On Fri, May 20, 2022 at 12:08:02PM +0100, Maciej W. Rozycki wrote:
> On Thu, 19 May 2022, Paul Cercueil wrote:
>
> > About the real issue: Maybe the last "addu" in csum_tcpudp_nofold() should be
> > a "daddu"?
>
> It doesn't matter whether there is ADDU or DADDU anymore, given that
> commit 198688edbf77 ("MIPS: Fix inline asm input/output type mismatch in
> checksum.h used with Clang") has added a redundant 32-bit sign-extension
> operation right afterwards, regressing size/performance by an extra
> instruction in this performance-critical code as a workaround for a stupid
> compiler.
Shouldn't this concern have been addressed by Tiezhu's follow up?
https://lore.kernel.org/1617359240-16609-1-git-send-email-yangtiezhu@xxxxxxxxxxx/
I guess it never got picked up?
Cheers,
Nathan
