drbd: announce FLUSH/FUA capability to upper layers

In 8.4, we may have bios spanning two activity log extents.
Fixup drbd_al_begin_io() and drbd_al_complete_io() to deal with zero sized bios.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>

4 files changed, 26 insertions, 10 deletions

diff --git a/drivers/block/drbd/drbd_actlog.c b/drivers/block/drbd/drbd_actlog.c
index 209b2e063b92..e81085795ec4 100644
--- a/drivers/block/drbd/drbd_actlog.c
+++ b/drivers/block/drbd/drbd_actlog.c
@@ -248,11 +248,12 @@ void drbd_al_begin_io(struct drbd_conf *mdev, struct drbd_interval *i)
         /* for bios crossing activity log extent boundaries,
          * we may need to activate two extents in one go */
         unsigned first = i->sector >> (AL_EXTENT_SHIFT-9);
-        unsigned last = (i->sector + (i->size >> 9) - 1) >> (AL_EXTENT_SHIFT-9);
+        unsigned last = i->size == 0 ? first : (i->sector + (i->size >> 9) - 1) >> (AL_EXTENT_SHIFT-9);
         unsigned enr;
         bool locked = false;
 
 
+        D_ASSERT(first <= last);
         D_ASSERT(atomic_read(&mdev->local_cnt) > 0);
 
         for (enr = first; enr <= last; enr++)
@@ -305,11 +306,12 @@ void drbd_al_complete_io(struct drbd_conf *mdev, struct drbd_interval *i)
         /* for bios crossing activity log extent boundaries,
          * we may need to activate two extents in one go */
         unsigned first = i->sector >> (AL_EXTENT_SHIFT-9);
-        unsigned last = (i->sector + (i->size >> 9) - 1) >> (AL_EXTENT_SHIFT-9);
+        unsigned last = i->size == 0 ? first : (i->sector + (i->size >> 9) - 1) >> (AL_EXTENT_SHIFT-9);
         unsigned enr;
         struct lc_element *extent;
         unsigned long flags;
 
+        D_ASSERT(first <= last);
         spin_lock_irqsave(&mdev->al_lock, flags);
 
         for (enr = first; enr <= last; enr++) {
@@ -756,7 +758,11 @@ int __drbd_set_out_of_sync(struct drbd_conf *mdev, sector_t sector, int size,
         unsigned int enr, count = 0;
         struct lc_element *e;
 
-        if (size <= 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_BIO_SIZE) {
+        /* this should be an empty REQ_FLUSH */
+        if (size == 0)
+                return 0;
+
+        if (size < 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_BIO_SIZE) {
                 dev_err(DEV, "sector: %llus, size: %d\n",
                         (unsigned long long)sector, size);
                 return 0;
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index f2af74d06860..85d95ec405ef 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -2640,6 +2640,7 @@ enum drbd_ret_code conn_new_minor(struct drbd_tconn *tconn, unsigned int minor,
         q->backing_dev_info.congested_data = mdev;
 
         blk_queue_make_request(q, drbd_make_request);
+        blk_queue_flush(q, REQ_FLUSH | REQ_FUA);
         /* Setting the max_hw_sectors to an odd value of 8kibyte here
            This triggers a max_bio_size message upon first attach or connect */
         blk_queue_max_hw_sectors(q, DRBD_MAX_BIO_SIZE_SAFE >> 8);
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 8fddec96dfbe..a6f0b2938363 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -295,6 +295,9 @@ static void drbd_free_pages(struct drbd_conf *mdev, struct page *page, int is_ne
         atomic_t *a = is_net ? &mdev->pp_in_use_by_net : &mdev->pp_in_use;
         int i;
 
+        if (page == NULL)
+                return;
+
         if (drbd_pp_vacant > (DRBD_MAX_BIO_SIZE/PAGE_SIZE) * minor_count)
                 i = page_chain_free(page);
         else {
@@ -331,7 +334,7 @@ drbd_alloc_peer_req(struct drbd_conf *mdev, u64 id, sector_t sector,
                     unsigned int data_size, gfp_t gfp_mask) __must_hold(local)
 {
         struct drbd_peer_request *peer_req;
-        struct page *page;
+        struct page *page = NULL;
         unsigned nr_pages = (data_size + PAGE_SIZE -1) >> PAGE_SHIFT;
 
         if (drbd_insert_fault(mdev, DRBD_FAULT_AL_EE))
@@ -344,9 +347,11 @@ drbd_alloc_peer_req(struct drbd_conf *mdev, u64 id, sector_t sector,
                 return NULL;
         }
 
-        page = drbd_alloc_pages(mdev, nr_pages, (gfp_mask & __GFP_WAIT));
-        if (!page)
-                goto fail;
+        if (data_size) {
+                page = drbd_alloc_pages(mdev, nr_pages, (gfp_mask & __GFP_WAIT));
+                if (!page)
+                        goto fail;
+        }
 
         drbd_clear_interval(&peer_req->i);
         peer_req->i.size = data_size;
@@ -1513,8 +1518,6 @@ read_in_block(struct drbd_conf *mdev, u64 id, sector_t sector,
                 data_size -= dgs;
         }
 
-        if (!expect(data_size != 0))
-                return NULL;
         if (!expect(IS_ALIGNED(data_size, 512)))
                 return NULL;
         if (!expect(data_size <= DRBD_MAX_BIO_SIZE))
@@ -1537,6 +1540,9 @@ read_in_block(struct drbd_conf *mdev, u64 id, sector_t sector,
         if (!peer_req)
                 return NULL;
 
+        if (!data_size)
+                return peer_req;
+
         ds = data_size;
         page = peer_req->pages;
         page_chain_for_each(page) {
@@ -2199,6 +2205,10 @@ static int receive_Data(struct drbd_tconn *tconn, struct packet_info *pi)
 
         dp_flags = be32_to_cpu(p->dp_flags);
         rw |= wire_flags_to_bio(mdev, dp_flags);
+        if (peer_req->pages == NULL) {
+                D_ASSERT(peer_req->i.size == 0);
+                D_ASSERT(dp_flags & DP_FLUSH);
+        }
 
         if (dp_flags & DP_MAY_SET_IN_SYNC)
                 peer_req->flags |= EE_MAY_SET_IN_SYNC;
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index e215dce4c694..8323449fbbab 100644
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -1097,7 +1097,6 @@ int drbd_make_request(struct request_queue *q, struct bio *bio)
         /*
          * what we "blindly" assume:
          */
-        D_ASSERT(bio->bi_size > 0);
         D_ASSERT(IS_ALIGNED(bio->bi_size, 512));
 
         inc_ap_bio(mdev);