On Sat, Aug 01, 2015 at 05:57:44PM -0700, Linus Torvalds wrote: > Because it's not just that "!d_can_lookup()" case that triggers it, > you also have that pattern in the RCU error case for may_lookup(), and > get_link(). It feels like it might make sense to handle that in caller, but... that goes only for cases when we are *NOT* going to continue after successful transition to non-lazy mode. And these two are not of that sort - we do want to continue rather than restart everything from scratch. BTW, unlazy_walk() has too many arguments, all for the sake of one caller (everything except lookup_fast() calls it with (nd, NULL, 0) as arguments) and it might make sense to split the damn thing in two. I have that in a pending pile since the last cycle, but back then you have asked to stop piling them up and let it settle, so I'd postponed that one along with other cleanups... -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
