Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On 04/28/2015 04:43 AM, Christoph Hellwig wrote:
> This seems to lack support for QUEUE_FLAG_SG_GAPS to work around
> the retarded PPR format in the NVMe driver. I'm actually not
> sure we have a way to reproduce them for BLOCK_PC requests, but I think
> we should make sure to properly handle them.
How about below incremental patch?
block/bio.c | 8 --------
block/blk-merge.c | 8 ++++++++
2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/block/bio.c b/block/bio.c
index ae31cdb..3f6bd9a 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -743,14 +743,6 @@ int bio_add_pc_page(struct request_queue *q, struct bio *bio, struct page
bio->bi_iter.bi_size += len;
goto done;
}
-
- /*
- * If the queue doesn't support SG gaps and adding this
- * offset would create a gap, disallow it.
- */
- if (q->queue_flags & (1 << QUEUE_FLAG_SG_GAPS) &&
- bvec_gap_to_prev(prev, offset))
- return 0;
}
if (bio->bi_vcnt >= bio->bi_max_vecs)
diff --git a/block/blk-merge.c b/block/blk-merge.c
index 9d565a0..32f6d6c 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -78,6 +78,14 @@ static struct bio *blk_bio_segment_split(struct request_queue *q,
if (sectors > queue_max_sectors(q))
goto split;
+ /*
+ * If the queue doesn't support SG gaps and adding this
+ * offset would create a gap, disallow it.
+ */
+ if (q->queue_flags & (1 << QUEUE_FLAG_SG_GAPS) &&
+ prev && bvec_gap_to_prev(&bvprv, bv.bv_offset))
+ goto split;
+
if (prev && blk_queue_cluster(q)) {
if (seg_size + bv.bv_len > queue_max_segment_size(q))
goto new_segment;