AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
54,851 Commits 3 Branches 0 Tags
a2cedb7fe5ccb4c0bbd911a661672e7d581e95e4
Commit Graph

54851 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Alex Voicu af0ae5421c This implements the trivial change needed to move back from the hip{Something}_{x, y, z} macros to the natural CUDA syntax of Something.{x, y, z}. This is contained in lines 384-404 in hip_runtime.h. All of the other changes have to do with changing unit tests to use this syntax. The macros are retained for backwards compatibility. 
[ROCm/clr commit: 0cc921f103]
 2017-11-19 01:54:12 +00:00
Alex Voicu 0cc921f103 This implements the trivial change needed to move back from the hip{Something}_{x, y, z} macros to the natural CUDA syntax of Something.{x, y, z}. This is contained in lines 384-404 in hip_runtime.h. All of the other changes have to do with changing unit tests to use this syntax. The macros are retained for backwards compatibility. 2017-11-19 01:54:12 +00:00
Alex Voicu cffd0e14eb This implements the trivial change needed to move back from the hip{Something}_{x, y, z} macros to the natural CUDA syntax of Something.{x, y, z}. This is contained in lines 384-404 in hip_runtime.h. All of the other changes have to do with changing unit tests to use this syntax. The macros are retained for backwards compatibility. 2017-11-19 01:54:12 +00:00
Alex Voicu a31109a290 This actually (tries) to do the right thing all the way, by using memcpy for bitcasting, and not rely on undefined behaviour of a different flavour as a substitute for the original undefined behaviour. Note that the compiler will (should) optimise down to the same emitted code, since this is a pattern it understands. 
[ROCm/clr commit: 07c9337f57]
 2017-11-18 01:16:31 +00:00
Alex Voicu 07c9337f57 This actually (tries) to do the right thing all the way, by using memcpy for bitcasting, and not rely on undefined behaviour of a different flavour as a substitute for the original undefined behaviour. Note that the compiler will (should) optimise down to the same emitted code, since this is a pattern it understands. 2017-11-18 01:16:31 +00:00
Alex Voicu 6fa7adf077 This actually (tries) to do the right thing all the way, by using memcpy for bitcasting, and not rely on undefined behaviour of a different flavour as a substitute for the original undefined behaviour. Note that the compiler will (should) optimise down to the same emitted code, since this is a pattern it understands. 2017-11-18 01:16:31 +00:00
Alex Voicu 5f3adc4124 This fixes some outright quaint choices made when implementing HIP's bitwise conversion functions, by using simple reinterpret_casts, as is idiomatic. These functions are supposed to be re-entrant, correct and efficient. Sadly, they were neither: they hid a massive race condition against a value stored in global memory, which means that they were also unreasonably slow if they ever managed to be correct, and relied on union based type punning which is in a grey area of the standard. It is difficult to ascertain what may have been the reason for coming up with this quirky solution. 
[ROCm/clr commit: d7d77cfbe1]
 2017-11-17 16:00:28 +00:00
Alex Voicu d7d77cfbe1 This fixes some outright quaint choices made when implementing HIP's bitwise conversion functions, by using simple reinterpret_casts, as is idiomatic. These functions are supposed to be re-entrant, correct and efficient. Sadly, they were neither: they hid a massive race condition against a value stored in global memory, which means that they were also unreasonably slow if they ever managed to be correct, and relied on union based type punning which is in a grey area of the standard. It is difficult to ascertain what may have been the reason for coming up with this quirky solution. 2017-11-17 16:00:28 +00:00
Alex Voicu 153878e368 This fixes some outright quaint choices made when implementing HIP's bitwise conversion functions, by using simple reinterpret_casts, as is idiomatic. These functions are supposed to be re-entrant, correct and efficient. Sadly, they were neither: they hid a massive race condition against a value stored in global memory, which means that they were also unreasonably slow if they ever managed to be correct, and relied on union based type punning which is in a grey area of the standard. It is difficult to ascertain what may have been the reason for coming up with this quirky solution. 2017-11-17 16:00:28 +00:00
foreman c2091ac9ae P4 to Git Change 1483566 by gandryey@gera-lnx-rcf on 2017/11/16 18:12:27 
SWDEV-79445 - Remove unused fields

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_gl_amd.hpp#19 edit


[ROCm/clr commit: d0a0f6575e]
 2017-11-16 18:22:09 -05:00
foreman d0a0f6575e P4 to Git Change 1483566 by gandryey@gera-lnx-rcf on 2017/11/16 18:12:27 
SWDEV-79445 - Remove unused fields

Affected files ...

... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_gl_amd.hpp#19 edit
 2017-11-16 18:22:09 -05:00
Alex Voicu 3b413e7ee7 Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe 
[ROCm/clr commit: 45ac5b70df]
 2017-11-16 23:20:15 +00:00
Alex Voicu 45ac5b70df Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe 2017-11-16 23:20:15 +00:00
Alex Voicu f93859cdc2 Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe 2017-11-16 23:20:15 +00:00
Evgeny c5e240bdb5 tool sources comments 2017-11-16 12:50:14 -06:00
Evgeny eebfea0de4 tool sources comments 
[ROCm/rocprofiler commit: c5e240bdb5]
 2017-11-16 12:50:14 -06:00
Chris Freehill 651ae1bf70 Device ID/family corrections for gfx9xx 
Change-Id: Icb25fbbaeb99ce886a2852b48d02875ee0f197a2
 2017-11-16 07:27:54 -05:00
Chris Freehill ed0537ed67 Device ID/family corrections for gfx9xx 
Change-Id: Icb25fbbaeb99ce886a2852b48d02875ee0f197a2


[ROCm/ROCR-Runtime commit: 651ae1bf70]
 2017-11-16 07:27:54 -05:00
foreman b540df1727 P4 to Git Change 1483131 by johtaylo@johtaylo-jtincrementor2-increment on 2017/11/16 03:00:04 
SWDEV-2 - Change OpenCL version number from 2539 to 2540.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2287 edit


[ROCm/clr commit: d0c3f25bf9]
 2017-11-16 03:07:49 -05:00
foreman d0c3f25bf9 P4 to Git Change 1483131 by johtaylo@johtaylo-jtincrementor2-increment on 2017/11/16 03:00:04 
SWDEV-2 - Change OpenCL version number from 2539 to 2540.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2287 edit
 2017-11-16 03:07:49 -05:00
Rahul Garg b0461a2176 Fixed test case for GFX900 
[ROCm/clr commit: cfc9f87465]
 2017-11-16 09:34:52 +05:30
Rahul Garg cfc9f87465 Fixed test case for GFX900 2017-11-16 09:34:52 +05:30
Rahul Garg 9af0f9cbc1 Fixed test case for GFX900 2017-11-16 09:34:52 +05:30
Evgeny 68a89bf87d adding completion handler 2017-11-15 20:59:24 -06:00
Evgeny 517c18325b adding completion handler 
[ROCm/rocprofiler commit: 68a89bf87d]
 2017-11-15 20:59:24 -06:00
foreman 35fa4b72f3 P4 to Git Change 1483021 by lmoriche@lmoriche_opencl_dev2 on 2017/11/15 21:49:30 
SWDEV-118564 - [OCL-LC-ROCm] Disable HSAILProgram and HSAILKernel if the compiler library is not available.

Affected files ...

... //depot/stg/opencl/drivers/opencl/opencldefs#222 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.cpp#214 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/roccompiler.cpp#39 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#76 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.cpp#31 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.hpp#18 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#76 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.hpp#31 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/program.cpp#89 edit


[ROCm/clr commit: 5ec0d059a3]
 2017-11-15 21:58:49 -05:00
foreman 5ec0d059a3 P4 to Git Change 1483021 by lmoriche@lmoriche_opencl_dev2 on 2017/11/15 21:49:30 
SWDEV-118564 - [OCL-LC-ROCm] Disable HSAILProgram and HSAILKernel if the compiler library is not available.

Affected files ...

... //depot/stg/opencl/drivers/opencl/opencldefs#222 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/device.cpp#214 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/roccompiler.cpp#39 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#76 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.cpp#31 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.hpp#18 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#76 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.hpp#31 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/program.cpp#89 edit
 2017-11-15 21:58:49 -05:00
foreman 0c8ce3dd18 P4 to Git Change 1482981 by skeely@skeely_HSA_linux on 2017/11/15 18:17:38 
SWDEV-136633 - Remove correlation handle.

	Not used by debugger or profiler, dead code.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#47 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#15 edit


[ROCm/clr commit: 88783fe59b]
 2017-11-15 18:24:51 -05:00
foreman 88783fe59b P4 to Git Change 1482981 by skeely@skeely_HSA_linux on 2017/11/15 18:17:38 
SWDEV-136633 - Remove correlation handle.

	Not used by debugger or profiler, dead code.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#47 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#15 edit
 2017-11-15 18:24:51 -05:00
foreman 787941237c P4 to Git Change 1482914 by wchau@wchau_OCL_boltzmann on 2017/11/15 16:12:45 
SWDEV-137374 - Collecting counters from some TCC, TCP and TA block instances causes kernel to hang
	- fixed by using the new AQLProfile feature to obtain precise size for command/data buffers

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/roccounters.cpp#2 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#46 edit


[ROCm/clr commit: 0e9cee7205]
 2017-11-15 16:22:18 -05:00
foreman 0e9cee7205 P4 to Git Change 1482914 by wchau@wchau_OCL_boltzmann on 2017/11/15 16:12:45 
SWDEV-137374 - Collecting counters from some TCC, TCP and TA block instances causes kernel to hang
	- fixed by using the new AQLProfile feature to obtain precise size for command/data buffers

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/roccounters.cpp#2 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#46 edit
 2017-11-15 16:22:18 -05:00
Rahul Garg c3ba309562 Fixed half2 issue on gfx900 
[ROCm/clr commit: 20947f80fb]
 2017-11-15 18:52:59 +05:30
Rahul Garg 20947f80fb Fixed half2 issue on gfx900 2017-11-15 18:52:59 +05:30
Rahul Garg fef496d4f1 Fixed half2 issue on gfx900 2017-11-15 18:52:59 +05:30
Rahul Garg 0f8296b0c2 Removed redundant desc variable 
[ROCm/clr commit: e2e499461a]
 2017-11-15 18:28:27 +05:30
Rahul Garg e2e499461a Removed redundant desc variable 2017-11-15 18:28:27 +05:30
Rahul Garg ae1eb7a03a Removed redundant desc variable 2017-11-15 18:28:27 +05:30
Rahul Garg 657aa51d5d -Fixed texture driver API sample 
-Added hipTexRefSetAddress and hipTexRefSetAddress2D APIs
 2017-11-15 18:23:28 +05:30
Rahul Garg a0f838a75e -Fixed texture driver API sample 
-Added hipTexRefSetAddress and hipTexRefSetAddress2D APIs


[ROCm/clr commit: edbfeeb141]
 2017-11-15 18:23:28 +05:30
Rahul Garg edbfeeb141 -Fixed texture driver API sample 
-Added hipTexRefSetAddress and hipTexRefSetAddress2D APIs
 2017-11-15 18:23:28 +05:30
Rahul Garg 4b19c2aa0c -Fixed texture driver API sample 
-Added hipTexRefSetAddress and hipTexRefSetAddress2D APIs
 2017-11-15 18:23:28 +05:30
Phaneendr-kumar Lanka 3bfe30e22c [nvccTests]Enabled inline_asm_vadd on nvcc 
[ROCm/clr commit: 9fa3ed4f6f]
 2017-11-14 16:37:59 +05:30
Phaneendr-kumar Lanka 9fa3ed4f6f [nvccTests]Enabled inline_asm_vadd on nvcc 2017-11-14 16:37:59 +05:30
Phaneendr-kumar Lanka 18f6e31d1d [nvccTests]Enabled inline_asm_vadd on nvcc 2017-11-14 16:37:59 +05:30
foreman c56619ed03 P4 to Git Change 1482174 by johtaylo@johtaylo-jtincrementor2-increment on 2017/11/14 03:00:05 
SWDEV-2 - Change OpenCL version number from 2538 to 2539.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2286 edit


[ROCm/clr commit: 0e891db6e4]
 2017-11-14 03:13:39 -05:00
foreman 0e891db6e4 P4 to Git Change 1482174 by johtaylo@johtaylo-jtincrementor2-increment on 2017/11/14 03:00:05 
SWDEV-2 - Change OpenCL version number from 2538 to 2539.

Affected files ...

... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2286 edit
 2017-11-14 03:13:39 -05:00
Rahul Garg c154921e10 Texture code reorganized 
[ROCm/clr commit: e536eb038d]
 2017-11-14 11:09:35 +05:30
Rahul Garg e536eb038d Texture code reorganized 2017-11-14 11:09:35 +05:30
Rahul Garg 63680edd30 Texture code reorganized 2017-11-14 11:09:35 +05:30
Evgeny 838b42153b Adding trace data output file, trace_data.sqtt 2017-11-13 17:03:53 -06:00