[PATCH] D90661: [NFC][AMDGPU] Minor editorial improvements to AMDGPUUsage.rst
Tony Tye via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Mon Nov 2 18:28:58 PST 2020
t-tye created this revision.
t-tye added a reviewer: scott.linder.
Herald added subscribers: llvm-commits, kerbowa, jfb, tpr, dstuttard, yaxunl, nhaehnle, jvesely, kzhuravl.
Herald added a project: LLVM.
t-tye requested review of this revision.
Herald added a subscriber: wdng.
Repository:
rG LLVM Github Monorepo
https://reviews.llvm.org/D90661
Files:
llvm/docs/AMDGPUUsage.rst
Index: llvm/docs/AMDGPUUsage.rst
===================================================================
--- llvm/docs/AMDGPUUsage.rst
+++ llvm/docs/AMDGPUUsage.rst
@@ -4306,7 +4306,7 @@
Scalar memory operations are only used to access memory that is proven to not
change during the execution of the kernel dispatch. This includes constant
-address space and global address space for program scope const variables.
+address space and global address space for program scope ``const`` variables.
Therefore, the kernel machine code does not have to maintain the scalar cache to
ensure it is coherent with the vector caches. The scalar and vector caches are
invalidated between kernel dispatches by CP since constant address space data
@@ -4380,8 +4380,7 @@
------------------------------------------------------------------------------------
load atomic unordered *any* *any* *Same as non-atomic*.
store atomic unordered *any* *any* *Same as non-atomic*.
- atomicrmw unordered *any* *any* *Same as monotonic
- atomic*.
+ atomicrmw unordered *any* *any* *Same as monotonic atomic*.
**Monotonic Atomic**
------------------------------------------------------------------------------------
load atomic monotonic - singlethread - global 1. buffer/global/ds/flat_load
@@ -5559,7 +5558,7 @@
Scalar memory operations are only used to access memory that is proven to not
change during the execution of the kernel dispatch. This includes constant
-address space and global address space for program scope const variables.
+address space and global address space for program scope ``const`` variables.
Therefore, the kernel machine code does not have to maintain the scalar cache to
ensure it is coherent with the vector caches. The scalar and vector caches are
invalidated between kernel dispatches by CP since constant address space data
@@ -5653,8 +5652,7 @@
------------------------------------------------------------------------------------
load atomic unordered *any* *any* *Same as non-atomic*.
store atomic unordered *any* *any* *Same as non-atomic*.
- atomicrmw unordered *any* *any* *Same as monotonic
- atomic*.
+ atomicrmw unordered *any* *any* *Same as monotonic atomic*.
**Monotonic Atomic**
------------------------------------------------------------------------------------
load atomic monotonic - singlethread - global 1. buffer/global/flat_load
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D90661.302445.patch
Type: text/x-patch
Size: 2724 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20201103/c067ad22/attachment.bin>
More information about the llvm-commits
mailing list