[libclc] r332324 - rootn: Use denormal path only

Author: jvesely Date: Mon May 14 21:22:43 2018 New Revision: 332324 URL: http://llvm.org/viewvc/llvm-project?rev=332324&view=rev Log: rootn: Use denormal path only It's OK to either flush to 0 or return denormal result if the device does not support denormals. See sec 7.2 and 7.5.3 of OCL specs U

[libclc] r332677 - Add initial support for half precision builtins

Author: jvesely Date: Thu May 17 15:55:30 2018 New Revision: 332677 URL: http://llvm.org/viewvc/llvm-project?rev=332677&view=rev Log: Add initial support for half precision builtins v2: fix fmax implementation use consistent checks for __CLC_FP_SIZE add missing TODOs fix whitespace in

[libclc] r314111 - prepare_builtins: Fix compile breakage with older LLVM

Author: jvesely Date: Mon Sep 25 09:04:37 2017 New Revision: 314111 URL: http://llvm.org/viewvc/llvm-project?rev=314111&view=rev Log: prepare_builtins: Fix compile breakage with older LLVM Fixes r314050 reviewer: Tom Stellard Signed-off-by: Jan Vesely Modified: libclc/trunk/utils/prepare-

[libclc] r314112 - Rework atomic ops to use clang builtins rather than llvm asm

Author: jvesely Date: Mon Sep 25 09:07:34 2017 New Revision: 314112 URL: http://llvm.org/viewvc/llvm-project?rev=314112&view=rev Log: Rework atomic ops to use clang builtins rather than llvm asm reviewer: Aaron Watry Signed-off-by: Jan Vesely Added: libclc/trunk/generic/lib/atomic/atomic_a

[libclc] r314261 - Add missing HAVE_LLVM define to fix build with latest llvm

Author: jvesely Date: Tue Sep 26 16:15:54 2017 New Revision: 314261 URL: http://llvm.org/viewvc/llvm-project?rev=314261&view=rev Log: Add missing HAVE_LLVM define to fix build with latest llvm Broken since r314111 V2: pointed out by Jan Vesely - Use format() instead of % formating Patch-by:

[libclc] r314547 - travis: Check built libraries on llvm-3.9

Author: jvesely Date: Fri Sep 29 12:06:50 2017 New Revision: 314547 URL: http://llvm.org/viewvc/llvm-project?rev=314547&view=rev Log: travis: Check built libraries on llvm-3.9 Signed-off-by: Jan Vesely Acked-by: Aaron Watry Modified: libclc/trunk/.travis.yml Modified: libclc/trunk/.travis

[libclc] r314544 - travis: add build using llvm-3.9

Author: jvesely Date: Fri Sep 29 12:06:45 2017 New Revision: 314544 URL: http://llvm.org/viewvc/llvm-project?rev=314544&view=rev Log: travis: add build using llvm-3.9 Signed-off-by: Jan Vesely Acked-by: Aaron Watry Modified: libclc/trunk/.travis.yml Modified: libclc/trunk/.travis.yml URL:

[libclc] r314545 - geometric: geometric functions are only supported for vector lengths <=4

Author: jvesely Date: Fri Sep 29 12:06:47 2017 New Revision: 314545 URL: http://llvm.org/viewvc/llvm-project?rev=314545&view=rev Log: geometric: geometric functions are only supported for vector lengths <=4 Signed-off-by: Jan Vesely Reviewed-by: Aaron Watry Modified: libclc/trunk/generic/i

[libclc] r314548 - Fix amdgcn-amdhsa on llvm-3.9

Author: jvesely Date: Fri Sep 29 12:06:52 2017 New Revision: 314548 URL: http://llvm.org/viewvc/llvm-project?rev=314548&view=rev Log: Fix amdgcn-amdhsa on llvm-3.9 Signed-off-by: Jan Vesely Acked-by: Aaron Watry Added: libclc/trunk/amdgcn-amdhsa/lib/OVERRIDES_3.9 libclc/trunk/amdgcn-am

[libclc] r314546 - Add script to check for unresolved function calls

Author: jvesely Date: Fri Sep 29 12:06:48 2017 New Revision: 314546 URL: http://llvm.org/viewvc/llvm-project?rev=314546&view=rev Log: Add script to check for unresolved function calls v2: add shell shebang improve error checks and reporting v3: fix typo Signed-off-by: Jan Vesely Reviewed-by

[libclc] r314543 - Restore support for llvm-3.9

Author: jvesely Date: Fri Sep 29 12:06:41 2017 New Revision: 314543 URL: http://llvm.org/viewvc/llvm-project?rev=314543&view=rev Log: Restore support for llvm-3.9 Signed-off-by: Jan Vesely Acked-by: Aaron Watry Added: libclc/trunk/amdgcn/lib/OVERRIDES_3.9 libclc/trunk/amdgcn/lib/SOURCE

[libclc] r314703 - integer/sub_sat: Use clang builtin instead of llvm asm

Author: jvesely Date: Mon Oct 2 11:39:03 2017 New Revision: 314703 URL: http://llvm.org/viewvc/llvm-project?rev=314703&view=rev Log: integer/sub_sat: Use clang builtin instead of llvm asm reviewer: Tom Stellard Signed-off-by: Jan Vesely Removed: libclc/trunk/generic/lib/integer/sub_sat_if

[libclc] r314701 - integer/clz: Use clang builtin instead of llvm asm

Author: jvesely Date: Mon Oct 2 11:38:57 2017 New Revision: 314701 URL: http://llvm.org/viewvc/llvm-project?rev=314701&view=rev Log: integer/clz: Use clang builtin instead of llvm asm The generated llvm IR mostly identical. char/uchar case is a bit worse. reviewer: Tom Stellard Signed-off-by:

[libclc] r314702 - integer/add_sat: Use clang builtin instead of llvm asm

Author: jvesely Date: Mon Oct 2 11:39:00 2017 New Revision: 314702 URL: http://llvm.org/viewvc/llvm-project?rev=314702&view=rev Log: integer/add_sat: Use clang builtin instead of llvm asm reviewer: Tom Stellard Signed-off-by: Jan Vesely Removed: libclc/trunk/generic/lib/integer/add_sat_if

[libclc] r315018 - configure: Fix handling of directories with compats only source lists

Author: jvesely Date: Thu Oct 5 13:16:28 2017 New Revision: 315018 URL: http://llvm.org/viewvc/llvm-project?rev=315018&view=rev Log: configure: Fix handling of directories with compats only source lists Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modified: libclc/trunk/configure.py

[libclc] r315170 - ldexp: Fix double precision function return type

Author: jvesely Date: Sat Oct 7 23:56:14 2017 New Revision: 315170 URL: http://llvm.org/viewvc/llvm-project?rev=315170&view=rev Log: ldexp: Fix double precision function return type Fixes ~1200 external calls from nvtpx library. Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modified:

[libclc] r315189 - math/nextafter: Use custom declaration inc file

Author: jvesely Date: Sun Oct 8 12:33:55 2017 New Revision: 315189 URL: http://llvm.org/viewvc/llvm-project?rev=315189&view=rev Log: math/nextafter: Use custom declaration inc file Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Added: libclc/trunk/generic/include/clc/math/nextafter.inc

[libclc] r315191 - ptx: Use __clc_nextafter to implement nextafter

Author: jvesely Date: Sun Oct 8 12:34:00 2017 New Revision: 315191 URL: http://llvm.org/viewvc/llvm-project?rev=315191&view=rev Log: ptx: Use __clc_nextafter to implement nextafter using clang builtin results in external library call Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Added:

[libclc] r315190 - Do not include clc_nextafter header globally

Author: jvesely Date: Sun Oct 8 12:33:58 2017 New Revision: 315190 URL: http://llvm.org/viewvc/llvm-project?rev=315190&view=rev Log: Do not include clc_nextafter header globally Drop unused clc/math/clc_nextafter.h header Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Removed: libclc/t

[libclc] r315188 - math/binary_decl.inc: Do not declare mixed float/double functions

Author: jvesely Date: Sun Oct 8 12:33:53 2017 New Revision: 315188 URL: http://llvm.org/viewvc/llvm-project?rev=315188&view=rev Log: math/binary_decl.inc: Do not declare mixed float/double functions fmin/fmax only need vector/scalar mix Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modifi

[libclc] r315193 - travis: Make sure we report failure even if only earlier checked files fail

Author: jvesely Date: Sun Oct 8 13:07:58 2017 New Revision: 315193 URL: http://llvm.org/viewvc/llvm-project?rev=315193&view=rev Log: travis: Make sure we report failure even if only earlier checked files fail for loop would only report status of the last command v2: return '1' call test inst

[libclc] r315192 - check_external_calls.sh: Print number of calls in tested file.

Author: jvesely Date: Sun Oct 8 13:07:56 2017 New Revision: 315192 URL: http://llvm.org/viewvc/llvm-project?rev=315192&view=rev Log: check_external_calls.sh: Print number of calls in tested file. Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modified: libclc/trunk/check_external_calls.

[libclc] r315341 - Make image builtins r600/llvm-3.9 only

Author: jvesely Date: Tue Oct 10 11:10:21 2017 New Revision: 315341 URL: http://llvm.org/viewvc/llvm-project?rev=315341&view=rev Log: Make image builtins r600/llvm-3.9 only The implementation uses r600 sepcific intrinsics LLVM-4 switched to _ro_t and _rw_t image types Portions of the code can be

[libclc] r315343 - travis: enable checks of nvptx libraries

Author: jvesely Date: Tue Oct 10 11:10:25 2017 New Revision: 315343 URL: http://llvm.org/viewvc/llvm-project?rev=315343&view=rev Log: travis: enable checks of nvptx libraries Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modified: libclc/trunk/.travis.yml Modified: libclc/trunk/.travis

[libclc] r315342 - travis: Enable external function call checks on llvm-{4, 5}

Author: jvesely Date: Tue Oct 10 11:10:24 2017 New Revision: 315342 URL: http://llvm.org/viewvc/llvm-project?rev=315342&view=rev Log: travis: Enable external function call checks on llvm-{4,5} Reviewer: Aaron Watry Signed-off-by: Jan Vesely Modified: libclc/trunk/.travis.yml Modified: libc

r316181 - AMDGPU: Parse r600 CPU name early and expose FMAF capability

Author: jvesely Date: Thu Oct 19 13:40:13 2017 New Revision: 316181 URL: http://llvm.org/viewvc/llvm-project?rev=316181&view=rev Log: AMDGPU: Parse r600 CPU name early and expose FMAF capability Improve amdgcn macro test Differential Revision: https://reviews.llvm.org/D38667 Modified: cfe/tr

[libclc] r316238 - r600: Add missing datalayout to .ll files

Author: jvesely Date: Fri Oct 20 14:00:31 2017 New Revision: 316238 URL: http://llvm.org/viewvc/llvm-project?rev=316238&view=rev Log: r600: Add missing datalayout to .ll files Signed-off-by: Jan Vesely Acked-by: Aaron Watry Modified: libclc/trunk/r600/lib/synchronization/barrier_impl.ll

[libclc] r316239 - amdgcn: Add missing datalayout info to .ll files

Author: jvesely Date: Fri Oct 20 14:10:18 2017 New Revision: 316239 URL: http://llvm.org/viewvc/llvm-project?rev=316239&view=rev Log: amdgcn: Add missing datalayout info to .ll files Signed-off-by: Jan Vesely Acked-by: Aaron Watry Modified: libclc/trunk/amdgcn/lib/cl_khr_int64_extended_ato

[libclc] r316290 - shared: Implement aligned vector loads (vloada_half)

Author: jvesely Date: Sun Oct 22 07:21:56 2017 New Revision: 316290 URL: http://llvm.org/viewvc/llvm-project?rev=316290&view=rev Log: shared: Implement aligned vector loads (vloada_half) Passes newly posted piglits on turks and carrizo v2: add scalar vloada_half v3: fix typo Reviewer: Aaron Watr

[libclc] r316291 - shared: Implement aligned vector stores (vstorea_half)

Author: jvesely Date: Sun Oct 22 07:21:59 2017 New Revision: 316291 URL: http://llvm.org/viewvc/llvm-project?rev=316291&view=rev Log: shared: Implement aligned vector stores (vstorea_half) Float version passes newly posted piglit tests on turks, float and double pass on carrizo. v2: scalar vstor

[libclc] r316588 - math: Implement native_log10

Author: jvesely Date: Wed Oct 25 09:49:22 2017 New Revision: 316588 URL: http://llvm.org/viewvc/llvm-project?rev=316588&view=rev Log: math: Implement native_log10 Use llvm instrinsic by default Provide amdgpu workaround v2: drop old amd copyrights Reviewer: Aaron Watry Reviewed-by: Vedran Milet

[libclc] r316587 - amdgpu/math: Don't use llvm instrinsic for native_log

Author: jvesely Date: Wed Oct 25 09:49:17 2017 New Revision: 316587 URL: http://llvm.org/viewvc/llvm-project?rev=316587&view=rev Log: amdgpu/math: Don't use llvm instrinsic for native_log AMDGPU targets don't have insturction for it, so it'll be expanded to C * log2 anyway. v2: use native_log2 i

[libclc] r338898 - amdgcn: Use __constant AS for amdgcn builtins.

Author: jvesely Date: Fri Aug 3 08:14:08 2018 New Revision: 338898 URL: http://llvm.org/viewvc/llvm-project?rev=338898&view=rev Log: amdgcn: Use __constant AS for amdgcn builtins. Fixes build after clang r338707. Reviewer: matthew.arsena...@amd.com Signed-off-by: Jan Vesely Modified: libcl

[libclc] r324371 - vstore_half: Add support for custom rounding functions

Author: jvesely Date: Tue Feb 6 10:44:39 2018 New Revision: 324371 URL: http://llvm.org/viewvc/llvm-project?rev=324371&view=rev Log: vstore_half: Add support for custom rounding functions Add another layer of indirection This will be used for specific rounding modes Reviewer: Jeroen Ketema Sig

[libclc] r324372 - vstore_half: Consolidate declarations

Author: jvesely Date: Tue Feb 6 10:44:41 2018 New Revision: 324372 URL: http://llvm.org/viewvc/llvm-project?rev=324372&view=rev Log: vstore_half: Consolidate declarations Add support for rounding suffix Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modified: libclc/trunk/generic/incl

[libclc] r324373 - Add vstore_half_rtz implementation

Author: jvesely Date: Tue Feb 6 10:44:43 2018 New Revision: 324373 URL: http://llvm.org/viewvc/llvm-project?rev=324373&view=rev Log: Add vstore_half_rtz implementation Passes CTS on carrizo Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modified: libclc/trunk/generic/include/clc/share

[libclc] r324370 - vstore_half: Make sure the helper function is always inline

Author: jvesely Date: Tue Feb 6 10:44:35 2018 New Revision: 324370 URL: http://llvm.org/viewvc/llvm-project?rev=324370&view=rev Log: vstore_half: Make sure the helper function is always inline Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modified: libclc/trunk/generic/lib/shared/vsto

[libclc] r324374 - Add vstore_half_rtn implementation

Author: jvesely Date: Tue Feb 6 10:44:45 2018 New Revision: 324374 URL: http://llvm.org/viewvc/llvm-project?rev=324374&view=rev Log: Add vstore_half_rtn implementation Passes CTS on carrizo Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modified: libclc/trunk/generic/include/clc/share

[libclc] r324375 - Add vstore_half_rtp implementation

Author: jvesely Date: Tue Feb 6 10:44:47 2018 New Revision: 324375 URL: http://llvm.org/viewvc/llvm-project?rev=324375&view=rev Log: Add vstore_half_rtp implementation Passes CTS on carrizo Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modified: libclc/trunk/generic/include/clc/share

[libclc] r324376 - Add vstore_half_rte implementation

Author: jvesely Date: Tue Feb 6 10:44:50 2018 New Revision: 324376 URL: http://llvm.org/viewvc/llvm-project?rev=324376&view=rev Log: Add vstore_half_rte implementation Passes CTS on carrizo Reviewer: Jeroen Ketema Signed-off-by: Jan Vesely Modified: libclc/trunk/generic/include/clc/share

[libclc] r325054 - amdgpu/half_sqrt: Switch implementation to native_sqrt

Author: jvesely Date: Tue Feb 13 14:09:33 2018 New Revision: 325054 URL: http://llvm.org/viewvc/llvm-project?rev=325054&view=rev Log: amdgpu/half_sqrt: Switch implementation to native_sqrt Reviewer: Tom Stellard Signed-off-by: Jan Vesely Added: libclc/trunk/amdgpu/lib/math/half_sqrt.cl Mod

[libclc] r325056 - amdgpu/half_exp10: Switch implementation to native_exp10

Author: jvesely Date: Tue Feb 13 14:09:37 2018 New Revision: 325056 URL: http://llvm.org/viewvc/llvm-project?rev=325056&view=rev Log: amdgpu/half_exp10: Switch implementation to native_exp10 Reviewer: Tom Stellard Signed-off-by: Jan Vesely Added: libclc/trunk/amdgpu/lib/math/half_exp10.cl

[libclc] r325057 - amdgpu/half_exp2: Switch implementation to native_exp2

Author: jvesely Date: Tue Feb 13 14:09:38 2018 New Revision: 325057 URL: http://llvm.org/viewvc/llvm-project?rev=325057&view=rev Log: amdgpu/half_exp2: Switch implementation to native_exp2 Reviewer: Tom Stellard Signed-off-by: Jan Vesely Added: libclc/trunk/amdgpu/lib/math/half_exp2.cl Mod

[libclc] r325059 - amdgpu/half_log10: Switch implementation to native_log10

Author: jvesely Date: Tue Feb 13 14:09:42 2018 New Revision: 325059 URL: http://llvm.org/viewvc/llvm-project?rev=325059&view=rev Log: amdgpu/half_log10: Switch implementation to native_log10 Reviewer: Tom Stellard Signed-off-by: Jan Vesely Added: libclc/trunk/amdgpu/lib/math/half_log10.cl

[libclc] r325053 - amdgpu/half_rsqrt: Switch implementation to native_rsqrt

Author: jvesely Date: Tue Feb 13 14:09:31 2018 New Revision: 325053 URL: http://llvm.org/viewvc/llvm-project?rev=325053&view=rev Log: amdgpu/half_rsqrt: Switch implementation to native_rsqrt Reviewer: Tom Stellard Signed-off-by: Jan Vesely Added: libclc/trunk/amdgpu/lib/math/half_native_un

[libclc] r325058 - amdgpu/half_log: Switch implementation to native_log

Author: jvesely Date: Tue Feb 13 14:09:41 2018 New Revision: 325058 URL: http://llvm.org/viewvc/llvm-project?rev=325058&view=rev Log: amdgpu/half_log: Switch implementation to native_log Reviewer: Tom Stellard Signed-off-by: Jan Vesely Added: libclc/trunk/amdgpu/lib/math/half_log.cl Modifi

[libclc] r325055 - amdgpu/half_exp: Switch implementation to native_exp

Author: jvesely Date: Tue Feb 13 14:09:35 2018 New Revision: 325055 URL: http://llvm.org/viewvc/llvm-project?rev=325055&view=rev Log: amdgpu/half_exp: Switch implementation to native_exp Reviewer: Tom Stellard Signed-off-by: Jan Vesely Added: libclc/trunk/amdgpu/lib/math/half_exp.cl Modifi

[libclc] r325060 - amdgpu/half_log2: Switch implementation to native_log2

Author: jvesely Date: Tue Feb 13 14:09:44 2018 New Revision: 325060 URL: http://llvm.org/viewvc/llvm-project?rev=325060&view=rev Log: amdgpu/half_log2: Switch implementation to native_log2 Reviewer: Tom Stellard Signed-off-by: Jan Vesely Added: libclc/trunk/amdgpu/lib/math/half_log2.cl Mod

[libclc] r325061 - amdgpu/half_recip: Switch implementation to native_recip

Author: jvesely Date: Tue Feb 13 14:09:46 2018 New Revision: 325061 URL: http://llvm.org/viewvc/llvm-project?rev=325061&view=rev Log: amdgpu/half_recip: Switch implementation to native_recip Reviewer: Tom Stellard Signed-off-by: Jan Vesely Added: libclc/trunk/amdgpu/lib/math/half_recip.cl

Re: r325203 - Reapply r325193

On Thu, 2018-02-15 at 02:37 +, Konstantin Zhuravlyov via cfe- commits wrote: > Author: kzhuravl > Date: Wed Feb 14 18:37:04 2018 > New Revision: 325203 > > URL: http://llvm.org/viewvc/llvm-project?rev=325203&view=rev > Log: > Reapply r325193 Ca you please not do that ? Is it that difficul

[libclc] r325863 - amdgcn: Fix datalayout after clang r324101

Author: jvesely Date: Thu Feb 22 23:36:39 2018 New Revision: 325863 URL: http://llvm.org/viewvc/llvm-project?rev=325863&view=rev Log: amdgcn: Fix datalayout after clang r324101 r324101 switched around AS numbering Acked-by: Aaron Watry Signed-off-by: Jan Vesely Added: libclc/trunk/amdgcn/

[libclc] r325865 - amdgcn: Fix datalayout after addition of 32bit const AS in r324747

Author: jvesely Date: Thu Feb 22 23:36:54 2018 New Revision: 325865 URL: http://llvm.org/viewvc/llvm-project?rev=325865&view=rev Log: amdgcn: Fix datalayout after addition of 32bit const AS in r324747 Acked-by: Aaron Watry Signed-off-by: Jan Vesely Modified: libclc/trunk/amdgcn/lib/cl_khr_

[libclc] r325864 - r600: Fix datalayout after clang r324101

Author: jvesely Date: Thu Feb 22 23:36:51 2018 New Revision: 325864 URL: http://llvm.org/viewvc/llvm-project?rev=325864&view=rev Log: r600: Fix datalayout after clang r324101 r324101 switched around AS numbering Acked-by: Aaron Watry Signed-off-by: Jan Vesely Added: libclc/trunk/r600/lib/

[libclc] r325866 - amdgcn: Fix build after GDS/const AS swap in r325030

Author: jvesely Date: Thu Feb 22 23:37:01 2018 New Revision: 325866 URL: http://llvm.org/viewvc/llvm-project?rev=325866&view=rev Log: amdgcn: Fix build after GDS/const AS swap in r325030 Acked-by: Aaron Watry Signed-off-by: Jan Vesely Added: libclc/trunk/amdgcn-amdhsa/lib/OVERRIDES_4.0

[libclc] r325867 - utils: Adapt to llvm r325155

Author: jvesely Date: Thu Feb 22 23:37:03 2018 New Revision: 325867 URL: http://llvm.org/viewvc/llvm-project?rev=325867&view=rev Log: utils: Adapt to llvm r325155 r325155 ("Pass a reference to a module to the bitcode writer.") changed bit writer interface from pointer to reference Reviewer: Aaro

[libclc] r304556 - generic: add missing get_work_dim include

Author: jvesely Date: Fri Jun 2 10:58:35 2017 New Revision: 304556 URL: http://llvm.org/viewvc/llvm-project?rev=304556&view=rev Log: generic: add missing get_work_dim include Fixes few piglits since clang r304193 Signed-off-by: Jan Vesely Reviewed-by: Aaron Watry Modified: libclc/trunk/g

[libclc] r326816 - Move cl_khr_fp64 exntension enablement to gentype include lists

Author: jvesely Date: Tue Mar 6 09:48:35 2018 New Revision: 326816 URL: http://llvm.org/viewvc/llvm-project?rev=326816&view=rev Log: Move cl_khr_fp64 exntension enablement to gentype include lists This will make adding cl_khr_fp16 support easier Reviewed-by: Aaron Watry Signed-off-by: Jan Vese

[libclc] r326817 - maxmag: Condition variable needs to be the same bitwidth as operands

Author: jvesely Date: Tue Mar 6 09:48:38 2018 New Revision: 326817 URL: http://llvm.org/viewvc/llvm-project?rev=326817&view=rev Log: maxmag: Condition variable needs to be the same bitwidth as operands No changes wrt CTS Reviewed-by: Aaron Watry Signed-off-by: Jan Vesely Modified: libclc

[libclc] r326818 - minmag: Condition variable needs to be the same bitwidth as operands

Author: jvesely Date: Tue Mar 6 09:48:40 2018 New Revision: 326818 URL: http://llvm.org/viewvc/llvm-project?rev=326818&view=rev Log: minmag: Condition variable needs to be the same bitwidth as operands No changes wrt CTS Reviewed-by: Aaron Watry Signed-off-by: Jan Vesely Modified: libclc

[libclc] r326819 - select: Add vector implementation

Author: jvesely Date: Tue Mar 6 09:48:43 2018 New Revision: 326819 URL: http://llvm.org/viewvc/llvm-project?rev=326819&view=rev Log: select: Add vector implementation Passes CTS on Carrizo Reviewed-by: Aaron Watry Signed-off-by: Jan Vesely Added: libclc/trunk/generic/include/clc/relation

[libclc] r326820 - frexp: Reuse types provided by gentype.inc

Author: jvesely Date: Tue Mar 6 09:48:45 2018 New Revision: 326820 URL: http://llvm.org/viewvc/llvm-project?rev=326820&view=rev Log: frexp: Reuse types provided by gentype.inc v2: Use select instead of bitselect to consolidate scalar and vector versions Passes CTS on Carrizo Reviewed-by: Aaron

[libclc] r326821 - lgamma_r: Move code from .inc to .cl file

Author: jvesely Date: Tue Mar 6 09:48:47 2018 New Revision: 326821 URL: http://llvm.org/viewvc/llvm-project?rev=326821&view=rev Log: lgamma_r: Move code from .inc to .cl file Reviewed-by: Aaron Watry Signed-off-by: Jan Vesely Modified: libclc/trunk/generic/lib/math/lgamma_r.cl libclc/

[libclc] r327042 - popcount: Provide function implementation rather than intrinsic redirect

Author: jvesely Date: Thu Mar 8 10:58:00 2018 New Revision: 327042 URL: http://llvm.org/viewvc/llvm-project?rev=327042&view=rev Log: popcount: Provide function implementation rather than intrinsic redirect amdgcn will need to override this Reviewer: Aaron Watry Signed-off-by: Jan Vesely Adde

[libclc] r327043 - integer/gentype: Add __CLC_VECSIZE macro

Author: jvesely Date: Thu Mar 8 10:58:05 2018 New Revision: 327043 URL: http://llvm.org/viewvc/llvm-project?rev=327043&view=rev Log: integer/gentype: Add __CLC_VECSIZE macro Reviewer: Aaron Watry Signed-off-by: Jan Vesely Modified: libclc/trunk/generic/include/clc/integer/gentype.inc Mod

[libclc] r327044 - amdgcn, popcount: Workaround broken llvm.ctpop intrinsic on some GCN ASICs

Author: jvesely Date: Thu Mar 8 10:58:07 2018 New Revision: 327044 URL: http://llvm.org/viewvc/llvm-project?rev=327044&view=rev Log: amdgcn,popcount: Workaround broken llvm.ctpop intrinsic on some GCN ASICs This is only really needed for VI+ ASICs. However, llvm would cast the value to i32 for o

[libclc] r327075 - amdgcn/fmin: fcanonicalize operands

Author: jvesely Date: Thu Mar 8 15:00:58 2018 New Revision: 327075 URL: http://llvm.org/viewvc/llvm-project?rev=327075&view=rev Log: amdgcn/fmin: fcanonicalize operands v_min instruction needs canonicalized operands. Passes CTS on carrizo Reviewer: Aaron Watry Signed-off-by: Jan Vesely Added

[libclc] r327076 - amdgcn/fmax: fcanonicalize operands

Author: jvesely Date: Thu Mar 8 15:01:01 2018 New Revision: 327076 URL: http://llvm.org/viewvc/llvm-project?rev=327076&view=rev Log: amdgcn/fmax: fcanonicalize operands v_max instruction needs canonicalized operands. Passes CTS on carrizo Reviewer: Aaron Watry Signed-off-by: Jan Vesely Added

[libclc] r327323 - travis: Add build using llvm-6

Author: jvesely Date: Mon Mar 12 12:46:48 2018 New Revision: 327323 URL: http://llvm.org/viewvc/llvm-project?rev=327323&view=rev Log: travis: Add build using llvm-6 Acked-by: Aaron Watry Signed-off-by: Jan Vesely Modified: libclc/trunk/.travis.yml Modified: libclc/trunk/.travis.yml URL:

[libclc] r327324 - nan: Implement

Author: jvesely Date: Mon Mar 12 12:46:52 2018 New Revision: 327324 URL: http://llvm.org/viewvc/llvm-project?rev=327324&view=rev Log: nan: Implement Passes CTS on carrizo and turks Reviewer: Aaron Watry Signed-off-by: Jan Vesely Added: libclc/trunk/generic/include/clc/math/nan.h libcl

[libclc] r327818 - remainder: Port from amd builtins

Author: jvesely Date: Sun Mar 18 18:01:10 2018 New Revision: 327818 URL: http://llvm.org/viewvc/llvm-project?rev=327818&view=rev Log: remainder: Port from amd builtins Mostly ported from amd_builtins, uses only denormal path for fp32. Passes CTS on carrizo and turks Reviewer: Aaron Watry Signed

[libclc] r334226 - math/fma: Add fp32 software implementation

Author: jvesely Date: Thu Jun 7 13:27:43 2018 New Revision: 334226 URL: http://llvm.org/viewvc/llvm-project?rev=334226&view=rev Log: math/fma: Add fp32 software implementation Passes CTS on carrizo (when forced to use sw fma) and turks. Reviewer: Tom Stellard Signed-off-by: Jan Vesely Added:

[libclc] r334227 - r600/fmax: Flush denormals before calling builtin.

Author: jvesely Date: Thu Jun 7 13:27:56 2018 New Revision: 334227 URL: http://llvm.org/viewvc/llvm-project?rev=334227&view=rev Log: r600/fmax: Flush denormals before calling builtin. Same reason as amdgcn. Fixes fmax, maxmag CTS on turks. Reviewer: Tom Stellard Signed-off-by: Jan Vesely Adde

[libclc] r334228 - r600/fmin: Flush denormals before calling builtin.

Author: jvesely Date: Thu Jun 7 13:27:58 2018 New Revision: 334228 URL: http://llvm.org/viewvc/llvm-project?rev=334228&view=rev Log: r600/fmin: Flush denormals before calling builtin. Same reason as amdgcn. Fixes fmin, minmag CTS on turks. Reviewer: Tom Stellard Signed-off-by: Jan Vesely Adde

[libclc] r335277 - atom: Consolidate cl_khr_int64_{base, extended}_atomics declarations

Author: jvesely Date: Thu Jun 21 12:27:23 2018 New Revision: 335277 URL: http://llvm.org/viewvc/llvm-project?rev=335277&view=rev Log: atom: Consolidate cl_khr_int64_{base,extended}_atomics declarations Reviewed-By: Aaron Watry Tested-By: Aaron Watry Signed-off-by: Jan Vesely Added: libclc

[libclc] r335275 - atomic: Cleanup atomic_cmpxchg header

Author: jvesely Date: Thu Jun 21 12:27:12 2018 New Revision: 335275 URL: http://llvm.org/viewvc/llvm-project?rev=335275&view=rev Log: atomic: Cleanup atomic_cmpxchg header It's easier to just list the four function declarations Reviewed-By: Aaron Watry Tested-By: Aaron Watry Signed-off-by: Jan

[libclc] r335276 - atom: Consolidate cl_khr_{local, global}_int32_{base, extended}_atomics declarations

Author: jvesely Date: Thu Jun 21 12:27:18 2018 New Revision: 335276 URL: http://llvm.org/viewvc/llvm-project?rev=335276&view=rev Log: atom: Consolidate cl_khr_{local,global}_int32_{base,extended}_atomics declarations Reviewed-By: Aaron Watry Tested-By: Aaron Watry Signed-off-by: Jan Vesely A

[libclc] r335278 - atomic: Provide function implementation of atomic_{dec, inc}

Author: jvesely Date: Thu Jun 21 12:27:26 2018 New Revision: 335278 URL: http://llvm.org/viewvc/llvm-project?rev=335278&view=rev Log: atomic: Provide function implementation of atomic_{dec,inc} Reviewed-By: Aaron Watry Tested-By: Aaron Watry Signed-off-by: Jan Vesely Added: libclc/trunk/g

[libclc] r335274 - atomic: Move define cleanup to shared include

Author: jvesely Date: Thu Jun 21 12:27:07 2018 New Revision: 335274 URL: http://llvm.org/viewvc/llvm-project?rev=335274&view=rev Log: atomic: Move define cleanup to shared include Reviewed-By: Aaron Watry Tested-By: Aaron Watry Signed-off-by: Jan Vesely Modified: libclc/trunk/generic/incl

[libclc] r335279 - atom: Consolidate cl_khr_{local, global}_int32_{base, extended}_atomics implementation

Author: jvesely Date: Thu Jun 21 12:27:33 2018 New Revision: 335279 URL: http://llvm.org/viewvc/llvm-project?rev=335279&view=rev Log: atom: Consolidate cl_khr_{local,global}_int32_{base,extended}_atomics implementation These are just atomic_* wrappers. Switch inc, dec to use atomic_* wrappers as

[libclc] r335280 - atom: Use volatile pointers for cl_khr_{global, local}_int32_{base, extended}_atomics

Author: jvesely Date: Thu Jun 21 12:27:39 2018 New Revision: 335280 URL: http://llvm.org/viewvc/llvm-project?rev=335280&view=rev Log: atom: Use volatile pointers for cl_khr_{global,local}_int32_{base,extended}_atomics int64 versions were switched to volatile pointers in cl1.1 cl1.1 also renamed

Re: r345866 - Reapply Logging: make os_log buffer size an integer constant expression.

Hi, this patch seems to cause a build failure: /usr/bin/ld: CMakeFiles/clangAST.dir/OSLog.cpp.o: in function `(anonymous namespace)::OSLogFormatStringHandler::~OSLogFormatStringHandler()': /home/jvesely/llvm/tools/clang/lib/AST/OSLog.cpp:18: undefined reference to `clang::analyze_format_string::F

Re: r345866 - Reapply Logging: make os_log buffer size an integer constant expression.

On Thu, 2018-11-01 at 18:40 +, Tim Northover via cfe-commits wrote: > On 1 Nov 2018, at 18:36, Jan Vesely wrote: > > this patch seems to cause a build failure: > > Ugh. Reverted as r345871. > > Do you know what linker and other options you’re using? The only > failure I’ve been able to repr

[libclc] r346076 - r600: Convert get_global_size to clc

Author: jvesely Date: Sat Nov 3 17:35:08 2018 New Revision: 346076 URL: http://llvm.org/viewvc/llvm-project?rev=346076&view=rev Log: r600: Convert get_global_size to clc Signed-off-by: Jan Vesely Reviewer: Aaron Watry Added: libclc/trunk/r600/lib/workitem/get_global_size.cl Removed: li

[libclc] r346075 - r600: Convert get_local_size to clc

Author: jvesely Date: Sat Nov 3 17:35:03 2018 New Revision: 346075 URL: http://llvm.org/viewvc/llvm-project?rev=346075&view=rev Log: r600: Convert get_local_size to clc Signed-off-by: Jan Vesely Reviewer: Aaron Watry Added: libclc/trunk/r600/lib/workitem/get_local_size.cl Removed: libc

[libclc] r346078 - r600: Convert barrier to clc

Author: jvesely Date: Sat Nov 3 17:35:15 2018 New Revision: 346078 URL: http://llvm.org/viewvc/llvm-project?rev=346078&view=rev Log: r600: Convert barrier to clc Signed-off-by: Jan Vesely Reviewer: Aaron Watry Added: libclc/trunk/r600/lib/synchronization/barrier.cl Removed: libclc/trun

[libclc] r346077 - r600: Convert get_num_groups to clc

Author: jvesely Date: Sat Nov 3 17:35:12 2018 New Revision: 346077 URL: http://llvm.org/viewvc/llvm-project?rev=346077&view=rev Log: r600: Convert get_num_groups to clc Signed-off-by: Jan Vesely Reviewer: Aaron Watry Added: libclc/trunk/r600/lib/workitem/get_num_groups.cl Removed: libc

[libclc] r346080 - amdgcn: Convert get_global_size to clc

Author: jvesely Date: Sat Nov 3 17:39:20 2018 New Revision: 346080 URL: http://llvm.org/viewvc/llvm-project?rev=346080&view=rev Log: amdgcn: Convert get_global_size to clc Signed-off-by: Jan Vesely Reviewed-By: Aaron Watry Added: libclc/trunk/amdgcn/lib/workitem/get_global_size.cl Removed

[libclc] r346079 - amdgcn: Convert get_local_size to clc

Author: jvesely Date: Sat Nov 3 17:39:16 2018 New Revision: 346079 URL: http://llvm.org/viewvc/llvm-project?rev=346079&view=rev Log: amdgcn: Convert get_local_size to clc Signed-off-by: Jan Vesely Reviewed-By: Aaron Watry Added: libclc/trunk/amdgcn/lib/workitem/get_local_size.cl Removed:

[libclc] r346082 - amdgcn: Move __clc_amdgcn_s_waitcnt definition to clc file

Author: jvesely Date: Sat Nov 3 17:39:27 2018 New Revision: 346082 URL: http://llvm.org/viewvc/llvm-project?rev=346082&view=rev Log: amdgcn: Move __clc_amdgcn_s_waitcnt definition to clc file Signed-off-by: Jan Vesely Reviewed-By: Aaron Watry Removed: libclc/trunk/amdgcn/lib/mem_fence/wai

[libclc] r346083 - amdgcn-amdhsa: Convert get_{global, local}_size to clc for all llvm versions

Author: jvesely Date: Sat Nov 3 17:39:30 2018 New Revision: 346083 URL: http://llvm.org/viewvc/llvm-project?rev=346083&view=rev Log: amdgcn-amdhsa: Convert get_{global,local}_size to clc for all llvm versions Signed-off-by: Jan Vesely Acked-By: Aaron Watry Removed: libclc/trunk/amdgcn-amd

[libclc] r346081 - amdgcn: Convert get_num_groups to clc

Author: jvesely Date: Sat Nov 3 17:39:25 2018 New Revision: 346081 URL: http://llvm.org/viewvc/llvm-project?rev=346081&view=rev Log: amdgcn: Convert get_num_groups to clc Signed-off-by: Jan Vesely Reviewed-By: Aaron Watry Added: libclc/trunk/amdgcn/lib/workitem/get_num_groups.cl Removed:

[libclc] r346084 - travis: Check tahiti-amdgcn-mesa-mesa3d.bc

Author: jvesely Date: Sat Nov 3 17:54:43 2018 New Revision: 346084 URL: http://llvm.org/viewvc/llvm-project?rev=346084&view=rev Log: travis: Check tahiti-amdgcn-mesa-mesa3d.bc Signed-off-by: Jan Vesely Reviewed-By: Aaron Watry Modified: libclc/trunk/.travis.yml Modified: libclc/trunk/.tr

[libclc] r346085 - configure: Provide symlink for amdgcn-mesa3d instead of configure hack

Author: jvesely Date: Sat Nov 3 17:54:45 2018 New Revision: 346085 URL: http://llvm.org/viewvc/llvm-project?rev=346085&view=rev Log: configure: Provide symlink for amdgcn-mesa3d instead of configure hack Signed-off-by: Jan Vesely Reviewed-By: Aaron Watry Added: libclc/trunk/amdgcn-mesa3d

[libclc] r346086 - Remove redundant OVERRRIDES file

Author: jvesely Date: Sat Nov 3 17:54:46 2018 New Revision: 346086 URL: http://llvm.org/viewvc/llvm-project?rev=346086&view=rev Log: Remove redundant OVERRRIDES file Signed-off-by: Jan Vesely Reviewed-By: Aaron Watry Removed: libclc/trunk/amdgpu/lib/OVERRIDES Removed: libclc/trunk/amdgpu

[libclc] r346597 - r600: Add datalayout to image builtin implementation

Author: jvesely Date: Sat Nov 10 13:43:40 2018 New Revision: 346597 URL: http://llvm.org/viewvc/llvm-project?rev=346597&view=rev Log: r600: Add datalayout to image builtin implementation Signed-off-by: Jan Vesely Reviewer: Aaron Watry Modified: libclc/trunk/r600/lib/image/get_image_attribut

[libclc] r347665 - amdgcn: Consolidate atomic minmax helpers

Author: jvesely Date: Tue Nov 27 08:01:13 2018 New Revision: 347665 URL: http://llvm.org/viewvc/llvm-project?rev=347665&view=rev Log: amdgcn: Consolidate atomic minmax helpers Removes most overrides Signed-off-by: Jan Vesely Reviewer: Aaron Watry Removed: libclc/trunk/amdgcn/lib/OVERRIDES

[libclc] r347664 - configure: Add target specific asm rule.

Author: jvesely Date: Tue Nov 27 08:01:10 2018 New Revision: 347664 URL: http://llvm.org/viewvc/llvm-project?rev=347664&view=rev Log: configure: Add target specific asm rule. Run the file through target specific preprocessing stage. Signed-off-by: Jan Vesely Reviewer: Aaron Watry Modified:

[libclc] r347663 - configure: provide llvm_as helper variable

Author: jvesely Date: Tue Nov 27 08:01:01 2018 New Revision: 347663 URL: http://llvm.org/viewvc/llvm-project?rev=347663&view=rev Log: configure: provide llvm_as helper variable Signed-off-by: Jan Vesely Reviewer: Aaron Watry Modified: libclc/trunk/configure.py Modified: libclc/trunk/config

[libclc] r347666 - r600: Remove empty OVERRIDES file

Author: jvesely Date: Tue Nov 27 08:01:16 2018 New Revision: 347666 URL: http://llvm.org/viewvc/llvm-project?rev=347666&view=rev Log: r600: Remove empty OVERRIDES file Signed-off-by: Jan Vesely Reviewer: Aaron Watry Removed: libclc/trunk/r600/lib/OVERRIDES Removed: libclc/trunk/r600/lib/OV

[libclc] r347667 - Add cmake build system

Author: jvesely Date: Tue Nov 27 08:07:19 2018 New Revision: 347667 URL: http://llvm.org/viewvc/llvm-project?rev=347667&view=rev Log: Add cmake build system Add cmake support for CLC and ll asm language, the latter includes clang preprocessing stage. Add ctests to check for external function call

  1   2   3   4   >