[llvm] [NVPTX] Customize getScalarizationOverhead (PR #128077)

[Justin Fargnoli via llvm-commits]

================
@@ -1,59 +1,123 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=nvptx64-nvidia-cuda -mcpu=sm_70 | FileCheck %s
-; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=nvptx64-nvidia-cuda -mcpu=sm_40 | FileCheck %s -check-prefix=NOVECTOR
+; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=nvptx64-nvidia-cuda -mcpu=sm_90 | FileCheck %s -check-prefixes=VECTOR,SM90
+; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=nvptx64-nvidia-cuda -mcpu=sm_80 | FileCheck %s -check-prefixes=VECTOR,SM80
+; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=nvptx64-nvidia-cuda -mcpu=sm_70 | FileCheck %s -check-prefixes=VECTOR,SM70
+; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=nvptx64-nvidia-cuda -mcpu=sm_50 | FileCheck %s -check-prefixes=NOVECTOR,SM50
----------------
justinfargnoli wrote:

```suggestion
; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=nvptx64-nvidia-cuda -mcpu=sm_90 | FileCheck %s -check-prefix=VECTOR
; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=nvptx64-nvidia-cuda -mcpu=sm_80 | FileCheck %s -check-prefix=VECTOR
; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=nvptx64-nvidia-cuda -mcpu=sm_70 | FileCheck %s -check-prefix=VECTOR
; RUN: opt < %s -passes=slp-vectorizer -S -mtriple=nvptx64-nvidia-cuda -mcpu=sm_50 | FileCheck %s -check-prefix=NOVECTOR
```

https://github.com/llvm/llvm-project/pull/128077