3.16.51-rc1 review patch. If anyone has any objections, please let me know. ------------------ From: Tang Junhui <tang.junhui@xxxxxxxxxx> commit 69daf03adef5f7bc13e0ac86b4b8007df1767aab upstream. Since bypassed IOs use no bucket, so do not subtract sectors_to_gc to trigger gc thread. Signed-off-by: tang.junhui <tang.junhui@xxxxxxxxxx> Acked-by: Coly Li <colyli@xxxxxxx> Reviewed-by: Eric Wheeler <bcache@xxxxxxxxxxxxxxxxxx> Reviewed-by: Christoph Hellwig <hch@xxxxxx> Signed-off-by: Jens Axboe <axboe@xxxxxxxxx> [bwh: Backported to 3.16: adjust context] Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx> --- drivers/md/bcache/request.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) --- a/drivers/md/bcache/request.c +++ b/drivers/md/bcache/request.c @@ -193,14 +193,14 @@ static void bch_data_insert_start(struct struct data_insert_op *op = container_of(cl, struct data_insert_op, cl); struct bio *bio = op->bio, *n; + if (op->bypass) + return bch_data_invalidate(cl); + if (atomic_sub_return(bio_sectors(bio), &op->c->sectors_to_gc) < 0) { set_gc_sectors(op->c); wake_up_gc(op->c); } - if (op->bypass) - return bch_data_invalidate(cl); - /* * Journal writes are marked REQ_FLUSH; if the original write was a * flush, it'll wait on the journal write.