[SYCL] Remove fallback assertions

[steffenlarsen]

This commit removes the fallback assertion implementation for SYCL kernels. The expected behavior after this is that backends that do not support native asserts, as reported through the ext_oneapi_native_assert aspect, will ignore assertions in kernel code.

This commit also makes the L0 adapter report support for native assert, as it was incorrectly reporting false before.

[npmiller]

I believe these two tests also rely on the fallback asserts:

• https://github.com/intel/llvm/blob/sycl/sycl/test-e2e/DiscardEvents/discard_events_using_assert.cpp
• https://github.com/intel/llvm/blob/sycl/sycl/test-e2e/DiscardEvents/discard_events_using_assert_ndebug.cpp

Or at least they used to, I'm not sure the current checks represent that intent, and they never had the macros defined. But you can see the comment for the cuda/hip unsupported. And fallback is really the only case that would make sense to test for discard events with assert since it adds extra commands that might add/use events. I've been meaning to dig further into them to re-enable them for CUDA and HIP, but if we're dropping fallback we might be able to just delete these eventually.

[gmlueck]

What is our plan for the assert feature once the environment variable is removed? Will assert statements simply be silently ignored if they appear in a kernel that is submitted to a device without native support? If that is the case, I think we need an update to the sycl_ext_oneapi_assert specification.

Or, will assert be treated like other device-optional features, and the implementation will throw an exception if the device does not support it? In this case, we would also need an update to sycl_ext_oneapi_assert.

[jbrodman]

I believe these two tests also rely on the fallback asserts:

• https://github.com/intel/llvm/blob/sycl/sycl/test-e2e/DiscardEvents/discard_events_using_assert.cpp
• https://github.com/intel/llvm/blob/sycl/sycl/test-e2e/DiscardEvents/discard_events_using_assert_ndebug.cpp

Or at least they used to, I'm not sure the current checks represent that intent, and they never had the macros defined. But you can see the comment for the cuda/hip unsupported. And fallback is really the only case that would make sense to test for discard events with assert since it adds extra commands that might add/use events. I've been meaning to dig further into them to re-enable them for CUDA and HIP, but if we're dropping fallback we might be able to just delete these eventually.

We'd ultimately like to get rid of DiscardEvents too...

[steffenlarsen]

What is our plan for the assert feature once the environment variable is removed? Will assert statements simply be silently ignored if they appear in a kernel that is submitted to a device without native support? If that is the case, I think we need an update to the sycl_ext_oneapi_assert specification.

This seems like the safest solution to me, as it means users can still run their kernels on devices where they would run before, even if we can't do the asserts in there for them. I've updated the extension.

[aelovikov-intel]

That should go before line 1531, I think.

[steffenlarsen]

Maybe we could, but line 1531 seems to do a quick check of a cast of the pointer, so it is a little inconsequential if we do this before or after. I think the benefit of having it here is if there is some path where the decompression fails and changes Img to nullptr.

[bader]

Do we have a tracker to remove this warning in the future?

[steffenlarsen]

Sure! CMPLRLLVM-70220

[jbrodman]

Please get your reviews in so we can finally merge this.

[steffenlarsen]

Ping @intel/dpcpp-doc-reviewers & @intel/dpcpp-esimd-reviewers & @intel/unified-runtime-reviewers-level-zero

[sarnex]

esimd lgtm!

[steffenlarsen]

@EuphoricThinking / @nrspruit - Would you mind having a look at this, pretty please?

[kswiecicki]

If the only change from L0 adapter side is misreporting native assert support fix, then LGTM.

[github-actions]

@intel/llvm-gatekeepers please consider merging