wangpc-pp wrote:

> > Hi @asb, @topperc and @preames, I'm a bit hesitant to take this fix into 
> > the release branch because it seems a bit large for what is essentially the 
> > last release of the 21.x branch. After this there will be no more releases, 
> > so no chances to fix any issues that may arise. The fact that one was 
> > already found when backporting the fix makes me even more worried. As the 
> > owners of the RISCV backend, do any of you have any thoughts on how risky 
> > it would be to take this fix in 21.x or should we just wait for the 22.x 
> > branch for the fix?
> 
> @lukel97 @wangpc-pp Is the change to RISCVInstrInfoVPseudos.td enough to fix 
> the bug? Was the RISCVVectorPeephole.cpp part for correctness or to prevent 
> regressions?

But I think the regression can be large. :-(

https://github.com/llvm/llvm-project/pull/170604
_______________________________________________
llvm-branch-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits

Reply via email to