[PATCH] D51542: [X86] Remove wrong ReadAdvance from multiclass sse_fp_unop_s

[Andrea Di Biagio via Phabricator via llvm-commits]

andreadb created this revision.
andreadb added reviewers: RKSimon, courbet, craig.topper, lebedev.ri, mattd.
Herald added a subscriber: gbedwell.

A ReadAdvance was incorrectly added to the list of SchedReadWrite for the following opcodes:

  sqrtss
  sqrtsd
  rsqrtss
  rcpss

As a consequence, a wrong operand latency was computed for the register operand used by the folded load.

This patch removes the wrong ReadAdvance, and updates the llvm-mca test cases. Now the llvm-mca timeline report shows correct timings for those unary fp SSE1/SSE2 instructions.

Please let me know if okay to commit.

Thanks,
-Andrea


https://reviews.llvm.org/D51542

Files:
  lib/Target/X86/X86InstrSSE.td
  test/tools/llvm-mca/X86/sqrt-rsqrt-rcp-memop.s

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D51542.163533.patch
Type: text/x-patch
Size: 13485 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180831/27a02e9c/attachment.bin>