block: Honor BDRV_REQ_FUA during write_zeroes

The block layer has a couple of cases where it can lose
Force Unit Access semantics when writing a large block of
zeroes, such that the request returns before the zeroes
have been guaranteed to land on underlying media.

SCSI does not support FUA during WRITESAME(10/16); FUA is only
supported if it falls back to WRITE(10/16).  But where the
underlying device is new enough to not need a fallback, it
means that any upper layer request with FUA semantics was
silently ignoring BDRV_REQ_FUA.

Conversely, NBD has situations where it can support FUA but not
ZERO_WRITE; when that happens, the generic block layer fallback
to bdrv_driver_pwritev() (or the older bdrv_co_writev() in qemu
2.6) was losing the FUA flag.

The problem of losing flags unrelated to ZERO_WRITE has been
latent in bdrv_co_do_write_zeroes() since commit aa7bfbff, but
back then, it did not matter because there was no FUA flag.  It
became observable when commit 93f5e6d8 paved the way for flags
that can impact correctness, when we should have been using
bdrv_co_writev_flags() with modified flags.  Compare to commit
9eeb6dd, which got flag manipulation right in
bdrv_co_do_zero_pwritev().

Symptoms: I tested with qemu-io with default writethrough cache
(which is supposed to use FUA semantics on every write), and
targetted an NBD client connected to a server that intentionally
did not advertise NBD_FLAG_SEND_FUA.  When doing 'write 0 512',
the NBD client sent two operations (NBD_CMD_WRITE then
NBD_CMD_FLUSH) to get the fallback FUA semantics; but when doing
'write -z 0 512', the NBD client sent only NBD_CMD_WRITE.

The fix is do to a cleanup bdrv_co_flush() at the end of the
operation if any step in the middle relied on a BDS that does
not natively support FUA for that step (note that we don't
need to flush after every operation, if the operation is broken
into chunks based on bounce-buffer sizing).  Each BDS gains a
new flag .supported_zero_flags, which parallels the use of
.supported_write_flags but only when accessing a zero write
operation (the flags MUST be different, because of SCSI having
different semantics based on WRITE vs. WRITESAME; and also
because BDRV_REQ_MAY_UNMAP only makes sense on zero writes).

Also fix some documentation to describe -ENOTSUP semantics,
particularly since iscsi depends on those semantics.

Down the road, we may want to add a driver where its
.bdrv_co_pwritev() honors all three of BDRV_REQ_FUA,
BDRV_REQ_ZERO_WRITE, and BDRV_REQ_MAY_UNMAP, and advertise
this via bs->supported_write_flags for blocks opened by that
driver; such a driver should NOT supply .bdrv_co_write_zeroes
nor .supported_zero_flags.  But none of the drivers touched
in this patch want to do that (the act of writing zeroes is
different enough from normal writes to deserve a second
callback).

Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

1 files changed, 25 insertions, 3 deletions

diff --git a/block/io.c b/block/io.c
index 1fb7afeaf0..cd6d71a503 100644
--- a/block/io.c
+++ b/block/io.c
@@ -652,7 +652,8 @@ int bdrv_write_zeroes(BlockDriverState *bs, int64_t sector_num,
  * Completely zero out a block device with the help of bdrv_write_zeroes.
  * The operation is sped up by checking the block status and only writing
  * zeroes to the device if they currently do not return zeroes. Optional
- * flags are passed through to bdrv_write_zeroes (e.g. BDRV_REQ_MAY_UNMAP).
+ * flags are passed through to bdrv_write_zeroes (e.g. BDRV_REQ_MAY_UNMAP,
+ * BDRV_REQ_FUA).
  *
  * Returns < 0 on error, 0 on success. For error codes see bdrv_write().
  */
@@ -1160,6 +1161,7 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
     QEMUIOVector qiov;
     struct iovec iov = {0};
     int ret = 0;
+    bool need_flush = false;
 
     int max_write_zeroes = MIN_NON_ZERO(bs->bl.max_write_zeroes,
                                         BDRV_REQUEST_MAX_SECTORS);
@@ -1192,13 +1194,29 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
         ret = -ENOTSUP;
         /* First try the efficient write zeroes operation */
         if (drv->bdrv_co_write_zeroes) {
-            ret = drv->bdrv_co_write_zeroes(bs, sector_num, num, flags);
+            ret = drv->bdrv_co_write_zeroes(bs, sector_num, num,
+                                            flags & bs->supported_zero_flags);
+            if (ret != -ENOTSUP && (flags & BDRV_REQ_FUA) &&
+                !(bs->supported_zero_flags & BDRV_REQ_FUA)) {
+                need_flush = true;
+            }
+        } else {
+            assert(!bs->supported_zero_flags);
         }
 
         if (ret == -ENOTSUP) {
             /* Fall back to bounce buffer if write zeroes is unsupported */
             int max_xfer_len = MIN_NON_ZERO(bs->bl.max_transfer_length,
                                             MAX_WRITE_ZEROES_BOUNCE_BUFFER);
+            BdrvRequestFlags write_flags = flags & ~BDRV_REQ_ZERO_WRITE;
+
+            if ((flags & BDRV_REQ_FUA) &&
+                !(bs->supported_write_flags & BDRV_REQ_FUA)) {
+                /* No need for bdrv_driver_pwrite() to do a fallback
+                 * flush on each chunk; use just one at the end */
+                write_flags &= ~BDRV_REQ_FUA;
+                need_flush = true;
+            }
             num = MIN(num, max_xfer_len);
             iov.iov_len = num * BDRV_SECTOR_SIZE;
             if (iov.iov_base == NULL) {
@@ -1212,7 +1230,8 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
             qemu_iovec_init_external(&qiov, &iov, 1);
 
             ret = bdrv_driver_pwritev(bs, sector_num * BDRV_SECTOR_SIZE,
-                                      num * BDRV_SECTOR_SIZE, &qiov, 0);
+                                      num * BDRV_SECTOR_SIZE, &qiov,
+                                      write_flags);
 
             /* Keep bounce buffer around if it is big enough for all
              * all future requests.
@@ -1228,6 +1247,9 @@ static int coroutine_fn bdrv_co_do_write_zeroes(BlockDriverState *bs,
     }
 
 fail:
+    if (ret == 0 && need_flush) {
+        ret = bdrv_co_flush(bs);
+    }
     qemu_vfree(iov.iov_base);
     return ret;
 }