CMakeLists.txt does not set up the DEBUG macro correctly to mean
!NDEBUG, so, as a workaround, replace all uses of ifdef NDEBUG with
ifndef DEBUG in the library sources.
Change-Id: I408adb36d1a2310fb894a486574469662ebb27cd
[ROCm/rocm_smi_lib commit: 9f87197d8d]
CMakeLists.txt does not set up the DEBUG macro correctly to mean
!NDEBUG, so, as a workaround, replace all uses of ifdef NDEBUG with
ifndef DEBUG in the library sources.
Change-Id: I408adb36d1a2310fb894a486574469662ebb27cd
The -Wl,--build-id option is added for address sanitizer build
Change-Id: I0d75bc8e6169010c460e62e51708828e75de478e
[ROCm/rocm_smi_lib commit: 7b69dde24f]
Minimum queue size was not enforced at the Agent level. Minimum
size should be one page to give unifority across all asics.
Change-Id: I26394f79458d09fbceb79fc8aaf495e2c26a8ff3
On gfx90a only a reduced number of CUs must be used for cooperative
dispatches due to CWSR and launcher interactions with asymetric
harvest. We must use one fewer CUs per SE than the lowest count of
CUs on any SE.
Also adds env var HSA_COOP_CU_COUNT which enables the cooperative
CU count computation. Set to 1 to enable the new computation.
This is an opt-in feature that will become enabled by default (opt-out)
in a future release.
Change-Id: Ifbb75ced3bbc15876eef44922c6a4f6fde8c4c28
Corrections have been made in libhsakmt, and corresponding changes are required here as well.
Signed-off-by: Chen Gong <curry.gong@amd.com>
Change-Id: Ib697ce25278c2c5ac6ef0206930ec285f46c60d1
Minimum queue size was not enforced at the Agent level. Minimum
size should be one page to give unifority across all asics.
Change-Id: I26394f79458d09fbceb79fc8aaf495e2c26a8ff3
[ROCm/ROCR-Runtime commit: a6742209f7]
Adds asserts for invalid dispatch dims and scratch requests that
don't actually use scratch.
Change-Id: I6e6eef3f17dc38adaf96550fa55bd8625868efa3
[ROCm/ROCR-Runtime commit: a65f3f5b71]
On gfx90a only a reduced number of CUs must be used for cooperative
dispatches due to CWSR and launcher interactions with asymetric
harvest. We must use one fewer CUs per SE than the lowest count of
CUs on any SE.
Also adds env var HSA_COOP_CU_COUNT which enables the cooperative
CU count computation. Set to 1 to enable the new computation.
This is an opt-in feature that will become enabled by default (opt-out)
in a future release.
Change-Id: Ifbb75ced3bbc15876eef44922c6a4f6fde8c4c28
[ROCm/ROCR-Runtime commit: 37942c982a]
Corrections have been made in libhsakmt, and corresponding changes are required here as well.
Signed-off-by: Chen Gong <curry.gong@amd.com>
Change-Id: Ib697ce25278c2c5ac6ef0206930ec285f46c60d1
[ROCm/ROCR-Runtime commit: dec63b4f15]
When OCL failed to obtain function pointer from GL, we should not call it.
Change-Id: I50f69d270ce445386906a286e44c4e8c83722302
[ROCm/clr commit: 15101e704b]
ajay
Laurent Morichetti
Bill(Shuzhou) Liu
Chauncey Hui
Wenkai Du
Sreekant Somasekharan
Stanley Tsang
Wenkai Du
Roopa Malavally
Sean Keely
Chen Gong
Jeremy Newton
Nirmal Unnikrishnan
Alex Xie
Payam