Raid0 can't accept I/O requests that span a chunk boundary, as each
chunk goes to a different underlying queue.
The appended patch registers a `mergeable' function with the device
queue, to prevent requests being merged if the merged request would
span a chunk boundary.
I'm assuming that the most restrictive queueing parameters are those
of the raid0 code itself; if that isn't the case, then extra code has
to be added to reflect the most limiting case from the underlying
devices up into the raid0 device's queue.
If raid0 is on top of SCSI or ATA discs, or loopback-mounted
files, the assumption will almost always hold.
diff -Nur --exclude=SCCS --exclude=BitKeeper --exclude=ChangeSet linux-2.5-import/drivers/md/raid0.c linux-2.5-mdfix/drivers/md/raid0.c
--- linux-2.5-import/drivers/md/raid0.c Mon Oct 14 10:32:35 2002
+++ linux-2.5-mdfix/drivers/md/raid0.c Mon Oct 14 11:46:53 2002
@@ -162,6 +162,29 @@
return 1;
}
+/**
+ * raid0_mergeable_bvec -- tell bio layer if a two requests can be merged
+ * @q: request queue
+ * @bio: the buffer head that's been built up so far
+ * @biovec: the request that could be merged to it.
+ *
+ * Return 1 if the merge is not permitted (because the
+ * result would cross a chunk boundary), 0 otherwise.
+ */
+static int raid0_mergeable_bvec(request_queue_t *q, struct bio *bio, struct bio_vec *biovec)
+{
+ mddev_t *mddev = q->queuedata;
+ sector_t block;
+ unsigned int chunk_size;
+ unsigned int bio_sz;
+
+ chunk_size = mddev->chunk_size >> 10;
+ block = bio->bi_sector >> 1;
+ bio_sz = (bio->bi_size + biovec->bv_len) >> 10;
+
+ return chunk_size < ((block & (chunk_size - 1)) + bio_sz);
+}
+
static int raid0_run (mddev_t *mddev)
{
unsigned cur=0, i=0, nb_zone;
@@ -233,6 +256,8 @@
conf->hash_table[i++].zone1 = conf->strip_zone + cur;
size -= (conf->smallest->size - zone0_size);
}
+ blk_queue_max_sectors(&mddev->queue, mddev->chunk_size >> 9);
+ blk_queue_merge_bvec(&mddev->queue, raid0_mergeable_bvec);
return 0;
out_free_zone_conf:
@@ -262,13 +287,6 @@
return 0;
}
-/*
- * FIXME - We assume some things here :
- * - requested buffers NEVER bigger than chunk size,
- * - requested buffers NEVER cross stripes limits.
- * Of course, those facts may not be valid anymore (and surely won't...)
- * Hey guys, there's some work out there ;-)
- */
static int raid0_make_request (request_queue_t *q, struct bio *bio)
{
mddev_t *mddev = q->queuedata;
@@ -291,8 +309,8 @@
hash = conf->hash_table + x;
}
- /* Sanity check */
- if (chunk_size < (block & (chunk_size - 1)) + (bio->bi_size >> 10))
+ /* Sanity check -- queue functions should prevent this happening */
+ if (unlikely(chunk_size < (block & (chunk_size - 1)) + (bio->bi_size >> 10)))
goto bad_map;
if (!hash)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue Oct 15 2002 - 22:00:48 EST