Skip to content

[SYCL][Graph] Kernel property bugfix #12148

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Dec 12, 2023
Merged

[SYCL][Graph] Kernel property bugfix #12148

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 0 additions & 5 deletions sycl/source/detail/graph_impl.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -724,11 +724,6 @@ void exec_graph_impl::createCommandBuffers(
}

exec_graph_impl::~exec_graph_impl() {
WriteLock LockImpl(MGraphImpl->MMutex);

// clear all recording queue if not done before (no call to end_recording)
MGraphImpl->clearQueues();

const sycl::detail::PluginPtr &Plugin =
sycl::detail::getSyclObjImpl(MContext)->getPlugin();
MSchedule.clear();
Expand Down
5 changes: 3 additions & 2 deletions sycl/source/detail/scheduler/commands.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2488,8 +2488,9 @@ pi_int32 enqueueImpCommandBufferKernel(
}

if (Res != pi_result::PI_SUCCESS) {
throw sycl::exception(errc::invalid,
"Failed to add kernel to PI command-buffer");
const device_impl &DeviceImplem = *(DeviceImpl);
detail::enqueue_kernel_launch::handleErrorOrWarning(Res, DeviceImplem,
PiKernel, NDRDesc);
}

return Res;
Expand Down
6 changes: 4 additions & 2 deletions sycl/test-e2e/Graph/Explicit/work_group_size_prop.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,8 +5,10 @@
//
// CHECK-NOT: LEAK

// Note: failing negative test with CUDA & HIP in the original test
// UNSUPPORTED: cuda, hip
// Temporarily disabled for CUDA.
// XFAIL: cuda
// Note: failing negative test with HIP in the original test
// TODO: disable hip when HIP backend will be supported by Graph

#define GRAPH_E2E_EXPLICIT

Expand Down
2 changes: 0 additions & 2 deletions sycl/test-e2e/Graph/Inputs/work_group_size_prop.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -85,8 +85,6 @@ int test(queue &Queue, PropertiesT Props, KernelType KernelFunc) {
return 1;
}

return 0;

// Negative test case: Specify local size that does not match required size.
exp_ext::command_graph GraphN{Queue.get_context(), Queue.get_device()};
try {
Expand Down
6 changes: 4 additions & 2 deletions sycl/test-e2e/Graph/RecordReplay/work_group_size_prop.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,8 +5,10 @@
//
// CHECK-NOT: LEAK

// Note: failing negative test with CUDA & HIP in the original test
// UNSUPPORTED: cuda, hip
// Temporarily disabled for CUDA.
// XFAIL: cuda
// Note: failing negative test with HIP in the original test
// TODO: disable hip when HIP backend will be supported by Graph

#define GRAPH_E2E_RECORD_REPLAY

Expand Down