[PATCH] D85357: [mlir] [VectorOps] Add masked load/store operations to Vector dialect

[Aart Bik via Phabricator via llvm-commits]

aartbik marked an inline comment as done.
aartbik added a comment.

Thanks!



================
Comment at: mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp:174
+  base = rewriter.create<LLVM::BitcastOp>(loc, pType, base);
+  ptr = rewriter.create<LLVM::GEPOp>(loc, pType, base);
+  return success();
----------------
bkramer wrote:
> Does a GEP without indices actually do something?
It is a nop. It seemed a bit cleaner to make all memoperations use a GEPOp, but perhaps I should remove it for the non-index cases (and use the base directly). This part is subject to change anyway, since I plan to general this for the cases not handled above, and move this into core utilities perhaps.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D85357/new/

https://reviews.llvm.org/D85357