EPR #399808 - Fix the value of HSA image channel order for CL_RGB
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.cpp#188 edit
[ROCm/clr commit: 5f93384dbc]
EPR #010002 - Change OpenCL version number from 1595 to 1596.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1342 edit
[ROCm/clr commit: 8431455a87]
ECR #304775 - Bug 10112 - Raise default unroll threshold. The current default is 100, which is even lower than the LLVM default of 150. Increasing to 200 is a modest increase, and this should probably be even higher.
Affected files ...
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/OPTIONS.def#111 edit
[ROCm/clr commit: 2c5424663c]
EPR #402935 - Reset Resource::pinOffset_ if gslResource couldn't be created for pinned memory.
When the pinned memory to be created is too large, gslResource couldn't be created, and a local memory will be created instead. If pinOffset_ is NOT reset in this case, it will mess up future copying of the local memory.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuresource.cpp#187 edit
[ROCm/clr commit: 1e0a5f64f5]
EPR #010002 - Change OpenCL version number from 1594 to 1595.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1341 edit
[ROCm/clr commit: a7c60aeaed]
ECR #304775 - Device enqueuing
- Use atomic fetch for enqueue flags
- Switch to a multithreaded scheduler
- Add a workaround for Linux host_multi_queue failures. Linux has only 2 queues, but the test allocates multiple host queues and the same HW ring can be used
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpublit.cpp#106 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#449 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.hpp#127 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuschedcl.cpp#22 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.cpp#325 edit
[ROCm/clr commit: d2b905f18e]
EPR #010002 - Change OpenCL version number from 1593 to 1594.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1340 edit
[ROCm/clr commit: 1e8c506c75]
ECR #304775 - Device enqueuing
- Add L2 cache flush after the scheduler execution. Although CP has to work with L2 cache, it seems some functionality relies on direct memory access and without explicit L2 flush CP can pick old values in the template.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.cpp#324 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLContext.cpp#61 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLContext.h#35 edit
[ROCm/clr commit: 4599bd0d4a]
ECR #304775 - OpenCL version should be 2.0 for only CI+
ReviewBoardURL = http://ocltc.amd.com/reviews/r/5311/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#448 edit
[ROCm/clr commit: db6baeefb6]
EPR #010002 - Change OpenCL version number from 1592 to 1593.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1339 edit
[ROCm/clr commit: 886bdae624]
EPR #010002 - Change OpenCL version number from 1591 to 1592.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1338 edit
[ROCm/clr commit: 55e7df9dd4]
EPR #010002 - Change OpenCL version number from 1590 to 1591.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1337 edit
[ROCm/clr commit: 645f5858fd]
EPR #010002 - Change OpenCL version number from 1589 to 1590.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1336 edit
[ROCm/clr commit: 95e9092dca]
EPR #010002 - Change OpenCL version number from 1588 to 1589.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1335 edit
[ROCm/clr commit: cb5a679043]
EPR #389586 - Add workaround for VI SPI SGPR initialization hardware bug for HSAIL path.
There is a hardware bug in VI (UBTS502672) which requires a workaround. Compute shaders need to tell shader compiler the available sGPR is 78 and set sGPUR usage in the compiled ISA to be 94. It has been done in AMDIL path but not done in HSAIL path. This change will apply the workaround to HSAIL path.
Affected files ...
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/scwrapper/SI/devStateSI.cpp#16 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/scwrapper/SI/devStateSI.h#11 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/scwrapper/SI/scCompileSI.cpp#41 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/OPTIONS.def#109 edit
[ROCm/clr commit: 1abfb8b305]
EPR #010002 - Change OpenCL version number from 1587 to 1588.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1334 edit
[ROCm/clr commit: 62c6c2d64e]
ECR #304775 - Device enqueuing
- Switch to the single thread scheduler for now(the current version isn't friendly for single thread). Hopefully it's a temporary solution until synchronization issue with multithreaded scheduler will be identified.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpublit.cpp#104 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuschedcl.cpp#20 edit
[ROCm/clr commit: 6314b334ba]
EPR #394069 - [CQE OCL][ISV][QR][G] Debugger CAL path no longer works due to OCL CL # 1003498. Fix ACL path to support debugger
ReviewBoardURL=http://ocltc.amd.com/reviews/r/5245/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpubinary.cpp#57 edit
[ROCm/clr commit: 7ca4c75217]
EPR #010002 - Change OpenCL version number from 1586 to 1587.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1333 edit
[ROCm/clr commit: 770224c321]
EPR #010002 - Change OpenCL version number from 1585 to 1586.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1332 edit
[ROCm/clr commit: 17c7b66e4e]
ECR #304775 - Fix valgrind use of uninitialized errors, modernize by using unique_ptr, and use existing function to extract filename from path.
Should probably also fix the static constructors this is called from.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/os/os_posix.cpp#37 edit
[ROCm/clr commit: a901b12dd1]
ECR #377625 - Allow function call for function with internal linkage.
Internal linkage correponds to static function in C. The function could be large and should be allowed to be not inlined.
Affected files ...
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/linker.cpp#106 edit
[ROCm/clr commit: 00cd00cbee]
EPR #010002 - Change OpenCL version number from 1584 to 1585.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1331 edit
[ROCm/clr commit: 1729d48d6c]
EPR #010002 - Change OpenCL version number from 1583 to 1584.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1330 edit
[ROCm/clr commit: 8aa5ad7859]
EPR #010002 - Change OpenCL version number from 1582 to 1583.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1329 edit
[ROCm/clr commit: caa8f05683]
EPR #304775 - temporarily disable the SVM fine_grained_buffer support for OpenCL 2.0 on discrete GPUs, because the feature is supposed to release in 14.50. After the 14.40 is branched, we will enable it again on stg.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#445 edit
[ROCm/clr commit: 18d6efdf2c]
ECR #304775 - Device enqueuing
- Add printing of the waiting events
- Add early exit in the scheduler if nothing to launch
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuschedcl.cpp#19 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuvirtual.cpp#321 edit
[ROCm/clr commit: 977f314094]
EPR #010002 - Change OpenCL version number from 1581 to 1582.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1328 edit
[ROCm/clr commit: 19cae20ac8]
EPR #397491 - Disable platform atomics temporarily until AFE which will be done on July 8.
Modify the flag of GPU_ENABLE_HIGH_PERFORMANCE_STATE to use it for platform atomics because GPU_ENABLE_HIGH_PERFORMANCE_STATE is not necessary for high clock anymore.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpusettings.cpp#271 edit
... //depot/stg/opencl/drivers/opencl/runtime/utils/flags.hpp#206 edit
[ROCm/clr commit: 2677e086a7]
foreman