[PATCH] D67572: [SVFS] The Search Vector Function System.

Francesco Petrogalli via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Mon Nov 4 13:40:24 PST 2019


fpetrogalli marked an inline comment as done.
fpetrogalli added inline comments.


================
Comment at: llvm/include/llvm/Analysis/VectorUtils.h:230
+    const VFShape TLIShape = {VF, false /*isScalable*/,
+                              VFISAKind::LLVM_INTERNAL_TLI, Parameters};
+    for (const auto &Info : Mappings)
----------------
The `VFISAKind::LLVM_INTERNAL_TLI` value, at the moment, is restricting the use of the SVFS to only those functions that are listed in the TLI. This makes me wonder whether it would be better to move the VFISAKind from the VFShape to the VFInfo. This could be justified under the assumption that a scalar function can be mapped to multiple vector functions with the same vectorization shape (same `FunctionType`),  just with a different underlying ISA.

Any thoughts?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D67572/new/

https://reviews.llvm.org/D67572





More information about the llvm-commits mailing list