[PATCH] D121437: [AMDGPU] Add s_nop WaitStates between neighboring mfma

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp:44
+    cl::desc(
+        "Percent of wait states between neighboring MFMA to pad with s_nops"));
+
----------------
I'm not sure what a percentage means here


================
Comment at: llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp:270-271
+                  TSchedModel.getWriteProcResEnd(SC))) {
+    if (!strcmp(TSchedModel.getProcResource(PRE.ProcResourceIdx)->Name,
+                "HWXDL")) {
+      return PRE.Cycles;
----------------
Why does this need to parse a name?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D121437/new/

https://reviews.llvm.org/D121437