-
Notifications
You must be signed in to change notification settings - Fork 130
[SYCL][ESIMD] Introduce test to validate thread ID API #1644
Conversation
Complementary compiler PR: intel/llvm#8569 |
SYCL/ESIMD/thread_id_test.cpp
Outdated
sycl::nd_range<1> Range(GlobalRange, LocalRange); | ||
|
||
{ | ||
auto e = queue.submit([&](sycl::handler &cgh) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
minor comment: e is unused
auto e = queue.submit([&](sycl::handler &cgh) { | |
queue.submit([&](sycl::handler &cgh) { |
shared_vector<int32_t> vector_output_thread_id(SIZE, -1, allocator); | ||
shared_vector<int32_t> vector_output_subdevice_id(SIZE, -1, allocator); | ||
|
||
auto GlobalRange = sycl::range<1>(SIZE); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Minor comment: The test has great mix of 2 naming styles: 'GlobalRange' and 'vector_output_subdevice_id_ptr'
It is a common problem for tests/sycl/esimd, but for new tests it would be nice to choose one style.
SYCL/ESIMD/thread_id_test.cpp
Outdated
// UNSUPPORTED: cuda || hip | ||
// RUN: %clangxx -fsycl %s -o %t.out | ||
// RUN: %GPU_RUN_PLACEHOLDER %t.out | ||
// XFAIL: gpu |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Does it really fail with esimd-emulator?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@fineg74 - This test probably passes with esimd_emulator (also coded as 'gpu'), but it is supposed to fail with this XFAIL.
/verify with intel/llvm#8569 |
…e-topic CMPLRTST-19973: skip run matrix_amxint8* on non-avx512 machine.
No description provided.