65a90c55e7
Checkin to fix bugs in math functions.
...
This change fixes the following bugs that were discovered while debuggnig TF unit test failures (cwise_ops_test)
1. __hisinf and __hisnan routines
Both had incorrect implementations.
2. abs
A "long long" (64bit int) version was missing, resulting in the 32bit version being used for 64bit ints (which resulted in incorrect results, when the value passed in was outside the 32bit int range)
3. lgamma
We seemed to have a custom version for the 'double' datatype (which was giving incorrect results). Replaced it with a call to the 'double' version of the underlying 'hc::precision_math::lgamma'
2018-04-24 18:10:07 +00:00
1149f88413
Remove unsupported targets gfx801 and gfx802
...
These architectures are no longer supported. Should not compile for them.
2018-04-23 15:25:35 -04:00
14a95fc231
Merge pull request #413 from emankov/master
...
[HIPIFY] undo apply .clangformat to HIPIFY source files
2018-04-23 20:29:43 +03:00
ad28f72ca9
[HIPIFY] undo apply .clangformat to HIPIFY source files
2018-04-23 20:22:15 +03:00
a68b9b6bf6
Added deprecation note in cuCtx section
2018-04-20 22:43:58 +05:30
bc61260168
Moved deprecated macro to the end
2018-04-20 21:57:23 +05:30
74faa61d52
SWDEV-141024
2018-04-20 17:40:00 +05:30
89674c54a7
Mark hipCtx APIs as deprecated
2018-04-20 17:34:48 +05:30
4770abf00e
Merge pull request #404 from gargrahul/hipMemset2DAsync_support
...
Added hipMemset2DAsync support
2018-04-18 14:19:41 +05:30
a898ccfb3f
Resolve tex1D float to int conversion issue
2018-04-18 14:15:18 +05:30
fcc0866681
Added hipMemset2DAsync support
2018-04-17 18:27:27 +05:30
72ff4c5cc4
Merge pull request #400 from gargrahul/hipModule_cleanup
...
hip_module code cleanup
2018-04-17 09:00:15 +05:30
a004c8dfa2
Clean up triple_to_hsa_isa naming ( #403 )
2018-04-17 06:47:55 +05:30
c23898f49b
hip_module code cleanup
...
-Fixed missing ihipLogStatus in hipModuleLoad()
-Fixed some ihipXXX functions
2018-04-16 15:35:04 +05:30
21382e361c
Fix bug when using older rocr naming ( #399 )
2018-04-14 15:47:46 +05:30
199cff2365
Fix new gfx target naming convention ( #397 )
...
This was introduced in recent xnack changes naming changes which affected our offload bundler. This patch will fix issues in HIP samples, ROCR tests, and others.
2018-04-13 16:24:39 +05:30
2ab05a2069
Merge pull request #395 from lmoriche/master
...
Add missing __device__ function attributes.
2018-04-12 14:51:32 +05:30
de8eae68e9
Merge pull request #393 from mangupta/cmake_update_deps
...
[cmake] Update target build dependencies
2018-04-12 14:36:49 +05:30
234488eba2
Merge pull request #394 from gargrahul/hipMemset_refactoring
...
hipMemset refactoring
2018-04-12 14:01:49 +05:30
2ee80397f2
Correct missed ihipMemsetCopyDataType change
2018-04-12 10:27:19 +05:30
2658963f5b
Changed ihipMemsetCopyDataType to ihipMemsetDataType
2018-04-12 09:29:22 +05:30
8dcfbf5bee
Add missing __device__ function attributes.
2018-04-11 09:29:37 -07:00
88073a17c1
Fix hipMemset stream resolution
2018-04-11 19:01:53 +05:30
40846f6f8e
hipMemset refactoring
2018-04-11 15:58:48 +05:30
ebc52e5bc0
[cmake] Update target build dependencies
...
Change-Id: Ib1c34135c0d04f88daa695e7a90272225d7e64b0
2018-04-11 15:30:13 +05:30
a39a96dcc9
Merge pull request #391 from lmoriche/master
...
Create one hipModule_t per device
2018-04-11 13:06:58 +05:30
f4cde23be3
Create one hipModule_t per device
2018-04-10 08:57:34 -07:00
2d1d79cb22
Merge pull request #390 from mangupta/hipmemcpy_fix
...
hipMemcpyAsync returns success when trying to copy 0 bytes
2018-04-09 14:02:36 +05:30
d945c85fa7
Merge pull request #388 from lmoriche/master
...
Add support for clang offload bundles and <<<>>> kernel launch.
2018-04-09 12:40:28 +05:30
445de66560
hipMemcpyAsync returns success when trying to copy 0 bytes
...
Change-Id: I4c0ee7ccc7563e2df657b50356cdd7fec9a1ef15
2018-04-09 12:39:44 +05:30
04f116a9e8
[ci] Disable nvcc CI builds
2018-04-09 12:23:56 +05:30
e5b14cd918
Merge pull request #389 from scchan/adjust_texture_hcc_workweek
...
adjust the hcc work week for the texture fix
2018-04-09 10:54:16 +05:30
5df923cdfa
adjust the hcc work week for the texture fix
2018-04-06 15:46:55 -04:00
7f03ff1012
Don't include __clang_cuda_builtin_vars.h
2018-04-05 16:27:05 -07:00
323caee8d2
Add support for clang offload bundles and <<<>>> kernel launch.
2018-04-04 13:32:33 -07:00
60faf90b62
Merge pull request #387 from mangupta/cleanup_bin
...
Cleanup bin directory
2018-04-04 15:06:57 +05:30
efac735621
Cleanup bin directory
...
Remove hccgenco.sh and hsa-api-filter-cxl.txt
Change-Id: I5942aec911a2fe6c871fe0eb5f44a0b243552b79
2018-04-04 14:31:40 +05:30
13b65efae0
Merge pull request #383 from gargrahul/remove_hcc_check_texref_sample
...
Remove textureObj kernel argument for HIP/HCC path
2018-04-04 10:35:58 +05:30
2021414e98
Merge pull request #384 from yxsamliu/patch-1
...
Update HIP language spec to support both <<< >>> kernel launching mec…
2018-04-04 10:35:34 +05:30
4826f84d5c
Merge pull request #385 from gargrahul/fix_texture_trsf_flag
...
Fix texture driver api TRFS flags
2018-04-04 10:34:09 +05:30
1bfdfecb59
Merge pull request #386 from ROCm-Developer-Tools/mangupta-patch-4
...
[ci] Re-enable testing against hcc_1.7
2018-04-04 10:01:41 +05:30
11eaac9cf4
[ci] Re-enable testing against hcc_1.7
2018-04-03 14:04:33 +05:30
414cb24351
Fix texture driver api TRFS flags
2018-04-02 21:47:20 +05:30
0e1ae245d2
Update HIP language spec to support both <<< >>> kernel launching mechanism and hipLaunchKernel
2018-04-02 11:58:32 -04:00
35bc23ecc3
Remove textureObj kernel argument for HIP/HCC path
2018-03-28 15:36:57 +05:30
c1505becef
Merge pull request #382 from scchan/fix_texture_addr_space
...
Change constant address space to 4
2018-03-27 07:48:50 +05:30
53d9cce9f9
Change constant address space to 4
...
when compiling with a newer hcc. This is
due to a recent change to address space
mapping in the amdgpu compiler backend.
2018-03-26 18:13:59 -04:00
a95ff17549
Merge pull request #379 from lmoriche/master
...
Add HIP_KERNEL_NAME/HIP_SYMBOL definitions for HIP-clang, and rename …
2018-03-26 15:38:02 +05:30
e8cc7c2e40
[ci] Get CI builds to work again ( #377 )
...
* [ci] Enable tests against HCC from ROCm1.7 as well
* [ci] rocm 1.7 requires nodes with dkms label
* [ci] Temporarily drop hcc_1_6 and hcc_1_7 builds
2018-03-26 15:21:45 +05:30
8f4c368dce
Merge remote-tracking branch 'upstream/master'
2018-03-21 11:17:52 -07:00
Deven Desai
Aaron Enye Shi
Evgeny Mankov
Evgeny Mankov
Rahul Garg
Lakhan Singh
Maneesh Gupta
Laurent Morichetti
Siu Chi Chan
Yaxun (Sam) Liu