On Tue, 2017-05-02 at 16:41 +0200, Jan Kara wrote: > So I'm also not aware of any particular breakage this would cause. However > logically the freeing of request mempools really belongs to > blk_release_queue() so it seems a bit dumb to move blk_exit_rl() just > because SCSI stores the fact from which slab cache it has allocated the > sense buffer in a structure (shost) that it frees under its hands by the > time blk_release_queue() is called. :-| Hello Jan, My concern when I wrote my previous e-mail was that I didn't want to add a scsi_host_get() / scsi_host_put() pair to the hot path in the SCSI core. But I just realized that scsi_init_rq() and scsi_exit_rq() are not in the hot path so adding a scsi_host_get() / scsi_host_put() pair should work fine. I will post a patch. Bart.
