SWDEV-2 - Change OpenCL version number from 2874 to 2875.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2622 edit
SWDEV-145570 - Check host_context when matching GPU device.
- In CL#1766264, `host_context` is introduced for mGPU support. Need to
match that context specially when trying to match GPU device context.
The following tests passed:
$ python test_dataloader.py TestDictDataLoader.test_pin_memory
.
----------------------------------------------------------------------
Ran 1 test in 0.004s
OK
$ python test_dataloader.py TestDataLoader.test_sequential_pin_memory
.
----------------------------------------------------------------------
Ran 1 test in 0.063s
OK
$ python test_dataloader.py TestDataLoader.test_shuffle_pin_memory
.
----------------------------------------------------------------------
Ran 1 test in 0.174s
OK
$ python test_dataloader.py TestStringDataLoader.test_shuffle_pin_memory
.
----------------------------------------------------------------------
Ran 1 test in 0.104s
OK
$ python test_torch.py TestTorch.test_pin_memory
.
----------------------------------------------------------------------
Ran 1 test in 0.124s
OK
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#52 edit
+ Fix the error in cmake: clang 8.0.0 supports CUDA 10.0, not 10.1; only trunk clang 9.0.0 supports CUDA 10.1.
+ Update Readme:
1. clang's bug 36384 is gone, so latest stable release for Linux is clang 8.0.0, which supports CUDA up to 10.0;
2. update dependencies, testing and building notes.
Due to a HCC bug in rocm-head, the test fails to compile. Disabling the test until the issue is resolved.
Change-Id: Ib4e7baf0b9c2cb5f5dbe38e6dd2bab894d28886a
SWDEV-2 - Change OpenCL version number from 2873 to 2874.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2621 edit
SWDEV-144570 - Fix pointer attribute query.
- For memory not registered with runtime, return
`hipErrorInvalidValue`. That's the behavior expected to check whether
a host buffer is pinned.
- Return `hipErrorInvalidDevice` in case a registered memory object
cannot find its matching device.
RB: http://ocltc.amd.com/reviews/r/17094/
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#51 edit
SWDEV-145570 - [HIP] Use a context with all devices in system for host register
hipHostRegister and hipMemcpy 0x10 and 0x20 fail in mGPU systems because
we only register the memory on the current device. But in HIP, the registering
needs to happen on all devices.
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/hip/hip_context.cpp#17 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_internal.hpp#26 edit
... //depot/stg/opencl/drivers/opencl/api/hip/hip_memory.cpp#50 edit
SWDEV-2 - Change OpenCL version number from 2872 to 2873.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2620 edit
SWDEV-145570 - Disable subwindow copy temporarily to see if TC random hangs go away
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocblit.cpp#30 edit
SWDEV-2 - Change OpenCL version number from 2871 to 2872.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2619 edit
SWDEV-79445 - OCL generic changes and code clean-up
- More changes to make sure runtime and LC could be built separately
Affected files ...
... //depot/stg/opencl/drivers/opencl/Makefile#71 edit
... //depot/stg/opencl/drivers/opencl/compiler/Makefile#73 edit
... //depot/stg/opencl/drivers/opencl/library/build/Makefile.library#78 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/devprogram.cpp#38 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#88 edit
SWDEV-132899 - [OCL] set lcWavefrontSize64_ on ROCm for all ASICs except gfx10+ so that we Link wavefrontsize64 correctly
ReviewBoardURL = http://ocltc.amd.com/reviews/r/17081/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#43 edit
SWDEV-2 - Change OpenCL version number from 2870 to 2871.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2618 edit
SWDEV-183848 - Update OpenCL perfcounter blocks to match the block lists used by GPUPerfAPI (which is in turn used by RCP and CodeXL)
- The PAL block index for the PA_SC block was incorrect in all hw generations (in PAL, GpuBlock::SC has ordinal 4, not 3)
- In GFX9, the MC and SRBM blocks are not supported
- In GFX10, the following changes are made:
-- There are 8 CB instances per SE
-- There are 8 DB instances per SE
-- There are 2 PA_SU instances per SE
-- There are 4 PS_SC instances per SE
-- There are 4 RMI instances per SE
-- There are 2 GL1A instances per SE
-- There are 2 GL1C instances per SE
-- There are 8 GL1CG instances per SE
-- There are 4 global GL2A instances
-- There are 16 global GL2A instances
-- There are 4 global CHC instances
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palcounters.cpp#19 edit
SWDEV-2 - Change OpenCL version number from 2869 to 2870.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2617 edit
SWDEV-2 - Change OpenCL version number from 2868 to 2869.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2616 edit
__hipRegisterFunction is called during by .init functions during program initialization.
It calls hipModuleGetFunction to locate kernel symbol in code objects. hipModuleGetFunction
assumes current device when locating kernel symbols. This works for HCC but not for hip-clang,
since hip-clang needs to locate kernel symbols for different devices without switching
between devices.
This patch introduces a new hsa agent parameter to ihipModuleGetFunction, which allows
__hipRegisterFunction to choose the correct hsa agent when locating kernel symbols. By
default it uses this_agent(), therefore this patch has no impact on HCC.
SWDEV-2 - Change OpenCL version number from 2867 to 2868.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#2615 edit
Yaxun (Sam) Liu
Jenkins
foreman
Maneesh Gupta
Evgeny Mankov
David Salinas
Siu Chi Chan
Anthony Cowley