foreman
2669fce89a
P4 to Git Change 1233743 by bcrane@MSDNBCRANE99-OCL on 2016/02/02 20:45:13
...
SWDEV-86253 - [QR] 6 to 7% performance drop is observed in BasemarkCL test
CL1226184 adds serialization to LHIO, because pxproxy accesses global state, without locks, leading to crash, etc.
To fix OpenCL perf regressions, allow unserialized access to pxproxy. It may be that OpenCL design leads away from racy behavior, so it may be safe. This was checked in to gather information. What software gets perf drops? Does any software get fixed?
The dangerous functions are CreateDevice, CreateContext, because the handles from the OS are saved in global caches, and if those global caches are modified in one thread, while other threads are looking-up from the caches - that is a problem.
TC: http://ocltc:8111/viewModification.html?modId=66278&personal=true
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#158 edit
[ROCm/clr commit: bd54838ca7 ]
2016-02-02 20:52:50 -05:00
foreman
bd54838ca7
P4 to Git Change 1233743 by bcrane@MSDNBCRANE99-OCL on 2016/02/02 20:45:13
...
SWDEV-86253 - [QR] 6 to 7% performance drop is observed in BasemarkCL test
CL1226184 adds serialization to LHIO, because pxproxy accesses global state, without locks, leading to crash, etc.
To fix OpenCL perf regressions, allow unserialized access to pxproxy. It may be that OpenCL design leads away from racy behavior, so it may be safe. This was checked in to gather information. What software gets perf drops? Does any software get fixed?
The dangerous functions are CreateDevice, CreateContext, because the handles from the OS are saved in global caches, and if those global caches are modified in one thread, while other threads are looking-up from the caches - that is a problem.
TC: http://ocltc:8111/viewModification.html?modId=66278&personal=true
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#158 edit
2016-02-02 20:52:50 -05:00
Peng Sun
4afe96bf21
Finish all TODO for error code
...
[ROCm/clr commit: b20e02ae58 ]
2016-02-02 17:39:46 -06:00
Peng Sun
b20e02ae58
Finish all TODO for error code
2016-02-02 17:39:46 -06:00
Peng Sun
8b74333204
Finish all TODO for error code
2016-02-02 17:39:46 -06:00
scchan
a9745be3f4
add inline attribute to shfl functions
...
[ROCm/clr commit: 63a6bce3d9 ]
2016-02-02 12:53:17 -06:00
scchan
63a6bce3d9
add inline attribute to shfl functions
2016-02-02 12:53:17 -06:00
scchan
265c42500f
add inline attribute to shfl functions
2016-02-02 12:53:17 -06:00
Ben Sander
0d62f7767a
Merge branch 'master' of https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP
...
[ROCm/clr commit: 714bfcbff6 ]
2016-02-02 10:05:44 -06:00
Ben Sander
714bfcbff6
Merge branch 'master' of https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP
2016-02-02 10:05:44 -06:00
Ben Sander
6175d665fc
Merge branch 'master' of https://github.com/GPUOpen-ProfessionalCompute-Tools/HIP
2016-02-02 10:05:44 -06:00
Ben Sander
71bfa20508
Remove warning on ballot/any/all and pop/clz.
...
Since these are supported in HIP no reason to emit warnings.
[ROCm/clr commit: 182296ce59 ]
2016-02-02 10:02:48 -06:00
Ben Sander
182296ce59
Remove warning on ballot/any/all and pop/clz.
...
Since these are supported in HIP no reason to emit warnings.
2016-02-02 10:02:48 -06:00
Ben Sander
3898f6af3c
Remove warning on ballot/any/all and pop/clz.
...
Since these are supported in HIP no reason to emit warnings.
2016-02-02 10:02:48 -06:00
streamhsa
c832718f42
Adjusted the value of __any as per CUDA -sandeep
...
[ROCm/clr commit: b14c890851 ]
2016-02-02 15:25:42 +05:30
streamhsa
b14c890851
Adjusted the value of __any as per CUDA -sandeep
2016-02-02 15:25:42 +05:30
streamhsa
974d491902
Adjusted the value of __any as per CUDA -sandeep
2016-02-02 15:25:42 +05:30
streamhsa
fa391f72c9
ADDED Support for __ffs() and __ffsll() having signed input -sandeep
...
[ROCm/clr commit: a7c0be6e4b ]
2016-02-02 15:05:46 +05:30
streamhsa
a7c0be6e4b
ADDED Support for __ffs() and __ffsll() having signed input -sandeep
2016-02-02 15:05:46 +05:30
streamhsa
23904df99b
ADDED Support for __ffs() and __ffsll() having signed input -sandeep
2016-02-02 15:05:46 +05:30
streamhsa
75643c8c00
Added test for ballot and removing HIP_FUNCTION from hipSampleAtomicsTest.cpp -sandeep
...
[ROCm/clr commit: e5a491f3c8 ]
2016-02-02 14:50:55 +05:30
streamhsa
e5a491f3c8
Added test for ballot and removing HIP_FUNCTION from hipSampleAtomicsTest.cpp -sandeep
2016-02-02 14:50:55 +05:30
streamhsa
2b02fe082f
Added test for ballot and removing HIP_FUNCTION from hipSampleAtomicsTest.cpp -sandeep
2016-02-02 14:50:55 +05:30
Jack Chung
627c9622c8
Merge branch 'privatestaging' of github.com:AMDComputeLibraries/HIP-privatestaging into privatestaging
...
[ROCm/clr commit: e62a9e1bb7 ]
2016-02-02 16:28:02 +08:00
Jack Chung
e62a9e1bb7
Merge branch 'privatestaging' of github.com:AMDComputeLibraries/HIP-privatestaging into privatestaging
2016-02-02 16:28:02 +08:00
Jack Chung
8551855f89
Merge branch 'privatestaging' of github.com:AMDComputeLibraries/HIP-privatestaging into privatestaging
2016-02-02 16:28:02 +08:00
Jack Chung
58e2e13f04
Suppress linker warnings in case HCC distribution contains OpenCL/SPIR symbols
...
[ROCm/clr commit: 8df395a006 ]
2016-02-02 16:27:42 +08:00
Jack Chung
8df395a006
Suppress linker warnings in case HCC distribution contains OpenCL/SPIR symbols
2016-02-02 16:27:42 +08:00
Jack Chung
1e7642c569
Suppress linker warnings in case HCC distribution contains OpenCL/SPIR symbols
2016-02-02 16:27:42 +08:00
foreman
d2c511fc38
P4 to Git Change 1233406 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/02/02 03:01:34
...
SWDEV-2 - Change OpenCL version number from 2023 to 2024.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1770 edit
[ROCm/clr commit: 99556aca1d ]
2016-02-02 03:13:00 -05:00
foreman
99556aca1d
P4 to Git Change 1233406 by johtaylo@johtaylo-JTBUILDER03-increment on 2016/02/02 03:01:34
...
SWDEV-2 - Change OpenCL version number from 2023 to 2024.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1770 edit
2016-02-02 03:13:00 -05:00
Andres Rodriguez
3797b56ec9
Bump version for bugfix release 1.8.1
...
Change-Id: I06701905592594221d26c075a8fe370b4cc92aff
2016-02-02 01:29:51 -05:00
Andres Rodriguez
cd849bc3e9
Bump version for bugfix release 1.8.1
...
Change-Id: I06701905592594221d26c075a8fe370b4cc92aff
[ROCm/ROCR-Runtime commit: 3797b56ec9 ]
2016-02-02 01:29:51 -05:00
scchan
4ca3c297be
adding shfl, shfl_up, shfl_down, shfl_xor intrinsics
...
[ROCm/clr commit: ed0a4fc1b7 ]
2016-02-01 23:55:31 -06:00
scchan
ed0a4fc1b7
adding shfl, shfl_up, shfl_down, shfl_xor intrinsics
2016-02-01 23:55:31 -06:00
scchan
04f3e3e598
adding shfl, shfl_up, shfl_down, shfl_xor intrinsics
2016-02-01 23:55:31 -06:00
Ben Sander
d1c1274f69
Merge pull request #10 from SethosII/patch-1
...
Update hip_faq.md based on Sethosll review. Closes #10
[ROCm/clr commit: 0c03868e90 ]
2016-02-01 22:01:17 -06:00
Ben Sander
0c03868e90
Merge pull request #10 from SethosII/patch-1
...
Update hip_faq.md based on Sethosll review. Closes #10
2016-02-01 22:01:17 -06:00
Ben Sander
b2b43d55cb
Merge pull request #10 from SethosII/patch-1
...
Update hip_faq.md based on Sethosll review. Closes #10
2016-02-01 22:01:17 -06:00
Maneesh Gupta
b5634c1e26
Add double and integer intrinsics to test
...
[ROCm/clr commit: 405ee35a04 ]
2016-02-01 16:00:45 +05:30
Maneesh Gupta
405ee35a04
Add double and integer intrinsics to test
2016-02-01 16:00:45 +05:30
Maneesh Gupta
861cba6f75
Add double and integer intrinsics to test
2016-02-01 16:00:45 +05:30
foreman
595aa5e804
P4 to Git Change 1232961 by emankov@em-hsa-amd on 2016/02/01 04:39:58
...
SWDEV-79309 - OpenCL Compiler Library improvements.
Minor Legacy Code Cleanup for OpenCL kernel caching feature implementation.
Testing: precheck-in
http://ocltc.amd.com:8111/viewModification.html?modId=66106&personal=true&buildTypeId=&tab=vcsModificationBuilds&show_all_builds=true
Reviewer: Stanislav Mekhanoshin
Affected files ...
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/v0_8/if_acl.cpp#90 edit
[ROCm/clr commit: 6dc767146f ]
2016-02-01 04:47:26 -05:00
foreman
6dc767146f
P4 to Git Change 1232961 by emankov@em-hsa-amd on 2016/02/01 04:39:58
...
SWDEV-79309 - OpenCL Compiler Library improvements.
Minor Legacy Code Cleanup for OpenCL kernel caching feature implementation.
Testing: precheck-in
http://ocltc.amd.com:8111/viewModification.html?modId=66106&personal=true&buildTypeId=&tab=vcsModificationBuilds&show_all_builds=true
Reviewer: Stanislav Mekhanoshin
Affected files ...
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/v0_8/if_acl.cpp#90 edit
2016-02-01 04:47:26 -05:00
Jack Chung
3e57a5abae
Disable sincosf which has trouble on hcc now.
...
[ROCm/clr commit: 518ef58652 ]
2016-02-01 17:42:37 +08:00
Jack Chung
518ef58652
Disable sincosf which has trouble on hcc now.
2016-02-01 17:42:37 +08:00
Jack Chung
93d1801a30
Disable sincosf which has trouble on hcc now.
2016-02-01 17:42:37 +08:00
Maneesh Gupta
8213a76631
Split math function tests into several smaller tests
...
[ROCm/clr commit: 97fb876c6a ]
2016-02-01 14:36:50 +05:30
Maneesh Gupta
97fb876c6a
Split math function tests into several smaller tests
2016-02-01 14:36:50 +05:30
Maneesh Gupta
4d9e8b8673
Split math function tests into several smaller tests
2016-02-01 14:36:50 +05:30