-
Notifications
You must be signed in to change notification settings - Fork 14.3k
[NVPTX] Fix ptxas
failures (NFC)
#125147
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[NVPTX] Fix ptxas
failures (NFC)
#125147
Conversation
@llvm/pr-subscribers-backend-nvptx Author: Justin Fargnoli (justinfargnoli) ChangesNote regarding lower-args.ll and kernel-param-align.ll: Full diff: https://github.com/llvm/llvm-project/pull/125147.diff 3 Files Affected:
diff --git a/llvm/test/CodeGen/NVPTX/griddepcontrol.ll b/llvm/test/CodeGen/NVPTX/griddepcontrol.ll
index fe15b3fe4afbd9..0bf9196aa2902f 100644
--- a/llvm/test/CodeGen/NVPTX/griddepcontrol.ll
+++ b/llvm/test/CodeGen/NVPTX/griddepcontrol.ll
@@ -1,6 +1,6 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
; RUN: llc < %s -mcpu=sm_90 -march=nvptx64 | FileCheck %s
-; RUN: %if ptxas-11.8 %{ llc < %s -mcpu=sm_90 -march=nvptx64 | %ptxas-verify %}
+; RUN: %if ptxas-11.8 %{ llc < %s -mcpu=sm_90 -march=nvptx64 | %ptxas-verify -arch=sm_90 %}
define void @griddepcontrol() {
; CHECK-LABEL: griddepcontrol(
diff --git a/llvm/test/CodeGen/NVPTX/kernel-param-align.ll b/llvm/test/CodeGen/NVPTX/kernel-param-align.ll
index 2889d2d89a8579..a56b85de801430 100644
--- a/llvm/test/CodeGen/NVPTX/kernel-param-align.ll
+++ b/llvm/test/CodeGen/NVPTX/kernel-param-align.ll
@@ -1,5 +1,5 @@
; RUN: llc < %s -mtriple=nvptx64 -mcpu=sm_60 | FileCheck %s
-; RUN: %if ptxas %{ llc < %s -mtriple=nvptx64 -mcpu=sm_60 | %ptxas-verify %}
+; RUN: %if ptxas %{ llc < %s -mtriple=nvptx64 -mcpu=sm_60 | %ptxas -arch=sm_60 - %}
%struct.Large = type { [16 x double] }
diff --git a/llvm/test/CodeGen/NVPTX/lower-args.ll b/llvm/test/CodeGen/NVPTX/lower-args.ll
index 269bba75dc5fb3..925497748348c5 100644
--- a/llvm/test/CodeGen/NVPTX/lower-args.ll
+++ b/llvm/test/CodeGen/NVPTX/lower-args.ll
@@ -1,8 +1,8 @@
; RUN: opt < %s -S -nvptx-lower-args --mtriple nvptx64-nvidia-cuda | FileCheck %s --check-prefixes COMMON,IR,IRC
; RUN: opt < %s -S -nvptx-lower-args --mtriple nvptx64-nvidia-nvcl | FileCheck %s --check-prefixes COMMON,IR,IRO
-; RUN: llc < %s -mcpu=sm_20 --mtriple nvptx64-nvidia-cuda | FileCheck %s --check-prefixes COMMON,PTX,PTXC
-; RUN: llc < %s -mcpu=sm_20 --mtriple nvptx64-nvidia-nvcl| FileCheck %s --check-prefixes COMMON,PTX,PTXO
-; RUN: %if ptxas %{ llc < %s -mcpu=sm_20 | %ptxas-verify %}
+; RUN: llc < %s -mcpu=sm_52 --mtriple nvptx64-nvidia-cuda | FileCheck %s --check-prefixes COMMON,PTX,PTXC
+; RUN: llc < %s -mcpu=sm_52 --mtriple nvptx64-nvidia-nvcl| FileCheck %s --check-prefixes COMMON,PTX,PTXO
+; RUN: %if ptxas %{ llc < %s -mcpu=sm_52 | %ptxas -arch=sm_52 - %}
target datalayout = "e-i64:64-i128:128-v16:16-v32:32-n16:32:64"
target triple = "nvptx64-nvidia-cuda"
@@ -145,4 +145,6 @@ define ptx_kernel void @ptr_as_int_aggr(ptr nocapture noundef readonly byval(%st
; Function Attrs: convergent nounwind
-declare dso_local ptr @escape(ptr) local_unnamed_addr
+define dso_local ptr @escape(ptr) local_unnamed_addr {
+ ret ptr %0
+}
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Seems fine to me.
From a quick search I don't see any other uses of standalone %ptxas
, always %ptxas-verify
. I don't expect we would know if the two tests ever worked with ptxas-verify
? I'd be curious about what changed if so.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the fix! LGTM!
Note: lower-args.ll and kernel-param-align.ll fail because
ptxas
doesn't support constant pointers in separate complication mode (-c
).