Hi all, Here's a quick patchset reworking writeback ioend completion processing into per-inode completion queues so that we don't have a thundering herd of unwritten/cow completion kworkers contending for the ILOCK that then tie up log reservation space and cause deadlocks. The third patch combines adjacent ioends when possible to reduce the overhead further. If you're going to start using this mess, you probably ought to just pull from my git trees, which are linked below. This is an extraordinary way to destroy everything. Enjoy! Comments and questions are, as always, welcome. --D kernel git tree: https://git.kernel.org/cgit/linux/kernel/git/djwong/xfs-linux.git/log/?h=merged-completions
