majianpeng [off-list ref] writes:

On 2012-07-27 22:21 Jeff Moyer [off-list ref] Wrote:

quoted

majianpeng [off-list ref] writes:

quoted

When exec bio_alloc, the bi_rw is zero.But after calling bio_add_page,
it will use bi_rw.
Fox example, in functiion __bio_add_page,it will call merge_bvec_fn().
The merge_bvec_fn of raid456 will use the bi_rw to judge the merge.

quoted

quoted

if ((bvm->bi_rw & 1) == WRITE)
	return biovec->bv_len; /* always allow writes to be mergeable */

Signed-off-by: Jianpeng Ma <redacted>

Good catch.  How did you find this?  Did you experience data corruption
as a result of this oversight, reduced performance due to missed merge
opportunities, or did you just notice it in reviewing the code?

Reviewed-by: Jeff Moyer <redacted>

Sorry for late to reply. When i analysed the performance of raid5, i found this bug.

OK, thanks.  In the future, it would be good to include that information
in the patch description.

Cheers,
Jeff