[PATCH] D128346: [AMDGPU] Report minimum scratch size in code object v5 and later by default

Matt Arsenault via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Mon Sep 26 08:04:26 PDT 2022


arsenm added inline comments.


================
Comment at: llvm/docs/AMDGPUUsage.rst:3557-3560
+:ref:`amdgpu-amdhsa-code-object-kernel-argument-metadata-map-table-v5` and in
+the scratch size reporting where only the minimum requirement is reported (the
+runtime is expected adjust this if the amdhsa_uses_dynamic_stack bit is set in
+the kernel descriptor).
----------------
abinavpp wrote:
> arsenm wrote:
> > This sentence doesn't read right at all. I don't know why this would be mentioned here and not just in the table with the description of the field
> In amdgpu-amdhsa-kernel-descriptor-v3-table? We could do that; It's under the section "Code Object V3 Kernel Descriptor", can we add V5+ details there?
Can mention the poor support for unknown sized stacks 


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D128346/new/

https://reviews.llvm.org/D128346



More information about the llvm-commits mailing list