[llvm] AMDGPU: Select VGPR MFMAs by default (PR #159493)

Matt Arsenault via llvm-commits llvm-commits at lists.llvm.org
Sun Nov 16 14:42:09 PST 2025


================
@@ -37,7 +37,7 @@ static cl::opt<bool, true> MFMAVGPRFormOpt(
     "amdgpu-mfma-vgpr-form",
     cl::desc("Whether to force use VGPR for Opc and Dest of MFMA. If "
              "unspecified, default to compiler heuristics"),
-    cl::location(SIMachineFunctionInfo::MFMAVGPRForm), cl::init(false),
+    cl::location(SIMachineFunctionInfo::MFMAVGPRForm), cl::init(true),
----------------
arsenm wrote:

But the system always was try to use the VGPR form. Any time we were emitting AGPR MFMAs was essentially an edge case or bug 

https://github.com/llvm/llvm-project/pull/159493


More information about the llvm-commits mailing list