Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1344365

[PATCH v2 2/4] block: check virt boundary in bio_will_gap()

From Ming Lei <ming.lei@canonical.com>
Newsgroups linux.kernel
Subject [PATCH v2 2/4] block: check virt boundary in bio_will_gap()
Date 2016-02-26 16:50 +0100
Message-ID <r6sOv-5pp-49@gated-at.bofh.it> (permalink)
References <r6sOt-5pp-9@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


In the following patch, the way for figuring out
the last bvec will be changed with a bit cost introduced,
so return immediately if the queue doesn't have virt
boundary limit. Actually most of devices have not
this limit.

Reviewed-by: Sagi Grimberg <sagig@mellanox.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
---
 include/linux/blkdev.h | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 4571ef1..cd06a41 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1372,6 +1372,13 @@ static inline void put_dev_sector(Sector p)
 	page_cache_release(p.v);
 }
 
+static inline bool __bvec_gap_to_prev(struct request_queue *q,
+				struct bio_vec *bprv, unsigned int offset)
+{
+	return offset ||
+		((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q));
+}
+
 /*
  * Check if adding a bio_vec after bprv with offset would create a gap in
  * the SG list. Most drivers don't care about this, but some do.
@@ -1381,18 +1388,17 @@ static inline bool bvec_gap_to_prev(struct request_queue *q,
 {
 	if (!queue_virt_boundary(q))
 		return false;
-	return offset ||
-		((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q));
+	return __bvec_gap_to_prev(q, bprv, offset);
 }
 
 static inline bool bio_will_gap(struct request_queue *q, struct bio *prev,
 			 struct bio *next)
 {
-	if (!bio_has_data(prev))
+	if (!bio_has_data(prev) || !queue_virt_boundary(q))
 		return false;
 
-	return bvec_gap_to_prev(q, &prev->bi_io_vec[prev->bi_vcnt - 1],
-				next->bi_io_vec[0].bv_offset);
+	return __bvec_gap_to_prev(q, &prev->bi_io_vec[prev->bi_vcnt - 1],
+				  next->bi_io_vec[0].bv_offset);
 }
 
 static inline bool req_gap_back_merge(struct request *req, struct bio *bio)
-- 
1.9.1

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH v2 0/4] block: fix bio_will_gap() Ming Lei <ming.lei@canonical.com> - 2016-02-26 16:50 +0100
  [PATCH v2 4/4] block: merge: get the 1st and last bvec via helpers Ming Lei <ming.lei@canonical.com> - 2016-02-26 16:50 +0100
  [PATCH v2 2/4] block: check virt boundary in bio_will_gap() Ming Lei <ming.lei@canonical.com> - 2016-02-26 16:50 +0100
  Re: [PATCH v2 0/4] block: fix bio_will_gap() Sagi Grimberg <sagig@dev.mellanox.co.il> - 2016-02-28 11:00 +0100

csiph-web