

Author: Matt Arsenault Date: 2020-03-06T11:46:55-08:00 New Revision: 00b2a9df452d25cf4a780d20bab7d557d76604ae URL: https://github.com/llvm/llvm-project/commit/00b2a9df452d25cf4a780d20bab7d557d76604ae DIFF: https://github.com/llvm/llvm-project/commit/00b2a9df452d25cf4a780d20bab7d557d76604ae.diff

[clang] a4e71f0 - Assume ieee behavior without denormal-fp-math attribute

Author: Matt Arsenault Date: 2020-03-07T12:10:56-05:00 New Revision: a4e71f01c08fbaeaccfe3e11cc08790432cc7e45 URL: https://github.com/llvm/llvm-project/commit/a4e71f01c08fbaeaccfe3e11cc08790432cc7e45 DIFF: https://github.com/llvm/llvm-project/commit/a4e71f01c08fbaeaccfe3e11cc08790432cc7e45.diff

[clang] a3c814d - Separately track input and output denormal mode

Author: Matt Arsenault Date: 2020-02-04T12:59:21-05:00 New Revision: a3c814d23497bc71b8ed53c35f773366aff02922 URL: https://github.com/llvm/llvm-project/commit/a3c814d23497bc71b8ed53c35f773366aff02922 DIFF: https://github.com/llvm/llvm-project/commit/a3c814d23497bc71b8ed53c35f773366aff02922.diff

[clang] fa7cd54 - clang: Guess at some platform FTZ/DAZ default settings

Author: Matt Arsenault Date: 2020-02-12T12:09:26-05:00 New Revision: fa7cd549d604bfd8f9dce5d649a19720cbc39cca URL: https://github.com/llvm/llvm-project/commit/fa7cd549d604bfd8f9dce5d649a19720cbc39cca DIFF: https://github.com/llvm/llvm-project/commit/fa7cd549d604bfd8f9dce5d649a19720cbc39cca.diff

[clang] 5dcffdf - Fix fp-model flag test failure on linux

Author: Matt Arsenault Date: 2020-02-12T10:14:29-08:00 New Revision: 5dcffdf58a33b51028f6b52e44c4af25b5d159aa URL: https://github.com/llvm/llvm-project/commit/5dcffdf58a33b51028f6b52e44c4af25b5d159aa DIFF: https://github.com/llvm/llvm-project/commit/5dcffdf58a33b51028f6b52e44c4af25b5d159aa.diff

[clang] 3833316 - Completely ignore strict FP model and denormal-fp-math interaction

Author: Matt Arsenault Date: 2020-02-12T13:26:46-05:00 New Revision: 38333164fc7dfc9d4cbf93e443e74809022ed78e URL: https://github.com/llvm/llvm-project/commit/38333164fc7dfc9d4cbf93e443e74809022ed78e DIFF: https://github.com/llvm/llvm-project/commit/38333164fc7dfc9d4cbf93e443e74809022ed78e.diff

[clang] 9b549f2 - AMDGPU: Update clang test

Author: Matt Arsenault Date: 2020-01-16T18:10:29-05:00 New Revision: 9b549f26fab6900b5328c0c6239fd77c7527159c URL: https://github.com/llvm/llvm-project/commit/9b549f26fab6900b5328c0c6239fd77c7527159c DIFF: https://github.com/llvm/llvm-project/commit/9b549f26fab6900b5328c0c6239fd77c7527159c.diff

r328657 - AMDGPU: Update datalayout for stack alignment

Author: arsenm Date: Tue Mar 27 12:26:51 2018 New Revision: 328657 URL: http://llvm.org/viewvc/llvm-project?rev=328657&view=rev Log: AMDGPU: Update datalayout for stack alignment Modified: cfe/trunk/lib/Basic/Targets/AMDGPU.cpp cfe/trunk/test/CodeGen/target-data.c cfe/trunk/test/CodeG

r328791 - Fix typo

Author: arsenm Date: Thu Mar 29 07:31:59 2018 New Revision: 328791 URL: http://llvm.org/viewvc/llvm-project?rev=328791&view=rev Log: Fix typo Modified: cfe/trunk/include/clang/Basic/AttrDocs.td cfe/trunk/test/CodeGenCXX/catch-undef-behavior.cpp Modified: cfe/trunk/include/clang/Basic/Att

r318098 - OpenCL: Assume inline asm is convergent

Author: arsenm Date: Mon Nov 13 14:40:55 2017 New Revision: 318098 URL: http://llvm.org/viewvc/llvm-project?rev=318098&view=rev Log: OpenCL: Assume inline asm is convergent Already done for CUDA. Modified: cfe/trunk/lib/CodeGen/CGStmt.cpp cfe/trunk/test/CodeGenOpenCL/convergent.cl Modif

[clang] ffe12e7 - clang: Handle MatrixType in hasFloatingRepresentation

Author: Matt Arsenault Date: 2023-03-15T01:13:08-04:00 New Revision: ffe12e765cb9ba77dc30ccec89fb163234b98541 URL: https://github.com/llvm/llvm-project/commit/ffe12e765cb9ba77dc30ccec89fb163234b98541 DIFF: https://github.com/llvm/llvm-project/commit/ffe12e765cb9ba77dc30ccec89fb163234b98541.diff

[clang] dd81810 - clang: Emit nofpclass(nan inf) for -ffinite-math-only

Author: Matt Arsenault Date: 2023-03-15T01:13:08-04:00 New Revision: dd8181055494d8f973dfaaba282bb1ae6eab4d40 URL: https://github.com/llvm/llvm-project/commit/dd8181055494d8f973dfaaba282bb1ae6eab4d40 DIFF: https://github.com/llvm/llvm-project/commit/dd8181055494d8f973dfaaba282bb1ae6eab4d40.diff

[clang] 8e00934 - clang: Use ptrmask for pointer alignment

Author: Matt Arsenault Date: 2023-03-16T07:16:41-04:00 New Revision: 8e009348e8a2e9c4577538eba6ca5c6cb286776f URL: https://github.com/llvm/llvm-project/commit/8e009348e8a2e9c4577538eba6ca5c6cb286776f DIFF: https://github.com/llvm/llvm-project/commit/8e009348e8a2e9c4577538eba6ca5c6cb286776f.diff

[clang] 8709bca - clang: Add __builtin_elementwise_fma

Author: Matt Arsenault Date: 2023-02-24T21:55:08-04:00 New Revision: 8709bcacfb3a06847b47bb6b47e8556db43f3a43 URL: https://github.com/llvm/llvm-project/commit/8709bcacfb3a06847b47bb6b47e8556db43f3a43 DIFF: https://github.com/llvm/llvm-project/commit/8709bcacfb3a06847b47bb6b47e8556db43f3a43.diff

[clang] b57e049 - clang/HIP: Use __builtin_fmaf16

Author: Matt Arsenault Date: 2023-06-18T13:13:49-04:00 New Revision: b57e049fe5ccf48bb105b96020901d2912de9588 URL: https://github.com/llvm/llvm-project/commit/b57e049fe5ccf48bb105b96020901d2912de9588 DIFF: https://github.com/llvm/llvm-project/commit/b57e049fe5ccf48bb105b96020901d2912de9588.diff

[clang] 85232b0 - HIP: Directly call isfinite builtins

Author: Matt Arsenault Date: 2023-06-19T08:48:48-04:00 New Revision: 85232b0ecbf817bc1d70ae602cf44cf6ea03c0e6 URL: https://github.com/llvm/llvm-project/commit/85232b0ecbf817bc1d70ae602cf44cf6ea03c0e6 DIFF: https://github.com/llvm/llvm-project/commit/85232b0ecbf817bc1d70ae602cf44cf6ea03c0e6.diff

[clang] f407a73 - clang/HIP: Remove __llvm_amdgcn_* wrapper hacks

Author: Matt Arsenault Date: 2023-06-19T08:55:10-04:00 New Revision: f407a7399575a6821940973c54754d42e72dd9ce URL: https://github.com/llvm/llvm-project/commit/f407a7399575a6821940973c54754d42e72dd9ce DIFF: https://github.com/llvm/llvm-project/commit/f407a7399575a6821940973c54754d42e72dd9ce.diff

[clang] 2a488b4 - clang: Add __builtin_elementwise_round

Author: Matt Arsenault Date: 2023-06-19T11:32:56-04:00 New Revision: 2a488b4443a9c0aa0f368aed901676508ced202f URL: https://github.com/llvm/llvm-project/commit/2a488b4443a9c0aa0f368aed901676508ced202f DIFF: https://github.com/llvm/llvm-project/commit/2a488b4443a9c0aa0f368aed901676508ced202f.diff

[clang] c448e1d - HIP: Directly call ldexp builtins

Author: Matt Arsenault Date: 2023-06-21T21:34:42-04:00 New Revision: c448e1dabeeb066c41d893fc8e7a3d8bde2056c8 URL: https://github.com/llvm/llvm-project/commit/c448e1dabeeb066c41d893fc8e7a3d8bde2056c8 DIFF: https://github.com/llvm/llvm-project/commit/c448e1dabeeb066c41d893fc8e7a3d8bde2056c8.diff

[clang] d7feba7 - HIP: Directly call trunc builtins

Author: Matt Arsenault Date: 2023-06-23T09:11:06-04:00 New Revision: d7feba74b649cb8aabc0cb09b9bef5d79659e31c URL: https://github.com/llvm/llvm-project/commit/d7feba74b649cb8aabc0cb09b9bef5d79659e31c DIFF: https://github.com/llvm/llvm-project/commit/d7feba74b649cb8aabc0cb09b9bef5d79659e31c.diff

[clang] 7ba9506 - clang: Fix elementwise value naming to match instruction

Author: Matt Arsenault Date: 2023-06-23T19:39:04-04:00 New Revision: 7ba95064856cd1be9a3b48b70caa0ec4a2cda0e4 URL: https://github.com/llvm/llvm-project/commit/7ba95064856cd1be9a3b48b70caa0ec4a2cda0e4 DIFF: https://github.com/llvm/llvm-project/commit/7ba95064856cd1be9a3b48b70caa0ec4a2cda0e4.diff

[clang] 9d84f8d - clang: Add __builtin_elementwise_rint and nearbyint

Author: Matt Arsenault Date: 2023-06-23T19:52:06-04:00 New Revision: 9d84f8dc948b3188fdec9a1a080eb6d845c2082d URL: https://github.com/llvm/llvm-project/commit/9d84f8dc948b3188fdec9a1a080eb6d845c2082d DIFF: https://github.com/llvm/llvm-project/commit/9d84f8dc948b3188fdec9a1a080eb6d845c2082d.diff

[clang] e248955 - clang: Add missing cases to elementwise builtin SemaCXX test

Author: Matt Arsenault Date: 2023-06-23T19:52:06-04:00 New Revision: e248955aeda50bd7e0910eabdabdf7bd617e649f URL: https://github.com/llvm/llvm-project/commit/e248955aeda50bd7e0910eabdabdf7bd617e649f DIFF: https://github.com/llvm/llvm-project/commit/e248955aeda50bd7e0910eabdabdf7bd617e649f.diff

[clang] 53d28b2 - OpenMP: Use generated checks and pragma declare target

Author: Matt Arsenault Date: 2023-06-26T14:45:09-04:00 New Revision: 53d28b2a71063b974ec058ced614d1c95d5584fe URL: https://github.com/llvm/llvm-project/commit/53d28b2a71063b974ec058ced614d1c95d5584fe DIFF: https://github.com/llvm/llvm-project/commit/53d28b2a71063b974ec058ced614d1c95d5584fe.diff

[clang] dee1f5b - OpenMP: Don't include stdbool.h in builtin headers

Author: Matt Arsenault Date: 2023-06-26T14:45:09-04:00 New Revision: dee1f5b32c3a6a5694c3bb2fbf68d162447a5970 URL: https://github.com/llvm/llvm-project/commit/dee1f5b32c3a6a5694c3bb2fbf68d162447a5970 DIFF: https://github.com/llvm/llvm-project/commit/dee1f5b32c3a6a5694c3bb2fbf68d162447a5970.diff

[clang] afa1fc4 - OpenMP: Add fma math test

Author: Matt Arsenault Date: 2023-06-26T14:47:56-04:00 New Revision: afa1fc47b669d8cbd480cb50a0d10aea8e1e4007 URL: https://github.com/llvm/llvm-project/commit/afa1fc47b669d8cbd480cb50a0d10aea8e1e4007 DIFF: https://github.com/llvm/llvm-project/commit/afa1fc47b669d8cbd480cb50a0d10aea8e1e4007.diff

[clang] 51bb9b1 - OpenMP: Add missing test coverage for nothrow new/delete

Author: Matt Arsenault Date: 2023-06-28T06:55:15-04:00 New Revision: 51bb9b12f87ea315f750490c39cd993507202861 URL: https://github.com/llvm/llvm-project/commit/51bb9b12f87ea315f750490c39cd993507202861 DIFF: https://github.com/llvm/llvm-project/commit/51bb9b12f87ea315f750490c39cd993507202861.diff

[clang] 0e17174 - OpenMP: Fix nothrow new/delete for amdgpu

Author: Matt Arsenault Date: 2023-06-28T06:55:16-04:00 New Revision: 0e17174d40b2d25a755d2ce9acf7c905f254746d URL: https://github.com/llvm/llvm-project/commit/0e17174d40b2d25a755d2ce9acf7c905f254746d DIFF: https://github.com/llvm/llvm-project/commit/0e17174d40b2d25a755d2ce9acf7c905f254746d.diff

[clang] 85bdea0 - clang: Use new frexp intrinsic for builtins and add f16 version

Author: Matt Arsenault Date: 2023-06-28T14:50:17-04:00 New Revision: 85bdea023f5116f789095b606554739403042a21 URL: https://github.com/llvm/llvm-project/commit/85bdea023f5116f789095b606554739403042a21 DIFF: https://github.com/llvm/llvm-project/commit/85bdea023f5116f789095b606554739403042a21.diff

[clang] bf8e92c - HIP: Use frexp builtins in math headers

Author: Matt Arsenault Date: 2023-06-28T15:06:54-04:00 New Revision: bf8e92c0e792cbe3c9cc50607a1e33c6912ffd0e URL: https://github.com/llvm/llvm-project/commit/bf8e92c0e792cbe3c9cc50607a1e33c6912ffd0e DIFF: https://github.com/llvm/llvm-project/commit/bf8e92c0e792cbe3c9cc50607a1e33c6912ffd0e.diff

[clang] fe65043 - HIP: Directly call floor builtins

Author: Matt Arsenault Date: 2023-06-28T16:57:53-04:00 New Revision: fe65043a67c85345887339f371cca1a0c735a639 URL: https://github.com/llvm/llvm-project/commit/fe65043a67c85345887339f371cca1a0c735a639 DIFF: https://github.com/llvm/llvm-project/commit/fe65043a67c85345887339f371cca1a0c735a639.diff

[clang] b15bf30 - Reapply "clang: Use new frexp intrinsic for builtins and add f16 version"

Author: Matt Arsenault Date: 2023-06-30T09:07:23-04:00 New Revision: b15bf305ca3e9ce63aaef7247d32fb3a75174531 URL: https://github.com/llvm/llvm-project/commit/b15bf305ca3e9ce63aaef7247d32fb3a75174531 DIFF: https://github.com/llvm/llvm-project/commit/b15bf305ca3e9ce63aaef7247d32fb3a75174531.diff

[clang] 40bb302 - clang: Add start of header test for __clang_hip_libdevice_declares

Author: Matt Arsenault Date: 2023-06-30T12:39:11-04:00 New Revision: 40bb302c451ec1a8f6a2b8238e0a56448b8e1a12 URL: https://github.com/llvm/llvm-project/commit/40bb302c451ec1a8f6a2b8238e0a56448b8e1a12 DIFF: https://github.com/llvm/llvm-project/commit/40bb302c451ec1a8f6a2b8238e0a56448b8e1a12.diff

[clang] 917eddf - OpenMP: Don't use target regions in library function test

Author: Matt Arsenault Date: 2023-06-30T15:04:03-04:00 New Revision: 917eddfdcb15bddf67a54ede1f1643d5fc83628d URL: https://github.com/llvm/llvm-project/commit/917eddfdcb15bddf67a54ede1f1643d5fc83628d DIFF: https://github.com/llvm/llvm-project/commit/917eddfdcb15bddf67a54ede1f1643d5fc83628d.diff

[clang] 14c44df - clang: Update tests after InstSimplify change

Author: Matt Arsenault Date: 2023-06-02T18:47:58-04:00 New Revision: 14c44dfbcf1d6f81c1cdaa90ed243b3d53147903 URL: https://github.com/llvm/llvm-project/commit/14c44dfbcf1d6f81c1cdaa90ed243b3d53147903 DIFF: https://github.com/llvm/llvm-project/commit/14c44dfbcf1d6f81c1cdaa90ed243b3d53147903.diff

[clang] 7cbbbc0 - clang: Rename misleading test name

Author: Matt Arsenault Date: 2023-02-15T08:32:57-04:00 New Revision: 7cbbbc0a045bf1597972dabd2e071786157c87bf URL: https://github.com/llvm/llvm-project/commit/7cbbbc0a045bf1597972dabd2e071786157c87bf DIFF: https://github.com/llvm/llvm-project/commit/7cbbbc0a045bf1597972dabd2e071786157c87bf.diff

[clang] 43fd46f - HIP: Directly call fma builtins

Author: Matt Arsenault Date: 2023-04-26T19:27:11-04:00 New Revision: 43fd46fda3c90b014e8a73c62f67af9543ea4d59 URL: https://github.com/llvm/llvm-project/commit/43fd46fda3c90b014e8a73c62f67af9543ea4d59 DIFF: https://github.com/llvm/llvm-project/commit/43fd46fda3c90b014e8a73c62f67af9543ea4d59.diff

[clang] bc37be1 - LangRef: Add "dynamic" option to "denormal-fp-math"

Author: Matt Arsenault Date: 2023-04-29T08:44:59-04:00 New Revision: bc37be1855773c1dcf8c6bf577a096a81fd58652 URL: https://github.com/llvm/llvm-project/commit/bc37be1855773c1dcf8c6bf577a096a81fd58652 DIFF: https://github.com/llvm/llvm-project/commit/bc37be1855773c1dcf8c6bf577a096a81fd58652.diff

[clang] 8a21ea1 - clang: Start emitting intrinsic for __builtin_ldexp*

Author: Matt Arsenault Date: 2023-06-06T17:07:19-04:00 New Revision: 8a21ea1d0a3883a8d0aa15440388f91e49da4d08 URL: https://github.com/llvm/llvm-project/commit/8a21ea1d0a3883a8d0aa15440388f91e49da4d08 DIFF: https://github.com/llvm/llvm-project/commit/8a21ea1d0a3883a8d0aa15440388f91e49da4d08.diff

[clang] 66d4dba - HIP: Directly call fmin/fmax builtins

Author: Matt Arsenault Date: 2023-06-08T13:49:40-04:00 New Revision: 66d4dbae02db2c0feb32ea844500f9758f078dd7 URL: https://github.com/llvm/llvm-project/commit/66d4dbae02db2c0feb32ea844500f9758f078dd7 DIFF: https://github.com/llvm/llvm-project/commit/66d4dbae02db2c0feb32ea844500f9758f078dd7.diff

[clang] b51ae6d - HIP: Directly call signbit builtins

Author: Matt Arsenault Date: 2023-06-08T15:22:43-04:00 New Revision: b51ae6d31f86de2e47b1912eb045ea37a0c5de23 URL: https://github.com/llvm/llvm-project/commit/b51ae6d31f86de2e47b1912eb045ea37a0c5de23 DIFF: https://github.com/llvm/llvm-project/commit/b51ae6d31f86de2e47b1912eb045ea37a0c5de23.diff

[clang] d14ac1d - HIP: Directly call isinf builtins

Author: Matt Arsenault Date: 2023-06-08T15:25:01-04:00 New Revision: d14ac1d11a5cb3994c63ecaaa5c950636a289fa1 URL: https://github.com/llvm/llvm-project/commit/d14ac1d11a5cb3994c63ecaaa5c950636a289fa1 DIFF: https://github.com/llvm/llvm-project/commit/d14ac1d11a5cb3994c63ecaaa5c950636a289fa1.diff

[clang] 31dfd2b - HIP: Directly call isnan builtin

Author: Matt Arsenault Date: 2023-06-08T16:08:24-04:00 New Revision: 31dfd2b99ce31ae076aa35867dbede052693a6c9 URL: https://github.com/llvm/llvm-project/commit/31dfd2b99ce31ae076aa35867dbede052693a6c9 DIFF: https://github.com/llvm/llvm-project/commit/31dfd2b99ce31ae076aa35867dbede052693a6c9.diff

[clang] 4524a31 - HIP: Directly call copysign builtins

Author: Matt Arsenault Date: 2023-06-08T16:16:11-04:00 New Revision: 4524a3159e8407d2b453293eff694f3a1534462d URL: https://github.com/llvm/llvm-project/commit/4524a3159e8407d2b453293eff694f3a1534462d DIFF: https://github.com/llvm/llvm-project/commit/4524a3159e8407d2b453293eff694f3a1534462d.diff

[clang] eccc89b - AMDGPU: Add llvm.amdgcn.log intrinsic

Author: Matt Arsenault Date: 2023-06-12T21:10:30-04:00 New Revision: eccc89b26cc24c917bea0baa72ad5cd2c8d8fa1a URL: https://github.com/llvm/llvm-project/commit/eccc89b26cc24c917bea0baa72ad5cd2c8d8fa1a DIFF: https://github.com/llvm/llvm-project/commit/eccc89b26cc24c917bea0baa72ad5cd2c8d8fa1a.diff

[clang] 28f3edd - AMDGPU: Add llvm.amdgcn.exp2 intrinsic

Author: Matt Arsenault Date: 2023-06-15T07:00:07-04:00 New Revision: 28f3edd2be2a27b9dc7739d137147007c5fd9e41 URL: https://github.com/llvm/llvm-project/commit/28f3edd2be2a27b9dc7739d137147007c5fd9e41 DIFF: https://github.com/llvm/llvm-project/commit/28f3edd2be2a27b9dc7739d137147007c5fd9e41.diff

[clang] 735eccc - OpenMP: Add a new test for constantexpr evaluation of math headers

Author: Matt Arsenault Date: 2023-06-15T11:46:19-04:00 New Revision: 735eccc02353827a29781d9a6a1270dbca91574e URL: https://github.com/llvm/llvm-project/commit/735eccc02353827a29781d9a6a1270dbca91574e DIFF: https://github.com/llvm/llvm-project/commit/735eccc02353827a29781d9a6a1270dbca91574e.diff

[clang] 009af3f - HIP: Add a few more cmath header tests

Author: Matt Arsenault Date: 2023-06-15T11:51:09-04:00 New Revision: 009af3fcd78378662891938fce96d6be17f0040c URL: https://github.com/llvm/llvm-project/commit/009af3fcd78378662891938fce96d6be17f0040c DIFF: https://github.com/llvm/llvm-project/commit/009af3fcd78378662891938fce96d6be17f0040c.diff

[clang] b84721d - clang/AMDGPU: Emit atomicrmw for atomic_inc/dec builtins

Author: Matt Arsenault Date: 2023-06-16T20:18:50-04:00 New Revision: b84721df636747a9955408a934e713192a9cfe21 URL: https://github.com/llvm/llvm-project/commit/b84721df636747a9955408a934e713192a9cfe21 DIFF: https://github.com/llvm/llvm-project/commit/b84721df636747a9955408a934e713192a9cfe21.diff

[clang] 0b01e3d - clang: Convert builtins test to opaque pointers

Author: Matt Arsenault Date: 2022-12-05T09:01:52-05:00 New Revision: 0b01e3d0ae1f81b7e823397ed5a452025e8a751b URL: https://github.com/llvm/llvm-project/commit/0b01e3d0ae1f81b7e823397ed5a452025e8a751b DIFF: https://github.com/llvm/llvm-project/commit/0b01e3d0ae1f81b7e823397ed5a452025e8a751b.diff

Re: [PATCH] D23536: Remove excessive padding from ImmOp and RegOp.

arsenm accepted this revision. arsenm added a reviewer: arsenm. arsenm added a comment. This revision is now accepted and ready to land. LGTM https://reviews.llvm.org/D23536 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.o

Re: [PATCH] D23536: Remove excessive padding from ImmOp and RegOp.

arsenm closed this revision. arsenm added a comment. r278844 https://reviews.llvm.org/D23536 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Re: [PATCH] D23361: [OpenCL] AMDGCN: Fix size_t type

arsenm added inline comments. Comment at: lib/CodeGen/CodeGenModule.h:664 @@ +663,3 @@ + /// Get integer type of the same size as a pointer type. + llvm::IntegerType *getIntPtrTy(llvm::PointerType *PT); + This has the same purpose as DataLayout::getIntPtrType, s

[libclc] r279343 - amdgcn: Correct return types to be size_t

Author: arsenm Date: Fri Aug 19 17:49:39 2016 New Revision: 279343 URL: http://llvm.org/viewvc/llvm-project?rev=279343&view=rev Log: amdgcn: Correct return types to be size_t Modified: libclc/trunk/amdgcn/lib/workitem/get_global_offset.cl libclc/trunk/amdgcn/lib/workitem/get_group_id.cl

[libclc] r279350 - amdgcn: Fix get_local_size IR return type

Author: arsenm Date: Fri Aug 19 19:01:21 2016 New Revision: 279350 URL: http://llvm.org/viewvc/llvm-project?rev=279350&view=rev Log: amdgcn: Fix get_local_size IR return type Added: libclc/trunk/amdgcn/lib/workitem/get_local_size.ll libclc/trunk/r600/lib/workitem/get_local_size.ll -

[libclc] r279359 - amdgpu: Fix default case value for get_local_size

Author: arsenm Date: Fri Aug 19 23:17:17 2016 New Revision: 279359 URL: http://llvm.org/viewvc/llvm-project?rev=279359&view=rev Log: amdgpu: Fix default case value for get_local_size Modified: libclc/trunk/amdgcn/lib/workitem/get_local_size.ll libclc/trunk/r600/lib/workitem/get_local_size

r279463 - AMDGPU: Handle structs directly in AMDGPUABIInfo

Author: arsenm Date: Mon Aug 22 14:25:59 2016 New Revision: 279463 URL: http://llvm.org/viewvc/llvm-project?rev=279463&view=rev Log: AMDGPU: Handle structs directly in AMDGPUABIInfo Structs are currently handled as pointer + byval, which makes AMDGPU LLVM backend generate incorrect code when stru

Re: [PATCH] D20168: [CodeGen] Handle structs directly in AMDGPUABIInfo

arsenm closed this revision. arsenm added a comment. r279463 https://reviews.llvm.org/D20168 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[libclc] r279644 - amdgcn: Fix return type for get_global_size

Author: arsenm Date: Wed Aug 24 12:52:04 2016 New Revision: 279644 URL: http://llvm.org/viewvc/llvm-project?rev=279644&view=rev Log: amdgcn: Fix return type for get_global_size Added: libclc/trunk/amdgcn/lib/workitem/get_global_size.ll libclc/trunk/r600/lib/workitem/get_global_size.ll

[libclc] r279656 - amdgcn: Also correct get_local_size type for HSA

Author: arsenm Date: Wed Aug 24 14:11:52 2016 New Revision: 279656 URL: http://llvm.org/viewvc/llvm-project?rev=279656&view=rev Log: amdgcn: Also correct get_local_size type for HSA Modified: libclc/trunk/amdgcn-amdhsa/lib/workitem/get_local_size.ll Modified: libclc/trunk/amdgcn-amdhsa/lib/w

[libclc] r279692 - Strip opencl.ocl.version metadata

Author: arsenm Date: Wed Aug 24 19:25:10 2016 New Revision: 279692 URL: http://llvm.org/viewvc/llvm-project?rev=279692&view=rev Log: Strip opencl.ocl.version metadata This should be uniqued when linking, but right now it creates a lot of metadata spam listing the same version. This should also pr

[libclc] r279723 - amdgcn: Fix return type of get_num_groups

Author: arsenm Date: Thu Aug 25 02:31:40 2016 New Revision: 279723 URL: http://llvm.org/viewvc/llvm-project?rev=279723&view=rev Log: amdgcn: Fix return type of get_num_groups Added: libclc/trunk/amdgcn/lib/workitem/get_num_groups.ll libclc/trunk/r600/lib/workitem/get_num_groups.ll -

Re: [PATCH] D23746: Basic/Targets.cpp: Add polaris10 and polaris11 gpus

arsenm added inline comments. Comment at: lib/Basic/Targets.cpp:1959 @@ -1959,1 +1958,3 @@ +GK_VOLCANIC_ISLANDS, +GK_ARCTIC_ISLANDS } GPU; tstellarAMD wrote: > We're trying to move to more descriptive GPU family names, so this should be > GK_GFX8 inste

[PATCH] D23953: OpenCL: Defining __ENDIAN_LITTLE__ and fix target endianness

arsenm created this revision. arsenm added a subscriber: cfe-commits. OpenCL requires __ENDIAN_LITTLE__ be set for little endian targets. The default for targets was also apparently big endian, so AMDGPU was incorrectly reported as big endian. Set this from the triple so targets don't have another

Re: [PATCH] D23953: OpenCL: Defining __ENDIAN_LITTLE__ and fix target endianness

arsenm updated this revision to Diff 69465. arsenm added a comment. Fix typo https://reviews.llvm.org/D23953 Files: include/clang/Basic/TargetInfo.h lib/Basic/TargetInfo.cpp lib/Basic/Targets.cpp lib/Frontend/InitPreprocessor.cpp test/Preprocessor/init.c Index: test/Preprocessor/init

Re: [PATCH] D23953: OpenCL: Defining __ENDIAN_LITTLE__ and fix target endianness

arsenm added a comment. ping https://reviews.llvm.org/D23953 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

r280786 - Fix whitespace issues

Author: arsenm Date: Wed Sep 7 02:07:59 2016 New Revision: 280786 URL: http://llvm.org/viewvc/llvm-project?rev=280786&view=rev Log: Fix whitespace issues ^M and extra space Modified: cfe/trunk/lib/CodeGen/TargetInfo.cpp Modified: cfe/trunk/lib/CodeGen/TargetInfo.cpp URL: http://llvm.org/v

r280787 - OpenCL: Defining __ENDIAN_LITTLE__ and fix target endianness

Author: arsenm Date: Wed Sep 7 02:08:02 2016 New Revision: 280787 URL: http://llvm.org/viewvc/llvm-project?rev=280787&view=rev Log: OpenCL: Defining __ENDIAN_LITTLE__ and fix target endianness OpenCL requires __ENDIAN_LITTLE__ be set for little endian targets. The default for targets was also ap

Re: [PATCH] D23953: OpenCL: Defining __ENDIAN_LITTLE__ and fix target endianness

arsenm closed this revision. arsenm added a comment. r280787 https://reviews.llvm.org/D23953 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[libclc] r280961 - Replace nextafter implementation

Author: arsenm Date: Thu Sep 8 11:37:56 2016 New Revision: 280961 URL: http://llvm.org/viewvc/llvm-project?rev=280961&view=rev Log: Replace nextafter implementation This one passes conformance. Modified: libclc/trunk/amdgpu/lib/math/nextafter.cl libclc/trunk/generic/lib/math/clc_nextaft

r299877 - Update for lifetime intrinsic signature change

Author: arsenm Date: Mon Apr 10 15:18:45 2017 New Revision: 299877 URL: http://llvm.org/viewvc/llvm-project?rev=299877&view=rev Log: Update for lifetime intrinsic signature change Modified: cfe/trunk/lib/CodeGen/CGDecl.cpp cfe/trunk/test/CodeGen/cleanup-destslot-simple.c cfe/trunk/tes

r299889 - Update for AllocaInst construction changes

Author: arsenm Date: Mon Apr 10 17:28:02 2017 New Revision: 299889 URL: http://llvm.org/viewvc/llvm-project?rev=299889&view=rev Log: Update for AllocaInst construction changes Modified: cfe/trunk/lib/CodeGen/CGCall.cpp cfe/trunk/lib/CodeGen/CGExpr.cpp Modified: cfe/trunk/lib/CodeGen/CGCa

r292636 - AMDGPU: Add builtin for getreg intrinsic

Author: arsenm Date: Fri Jan 20 13:24:22 2017 New Revision: 292636 URL: http://llvm.org/viewvc/llvm-project?rev=292636&view=rev Log: AMDGPU: Add builtin for getreg intrinsic Modified: cfe/trunk/include/clang/Basic/BuiltinsAMDGPU.def cfe/trunk/test/CodeGenOpenCL/builtins-amdgcn.cl cfe/

r292838 - AMDGPU: Update for changed subtarget feature name

Author: arsenm Date: Mon Jan 23 16:31:14 2017 New Revision: 292838 URL: http://llvm.org/viewvc/llvm-project?rev=292838&view=rev Log: AMDGPU: Update for changed subtarget feature name Modified: cfe/trunk/lib/Basic/Targets.cpp cfe/trunk/test/CodeGenOpenCL/denorms-are-zero.cl Modified: cfe/

r293600 - AMDGPU: Add builtin for fmed3 intrinsic

Author: arsenm Date: Mon Jan 30 21:42:07 2017 New Revision: 293600 URL: http://llvm.org/viewvc/llvm-project?rev=293600&view=rev Log: AMDGPU: Add builtin for fmed3 intrinsic Modified: cfe/trunk/include/clang/Basic/BuiltinsAMDGPU.def cfe/trunk/lib/CodeGen/CGBuiltin.cpp cfe/trunk/test/Co

r294278 - AMDGPU: Add a test checking alignments of emitted globals/allocas

Author: arsenm Date: Mon Feb 6 22:28:02 2017 New Revision: 294278 URL: http://llvm.org/viewvc/llvm-project?rev=294278&view=rev Log: AMDGPU: Add a test checking alignments of emitted globals/allocas Make sure the spec required type alignments are used in preparation for a possible change which ma

[libclc] r294608 - math: Add native_rsqrt builtin function

Author: arsenm Date: Thu Feb 9 12:39:26 2017 New Revision: 294608 URL: http://llvm.org/viewvc/llvm-project?rev=294608&view=rev Log: math: Add native_rsqrt builtin function Trivial define to rsqrt. Patch by Vedran Miletić Added: libclc/trunk/generic/include/clc/math/native_rsqrt.h Modified

r295556 - AMDGPU: Add gfx900 and gfx901 processors

Author: arsenm Date: Sat Feb 18 13:02:41 2017 New Revision: 295556 URL: http://llvm.org/viewvc/llvm-project?rev=295556&view=rev Log: AMDGPU: Add gfx900 and gfx901 processors Modified: cfe/trunk/lib/Basic/Targets.cpp cfe/trunk/test/Driver/r600-mcpu.cl Modified: cfe/trunk/lib/Basic/Targets

r295874 - AMDGPU: Add fmed3 half builtin

Author: arsenm Date: Wed Feb 22 14:55:59 2017 New Revision: 295874 URL: http://llvm.org/viewvc/llvm-project?rev=295874&view=rev Log: AMDGPU: Add fmed3 half builtin Added: cfe/trunk/test/CodeGenOpenCL/builtins-amdgcn-gfx9.cl cfe/trunk/test/SemaOpenCL/builtins-amdgcn-error-gfx9.cl Modified:

r297659 - AMDGPU: Make 0 the private nullptr value

Author: arsenm Date: Mon Mar 13 14:47:53 2017 New Revision: 297659 URL: http://llvm.org/viewvc/llvm-project?rev=297659&view=rev Log: AMDGPU: Make 0 the private nullptr value We can't actually pretend that 0 is valid for address space 0. r295877 added a workaround to stop allocating user objects t

[PATCH] D26863: [AMDGPU] Change frexp.exp builtin to return i16 for f16 input

arsenm accepted this revision. arsenm added a comment. This revision is now accepted and ready to land. LGTM https://reviews.llvm.org/D26863 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-c

[PATCH] D26196: AMDGPU: Translate null pointers in private and local addr space

arsenm added inline comments. Comment at: test/CodeGenOpenCL/amdgpu-nullptr.cl:49 +} + I think there need to be a lot more tests. A few I can think of: - Tests that compare to NULL instead of 0 - Pointer compares without the explicit 0, i.e. if (p)/if(!p)

[PATCH] D26196: AMDGPU: Translate null pointers in private and local addr space

arsenm added inline comments. Comment at: test/CodeGenOpenCL/amdgpu-nullptr.cl:30 + +// CHECK: icmp eq i8 addrspace(1)* %p, null +void cmp_global(global char* p) { Missing check-label https://reviews.llvm.org/D26196 __

[PATCH] D26196: AMDGPU: Translate null pointers in private and local addr space

arsenm added inline comments. Comment at: lib/CodeGen/TargetInfo.cpp:6957-6958 + + llvm::Constant *translateNullPtr(const CodeGen::CodeGenModule &CGM, + llvm::Constant *C) const override; }; I was thinking that addrspacecast (generic null) should be valid

[PATCH] D26196: AMDGPU: Translate null pointers in private and local addr space

arsenm added a comment. In https://reviews.llvm.org/D26196#586433, @rjmccall wrote: > I think there are a number of conceptual questions here, chiefly among them > whether an llvm::ConstantPointerNull in an address-space type should have the > correct representation for the target. If the deci

[PATCH] D26476: [AMDGPU] Add support for f16 builtin functions for VI+

arsenm added inline comments. Comment at: test/CodeGenOpenCL/builtins-amdgcn-error-f16-class.cl:1-9 +// REQUIRES: amdgpu-registered-target +// RUN: %clang_cc1 -triple amdgcn-unknown-amdhsa -target-cpu tahiti -verify -S -o - %s + +#pragma OPENCL EXTENSION cl_khr_fp16 : enable + +

[PATCH] D26476: [AMDGPU] Add support for f16 builtin functions for VI+

arsenm added inline comments. Comment at: test/CodeGenOpenCL/builtins-amdgcn-error-f16-class.cl:1-9 +// REQUIRES: amdgpu-registered-target +// RUN: %clang_cc1 -triple amdgcn-unknown-amdhsa -target-cpu tahiti -verify -S -o - %s + +#pragma OPENCL EXTENSION cl_khr_fp16 : enable + +

[PATCH] D26476: [AMDGPU] Add support for f16 builtin functions for VI+

arsenm accepted this revision. arsenm added a comment. This revision is now accepted and ready to land. LGTM https://reviews.llvm.org/D26476 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-c

Re: [PATCH] D10586: Add HSAIL support

arsenm updated this revision to Diff 35157. arsenm added a comment. Herald added a subscriber: rampitec. Update for trunk. Fix __builtin_hsail_gridsize/__builtin_hsail_gridsizel http://reviews.llvm.org/D10586 Files: include/clang/Basic/BuiltinsHSAIL.def include/clang/Basic/TargetBuiltins.h

Re: [PATCH] D10586: Add HSAIL support

arsenm updated this revision to Diff 31825. arsenm added a comment. Update for trunk http://reviews.llvm.org/D10586 Files: include/clang/Basic/BuiltinsHSAIL.def include/clang/Basic/TargetBuiltins.h include/clang/module.modulemap lib/Basic/Targets.cpp lib/CodeGen/CGBuiltin.cpp lib/Co

[libclc] r275874 - R600: Use new barrier intrinsic

Author: arsenm Date: Mon Jul 18 13:42:17 2016 New Revision: 275874 URL: http://llvm.org/viewvc/llvm-project?rev=275874&view=rev Log: R600: Use new barrier intrinsic Modified: libclc/trunk/r600/lib/synchronization/barrier_impl.ll Modified: libclc/trunk/r600/lib/synchronization/barrier_impl.ll

<    2   3   4   5   6   7   8   9   10   11   >