Messages in this thread | | | Date | Fri, 13 Feb 2015 11:27:46 +0100 | From | Peter Zijlstra <> | Subject | Re: RAID1 might_sleep() warning on 3.19-rc7 |
| |
On Fri, Feb 13, 2015 at 07:49:53PM +1100, NeilBrown wrote: > > Like said, that will still recursive call delayacct_blkio_*() and would > > increase nr_iowait for a second time; while arguably its still the same > > one io-wait instance. > > No it doesn't. There is no "blk_flush_plug" call between the > delayacct_blkio_*() calls.
Duh, clearly I needed to still wake up :/
> I've moved blk_flush_plug to the beginning of the function.
> I wondered if it really make sense to call blk_flush_plug with nr_iowait > elevated and delayacct_blkio active. blk_flush_plug() could call schedule() > for non-"io" reasons and maybe that could upset stuff???
Yeah, good question that. Lemme ponder that a bit.
> I don't really know. I'm happy with your version. I don't suppose anyone > else is paying attention and could give a third opinion....
:-)
| |