[PATCH] D110259: [SVE][Analysis] Tune the cost model according to the tune-cpu attribute

[David Sherwood via Phabricator via llvm-commits]

david-arm added inline comments.


================
Comment at: llvm/lib/Target/AArch64/AArch64Subtarget.cpp:153
   case NeoverseN1:
+    PrefFunctionLogAlignment = 4;
+    break;
----------------
sdesmalen wrote:
> nit: seems unrelated?
I had to split this out as a separate case to avoid the fallthrough into the NeoverseN2 case that's all.


================
Comment at: llvm/lib/Target/AArch64/AArch64Subtarget.h:282
   unsigned MaxSVEVectorSizeInBits;
+  unsigned VScaleForTuning = 2;
 
----------------
sdesmalen wrote:
> nit: This may be personal preference, but would `VScaleForCPU` be a more suitable name?
I don't mind really. I don't have a strong preference. I guess the reason I used VScaleForTuning was just to make it clear that this is not an indication of what the exact vscale will be at runtime, since it could be anything from 1 -> max vscale for the CPU. Rather it's just an indication of how we'd like to tune the vectoriser, optimisations and codegen.

However, happy to change it if you prefer it!


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D110259/new/

https://reviews.llvm.org/D110259