md: deal with merge_bvec_fn in component devices better.

commit 627a2d3c29427637f4c5d31ccc7fcbd8d312cd71 upstream.

If a component device has a merge_bvec_fn then as we never call it
we must ensure we never need to.  Currently this is done by setting
max_sector to 1 PAGE, however this does not stop a bio being created
with several sub-page iovecs that would violate the merge_bvec_fn.

So instead set max_phys_segments to 1 and set the segment boundary to the
same as a page boundary to ensure there is only ever one single-page
segment of IO requested at a time.

This can particularly be an issue when 'xen' is used as it is
known to submit multiple small buffers in a single bio.

Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

4 files changed, 43 insertions, 30 deletions

diff --git a/drivers/md/linear.c b/drivers/md/linear.c
index 00435bd20699..001317b50034 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -172,12 +172,14 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
                 disk_stack_limits(mddev->gendisk, rdev->bdev,
                                   rdev->data_offset << 9);
                 /* as we don't honour merge_bvec_fn, we must never risk
-                 * violating it, so limit ->max_sector to one PAGE, as
-                 * a one page request is never in violation.
+                 * violating it, so limit max_phys_segments to 1 lying within
+                 * a single page.
                  */
-                if (rdev->bdev->bd_disk->queue->merge_bvec_fn &&
-                    queue_max_sectors(mddev->queue) > (PAGE_SIZE>>9))
-                        blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9);
+                if (rdev->bdev->bd_disk->queue->merge_bvec_fn) {
+                        blk_queue_max_phys_segments(mddev->queue, 1);
+                        blk_queue_segment_boundary(mddev->queue,
+                                                   PAGE_CACHE_SIZE - 1);
+                }
 
                 conf->array_sectors += rdev->sectors;
                 cnt++;
diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c
index 32a662fc55c9..f9ee99f9a94c 100644
--- a/drivers/md/multipath.c
+++ b/drivers/md/multipath.c
@@ -301,14 +301,16 @@ static int multipath_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
                                           rdev->data_offset << 9);
 
                 /* as we don't honour merge_bvec_fn, we must never risk
-                 * violating it, so limit ->max_sector to one PAGE, as
-                 * a one page request is never in violation.
+                 * violating it, so limit ->max_phys_segments to one, lying
+                 * within a single page.
                  * (Note: it is very unlikely that a device with
                  * merge_bvec_fn will be involved in multipath.)
                  */
-                        if (q->merge_bvec_fn &&
-                            queue_max_sectors(q) > (PAGE_SIZE>>9))
-                                blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9);
+                        if (q->merge_bvec_fn) {
+                                blk_queue_max_phys_segments(mddev->queue, 1);
+                                blk_queue_segment_boundary(mddev->queue,
+                                                           PAGE_CACHE_SIZE - 1);
+                        }
 
                         conf->working_disks++;
                         mddev->degraded--;
@@ -476,9 +478,11 @@ static int multipath_run (mddev_t *mddev)
                 /* as we don't honour merge_bvec_fn, we must never risk
                  * violating it, not that we ever expect a device with
                  * a merge_bvec_fn to be involved in multipath */
-                if (rdev->bdev->bd_disk->queue->merge_bvec_fn &&
-                    queue_max_sectors(mddev->queue) > (PAGE_SIZE>>9))
-                        blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9);
+                if (rdev->bdev->bd_disk->queue->merge_bvec_fn) {
+                        blk_queue_max_phys_segments(mddev->queue, 1);
+                        blk_queue_segment_boundary(mddev->queue,
+                                                   PAGE_CACHE_SIZE - 1);
+                }
 
                 if (!test_bit(Faulty, &rdev->flags))
                         conf->working_disks++;
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index 77605cdceaf1..41ee9deed250 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -176,14 +176,15 @@ static int create_strip_zones(mddev_t *mddev)
                 disk_stack_limits(mddev->gendisk, rdev1->bdev,
                                   rdev1->data_offset << 9);
                 /* as we don't honour merge_bvec_fn, we must never risk
-                 * violating it, so limit ->max_sector to one PAGE, as
-                 * a one page request is never in violation.
+                 * violating it, so limit ->max_phys_segments to 1, lying within
+                 * a single page.
                  */
 
-                if (rdev1->bdev->bd_disk->queue->merge_bvec_fn &&
-                    queue_max_sectors(mddev->queue) > (PAGE_SIZE>>9))
-                        blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9);
-
+                if (rdev1->bdev->bd_disk->queue->merge_bvec_fn) {
+                        blk_queue_max_phys_segments(mddev->queue, 1);
+                        blk_queue_segment_boundary(mddev->queue,
+                                                   PAGE_CACHE_SIZE - 1);
+                }
                 if (!smallest || (rdev1->sectors < smallest->sectors))
                         smallest = rdev1;
                 cnt++;
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index d119b7b75e71..047c468e3d28 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -1155,13 +1155,17 @@ static int raid10_add_disk(mddev_t *mddev, mdk_rdev_t *rdev)
 
                         disk_stack_limits(mddev->gendisk, rdev->bdev,
                                           rdev->data_offset << 9);
-                        /* as we don't honour merge_bvec_fn, we must never risk
-                         * violating it, so limit ->max_sector to one PAGE, as
-                         * a one page request is never in violation.
+                        /* as we don't honour merge_bvec_fn, we must
+                         * never risk violating it, so limit
+                         * ->max_phys_segments to one lying with a single
+                         * page, as a one page request is never in
+                         * violation.
                          */
-                        if (rdev->bdev->bd_disk->queue->merge_bvec_fn &&
-                            queue_max_sectors(mddev->queue) > (PAGE_SIZE>>9))
-                                blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9);
+                        if (rdev->bdev->bd_disk->queue->merge_bvec_fn) {
+                                blk_queue_max_phys_segments(mddev->queue, 1);
+                                blk_queue_segment_boundary(mddev->queue,
+                                                           PAGE_CACHE_SIZE - 1);
+                        }
 
                         p->head_position = 0;
                         rdev->raid_disk = mirror;
@@ -2255,12 +2259,14 @@ static int run(mddev_t *mddev)
                 disk_stack_limits(mddev->gendisk, rdev->bdev,
                                   rdev->data_offset << 9);
                 /* as we don't honour merge_bvec_fn, we must never risk
-                 * violating it, so limit ->max_sector to one PAGE, as
-                 * a one page request is never in violation.
+                 * violating it, so limit max_phys_segments to 1 lying
+                 * within a single page.
                  */
-                if (rdev->bdev->bd_disk->queue->merge_bvec_fn &&
-                    queue_max_sectors(mddev->queue) > (PAGE_SIZE>>9))
-                        blk_queue_max_sectors(mddev->queue, PAGE_SIZE>>9);
+                if (rdev->bdev->bd_disk->queue->merge_bvec_fn) {
+                        blk_queue_max_phys_segments(mddev->queue, 1);
+                        blk_queue_segment_boundary(mddev->queue,
+                                                   PAGE_CACHE_SIZE - 1);
+                }
 
                 disk->head_position = 0;
         }