[PATCH] D57735: [X86] Add FPCW as a register and start using it as an implicit use on floating point instructions.

[Craig Topper via Phabricator via llvm-commits]

craig.topper added a comment.

In D57735#1388651 <https://reviews.llvm.org/D57735#1388651>, @RKSimon wrote:

> LGTM but a PR40206 test case as well would be useful


Both PR40529 and PR40206 are caused by a multiply instruction being moved between the fldcw. They aren't really different. The case I put in for pr40529 is the same reduced test case that Dimitry put in both bugs.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D57735/new/

https://reviews.llvm.org/D57735