Hi Linus,
Turns out that the fix for not hammering on the delayed work timer too
much caused a performance regression for BFQ, so let's revert the change
for now. I've got some ideas on how to fix it appropriately, but they
should wait for 5.17.
Please pull!
The following changes since commit aa97f6cdb7e92909e17c8ca63e622fcb81d57a57:
bcache: fix NULL pointer reference in cached_dev_detach_finish (2021-12-14 20:32:54 -0700)
are available in the Git repository at:
git://git.kernel.dk/linux-block.git tags/block-5.16-2021-12-19
for you to fetch changes up to 87959fa16cfbcf76245c11559db1940069621274:
Revert "block: reduce kblockd_mod_delayed_work_on() CPU consumption" (2021-12-19 07:58:44 -0700)
----------------------------------------------------------------
block-5.16-2021-12-19
----------------------------------------------------------------
Jens Axboe (1):
Revert "block: reduce kblockd_mod_delayed_work_on() CPU consumption"
block/blk-core.c | 2 --
1 file changed, 2 deletions(-)
--
Jens Axboe
