[PATCH] D122709: [RISCV] Enable cross basic block aware writevxrm insertion

[ShihPo Hung via Phabricator via llvm-commits]

arcbbb added a comment.

In D122709#3419837 <https://reviews.llvm.org/D122709#3419837>, @kito-cheng wrote:

> Hmm, I think that made the dynamic rounding mode become unexpected/random rounding mode, give following example, the behavior `vaddd` in `foo` is unpredictably, `vaddd` in foo will operated with current rounding mode if the bar is inlined, but it will become `rnu` once bar inlined, I really think we should restore that for those static rounding mode stuffs.
>
>   foo:
>     call bar
>     vaddd
>   
>   bar:
>     vaadd.rm rnu

The summary mentioned that this assumed that `vxrm` is call-clobbered,
so in this case after the bar() return , `vxrm` should be set explicitly before the next vaadd with dynamic round mode.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D122709/new/

https://reviews.llvm.org/D122709