[PATCH] D92296: [AARCH64] Improve accumulator forwarding for Cortex-A57 model

[Usman Nadeem via Phabricator via llvm-commits]

mnadeem created this revision.
mnadeem added reviewers: MatzeB, fhahn, javed.absar.
Herald added subscribers: llvm-commits, danielkiss, hiraditya, kristof.beyls.
Herald added a project: LLVM.
mnadeem requested review of this revision.

Added some accumulate op to accumulate op, and non-accumulate op
(e.g. mul, shift) to accumulate op forwarding according to the
Cortex A57 Software Optimization Guide.

Old model only had mla->mla forwarding.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D92296

Files:
  llvm/lib/Target/AArch64/AArch64SchedA57.td
  llvm/lib/Target/AArch64/AArch64SchedA57WriteRes.td
  llvm/test/CodeGen/AArch64/aarch64-misched-forwarding-A57.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D92296.308262.patch
Type: text/x-patch
Size: 14421 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20201130/e0719519/attachment.bin>