blk-throttle: Fix calculation of max number of WRITES to be dispatched

o Currently we try to dispatch more READS and less WRITES (75%, 25%) in one
  dispatch round. ummy pointed out that there is a bug in max_nr_writes
  calculation. This patch fixes it.

Reported-by: ummy y <yummylln@yahoo.com.cn>
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
This commit is contained in:
Vivek Goyal 2010-11-15 19:32:42 +01:00 committed by Jens Axboe
parent d69b78ba1d
commit c2f6805d47

View File

@ -645,7 +645,7 @@ static int throtl_dispatch_tg(struct throtl_data *td, struct throtl_grp *tg,
{
unsigned int nr_reads = 0, nr_writes = 0;
unsigned int max_nr_reads = throtl_grp_quantum*3/4;
unsigned int max_nr_writes = throtl_grp_quantum - nr_reads;
unsigned int max_nr_writes = throtl_grp_quantum - max_nr_reads;
struct bio *bio;
/* Try to dispatch 75% READS and 25% WRITES */