[llvm-branch-commits] [llvm] 72f1f08 - [NFC][AMDGPU] Clarify memory model support for volatile

Tony Tye via llvm-branch-commits llvm-branch-commits at lists.llvm.org
Mon Jan 11 12:05:12 PST 2021


Author: Tony Tye
Date: 2021-01-11T19:59:55Z
New Revision: 72f1f084670ed66362ecb7c4079ee372665ce842

URL: https://github.com/llvm/llvm-project/commit/72f1f084670ed66362ecb7c4079ee372665ce842
DIFF: https://github.com/llvm/llvm-project/commit/72f1f084670ed66362ecb7c4079ee372665ce842.diff

LOG: [NFC][AMDGPU] Clarify memory model support for volatile

Reorder the AMDGPUUage description of the memory model code sequences
for volatile so clear that it applies independent of the nontemporal
setting.

Differential Revision: https://reviews.llvm.org/D94358

Added: 
    

Modified: 
    llvm/docs/AMDGPUUsage.rst

Removed: 
    


################################################################################
diff  --git a/llvm/docs/AMDGPUUsage.rst b/llvm/docs/AMDGPUUsage.rst
index 81734dd30cd2..f661684d9b74 100644
--- a/llvm/docs/AMDGPUUsage.rst
+++ b/llvm/docs/AMDGPUUsage.rst
@@ -4732,7 +4732,12 @@ in table :ref:`amdgpu-amdhsa-memory-model-code-sequences-gfx6-gfx9-table`.
                                               - generic
                                               - private    1. buffer/global/flat_load
                                               - constant
-                                                         - volatile & !nontemporal
+                                                         - !volatile & nontemporal
+
+                                                           1. buffer/global/flat_load
+                                                              glc=1 slc=1
+
+                                                         - volatile
 
                                                            1. buffer/global/flat_load
                                                               glc=1
@@ -4750,17 +4755,17 @@ in table :ref:`amdgpu-amdhsa-memory-model-code-sequences-gfx6-gfx9-table`.
                                                               be reordered by
                                                               hardware.
 
-                                                         - !volatile & nontemporal
-
-                                                           1. buffer/global/flat_load
-                                                              glc=1 slc=1
-
      load         *none*       *none*         - local    1. ds_load
      store        *none*       *none*         - global   - !volatile & !nontemporal
                                               - generic
                                               - private    1. buffer/global/flat_store
                                               - constant
-                                                         - volatile & !nontemporal
+                                                         - !volatile & nontemporal
+
+                                                           1. buffer/global/flat_store
+                                                              glc=1 slc=1
+
+                                                         - volatile
 
                                                            1. buffer/global/flat_store
                                                            2. s_waitcnt vmcnt(0)
@@ -4777,11 +4782,6 @@ in table :ref:`amdgpu-amdhsa-memory-model-code-sequences-gfx6-gfx9-table`.
                                                               be reordered by
                                                               hardware.
 
-                                                         - !volatile & nontemporal
-
-                                                           1. buffer/global/flat_store
-                                                              glc=1 slc=1
-
      store        *none*       *none*         - local    1. ds_store
      **Unordered Atomic**
      ------------------------------------------------------------------------------------
@@ -6034,7 +6034,12 @@ table :ref:`amdgpu-amdhsa-memory-model-code-sequences-gfx10-table`.
                                               - generic
                                               - private    1. buffer/global/flat_load
                                               - constant
-                                                         - volatile & !nontemporal
+                                                         - !volatile & nontemporal
+
+                                                           1. buffer/global/flat_load
+                                                              slc=1
+
+                                                         - volatile
 
                                                            1. buffer/global/flat_load
                                                               glc=1 dlc=1
@@ -6052,17 +6057,17 @@ table :ref:`amdgpu-amdhsa-memory-model-code-sequences-gfx10-table`.
                                                               be reordered by
                                                               hardware.
 
-                                                         - !volatile & nontemporal
-
-                                                           1. buffer/global/flat_load
-                                                              slc=1
-
      load         *none*       *none*         - local    1. ds_load
      store        *none*       *none*         - global   - !volatile & !nontemporal
                                               - generic
                                               - private    1. buffer/global/flat_store
                                               - constant
-                                                         - volatile & !nontemporal
+                                                         - !volatile & nontemporal
+
+                                                            1. buffer/global/flat_store
+                                                               slc=1
+
+                                                         - volatile
 
                                                            1. buffer/global/flat_store
                                                            2. s_waitcnt vscnt(0)
@@ -6079,11 +6084,6 @@ table :ref:`amdgpu-amdhsa-memory-model-code-sequences-gfx10-table`.
                                                               be reordered by
                                                               hardware.
 
-                                                         - !volatile & nontemporal
-
-                                                            1. buffer/global/flat_store
-                                                               slc=1
-
      store        *none*       *none*         - local    1. ds_store
      **Unordered Atomic**
      ------------------------------------------------------------------------------------


        


More information about the llvm-branch-commits mailing list