Skip to content
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

[SYCL][E2E] Update is_compatible tests #17606

Merged
merged 4 commits into from
Mar 26, 2025
Merged

[SYCL][E2E] Update is_compatible tests #17606

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
36994e2
[SYCL][E2E] Update is_compatible tests
KornevNikita Mar 24, 2025
6d57c47
make pre-commit happier
KornevNikita Mar 25, 2025
3100ac2
also disable amdgcn test on win
KornevNikita Mar 25, 2025
250815d
fix amd test
KornevNikita Mar 26, 2025
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
10 changes: 5 additions & 5 deletions sycl/test-e2e/OptionalKernelFeatures/is_compatible/is_compatible_amdgcn.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
// REQUIRES: target-amd, opencl, gpu, cpu
// UNSUPPORTED: windows
// UNSUPPORTED-TRACKER: https://github.com/intel/llvm/issues/17515
// There are no ROCm libs on win machines, so the compilation fails.

// RUN: %clangxx -fsycl -Xsycl-target-backend=amdgcn-amd-amdhsa --offload-arch=gfx906 -fsycl-targets=amdgcn-amd-amdhsa %S/Inputs/is_compatible_with_env.cpp -o %t.out
// RUN: %clangxx -fsycl -fsycl-targets=amdgcn-amd-amdhsa -Xsycl-target-backend=amdgcn-amd-amdhsa --offload-arch=gfx1030 %S/Inputs/is_compatible_with_env.cpp -o %t.out
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

is there a necessity to specify --offload-arch? it is absent for cuda, maybe can be skipped here either.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I agree it may lead to a failure in the future if we change the hardware, but It looks like currently we don't specify the offload-arch automatically, need to extend the testing script. @intel/dpcpp-devops-reviewers am I missing something?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Perhaps you could use %{hip_arch_opts}
see this bit,

llvm/sycl/test-e2e/format.py

Lines 239 to 246 in 0f42c75

if "target-amd" in build_targets:
hip_arch_opts = (
" -Xsycl-target-backend=amdgcn-amd-amdhsa --offload-arch={}".format(
test.config.amd_arch
)
)
sycl_target_opts += hip_arch_opts
substitutions.append(("%{hip_arch_opts}", hip_arch_opts))

Perhaps this could also be guarded by REQUIRES: any-target-is-amd instead of doing UNSUPPORTED: windows?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Perhaps you could use %{hip_arch_opts}

As I understand hip_arch_opts works only with target-amd, so it won't be substituted otherwise.

this could also be guarded by REQUIRES: any-target-is-amd

the intention of this test is to be launched on other than amd platform to check if the is_compatible<kernel>(dev) call returns false if the kernel was built for another device.


// RUN: env ONEAPI_DEVICE_SELECTOR=hip:gpu %{run} %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:gpu %{run} not %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:cpu %{run} not %t.out
// RUN: %if !hip %{ not %} %{run} %t.out
8 changes: 4 additions & 4 deletions sycl/test-e2e/OptionalKernelFeatures/is_compatible/is_compatible_nvptx64.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
// REQUIRES: target-nvidia, opencl, gpu, cpu
// UNSUPPORTED: windows
// UNSUPPORTED-TRACKER: https://github.com/intel/llvm/issues/17515
// There are no CUDA libs on win machines, so the compilation fails.

// RUN: %clangxx -fsycl -fsycl-targets=nvptx64-nvidia-cuda %S/Inputs/is_compatible_with_env.cpp -o %t.out

// RUN: env ONEAPI_DEVICE_SELECTOR=cuda:gpu %{run} %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:gpu %{run} not %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:cpu %{run} not %t.out
// RUN: %if !cuda %{ not %} %{run} %t.out
12 changes: 7 additions & 5 deletions sycl/test-e2e/OptionalKernelFeatures/is_compatible/is_compatible_several_targets.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,9 @@
// REQUIRES: ocloc, any-device-is-level_zero, any-device-is-gpu, any-device-is-cpu
// REQUIRES: opencl-aot, ocloc

// RUN: %clangxx -fsycl -fsycl-targets=spir64_gen -Xsycl-target-backend=spir64_gen "-device *" %S/Inputs/is_compatible_with_env.cpp -o %t.out
// UNSUPPORTED: windows
// UNSUPPORTED-TRACKER: https://github.com/intel/llvm/issues/17515
// There is no CPU device on win yet, so opencl-aot fails to compile the kernel.

// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:cpu %{run-unfiltered-devices} not %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:gpu %{run-unfiltered-devices} %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=level_zero:gpu %{run-unfiltered-devices} %t.out
// RUN: %{run-aux} %clangxx -fsycl -fsycl-targets=spir64_x86_64,spir64_gen -Xsycl-target-backend=spir64_gen %gpu_aot_target_opts %S/Inputs/is_compatible_with_env.cpp -o %t.out

// RUN: %if !(level_zero || opencl) %{ not %} %{run} %t.out
6 changes: 1 addition & 5 deletions sycl/test-e2e/OptionalKernelFeatures/is_compatible/is_compatible_spir64.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,3 @@
// REQUIRES: cuda, opencl, gpu, cpu

// RUN: %clangxx -fsycl -fsycl-targets=spir64 %S/Inputs/is_compatible_with_env.cpp -o %t.out

// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:cpu %{run} %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:gpu %{run} %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=cuda:gpu %{run} not %t.out
// RUN: %if hip || cuda %{ not %} %{run} %t.out
8 changes: 3 additions & 5 deletions sycl/test-e2e/OptionalKernelFeatures/is_compatible/is_compatible_spir64_gen.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,5 @@
// REQUIRES: ocloc, gpu, level_zero, cpu
// REQUIRES: ocloc

// RUN: %clangxx -fsycl -fsycl-targets=spir64_gen -Xsycl-target-backend "-device *" %S/Inputs/is_compatible_with_env.cpp -o %t.out
// RUN: %clangxx -fsycl -fsycl-targets=spir64_gen -Xsycl-target-backend=spir64_gen "-device *" %S/Inputs/is_compatible_with_env.cpp -o %t.out

// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:gpu %{run} %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=level_zero:gpu %{run} %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:cpu %{run} not %t.out
// RUN: %if !(level_zero || opencl && gpu) %{ not %} %{run} %t.out
12 changes: 7 additions & 5 deletions sycl/test-e2e/OptionalKernelFeatures/is_compatible/is_compatible_spir64_x86_64.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,9 @@
// REQUIRES: opencl-aot, cpu, gpu, level_zero
// REQUIRES: opencl-aot

// RUN: %clangxx -fsycl -fsycl-targets=spir64_x86_64 %S/Inputs/is_compatible_with_env.cpp -o %t.out
// UNSUPPORTED: windows
// UNSUPPORTED-TRACKER: https://github.com/intel/llvm/issues/17515
// There is no CPU device on win yet, so opencl-aot fails to compile the kernel.

// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:cpu %{run} %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=opencl:gpu %{run} not %t.out
// RUN: env ONEAPI_DEVICE_SELECTOR=level_zero:gpu %{run} not %t.out
// RUN: %{run-aux} %clangxx -fsycl -fsycl-targets=spir64_x86_64 %S/Inputs/is_compatible_with_env.cpp -o %t.out

// RUN: %if !cpu %{ not %} %{run} %t.out
11 changes: 1 addition & 10 deletions sycl/test-e2e/OptionalKernelFeatures/is_compatible/is_compatible_with_aspects.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,3 @@
// requires: cpu, gpu, accelerator
// RUN: %{build} -Wno-error=incorrect-sub-group-size %O0 -o %t.out
// RUN: %{run} %t.out

Expand All @@ -7,11 +6,10 @@

[[sycl::device_has(sycl::aspect::cpu)]] void foo(){};
[[sycl::device_has(sycl::aspect::gpu)]] void bar(){};
[[sycl::device_has(sycl::aspect::accelerator)]] void baz(){};

class KernelCPU;
class KernelGPU;
class KernelACC;

class GoodWGSize;
class WrongReqWGSize;

Expand Down Expand Up @@ -40,13 +38,6 @@ int main() {
Compatible &= Dev.is_gpu();
Called = true;
}
if (sycl::is_compatible<KernelACC>(Dev)) {
Copy link
Contributor Author

@KornevNikita KornevNikita Mar 25, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Removing this since there is no chance now to launch this on fpga.

Q.submit(
[&](sycl::handler &h) { h.single_task<KernelACC>([=]() { baz(); }); });
Q.wait();
Compatible &= Dev.is_accelerator();
Called = true;
}

if (sycl::is_compatible<GoodWGSize>(Dev)) {
Q.submit([&](sycl::handler &h) {
Expand Down