On Tue, 2009-02-10 at 17:43 +0000, James Bottomley wrote:
> There's a current barrier to this: osd_initiator has also become a
> consumer of blk_rq_append_bio().
>
> It seems to be emulating block internals, so I think the fix is twofold:
>
> 1. adjust blk_rq_map_kern to call blk_rq_append_bio() instead of
> blk_rq_prep_bio() (with an extra failure path).
> 2. make osd_initiator simply call it for additions.
>
> I can code up a patch to see if it works.
So this is the patch to allow blk_rq_map_kern to append to requests,
which is what I think is needed.
Unfortunately unwinding osd_initator's bio dependence and putting it
back on data buffers looks to be a bit of a longer chore.
James
---
diff --git a/block/blk-map.c b/block/blk-map.c
index 25d15ff..1a1e26d 100644
--- a/block/blk-map.c
+++ b/block/blk-map.c
@@ -289,6 +289,7 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf,
int reading = rq_data_dir(rq) == READ;
int do_copy = 0;
struct bio *bio;
+ int ret;
if (len > (q->max_hw_sectors << 9))
return -EINVAL;
@@ -310,7 +311,13 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf,
if (do_copy)
rq->cmd_flags |= REQ_COPY_USER;
- blk_rq_bio_prep(q, rq, bio);
+ ret = blk_rq_append_bio(q, rq, bio);
+ if (unlikely(ret)) {
+ /* request is too big */
+ bio_put(bio);
+ return ret;
+ }
+
blk_queue_bounce(q, &rq->bio);
rq->buffer = rq->data = NULL;
return 0;
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
