On Wed, Oct 16, 2019 at 09:07:21AM +1100, Dave Chinner wrote: > > +/* > > + * We implement an immediate ioend submission policy here to avoid needing to > > + * chain multiple ioends and hence nest mempool allocations which can violate > > + * forward progress guarantees we need to provide. The current ioend we are > > + * adding blocks to is cached on the writepage context, and if the new block > > adding pages to ... , and if the new block mapping So reviewing this comment I disagree with the change. We add on a per- block basis to the ioend, not a per-page one. Similar for the second one it is the block that needs to append (which is defined by the mapping, but still..).
