On Fri, Mar 08, 2024 at 06:19:02PM +0800, Herbert Xu wrote: > > > diff --git a/arch/loongarch/crypto/crc32-loongarch.c b/arch/loongarch/crypto/crc32-loongarch.c > > index a49e507af38c..3eebea3a7b47 100644 > > --- a/arch/loongarch/crypto/crc32-loongarch.c > > +++ b/arch/loongarch/crypto/crc32-loongarch.c > > @@ -44,7 +44,6 @@ static u32 crc32_loongarch_hw(u32 crc_, const u8 *p, unsigned int len) > > > > CRC32(crc, value, w); > > p += sizeof(u32); > > - len -= sizeof(u32); > > } > > This makes no sense whatsoever. Please review this patch carefully > before you resubmit. Nevermind, I see what's going on now. Your reference to the lack of a CRC instruction utterly confused me. The fact that len -= is unnecessary has nothing to do with whether there is a CRC instruction. Please modify your patch description so that it is not needlessly confusing. You should simply state that len -= is unnecessary because you only test whether the relevant bit is set in len for the tail case. Thanks, -- Email: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
