[PATCH] D118229: [AMDGPUHSAMetadataStreamer] Do not assume ABI alignment for pointers

Wed Feb 2 08:57:16 PST 2022

arsenm added a comment.

In D118229#3290747 <https://reviews.llvm.org/D118229#3290747>, @yaxunl wrote:

> In D118229#3290659 <https://reviews.llvm.org/D118229#3290659>, @arsenm wrote:
>
>> In D118229#3290658 <https://reviews.llvm.org/D118229#3290658>, @kzhuravl wrote:
>>
>>> This change introduces the regression in OpenCL conformance test: basic - kernel_memory_alignment_local. Does it require any corresponding runtime changes?
>>
>> Is clang correctly emitting the align attribute on all these arguments?
>
> clang does not do anything special for alignment of pointer type kernel arg. It assumes the pointee alignment is default 1. https://godbolt.org/z/xs195rKoW
>
> Question is: What OpenCL spec says about kernel arg pointee alignment? @b-sumner @Anastasia

It should be the ABI type alignment as was used before

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D118229/new/

https://reviews.llvm.org/D118229