Udit Kumar Agarwal

LIT tests failures: Failed Tests (2): SYCL :: basic_tests/types.cpp SYCL :: basic_tests/vectors/constexpr-constructor.cpp

``` sycl/vector_preview.hpp:992:60: error: requested alignment is not an integer constant [2024-05-16T07:17:19.790Z] alignas(std::min((size_t)64, sizeof(DataType))) DataType m_Data; ```

~TODO: Add value_type and element_type in sycl::vec https://github.com/intel/llvm/issues/12954~

> I'm observing a similar problem with [Basic/submit_time.cpp](https://github.com/intel/llvm/blob/sycl/sycl/test-e2e/Basic/submit_time.cpp) on linux/CL. I've found you need a bit of system load and a lot of runs to reproduce but it's consistently do-able...

> > Overall, the PR LGTM. I can approve the changes once the corresponding UR PR gets merged and we accordingly update the UR tag. > > @uditagarwal97 please rest...

@JackAKirk Could this be related to https://github.com/intel/llvm/pull/13752?

@asudarsa The AMD runner went rogue, so we had to move the AMD testing to postcommit: https://github.com/intel/llvm/pull/14977 If you want to trigger post-commit testing (including HIP), you could create a...

> @asudarsa @uditagarwal97 hi guys, I guess we can close this one? Yes, it's not really a bug.

FYI, while enabling E2E tests on PVC (https://github.com/intel/llvm/pull/14720), I see a failure in `Matrix/SG32/element_wise_all_ops.cpp`, which could be related to PI2UR work. Similar to the failure in `Matrix/element_wise_all_ops.cpp` (https://github.com/intel/llvm/issues/14795).

For post-commit testing, your branch `sycl-devops-pr/...` has to be in intel/llvm. Like: https://github.com/intel/llvm/tree/sycl-devops-pr/udit/pvc_linux_runner Using a fork of intel/llvm won't work.