On Fri, Nov 13, 2020 at 04:13:37PM -0800, Linus Torvalds wrote: > On Fri, Nov 13, 2020 at 3:38 PM Darrick J. Wong <djwong@xxxxxxxxxx> wrote: > > > > Since the hack is unnecessary and causes thread race errors, just get > > rid of it completely. Pushing this kind of vfs change midway through a > > cycle makes me nervous, but a large enough number of the usual > > VFS/ext4/XFS/btrfs suspects have said this looks good and solves a real > > problem vector, so I'm sending this for your consideration instead of > > holding off until 5.11. > > Not a fan of the timing, but you make a good argument, and I love > seeing code removal. So I took it. Thanks! Admittedly this is super late because it didn't occur to me until after -rc1 that the periodic hangs I saw in for-next were related to lockdep being broken and weren't some other weird vfs/xfs breakage; and then I wanted to spin this patch through my internal testing systems for a week to convince myself that changing the freeze locking code wasn't totally nuts. :) --D > And once I took the real code change, the two cleanups looked like the > least of the problem, so I took them too. > > I ended up doing it all just as a single pull, since it seemed > pointless to make history more complicated just to separate out the > cleanups in a separate pull. > > Now I really hope this won't cause any problems, but it certainly > _looks_ harmless. > > Linus >
