Gráfico de commits

224 Commits

Autor SHA1 Mensaje Fecha
Maneesh Gupta 874af1bf7d Merge pull request #539 from ROCm-Developer-Tools/fix-tf-link
Add workaround to hipcc for build failure in tensorflow due to missin…
2018-07-04 11:08:52 +05:30
Maneesh Gupta 3b157546f9 Merge pull request #517 from ROCm-Developer-Tools/fix-hipcc-link
Let hipcc handle static library for hip-clang only if it contains bun…
2018-07-04 10:51:44 +05:30
Yaxun (Sam) Liu 632baba353 Add workaround to hipcc for build failure in tensorflow due to missing symbol __cpu_model
https://github.com/tensorflow/tensorflow/issues/9593
2018-06-25 18:14:51 -04:00
Aaron Enye Shi dc850cc3e7 HIPCC - Use clang if clang++ is not compiled 2018-06-21 18:12:55 +00:00
Yaxun (Sam) Liu 454d45e5bb Let hipcc handle library with extension lo for hip-clang 2018-06-19 10:51:08 -04:00
Yaxun (Sam) Liu fd03c8fc18 Fix handling of static library in hipcc for hip-clang 2018-06-17 20:11:10 -04:00
Yaxun (Sam) Liu d442a57002 Let hipcc handle static library for hip-clang only if it contains bundles 2018-06-17 20:11:10 -04:00
Yaxun (Sam) Liu 8d3e4b4475 Let hipcc handle library in linker response file for hip-clang 2018-06-08 11:14:26 -04:00
Maneesh Gupta 8ecb3eeb55 Merge pull request #466 from ROCm-Developer-Tools/feature_use_Float16
Feature use _Float16 and match CUDA __half behaviour.
2018-06-01 13:50:12 +05:30
Yaxun (Sam) Liu 3609e6f3b2 Drop --amdgpu-target= options for hip-clang
They are replaced by --cuda-gpu-arch= options elsewhere
2018-05-30 06:47:44 -04:00
Yaxun (Sam) Liu 9c43970a27 Add more function declarations to hip-clang 2018-05-25 16:18:46 -04:00
Alex Voicu f2a86f3e1c Remove vestigial inline LLVMIR. 2018-05-24 12:46:14 +01:00
Maneesh Gupta 43ac6a61e7 Merge pull request #454 from ROCm-Developer-Tools/hip-clang-hipcc
Let hipcc suport hip-clang
2018-05-21 09:51:42 +05:30
Yaxun (Sam) Liu 55f01cbf36 Let hipcc suport hip-clang 2018-05-17 14:40:15 -04:00
Siu Chi Chan 368affcea4 initial gfx906 support 2018-05-10 19:28:00 +00:00
Maneesh Gupta efac735621 Cleanup bin directory
Remove hccgenco.sh and hsa-api-filter-cxl.txt

Change-Id: I5942aec911a2fe6c871fe0eb5f44a0b243552b79
2018-04-04 14:31:40 +05:30
Siu Chi Chan 0501f6c582 On RHEL and CentOS, don't use libc++ since we are using the devtoolset-7 which has a newer g++ 2018-03-15 16:46:01 -04:00
Maneesh Gupta 72ec13ab86 Merge pull request #287 from ROCm-Developer-Tools/fix_llc_path
Change hipconfig to adopt new location of llc
2018-03-06 11:59:44 +05:30
Alex Voicu 696f0595a3 Change directory name to match HIP lowercase style. 2018-02-22 13:15:10 +00:00
Alex Voicu f9e7195451 Merge branch 'master' of https://github.com/ROCm-Developer-Tools/HIP into feature_purge_genco 2018-01-24 12:33:45 +00:00
Evgeny Mankov 87895b7d01 [HIPIFY][perl] Typo fix 2018-01-18 09:14:16 +03:00
Alex Voicu 4d0d4dc701 Merge branch 'master' of https://github.com/ROCm-Developer-Tools/HIP into feature_purge_genco 2017-12-14 13:50:49 +00:00
Phaneendr-kumar Lanka 7a3e0bc5d0 [nvccWarnings] Fix gpu deprecated targets warnings 2017-12-12 16:00:14 +05:30
Alex Voicu 4e0739c68a This introduces LipoProteinLipase (lpl), a simple tool for creating fat binaries. It represents a direct replacement of the creaky hccgenco.sh script, which had various issues. The format it uses is that of a code object bundle, generated by the Clang Offload Bundler. The output is always suffixed with the ".adipose" extension. It is shared with HCC. The hipcc script and associated tests are modified to use lpl. Help can be obtained by invoking lpl --help. A more computer-sciency / corporate friendly name is likely to be beneficial, which is a reason for choosing easily searchable/replaceable names such as lpl or adipose. 2017-12-08 04:22:57 +00:00
Wen-Heng (Jack) Chung 8946001f85 Change hipconfig to adopt new location of llc
llc is now placed under $HCC_HOME/bin, instead of $HCC_HOME/compiler/bin.
2017-12-06 19:30:10 +00:00
Ben Sander a431bf6ba2 Remove coordinate conversions (threadIdx.x -> hipThreadIdx_x)
Since these are now supported directly.
2017-12-02 07:44:54 -06:00
Maneesh Gupta 2d5e0274ef Bumped HIP version to 1.5 and updated summary in HIP release notes
Change-Id: I4b612da88236674fac2c585ab6891bdd495511b0
2017-11-29 13:56:17 +05:30
Rahul Garg 20947f80fb Fixed half2 issue on gfx900 2017-11-15 18:52:59 +05:30
Maneesh Gupta 473208368e Remove libsupc++ dependency from HIP
Change-Id: Ib843f99dfb3057b7c28029250c2449732ea7f17f
2017-11-02 12:29:09 +05:30
Siu Chi Chan cb454d5366 use libc++ on centos 7 and rhel 7.4 2017-10-31 18:14:07 +00:00
Maneesh Gupta 253c1bf890 Bump HIP version to 1.4 and update release notes
Change-Id: I3570dedb32fe5fec6f60565c587a3282a4c6c709
2017-10-06 10:26:11 +05:30
wsttiger bda788ca97 Remove the conversion of warpSize to be consistent with hipDeviceProp 2017-09-15 15:05:05 -05:00
Maneesh Gupta 01163d327b Merge pull request #155 from wsttiger/master
Modify hip headers such that they are more compatible with strict C compiler
2017-09-15 10:49:55 +05:30
Aaron Enye Shi de8c59ed00 hcc-config already sets -lc++ -lc++abi flags
Since hcc-config will set the flags for CentOS7 and RHEL 7.X, we don't need to set flags here.
2017-09-14 10:55:38 -04:00
Aaron Enye Shi bfff9bf005 Added support for RHEL 7.4 2017-09-13 14:21:43 -04:00
Aaron Enye Shi 4f7972b0ee Update hipcc 2017-09-12 17:50:44 -04:00
Aaron Enye Shi f7a2c019c6 Must explicitly add libc++ and libc++abi 2017-09-12 16:59:43 -04:00
Aaron Enye Shi 728f1f4c95 [CentOS7] Fix HIP sample hipInfo and other tests
Cent OS 7 has conflicts between its libc++ and libstdc++. Seems that we cannot use libstdc++ on Cent OS. This is related to SWDEV-131972 [ROCm CQE][Cent OS 7][G] Building any HIP sample giving an fatal error: 'bits/c++config.h'.
2017-09-08 11:46:10 -04:00
Maneesh Gupta 809a548bd4 Bump HIP version to 1.3 and update release notes
Change-Id: I53f89efca03472cb906bd56200a2bf2996453915
2017-08-16 11:18:51 +05:30
wsttiger 7933b3292d Add missing CUDA tags 2017-08-15 13:22:20 -05:00
Maneesh Gupta 8f6c150134 [hipcc] Fix amdgpu target selection logic
- Refactoring introduced a bug when user does not specify any target via
--amdgpu-target, but has an invalid target specified in
HCC_AMDGPU_TARGET. In this case the selection logic was defaulting to
gf803.
- Removed defaulting to any specific target if rocm_agent_enumerator
  fails. hipcc will report this and die if linking was required.

Change-Id: I76131867049fef92331807dd19a926406dcc1d02
2017-07-31 10:11:19 +05:30
Maneesh Gupta 2a2c7575eb [hipcc] Cleanup amdgpu target logic
Existing logic has a bug. If user specifies targetA via commandline
options, while enumerator returns targetB, hipcc will create a fatbin
containing targets targetA and targetB. enumerator should only be used
when no target is specified by user (commandline or env var).

Change-Id: I6da857f86860c0e671b5988cd858644a08f723b9
2017-07-28 16:18:15 +05:30
Ben Sander 702651fc85 Merge pull request #100 from weixingzhang/texture
HIP Texture Support
2017-07-26 19:28:15 -05:00
Aditya Atluri b10962a539 fixed device selection during compilation to use rocm_agent_enumerator
1. Changed hipcc to use rocm_agent_enumerator
2. Changed square sample test to use device variable
2017-07-21 15:50:12 -05:00
Weixing Zhang ccbd3b13fc HIP Texture Support 2017-07-17 15:16:12 -04:00
Sun, Peng c6dee2a266 fix typo in hipcc
Change-Id: I5f5fb3d8f1e5e8104dca45dc213c13ae9d98a583
2017-07-01 09:43:31 -05:00
Maneesh Gupta 4fa2090b78 Bump HIP base version to 1.2
Change-Id: I8ecc164afed4383f78579ed86a5c8c11a73b0780
2017-06-14 09:45:46 +05:30
Maneesh Gupta d2b90ad93c Merge branch hipify-updates into amd-develop
Change-Id: I13d8750027a2a8787e4eb2e1ed525cf69d14b805
2017-06-12 10:10:19 +05:30
Maneesh Gupta c5366a55f1 Initial implementation of hipify-cmakefile
Change-Id: Id365da9f887b5c3409639f000b430d093fd4f6b3
2017-06-12 09:57:17 +05:30
emankov c5f9758f4b [HIPIFY] annotation 2017-06-02 16:33:48 +03:00