[UR][SYCL] Add urCommandBufferAppendKernelLaunchWithArgsExp()#21110
[UR][SYCL] Add urCommandBufferAppendKernelLaunchWithArgsExp()#21110sarnex merged 6 commits intointel:syclfrom
Conversation
2cc6349 to
0dc6aa2
Compare
0dc6aa2 to
822f68f
Compare
822f68f to
cf7d715
Compare
cf7d715 to
ec70d72
Compare
ec70d72 to
a8e440b
Compare
a8e440b to
91cfcc5
Compare
73defce to
2adcdcb
Compare
|
The HIP test fails because of #18763 |
|
Rebased |
|
Please review |
|
Rebased |
|
Everyone - please review |
|
Rebased |
Signed-off-by: Lukasz Dorau <lukasz.dorau@intel.com>
Commit output of 'make generate'. Signed-off-by: Lukasz Dorau <lukasz.dorau@intel.com>
Signed-off-by: Lukasz Dorau <lukasz.dorau@intel.com>
Signed-off-by: Lukasz Dorau <lukasz.dorau@intel.com>
Signed-off-by: Lukasz Dorau <lukasz.dorau@intel.com>
|
Please review |
adamfidel
left a comment
There was a problem hiding this comment.
Besides a minor change to a comment, LGTM. I mainly looked at the graph changes and some of the L0 adapter.
| &redefinedCommandBufferAppendKernelLaunchWithArgsExpAfter); | ||
|
|
||
| ASSERT_NO_THROW(Graph.finalize(experimental::property::graph::updatable{})); | ||
| // GetInfo and AppendKernelLaunch should be called once each time a node is |
There was a problem hiding this comment.
| // GetInfo and AppendKernelLaunch should be called once each time a node is | |
| // GetInfo and AppendKernelLaunchWithArgs should be called once each time a node is |
|
Please review @intel/unified-runtime-reviewers-opencl @intel/dpcpp-nativecpu-reviewers @intel/llvm-reviewers-cuda |
|
One CI check fails because of hung tests detected on a CI machine. |
|
Please review @intel/unified-runtime-reviewers-opencl @intel/dpcpp-nativecpu-reviewers @intel/llvm-reviewers-cuda |
|
Please review @intel/llvm-reviewers-runtime |
|
@intel/llvm-gatekeepers please merge The SYCL Pre Commit on Windows / run_prebuilt_e2e_tests (Intel Arc Graphics with Level Zero) CI job keeps failing because of #20750 |
No description provided.