SWDEV-102733 - [OCL-LC-ROCm] Cmake build Write CMakeLists.txt to enable building with and without the DK environment
- Make it possible to build with libstdc++
Affected files ...
... //depot/stg/opencl/drivers/opencl/api/opencl/amdocl/cl_svm.cpp#19 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/loaders/elf/utils/libelf/memfile.cpp#3 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/appprofile.cpp#16 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/roccompiler.cpp#31 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.hpp#19 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#63 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.hpp#22 edit
... //depot/stg/opencl/drivers/opencl/runtime/platform/memory.cpp#126 edit
... //depot/stg/opencl/drivers/opencl/runtime/thread/semaphore.cpp#9 edit
SWDEV-116535 - Report CL_FP_DENORM for single fp config for gfx9 for LC for rocm/pal and force denorms on based on AMD_GPU_FORCE_SINGLE_FP_DENORM for rocm.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palsettings.cpp#23 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#47 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#62 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#16 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.hpp#7 edit
SWDEV-79278 - [OpenCL][GFXIP9 Bring up] Add the Vega10 DeviceInfo entry and the gfx90X control functions.
- modify getHsaDeviceId() to use HSA_AGENT_INFO_NAME to determine the HSA device ID.
- add back the codes to get the pci_id in the function
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#46 edit
SWDEV-79278 - [OpenCL][GFXIP9 Bring up] Add the Vega10 DeviceInfo entry and the gfx90X control functions.
- modify getHsaDeviceId() to use HSA_AGENT_INFO_NAME to determine the HSA device ID.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#45 edit
SWDEV-116959 - ROCm OpenCL: split -amdgpu-internalize-symbols and -amdgpu-early-inline-all
We shall not run internalize pass in FE, it breaks separate compilation.
Thus two options are split now.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palcompiler.cpp#14 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#38 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/roccompiler.cpp#30 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#61 edit
SWDEV-112401 - [OCLonROCm vs LinuxPro][Fiji][Perf] Around ~ 70% performance drop observed with OCL SDK - BlackScholes test
- Switch persistent emulation to device memory instead of system memory, since ROCr doesn't support visible heap without large bar. It's a potential tradeoff of persistent usage in the applications between an upload buffer or directly in a kernel.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmemory.cpp#11 edit
SWDEV-103424 - [ROCm CQE][OCL] OCLRuntime - OCLCreateBuffer tests are failing. The failure is due to AQL cannot support global size > 32bit range. Adding dispatch split support for ROCm, similar to that of GSL (CL#1159349), to resolve the issue.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.hpp#13 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#56 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#31 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.hpp#8 edit
... //depot/stg/opencl/drivers/opencl/tests/ocltst/module/runtime/OCLCreateBuffer.cpp#6 edit
SWDEV-112661 - [ROCm CQE][OCLonLC][QR][G] 80 failures observed with Images(Samplerless and Samplerless pitch) tests
- Make sure the pitch for image buffer is properly aligned to the reported device info
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmemory.cpp#10 edit
SWDEV-112171 - [ROCm CQE][OCLonLC][QR][G] System hangs/Failures observed with few WF conf tests, due to CL#1364923
- reenable HSA copy in ROCr, since OCL copy doesn't really fix the hang issue in integer_ops, but just delays it.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#30 edit
SWDEV-112352 - [ROCm CQE][OCLonLC][QR][G] Aborted/Segmentation fault is observed with "sub_buffers_read_write_dual_devices" of Buffers, due to CL#1364923
- Keep system memory for multiple devices in case of possible initialization on the second device
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#41 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmemory.cpp#9 edit
SWDEV-112171 - [ROCm CQE][OCLonLC][QR][G] System hangs/Failures observed with few WF conf tests, due to CL#1364923
- Disable ROCr copy buffer and use runtime kernel to avoid HW hangs in integer_ops.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/blit.hpp#3 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocblit.cpp#12 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#29 edit
SWDEV-112171 - [ROCm CQE][OCLonLC][QR][G] System hangs/Failures observed with few WF conf tests, due to CL#1364923
- Back out changelist 1367810. A hang was delayed, but still occured much later in the test.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocblit.cpp#11 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#40 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.hpp#16 edit
SWDEV-105835 - ROCm OpenCL: add -amdgpu-internalize-symbols to BE
The option -amdgpu-internalize-symbols allows to drop unused symbols from program,
functions and global variables. This saves compile time and object size, a lot in
case of a big program.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#33 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#55 edit
SWDEV-112171 - [ROCm CQE][OCLonLC][QR][G] System hangs/Failures observed with few WF conf tests, due to CL#1364923
- Make HSA copy call lock protected, due to a possible race condition and HW hangs in integer_ops long_math test
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocblit.cpp#10 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#39 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.hpp#15 edit
SWDEV-112171 - [ROCm CQE][OCLonLC][QR][G] System hangs/Failures observed with few WF conf tests, due to CL#1364923
- Add missing row and slice pitches to fix "clReadWriteImage Pitch" and "clFillImage pitch" tests
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocblit.cpp#9 edit
SWDEV-112273 - [ROCm CQE][OCLonLC][QR][G] 2 of 9 tests failed with mem_host_flags due to CL#1366459
- Keep system memory for images with UHP allocations
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#38 edit
SWDEV-112016 - [ROCm CQE][OCLonLC][QR][G] Seg fault observed with "sub_buffers_read_write" of Buffers, due to CL#1364923
- Take the host pointer for CPU access in bufferfill.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocblit.cpp#8 edit
SWDEV-94644 - Make sure we are processing metadata note entry with supported n_type. Update build log and fail for not supported metadata n_type. Use constants defined in AMDGPUPTNote.h
This change is needed for https://reviews.llvm.org/D29115
This change is required for CL 1366203
ReviewBoardURL: http://ocltc.amd.com/reviews/r/12223/
Testing: lightning conformance tests locally
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#31 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#54 edit
SWDEV-108854 - Support initialization of SGPR and VGPR usage from kernel header object. Changes to fix issue reported by SWDEV-108854
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#33 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.cpp#18 edit
SWDEV-110504 - OpenCL on ROCm: MAX_WORK_GROUP_SIZE is 256 but max work items per dimension is 1024
- Report Orca stack capabilities. OCL runtime currently can't report 1024 due to possible regressions in the apps.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocdevice.cpp#32 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocsettings.cpp#12 edit
SWDEV-108384 - Force map memory offset to 0
Copy what we do in ORCA and PAL backend: When mapping an image, default the offset to 0 instead of the map offset for the staging buffer.
ReviewBoardURL = http://ocltc.amd.com/reviews/r/12010/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmemory.cpp#5 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#25 edit
SWDEV-108384 - Send down row pitch and slice pitch
This fixes test case #8 and 9.
ReviewBoardURL = http://ocltc.amd.com/reviews/r/12009/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#24 edit
SWDEV-79445 - OCL generic changes and code clean-up
- Use switch/case instead of if/else to simplify the code tracing of the argument setup.
http://ocltc.amd.com/reviews/r/11991/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palkernel.cpp#28 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#22 edit
SWDEV-107442, SWDEV-108405 - Need to release GPU memory fence before releasing resources
When we call submitKernelInternal, we need to release GPU memory fence before release resources used in the kernel otherwise we can page fault.
ReviewBoardURL = http://ocltc.amd.com/reviews/r/11955/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocblit.cpp#5 edit
SWDEV-107568 - [ROCm CQE][OCL][CZ] Basic 2.0 conformance test giving Segmentation fault (core dumped) at "progvar_prog_scope_uninit"
- Detect if writable program scope variables are present in the program, and if so, insert barrier each dispatch of a kernel from this program.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.hpp#11 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#48 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.hpp#16 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#21 edit
SWDEV-102966 - Dump code object disassembly in OpenCL rocm device.
Invoke DumpExecutableAsText from driver library.
Update build to depend on some more LLVM libraries.
LLVM changes are included, but will come through amd-common.
Driver changes will come through ROCm-OpenCL-driver.
Testing: Run some SDK samples/test_basic with AMD_OCL_BUILD_OPTIONS_APPEND=-save-temps
Reviewed by: Laurent Morichetti, German Andryeyev.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#27 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.cpp#47 edit
SWDEV-107558 - Wrong path for map/unmap images
If mapMemory is allocated, we shouldn't use the read/writeImage path since they lock/unlock the host pointer which in this case, is mapped to GPU memory already.
ReviewBoardURL = http://ocltc.amd.com/reviews/r/11891/diff/
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocvirtual.cpp#20 edit
SWDEV-108024 - [ROCm CQE] Printf broken on ROCm , LC and HSAIL path
A zero-lenth printfBuffer is not an error. Even if a printfInfo is present, it does not mean that the kernel is going to call printf. Return true if offsetSize is 0.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprintf.cpp#5 edit
foreman