[Mlir-commits] [mlir] [mlir][Target] Make nvptxcompiler passed options high priority to keep consistent with ptxas behavior (PR #121036)

[llvmlistbot at llvm.org]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-mlir-llvm

Author: Zichen Lu (MikaOvO)

<details>
<summary>Changes</summary>

When using `ptxas` to do ptx->cubin, the options passed (via `cmd`) have higher priority than the gpuModule target (such as `opt-level`). When using `nvptxcompiler`, it is the opposite and we need to be consistent.

---
Full diff: https://github.com/llvm/llvm-project/pull/121036.diff


1 Files Affected:

- (modified) mlir/lib/Target/LLVM/NVVM/Target.cpp (+2-1) 


``````````diff
diff --git a/mlir/lib/Target/LLVM/NVVM/Target.cpp b/mlir/lib/Target/LLVM/NVVM/Target.cpp
index bca26e3a0e84a9..0f630f56af301e 100644
--- a/mlir/lib/Target/LLVM/NVVM/Target.cpp
+++ b/mlir/lib/Target/LLVM/NVVM/Target.cpp
@@ -484,7 +484,8 @@ NVPTXSerializer::compileToBinaryNVPTX(const std::string &ptxCode) {
   std::string optLevel = std::to_string(this->optLevel);
   std::pair<llvm::BumpPtrAllocator, SmallVector<const char *>> cmdOpts =
       targetOptions.tokenizeCmdOptions();
-  cmdOpts.second.append(
+  cmdOpts.second.insert(
+      cmdOpts.second.begin(),
       {"-arch", getTarget().getChip().data(), "--opt-level", optLevel.c_str()});
 
   // Create the compiler handle.

``````````

</details>


https://github.com/llvm/llvm-project/pull/121036