On 2022/05/16 8:36, Christoph Hellwig wrote:
> submit_bio uses some extremely convoluted checks and confusing comments
> to only account REQ_OP_READ/REQ_OP_WRITE comments. Just switch to the
> plain obvious checks instead.
>
> Signed-off-by: Christoph Hellwig <hch@xxxxxx>
> ---
> block/blk-core.c | 18 +++++-------------
> 1 file changed, 5 insertions(+), 13 deletions(-)
>
> diff --git a/block/blk-core.c b/block/blk-core.c
> index ee18b6a699bdf..48a58c24d452e 100644
> --- a/block/blk-core.c
> +++ b/block/blk-core.c
> @@ -893,19 +893,11 @@ void submit_bio(struct bio *bio)
> if (blkcg_punt_bio_submit(bio))
> return;
>
> - /*
> - * If it's a regular read/write or a barrier with data attached,
> - * go through the normal accounting stuff before submission.
> - */
> - if (bio_has_data(bio)) {
> - unsigned int count = bio_sectors(bio);
> -
> - if (op_is_write(bio_op(bio))) {
> - count_vm_events(PGPGOUT, count);
> - } else {
> - task_io_account_read(bio->bi_iter.bi_size);
> - count_vm_events(PGPGIN, count);
> - }
> + if (bio_op(bio) == REQ_OP_READ) {
> + task_io_account_read(bio->bi_iter.bi_size);
> + count_vm_events(PGPGIN, bio_sectors(bio));
> + } else if (bio_op(bio) == WRITE) {
I know it is the same value, but for consistency, wouldn't it be better to use
REQ_OP_WRITE here ?
> + count_vm_events(PGPGOUT, bio_sectors(bio));
> }
>
> /*
--
Damien Le Moal
Western Digital Research
