Hi Jens, Linus, Vivek, Jerome,
Do you remember following problem?
I try to fix a original problem, which is a wrong accounting of
hd_struct->in_flight, by completely rewriting a patch.
Here are patches for fixing the original problem.
[PATCH 1/2] Don't merge different partition's IOs
[PATCH 1/2] Change a argument of disk_map_sector_rcu() to
hd_part->start_sect from req->__sector
For more details, please refer to the header of each patch.
Any comments or questions are welcome.