David Addison
e73e5a239b
Added support for CUDA graph capture/replay (-G)
...
[ROCm/rccl-tests commit: e55ad3796d ]
2021-06-28 14:19:45 -07:00
David Addison
e55ad3796d
Added support for CUDA graph capture/replay (-G)
2021-06-28 14:19:45 -07:00
jujiang
748680e8a6
SWDEV-286322 - add .gitattribute
...
Change-Id: I0452a9c85bbb27adffebf4b1329d56a0f1d23ecd
[ROCm/clr commit: 2aa5689d7e ]
2021-06-28 14:16:13 -04:00
jujiang
2aa5689d7e
SWDEV-286322 - add .gitattribute
...
Change-Id: I0452a9c85bbb27adffebf4b1329d56a0f1d23ecd
2021-06-28 14:16:13 -04:00
David Addison
20b63cf465
Fixed formatting for bfloat16 support
...
[ROCm/rccl-tests commit: 526eacadf7 ]
2021-06-28 10:12:34 -07:00
David Addison
526eacadf7
Fixed formatting for bfloat16 support
2021-06-28 10:12:34 -07:00
David Addison
a41268e26e
Add support for ncclAvg operation
...
[ROCm/rccl-tests commit: cde7e769c1 ]
2021-06-28 09:41:58 -07:00
David Addison
cde7e769c1
Add support for ncclAvg operation
2021-06-28 09:41:58 -07:00
Anusha GodavarthySurya
7638199a50
SWDEV-290637 - Migrate cmake build from hip to hipamd
...
Change-Id: I31baef5a76ea0721927598615794437999ec195e
[ROCm/hip commit: 4903b0d834 ]
2021-06-28 06:23:41 -04:00
Anusha GodavarthySurya
4903b0d834
SWDEV-290637 - Migrate cmake build from hip to hipamd
...
Change-Id: I31baef5a76ea0721927598615794437999ec195e
2021-06-28 06:23:41 -04:00
Anusha GodavarthySurya
f3cc7d9db0
SWDEV-290637 - Migrate cmake build from hip to hipamd
...
Change-Id: I605fc633ba18df4fca95ceee83670600dd3f0b71
[ROCm/clr commit: e420815e75 ]
2021-06-28 06:23:33 -04:00
Anusha GodavarthySurya
e420815e75
SWDEV-290637 - Migrate cmake build from hip to hipamd
...
Change-Id: I605fc633ba18df4fca95ceee83670600dd3f0b71
2021-06-28 06:23:33 -04:00
changzhu
1a9604ad57
kfdtest: skip KFDSVMRangeTest.MigrateAccessInPlaceTest for gfx902 and gfx90c
...
Change-Id: I671440c212a07fdfdb1c4245b4551c6344eaedc6
Signed-off-by: changzhu <Changfeng.Zhu@amd.com >
2021-06-28 13:37:52 +08:00
changzhu
1d15f33b5c
kfdtest: skip KFDSVMRangeTest.MigrateAccessInPlaceTest for gfx902 and gfx90c
...
Change-Id: I671440c212a07fdfdb1c4245b4551c6344eaedc6
Signed-off-by: changzhu <Changfeng.Zhu@amd.com >
[ROCm/ROCR-Runtime commit: 1a9604ad57 ]
2021-06-28 13:37:52 +08:00
Sarbojit Sarkar
9e9b89c4c3
SWDEV-292449 - Bump version to 4.3
...
Change-Id: Ia425cdf0d9c030894a180fda5a786240e19dcbcc
[ROCm/hip commit: 341de7b12e ]
2021-06-26 05:53:45 -04:00
Sarbojit Sarkar
341de7b12e
SWDEV-292449 - Bump version to 4.3
...
Change-Id: Ia425cdf0d9c030894a180fda5a786240e19dcbcc
2021-06-26 05:53:45 -04:00
Jatin
4bcfffbfcf
SWDEV-245414 - fix nvcc CI breakage
...
Change-Id: I34ff533f34cfc2f60abefc0d8f6c6d8266d30f61
[ROCm/hipother commit: c0a0999d1a ]
2021-06-26 13:31:10 +05:30
Jatin
c0a0999d1a
SWDEV-245414 - fix nvcc CI breakage
...
Change-Id: I34ff533f34cfc2f60abefc0d8f6c6d8266d30f61
2021-06-26 13:31:10 +05:30
Jatin
52dd6c20c9
SWDEV-245414 - fix nvcc CI breakage
...
Change-Id: I34ff533f34cfc2f60abefc0d8f6c6d8266d30f61
[ROCm/clr commit: 19a30de0a6 ]
2021-06-26 13:31:10 +05:30
Jatin
19a30de0a6
SWDEV-245414 - fix nvcc CI breakage
...
Change-Id: I34ff533f34cfc2f60abefc0d8f6c6d8266d30f61
2021-06-26 13:31:10 +05:30
rachida
b54bfac5bf
SWDEV-291267 Script to merge mpi rocprof traces
...
Change-Id: I3d86276666efb067ccb62730e74a67e0cc7c4d42
2021-06-25 15:12:35 -04:00
rachida
61efc819c3
SWDEV-291267 Script to merge mpi rocprof traces
...
Change-Id: I3d86276666efb067ccb62730e74a67e0cc7c4d42
[ROCm/rocprofiler commit: b54bfac5bf ]
2021-06-25 15:12:35 -04:00
Philip Yang
c4d5ee28f0
libhsakmt: fix multiple threads register userptr race
...
Aperture locking is too fine-grained, it has race between find userptr
and allocate userptr object.
Change _fmm_allocate_device and fmm_allocate_memory_object to not take
the aperture lock, the callers take it, this implements an atomic find
userptr or allocate a new one.
Change-Id: I6773404e22c1f4382a211c5a9817df23c5534a2a
Signed-off-by: Philip Yang <Philip.Yang@amd.com >
2021-06-25 14:16:20 -04:00
Philip Yang
f94be2078c
libhsakmt: fix multiple threads register userptr race
...
Aperture locking is too fine-grained, it has race between find userptr
and allocate userptr object.
Change _fmm_allocate_device and fmm_allocate_memory_object to not take
the aperture lock, the callers take it, this implements an atomic find
userptr or allocate a new one.
Change-Id: I6773404e22c1f4382a211c5a9817df23c5534a2a
Signed-off-by: Philip Yang <Philip.Yang@amd.com >
[ROCm/ROCR-Runtime commit: c4d5ee28f0 ]
2021-06-25 14:16:20 -04:00
Bill(Shuzhou) Liu
a4dba3ba7e
AddressSanitizer report stack-use-after-scope
...
Fix the stack-use-after-scope error reported by the AddressSanitizer.
Bug: SWDEV-291913
Change-Id: I0ffd71af8679b8bff6c363096fafe75dffcf329e
[ROCm/rocm_smi_lib commit: 8c60dbebaa ]
2021-06-25 13:33:38 -04:00
Bill(Shuzhou) Liu
8c60dbebaa
AddressSanitizer report stack-use-after-scope
...
Fix the stack-use-after-scope error reported by the AddressSanitizer.
Bug: SWDEV-291913
Change-Id: I0ffd71af8679b8bff6c363096fafe75dffcf329e
2021-06-25 13:33:38 -04:00
Bill(Shuzhou) Liu
d9e824060c
AddressSanitizer report stack-use-after-scope
...
Fix the stack-use-after-scope error reported by the AddressSanitizer.
Bug: SWDEV-291913
Change-Id: I0ffd71af8679b8bff6c363096fafe75dffcf329e
[ROCm/amdsmi commit: 8c60dbebaa ]
2021-06-25 13:33:38 -04:00
Sarbojit Sarkar
0ff075e785
SWDEV-284435 - StreamOps mapping with cuda drv API
...
Change-Id: I6ed9196cfe59d9004338206f6dd8d6fb367b3ffa
[ROCm/hipother commit: ef4d479169 ]
2021-06-25 12:20:40 -04:00
Sarbojit Sarkar
ef4d479169
SWDEV-284435 - StreamOps mapping with cuda drv API
...
Change-Id: I6ed9196cfe59d9004338206f6dd8d6fb367b3ffa
2021-06-25 12:20:40 -04:00
Sarbojit Sarkar
b99ebf8d00
SWDEV-284435 - StreamOps mapping with cuda drv API
...
Change-Id: I6ed9196cfe59d9004338206f6dd8d6fb367b3ffa
[ROCm/clr commit: 7e3c17c023 ]
2021-06-25 12:20:40 -04:00
Sarbojit Sarkar
7e3c17c023
SWDEV-284435 - StreamOps mapping with cuda drv API
...
Change-Id: I6ed9196cfe59d9004338206f6dd8d6fb367b3ffa
2021-06-25 12:20:40 -04:00
Sarbojit Sarkar
dfc088013e
SWDEV-284435 - Fixed latest CUDA Drv support doc
...
Change-Id: Iba172236c4816fdc5c5caad763be0a9fd51e6f58
[ROCm/hip commit: f59c5548ed ]
2021-06-25 12:20:11 -04:00
Sarbojit Sarkar
f59c5548ed
SWDEV-284435 - Fixed latest CUDA Drv support doc
...
Change-Id: Iba172236c4816fdc5c5caad763be0a9fd51e6f58
2021-06-25 12:20:11 -04:00
Sarbojit Sarkar
035cbfa79e
SWDEV-261977 - Fixed function symbols
...
Change-Id: I70f568d12e66ae02efa16c48786a56ff10628a6b
[ROCm/clr commit: b30543e890 ]
2021-06-25 01:32:57 -04:00
Sarbojit Sarkar
b30543e890
SWDEV-261977 - Fixed function symbols
...
Change-Id: I70f568d12e66ae02efa16c48786a56ff10628a6b
2021-06-25 01:32:57 -04:00
Anusha GodavarthySurya
e5b673f2e2
SWDEV-290637 - Update tests to use current directory CMAKE_CURRENT_LIST_DIR instead of HIP_SRC_PATH to build from hipamd repo
...
Change-Id: Ide04d9f7b1fc43b475d985a6dce0aa1bd2ec00da
[ROCm/hip-tests commit: c749be3299 ]
2021-06-25 01:01:16 -04:00
Anusha GodavarthySurya
c749be3299
SWDEV-290637 - Update tests to use current directory CMAKE_CURRENT_LIST_DIR instead of HIP_SRC_PATH to build from hipamd repo
...
Change-Id: Ide04d9f7b1fc43b475d985a6dce0aa1bd2ec00da
2021-06-25 01:01:16 -04:00
Anusha GodavarthySurya
87014ed96e
SWDEV-290637 - Update tests to use current directory CMAKE_CURRENT_LIST_DIR instead of HIP_SRC_PATH to build from hipamd repo
...
Change-Id: Ide04d9f7b1fc43b475d985a6dce0aa1bd2ec00da
[ROCm/hip commit: b49118dc2f ]
2021-06-25 01:01:16 -04:00
Anusha GodavarthySurya
b49118dc2f
SWDEV-290637 - Update tests to use current directory CMAKE_CURRENT_LIST_DIR instead of HIP_SRC_PATH to build from hipamd repo
...
Change-Id: Ide04d9f7b1fc43b475d985a6dce0aa1bd2ec00da
2021-06-25 01:01:16 -04:00
German Andryeyev
03fa65d857
SWDEV-292018 - Make sure start event is done
...
HSA signal callback order is undefined. Make sure start event is
also done before calculating the elapsed time.
Change-Id: Ic69bfe336b20cd62ef35194261a5d0d234bc65ce
[ROCm/clr commit: 910d8bdfde ]
2021-06-24 20:02:54 -04:00
German Andryeyev
910d8bdfde
SWDEV-292018 - Make sure start event is done
...
HSA signal callback order is undefined. Make sure start event is
also done before calculating the elapsed time.
Change-Id: Ic69bfe336b20cd62ef35194261a5d0d234bc65ce
2021-06-24 20:02:54 -04:00
German Andryeyev
e7580989cc
SWDEV-290160 - Add blocking command for callbacks in IPC and graph
...
Change-Id: I1474e19db6ebf2559e1b2c8dd7d06139bd699835
[ROCm/clr commit: 52f5b0fa16 ]
2021-06-24 17:01:55 -04:00
German Andryeyev
52f5b0fa16
SWDEV-290160 - Add blocking command for callbacks in IPC and graph
...
Change-Id: I1474e19db6ebf2559e1b2c8dd7d06139bd699835
2021-06-24 17:01:55 -04:00
Aaron En Ye Shi
99eea2506a
SWDEV-286884 - Add CMake hiprtc-builtins library generation part 2
...
Add libhiprtc-builtins library to packaging/hip-rocclr.
Depends on part 1 patch to compute/ec/hipamd.
Change-Id: I366d7f1868c868f5d6b02363c7e78ff399e06b19
[ROCm/hip commit: fdbb8f10bd ]
2021-06-24 18:04:22 +00:00
Aaron En Ye Shi
fdbb8f10bd
SWDEV-286884 - Add CMake hiprtc-builtins library generation part 2
...
Add libhiprtc-builtins library to packaging/hip-rocclr.
Depends on part 1 patch to compute/ec/hipamd.
Change-Id: I366d7f1868c868f5d6b02363c7e78ff399e06b19
2021-06-24 18:04:22 +00:00
Aaron En Ye Shi
791087a6b8
SWDEV-286884 - Add CMake hiprtc-builtins library generation part 1
...
Add the CMake support for generating the hiprtc-builtins
library. Replaces the previous execution of shell script.
For backwards compatibility, link hiprtc pre-processed
object to amdhip64, and support versioning in name.
Add libcmt.lib to Windows link step, due to manually
generated .obj from llvm-mc.
Change-Id: I267be3cf4b241840b35f7f27a0b8659530108b0e
[ROCm/clr commit: c42308e2e5 ]
2021-06-24 14:34:47 +00:00
Aaron En Ye Shi
c42308e2e5
SWDEV-286884 - Add CMake hiprtc-builtins library generation part 1
...
Add the CMake support for generating the hiprtc-builtins
library. Replaces the previous execution of shell script.
For backwards compatibility, link hiprtc pre-processed
object to amdhip64, and support versioning in name.
Add libcmt.lib to Windows link step, due to manually
generated .obj from llvm-mc.
Change-Id: I267be3cf4b241840b35f7f27a0b8659530108b0e
2021-06-24 14:34:47 +00:00
Jatin Chaudhary
e3b0afc852
SWDEV-289409 - Add libstdc++fs to unit test exe
...
Change-Id: Id668cd450f032a57d880270188cecf6950b69f40
[ROCm/hip-tests commit: 55c711460b ]
2021-06-24 04:18:53 -07:00
Jatin Chaudhary
55c711460b
SWDEV-289409 - Add libstdc++fs to unit test exe
...
Change-Id: Id668cd450f032a57d880270188cecf6950b69f40
2021-06-24 04:18:53 -07:00
Jatin Chaudhary
7956f028ff
SWDEV-289409 - Add libstdc++fs to unit test exe
...
Change-Id: Id668cd450f032a57d880270188cecf6950b69f40
[ROCm/hip commit: 23f6977e49 ]
2021-06-24 04:18:53 -07:00