[llvm] [MCA] New option -scheduling-info (PR #130574)

[Min-Yih Hsu via llvm-commits]

================
@@ -141,6 +243,34 @@ void InstructionInfoView::collectData(
     IIVDEntry.mayLoad = MCDesc.mayLoad();
     IIVDEntry.mayStore = MCDesc.mayStore();
     IIVDEntry.hasUnmodeledSideEffects = MCDesc.hasUnmodeledSideEffects();
+
+    if (PrintFullInfo) {
+      // Get latency with bypass
+      IIVDEntry.Bypass =
+          IIVDEntry.Latency - MCSchedModel::getBypassDelayCycles(STI, SCDesc);
+      IIVDEntry.OpcodeName = (std::string)MCII.getName(Inst.getOpcode());
+      raw_string_ostream TempStream(IIVDEntry.Resources);
+      const MCWriteProcResEntry *Index = STI.getWriteProcResBegin(&SCDesc);
+      const MCWriteProcResEntry *Last = STI.getWriteProcResEnd(&SCDesc);
+      auto sep = "";
+      for (; Index != Last; ++Index) {
+        if (!Index->ReleaseAtCycle)
+          continue;
+        const MCProcResourceDesc *MCProc =
+            SM.getProcResource(Index->ProcResourceIdx);
+        if (Index->ReleaseAtCycle > 1) {
+          // Output ReleaseAtCycle between [] if not 1 (default)
+          // This is to be able to evaluate throughput.
+          // See getReciprocalThroughput in MCSchedule.cpp
+          TempStream << sep
+                     << format("%s[%d]", MCProc->Name, Index->ReleaseAtCycle);
+        } else {
+          TempStream << sep << format("%s", MCProc->Name);
+        }
+        sep = ",";
+      }
+      TempStream.flush();
----------------
mshockwave wrote:

`raw_string_ostream`, unlike `std::stringstream`, is recommended not to flush it manually.

https://github.com/llvm/llvm-project/pull/130574