[PATCH] Implement aarch64 neon instruction class AdvSIMD (by element) - Clang

Tim Northover t.p.northover at gmail.com
Sun Sep 29 10:59:53 PDT 2013

  Hi Jiangning,

  Mostly looks reasonable. I've just got one question.

Comment at: utils/TableGen/NeonEmitter.cpp:1571-1572
@@ -1503,3 +1570,4 @@
   case OpMulLane:
+  case OpMulLaneQ:
     s += "__a * " + SplatLane(nElts, "__b", "__c") + ";";
Why are these distinct? As far as I can see they are treated in exactly the same way by TableGen. Or am I missing something not shown by a simple grep; some endsWith('Q') weirdness or something?


More information about the llvm-commits mailing list