On 2/14/21 3:57 PM, Al Viro wrote: > On Sun, Feb 14, 2021 at 09:45:39AM -0700, Jens Axboe wrote: > >>>> +out3: >>>> + nd->depth = 0; // as we hadn't gotten to legitimize_links() >>>> out2: >>>> nd->path.mnt = NULL; >>>> out1: >>> >>> Alternatively, we could use the fact that legitimize_links() is not >>> called anywhere other than these two places and have LOOKUP_CACHED >>> checked there. As in >> >> Both fix the issue for me, just tested them. The second one seems >> cleaner to me, would probably be nice to have a comment on that in >> either the two callers or at least in legitimize_links() though. > > Hmm... Do you have anything on top of that branch? If you do, there's > no way to avoid leaving bisect hazard; if not, I'd rather fold a fix > into the broken commit... I do, that's basically the base of that series, -rc6 + that branch. So I'd prefer if you just apply the fixup, which I do think is pretty low risk even if it is a potential bisection pain point. But not really a huge one, in this case. That said, if you do want to rebase it, I can rebase mine. That's not the end of the world either. -- Jens Axboe
