caio/clang-p2996
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[AMDGPU] Replace dynamic VGPR feature with attribute (#133444)

Browse Source 
Use a function attribute (amdgpu-dynamic-vgpr) instead of a subtarget
feature, as requested in #130030.
This commit is contained in:
Diana Picus
2025-06-24 11:09:36 +02:00
committed by GitHub
parent 6cfa03f1f1
commit a201f8872a
34 changed files with 1205 additions and 199 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
llvm/docs/AMDGPUUsage.rst
View File

@@ -768,11 +768,6 @@ For example:
performant than code generated for XNACK replay
disabled.
dynamic-vgpr TODO Represents the "Dynamic VGPR" hardware mode, introduced in GFX12.
Waves launched in this mode may allocate or deallocate the VGPRs
using dedicated instructions, but may not send the DEALLOC_VGPRS
message.
=============== ============================ ==================================================
.. _amdgpu-target-id:
@@ -1764,6 +1759,15 @@ The AMDGPU backend supports the following LLVM IR attributes.
"amdgpu-promote-alloca-to-vector-vgpr-ratio" Ratio of VGPRs to budget for promoting alloca to vectors.
"amdgpu-dynamic-vgpr-block-size" Represents the size of a VGPR block in the "Dynamic VGPR" hardware mode,
introduced in GFX12.
A value of 0 (default) means that dynamic VGPRs are not enabled.
Valid values for GFX12+ are 16 and 32.
Waves launched in this mode may allocate or deallocate the VGPRs
using dedicated instructions, but may not send the DEALLOC_VGPRS
message. If a shader has this attribute, then all its callees must
match its value.
================================================ ==========================================================
Calling Conventions