* Fixed Dyninst TBB symbolic links + bump to v0.0.2
* hosttrace exe and library updates + submodule updates
- Updated dyninst submodule with TBB build ORIGIN rpath
- Updated timemory submodule
- Dyninst build with shared libs
- Dockerfile for building packaging
- Disable hidden viz in examples
- parallel-overhead max parallelism
- query_instr in hosttrace
- different file-line info format
- full module names
- minor fix to MPI support
- disable instrumention stack frames by default
- disable trap instrumentation by default
- updated hosttrace output file dumps
- removed cstdlib option
- dyninst DebugParsing option
- improved instrument_module function
- fixed some MPI support
- tweaked some testing parameters
After adjusting the memory usage, re-enable KFDEvictTest.BasicTest on
gfx906 and KFDEvictTest.QueueTest on GFX10
Change-Id: I401e679e447f3150241078154635f0b30692513d
Signed-off-by: Kent Russell <kent.russell@amd.com>
After adjusting the memory usage, re-enable KFDEvictTest.BasicTest on
gfx906 and KFDEvictTest.QueueTest on GFX10
Change-Id: I401e679e447f3150241078154635f0b30692513d
Signed-off-by: Kent Russell <kent.russell@amd.com>
[ROCm/ROCR-Runtime commit: 392e0e2afc]
We were hitting memory map errors and segfaults when trying to use 7/8
VRAM on certain cards (dmesg showing "Failed to map to gpu 0/1"), as the
original check didn't see if this would exceed the GTT size.
Allocate the smaller of 1/3 of SRAM or 7/16 of VRAM (7/8 / 2) for the
tests
Signed-off-by: Kent Russell <kent.russell@amd.com>
Change-Id: Ic11a5cee058535418eef903a28846e00e1839969
We were hitting memory map errors and segfaults when trying to use 7/8
VRAM on certain cards (dmesg showing "Failed to map to gpu 0/1"), as the
original check didn't see if this would exceed the GTT size.
Allocate the smaller of 1/3 of SRAM or 7/16 of VRAM (7/8 / 2) for the
tests
Signed-off-by: Kent Russell <kent.russell@amd.com>
Change-Id: Ic11a5cee058535418eef903a28846e00e1839969
[ROCm/ROCR-Runtime commit: d5d8c022f1]
Changed var_pattern in tblextr.py to include pattern like "name[0]"
Change-Id: Ibe1c512595cfbdcaca8fa5bddceb3f6a570caf43
[ROCm/rocprofiler commit: ff43ca1542]
Implementation to use a blit kernel to perform
a hipStreamWait/write instead of an AQL packet.
Change-Id: I462671ed5cec37144dfe97ff66439249196117c1
[ROCm/clr commit: cbb8d82bdb]
Implementation to use a blit kernel to perform
a hipStreamWait instead of an AQL packet.
Change-Id: I9d9bc1dd793c1454e91e68afb793460749e10d2e
[ROCm/clr commit: 4177d20877]
As part of the stack unification some of the application using
opencl seem to depend on libopencl.so provided from thirdparty
providers. For libopencl.so from thirdparty provider to find
libamdocl.so ldconfig path to {opencl install path}/lib is required
Change-Id: I8e12e463209a44177ee433644c75ef0781f1da2f
[ROCm/clr commit: b4b83e274e]
As part of the stack unification some of the application using
opencl seem to depend on libopencl.so provided from thirdparty
providers. For libopencl.so from thirdparty provider to find
libamdocl.so ldconfig path to {opencl install path}/lib is required
Change-Id: I8e12e463209a44177ee433644c75ef0781f1da2f
ATI_OS_WIN -> _WIN32
ATI_OS_LINUX -> __linux__
We should not rely on non-standard macros for platform detections.
Change-Id: If1d06e2e1187268df62a59609ea5496ab7eb709d
[ROCm/clr commit: c4aa6febe5]
ATI_OS_WIN -> _WIN32
ATI_OS_LINUX -> __linux__
We should not rely on non-standard macros for platform detections.
Change-Id: If1d06e2e1187268df62a59609ea5496ab7eb709d
The OCL_ICD_FILENAMES env var contains a list of OpenCL platforms that
the ICD should prepend to the list of reported platforms. Since the
built amdocl library will show up as the first platform, we can tell
ocltst to always run with platform 0. This way we guarantee that the
correct binary is being loaded.
Using this env var also makes programming LD_LIBRARY_PATH redundant
for Linux.
Change-Id: I8c1729e8b4dd1cd2df1a3da08c7b7addebd08b4f
[ROCm/clr commit: 6c9b07370c]
The OCL_ICD_FILENAMES env var contains a list of OpenCL platforms that
the ICD should prepend to the list of reported platforms. Since the
built amdocl library will show up as the first platform, we can tell
ocltst to always run with platform 0. This way we guarantee that the
correct binary is being loaded.
Using this env var also makes programming LD_LIBRARY_PATH redundant
for Linux.
Change-Id: I8c1729e8b4dd1cd2df1a3da08c7b7addebd08b4f
optind's starting value is 1 and it needs to be reset to 1 after each
getopt() invocation. Resetting it to 0 will cause getopt() to
reinitialize its internal state. This is only needed for some corner
cases. In our current usage, we should be resetting it to 1.
Change-Id: I68da7e41ee8b83b275e2a7bd660c13af569769f1
[ROCm/clr commit: 3d82092632]
optind's starting value is 1 and it needs to be reset to 1 after each
getopt() invocation. Resetting it to 0 will cause getopt() to
reinitialize its internal state. This is only needed for some corner
cases. In our current usage, we should be resetting it to 1.
Change-Id: I68da7e41ee8b83b275e2a7bd660c13af569769f1
SDMA firmware of gfx902 has a regression which causes cp hang
in kfdtest, blacklist related test cases.
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Change-Id: Ie75b72a952a29d6b1394c5fcb67ff9a5143b3b07
SDMA firmware of gfx902 has a regression which causes cp hang
in kfdtest, blacklist related test cases.
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Change-Id: Ie75b72a952a29d6b1394c5fcb67ff9a5143b3b07
[ROCm/ROCR-Runtime commit: 04b149b7fd]
Elena Sakhnovitch
Jonathan R. Madsen
Kent Russell
Ammar ELWazir
Chun Yang
Julia Jiang
Maneesh Gupta
Bill(Shuzhou) Liu
Chauncey Hui
Sarbojit Sarkar
Sourabh
Freddy Paul
Vladislav Sytchenko
Jason Tang
Yifan Zhang