On Mon, 11 Feb, at 05:30:57PM, Al Viro wrote: > On Mon, Feb 11, 2013 at 04:05:57PM +0000, Matt Fleming wrote: > > > Egads, man... > > > [snip the horror with copying the name] > > > > > > unsigned long hash = init_name_hash(); > > > const unsigned char *s = qstr->name; > > > int len = qstr->len; > > > > > > if (!efivarfs_valid_name(s, len)) > > > return -EINVAL; > > > while (len-- > GUID_LEN) > > > hash = partial_name_hash(*s++, hash); > > > /* GUID is case-insensitive. */ > > > while (len--) > > > hash = partial_name_hash(tolower(*s++), hash); > > > return end_name_hash(hash); > > > > Oh, wow, yes that is much nicer. I didn't realise we could build the > > hash incrementally like that. Very cool. > > ... except that it should end with qstr->hash = end_name_hash(hash); return 0; > instead. My apologies. Oh, efivarfs also needs to stop using d_alloc_name(), which calls full_name_hash(), because we now need to use our fs-specific hash function. Updated versions of both patches coming shortly. -- Matt Fleming, Intel Open Source Technology Center -- To unsubscribe from this list: send the line "unsubscribe linux-efi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
