Hi Jens,
I ran with your suggestion and DM now sees a ~7% improvement in hipri
bio polling with io_uring (using dm-linear on null_blk, IOPS went from
900K to 966K).
Christoph,
I tried to address your review of the previous set. Patch 1 and 2 can
obviously be folded but I left them split out for review purposes.
Feel free to see if these changes are meaningful for nvme's use.
Happy for either you to take on iterating on these block changes
further or you letting me know what changes you'd like made.
Thanks,
Mike
v2: add REQ_ALLOC_CACHE and move use of bio_alloc_percpu_cache to
bio_alloc_bioset
Mike Snitzer (4):
block: allow BIOSET_PERCPU_CACHE use from bio_alloc_clone
block: allow BIOSET_PERCPU_CACHE use from bio_alloc_bioset
dm: enable BIOSET_PERCPU_CACHE for dm_io bioset
dm: conditionally enable BIOSET_PERCPU_CACHE for bio-based dm_io bioset
block/bio.c | 67 +++++++++++++++++++++++++++++++----------------
block/blk.h | 7 -----
drivers/md/dm-table.c | 11 +++++---
drivers/md/dm.c | 10 +++----
drivers/md/dm.h | 4 +--
include/linux/bio.h | 9 +++++++
include/linux/blk_types.h | 4 ++-
7 files changed, 71 insertions(+), 41 deletions(-)
--
2.15.0
