[llvm] aebd338 - [AMDGPU] Fix user SGPR alloc order in docs (#119092)

[via llvm-commits]

Author: Austin Kerbow
Date: 2024-12-07T13:08:35-08:00
New Revision: aebd3389a9e694f7087d55e159186734d4559ca6

URL: https://github.com/llvm/llvm-project/commit/aebd3389a9e694f7087d55e159186734d4559ca6
DIFF: https://github.com/llvm/llvm-project/commit/aebd3389a9e694f7087d55e159186734d4559ca6.diff

LOG: [AMDGPU] Fix user SGPR alloc order in docs (#119092)

NFC. Preload kernarg SGPRs are allocated after the private segment size
SGPR. This patch updates AMDGPUUsage.rst to reflect this.

Added: 
    

Modified: 
    llvm/docs/AMDGPUUsage.rst

Removed: 
    


################################################################################
diff  --git a/llvm/docs/AMDGPUUsage.rst b/llvm/docs/AMDGPUUsage.rst
index 7dbfa8c085b91a..5c6034753eb4af 100644
--- a/llvm/docs/AMDGPUUsage.rst
+++ b/llvm/docs/AMDGPUUsage.rst
@@ -5756,9 +5756,6 @@ SGPR register initial state is defined in
      then       Flat Scratch Init          2      See
                 (enable_sgpr_flat_scratch         :ref:`amdgpu-amdhsa-kernel-prolog-flat-scratch`.
                 _init)
-     then       Preloaded Kernargs         N/A    See
-                (kernarg_preload_spec             :ref:`amdgpu-amdhsa-kernarg-preload`.
-                _length)
      then       Private Segment Size       1      The 32-bit byte size of a
                 (enable_sgpr_private              single work-item's memory
                 _segment_size)                    allocation. This is the
@@ -5779,6 +5776,9 @@ SGPR register initial state is defined in
                                                   may be needed for GFX9-GFX11 which
                                                   changes the meaning of the
                                                   Flat Scratch Init value.
+     then       Preloaded Kernargs         N/A    See
+                (kernarg_preload_spec             :ref:`amdgpu-amdhsa-kernarg-preload`.
+                _length)
      then       Work-Group Id X            1      32-bit work-group id in X
                 (enable_sgpr_workgroup_id         dimension of grid for
                 _X)                               wavefront.