3.16.51-rc1 review patch. If anyone has any objections, please let me know.
------------------ From: Tang Junhui <[email protected]> commit c81ffa32a214c84b08900fbc9d432187bd948eba upstream. Sequential write IOs were tested with bs=1M by FIO in writeback cache mode, these IOs were expected to be bypassed, but actually they did not. We debug the code, and find in check_should_bypass(): if (!congested && mode == CACHE_MODE_WRITEBACK && op_is_write(bio_op(bio)) && (bio-＞bi_opf & REQ_SYNC)) goto rescale that means, If in writeback mode, a write IO with REQ_SYNC flag will not be bypassed though it is a sequential large IO, It's not a correct thing to do actually, so this patch remove these codes. Signed-off-by: tang.junhui <[email protected]> Reviewed-by: Kent Overstreet <[email protected]> Reviewed-by: Eric Wheeler <[email protected]> Signed-off-by: Jens Axboe <[email protected]> [bwh: Backported to 3.16: deleted code is slightly different] Signed-off-by: Ben Hutchings <[email protected]> --- drivers/md/bcache/request.c | 6 ------ 1 file changed, 6 deletions(-) --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -395,12 +395,6 @@ static bool check_should_bypass(struct c if (!congested && !dc->sequential_cutoff) goto rescale; - if (!congested && - mode == CACHE_MODE_WRITEBACK && - (bio->bi_rw & REQ_WRITE) && - (bio->bi_rw & REQ_SYNC)) - goto rescale; - spin_lock(&dc->io_lock); hlist_for_each_entry(i, iohash(dc, bio->bi_iter.bi_sector), hash)
