Block layer patches:

- Fix resize (extending) of short overlays
 - nvme: introduce PMR support from NVMe 1.4 spec
 - qemu-storage-daemon: Fix non-string --object properties
 -----BEGIN PGP SIGNATURE-----
 
 iQJFBAABCAAvFiEE3D3rFZqa+V09dFb+fwmycsiPL9YFAl6q9BERHGt3b2xmQHJl
 ZGhhdC5jb20ACgkQfwmycsiPL9bgag//WlwtXgTRj2N/Gijav7d9ihoaUXs4Kza1
 UpP6MA4WHCPCNJlhkGyYFO4PeQP+BFoGt9+TxGfwVIP8Sv2FYXFY/ZdgrdCw5oHV
 6Zh7xpEcOvDnR5psEvaKHBwIjEWqHMOowm0rgpzf0AmJo8C419oO6Met7zkwKUgL
 gVNtB8RkKWXEoht+JrkhRd15gSAC5bFH1Ragh2ywYhwZuMk4lSZVASk1q/Zd2lX/
 e5a8uFEwC8rNbHJxtUMPZfgCjN0H/t3h1rovxpSp9Z9YL0SQ67qZcDQy9sbOXokN
 2UwIZZsFAZPeDsBoWn+Vy0GtNsYphcG6UOn4UG6404hxMzINZB3p4f6dRDJmssvT
 whTbtfczs5rXyeoQSUHpg6RyVo8sgNWxqfNjiXvSuWy5PtW/qVYPBeO7Bexb/rxl
 +yi0oU72o5AaYhS8FV2Uj1dEMLJSJkdF7558q+eKTL5sUnBOSC4xi2UMKBwhZ46q
 IdRwOrqGI0S23BQkymQR4hXrTsrPrYZAZJHAfzqckq1qqZHg85+RP/wo/Nq1ZykC
 xTJgzzIwBOAQUoqIgARtJY5rE6tmcww/T6nuVIIFglew80nzdf3Ga4kVYWHZk9Dz
 syCTQib7R3bxz0NSZrCkeynkzNrvAt+iGQrYs+RsHDWjSEzIzlYrqVuYxWLuEBSk
 CNbdkLcHi/w=
 =ad71
 -----END PGP SIGNATURE-----

Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging

Block layer patches:

- Fix resize (extending) of short overlays
- nvme: introduce PMR support from NVMe 1.4 spec
- qemu-storage-daemon: Fix non-string --object properties

# gpg: Signature made Thu 30 Apr 2020 16:51:45 BST
# gpg:                using RSA key DC3DEB159A9AF95D3D7456FE7F09B272C88F2FD6
# gpg:                issuer "kwolf@redhat.com"
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* remotes/kevin/tags/for-upstream:
  qemu-storage-daemon: Fix non-string --object properties
  qom: Factor out user_creatable_add_dict()
  nvme: introduce PMR support from NVMe 1.4 spec
  qcow2: Forward ZERO_WRITE flag for full preallocation
  iotests: Test committing to short backing file
  iotests: Filter testfiles out in filter_img_info()
  block: truncate: Don't make backing file data visible
  file-posix: Support BDRV_REQ_ZERO_WRITE for truncate
  raw-format: Support BDRV_REQ_ZERO_WRITE for truncate
  qcow2: Support BDRV_REQ_ZERO_WRITE for truncate
  block-backend: Add flags to blk_truncate()
  block: Add flags to bdrv(_co)_truncate()
  block: Add flags to BlockDriver.bdrv_co_truncate()
  qemu-iotests: allow qcow2 external discarded clusters to contain stale data
  qcow2: Add incompatibility note between backing files and raw external data files

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

block.c

@@ -548,7 +548,8 @@ static int64_t create_file_fallback_truncate(BlockBackend *blk,
     int64_t size;
     int ret;
 
-    ret = blk_truncate(blk, minimum_size, false, PREALLOC_MODE_OFF, &local_err);
+    ret = blk_truncate(blk, minimum_size, false, PREALLOC_MODE_OFF, 0,
+                       &local_err);
     if (ret < 0 && ret != -ENOTSUP) {
         error_propagate(errp, local_err);
         return ret;

block/block-backend.c

@@ -2137,14 +2137,14 @@ int blk_pwrite_compressed(BlockBackend *blk, int64_t offset, const void *buf,
 }
 
 int blk_truncate(BlockBackend *blk, int64_t offset, bool exact,
-                 PreallocMode prealloc, Error **errp)
+                 PreallocMode prealloc, BdrvRequestFlags flags, Error **errp)
 {
     if (!blk_is_available(blk)) {
         error_setg(errp, "No medium inserted");
         return -ENOMEDIUM;
     }
 
-    return bdrv_truncate(blk->root, offset, exact, prealloc, errp);
+    return bdrv_truncate(blk->root, offset, exact, prealloc, flags, errp);
 }
 
 int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf,

block/commit.c

@@ -133,7 +133,7 @@ static int coroutine_fn commit_run(Job *job, Error **errp)
     }
 
     if (base_len < len) {
-        ret = blk_truncate(s->base, len, false, PREALLOC_MODE_OFF, NULL);
+        ret = blk_truncate(s->base, len, false, PREALLOC_MODE_OFF, 0, NULL);
         if (ret) {
             goto out;
         }
@@ -458,7 +458,7 @@ int bdrv_commit(BlockDriverState *bs)
      * grow the backing file image if possible.  If not possible,
      * we must return an error */
     if (length > backing_length) {
-        ret = blk_truncate(backing, length, false, PREALLOC_MODE_OFF,
+        ret = blk_truncate(backing, length, false, PREALLOC_MODE_OFF, 0,
                            &local_err);
         if (ret < 0) {
             error_report_err(local_err);

block/crypto.c

@@ -115,7 +115,7 @@ static ssize_t block_crypto_init_func(QCryptoBlock *block,
      * which will be used by the crypto header
      */
     return blk_truncate(data->blk, data->size + headerlen, false,
-                        data->prealloc, errp);
+                        data->prealloc, 0, errp);
 }
 
 
@@ -299,7 +299,8 @@ static int block_crypto_co_create_generic(BlockDriverState *bs,
 
 static int coroutine_fn
 block_crypto_co_truncate(BlockDriverState *bs, int64_t offset, bool exact,
-                         PreallocMode prealloc, Error **errp)
+                         PreallocMode prealloc, BdrvRequestFlags flags,
+                         Error **errp)
 {
     BlockCrypto *crypto = bs->opaque;
     uint64_t payload_offset =
@@ -312,7 +313,7 @@ block_crypto_co_truncate(BlockDriverState *bs, int64_t offset, bool exact,
 
     offset += payload_offset;
 
-    return bdrv_co_truncate(bs->file, offset, exact, prealloc, errp);
+    return bdrv_co_truncate(bs->file, offset, exact, prealloc, 0, errp);
 }
 
 static void block_crypto_close(BlockDriverState *bs)

block/file-posix.c

@@ -702,6 +702,10 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
 #endif
 
     bs->supported_zero_flags = BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK;
+    if (S_ISREG(st.st_mode)) {
+        /* When extending regular files, we get zeros from the OS */
+        bs->supported_truncate_flags = BDRV_REQ_ZERO_WRITE;
+    }
     ret = 0;
 fail:
     if (filename && (bdrv_flags & BDRV_O_TEMPORARY)) {
@@ -2080,7 +2084,7 @@ raw_regular_truncate(BlockDriverState *bs, int fd, int64_t offset,
 
 static int coroutine_fn raw_co_truncate(BlockDriverState *bs, int64_t offset,
                                         bool exact, PreallocMode prealloc,
-                                        Error **errp)
+                                        BdrvRequestFlags flags, Error **errp)
 {
     BDRVRawState *s = bs->opaque;
     struct stat st;

block/file-win32.c

@@ -469,7 +469,7 @@ static void raw_close(BlockDriverState *bs)
 
 static int coroutine_fn raw_co_truncate(BlockDriverState *bs, int64_t offset,
                                         bool exact, PreallocMode prealloc,
-                                        Error **errp)
+                                        BdrvRequestFlags flags, Error **errp)
 {
     BDRVRawState *s = bs->opaque;
     LONG low, high;

block/gluster.c

@@ -1228,6 +1228,7 @@ static coroutine_fn int qemu_gluster_co_truncate(BlockDriverState *bs,
                                                  int64_t offset,
                                                  bool exact,
                                                  PreallocMode prealloc,
+                                                 BdrvRequestFlags flags,
                                                  Error **errp)
 {
     BDRVGlusterState *s = bs->opaque;

block/io.c

@@ -3339,7 +3339,8 @@ static void bdrv_parent_cb_resize(BlockDriverState *bs)
  * 'offset' bytes in length.
  */
 int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact,
-                                  PreallocMode prealloc, Error **errp)
+                                  PreallocMode prealloc, BdrvRequestFlags flags,
+                                  Error **errp)
 {
     BlockDriverState *bs = child->bs;
     BlockDriver *drv = bs->drv;
@@ -3393,10 +3394,40 @@ int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact,
         goto out;
     }
 
+    /*
+     * If the image has a backing file that is large enough that it would
+     * provide data for the new area, we cannot leave it unallocated because
+     * then the backing file content would become visible. Instead, zero-fill
+     * the new area.
+     *
+     * Note that if the image has a backing file, but was opened without the
+     * backing file, taking care of keeping things consistent with that backing
+     * file is the user's responsibility.
+     */
+    if (new_bytes && bs->backing) {
+        int64_t backing_len;
+
+        backing_len = bdrv_getlength(backing_bs(bs));
+        if (backing_len < 0) {
+            ret = backing_len;
+            error_setg_errno(errp, -ret, "Could not get backing file size");
+            goto out;
+        }
+
+        if (backing_len > old_size) {
+            flags |= BDRV_REQ_ZERO_WRITE;
+        }
+    }
+
     if (drv->bdrv_co_truncate) {
-        ret = drv->bdrv_co_truncate(bs, offset, exact, prealloc, errp);
+        if (flags & ~bs->supported_truncate_flags) {
+            error_setg(errp, "Block driver does not support requested flags");
+            ret = -ENOTSUP;
+            goto out;
+        }
+        ret = drv->bdrv_co_truncate(bs, offset, exact, prealloc, flags, errp);
     } else if (bs->file && drv->is_filter) {
-        ret = bdrv_co_truncate(bs->file, offset, exact, prealloc, errp);
+        ret = bdrv_co_truncate(bs->file, offset, exact, prealloc, flags, errp);
     } else {
         error_setg(errp, "Image format driver does not support resize");
         ret = -ENOTSUP;
@@ -3429,6 +3460,7 @@ typedef struct TruncateCo {
     int64_t offset;
     bool exact;
     PreallocMode prealloc;
+    BdrvRequestFlags flags;
     Error **errp;
     int ret;
 } TruncateCo;
@@ -3437,12 +3469,12 @@ static void coroutine_fn bdrv_truncate_co_entry(void *opaque)
 {
     TruncateCo *tco = opaque;
     tco->ret = bdrv_co_truncate(tco->child, tco->offset, tco->exact,
-                                tco->prealloc, tco->errp);
+                                tco->prealloc, tco->flags, tco->errp);
     aio_wait_kick();
 }
 
 int bdrv_truncate(BdrvChild *child, int64_t offset, bool exact,
-                  PreallocMode prealloc, Error **errp)
+                  PreallocMode prealloc, BdrvRequestFlags flags, Error **errp)
 {
     Coroutine *co;
     TruncateCo tco = {
@@ -3450,6 +3482,7 @@ int bdrv_truncate(BdrvChild *child, int64_t offset, bool exact,
         .offset     = offset,
         .exact      = exact,
         .prealloc   = prealloc,
+        .flags      = flags,
         .errp       = errp,
         .ret        = NOT_DONE,
     };

block/iscsi.c

@@ -2124,7 +2124,7 @@ static void iscsi_reopen_commit(BDRVReopenState *reopen_state)
 
 static int coroutine_fn iscsi_co_truncate(BlockDriverState *bs, int64_t offset,
                                           bool exact, PreallocMode prealloc,
-                                          Error **errp)
+                                          BdrvRequestFlags flags, Error **errp)
 {
     IscsiLun *iscsilun = bs->opaque;
     int64_t cur_length;

block/mirror.c

@@ -900,7 +900,7 @@ static int coroutine_fn mirror_run(Job *job, Error **errp)
 
         if (s->bdev_length > base_length) {
             ret = blk_truncate(s->target, s->bdev_length, false,
-                               PREALLOC_MODE_OFF, NULL);
+                               PREALLOC_MODE_OFF, 0, NULL);
             if (ret < 0) {
                 goto immediate_exit;
             }

block/nfs.c

@@ -755,7 +755,8 @@ static int64_t nfs_get_allocated_file_size(BlockDriverState *bs)
 
 static int coroutine_fn
 nfs_file_co_truncate(BlockDriverState *bs, int64_t offset, bool exact,
-                     PreallocMode prealloc, Error **errp)
+                     PreallocMode prealloc, BdrvRequestFlags flags,
+                     Error **errp)
 {
     NFSClient *client = bs->opaque;
     int ret;

block/parallels.c

@@ -203,7 +203,7 @@ static int64_t allocate_clusters(BlockDriverState *bs, int64_t sector_num,
         } else {
             ret = bdrv_truncate(bs->file,
                                 (s->data_end + space) << BDRV_SECTOR_BITS,
-                                false, PREALLOC_MODE_OFF, NULL);
+                                false, PREALLOC_MODE_OFF, 0, NULL);
         }
         if (ret < 0) {
             return ret;
@@ -493,7 +493,7 @@ static int coroutine_fn parallels_co_check(BlockDriverState *bs,
              * That means we have to pass exact=true.
              */
             ret = bdrv_truncate(bs->file, res->image_end_offset, true,
-                                PREALLOC_MODE_OFF, &local_err);
+                                PREALLOC_MODE_OFF, 0, &local_err);
             if (ret < 0) {
                 error_report_err(local_err);
                 res->check_errors++;
@@ -889,7 +889,7 @@ static void parallels_close(BlockDriverState *bs)
 
         /* errors are ignored, so we might as well pass exact=true */
         bdrv_truncate(bs->file, s->data_end << BDRV_SECTOR_BITS, true,
-                      PREALLOC_MODE_OFF, NULL);
+                      PREALLOC_MODE_OFF, 0, NULL);
     }
 
     g_free(s->bat_dirty_bmap);

block/qcow.c

@@ -480,7 +480,7 @@ static int get_cluster_offset(BlockDriverState *bs,
                     return -E2BIG;
                 }
                 ret = bdrv_truncate(bs->file, cluster_offset + s->cluster_size,
-                                    false, PREALLOC_MODE_OFF, NULL);
+                                    false, PREALLOC_MODE_OFF, 0, NULL);
                 if (ret < 0) {
                     return ret;
                 }
@@ -1035,7 +1035,7 @@ static int qcow_make_empty(BlockDriverState *bs)
             l1_length) < 0)
         return -1;
     ret = bdrv_truncate(bs->file, s->l1_table_offset + l1_length, false,
-                        PREALLOC_MODE_OFF, NULL);
+                        PREALLOC_MODE_OFF, 0, NULL);
     if (ret < 0)
         return ret;
 

block/qcow2-cluster.c

@@ -1795,7 +1795,7 @@ int qcow2_cluster_zeroize(BlockDriverState *bs, uint64_t offset,
     /* Caller must pass aligned values, except at image end */
     assert(QEMU_IS_ALIGNED(offset, s->cluster_size));
     assert(QEMU_IS_ALIGNED(end_offset, s->cluster_size) ||
-           end_offset == bs->total_sectors << BDRV_SECTOR_BITS);
+           end_offset >= bs->total_sectors << BDRV_SECTOR_BITS);
 
     /* The zero flag is only supported by version 3 and newer */
     if (s->qcow_version < 3) {

block/qcow2-refcount.c

@@ -2018,7 +2018,7 @@ static int check_refblocks(BlockDriverState *bs, BdrvCheckResult *res,
                 }
 
                 ret = bdrv_truncate(bs->file, offset + s->cluster_size, false,
-                                    PREALLOC_MODE_OFF, &local_err);
+                                    PREALLOC_MODE_OFF, 0, &local_err);
                 if (ret < 0) {
                     error_report_err(local_err);
                     goto resize_fail;

block/qcow2.c

@@ -1726,6 +1726,7 @@ static int coroutine_fn qcow2_do_open(BlockDriverState *bs, QDict *options,
 
     bs->supported_zero_flags = header.version >= 3 ?
                                BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK : 0;
+    bs->supported_truncate_flags = BDRV_REQ_ZERO_WRITE;
 
     /* Repair image if dirty */
     if (!(flags & (BDRV_O_CHECK | BDRV_O_INACTIVE)) && !bs->read_only &&
@@ -3095,7 +3096,7 @@ static int coroutine_fn preallocate_co(BlockDriverState *bs, uint64_t offset,
             mode = PREALLOC_MODE_OFF;
         }
         ret = bdrv_co_truncate(s->data_file, host_offset + cur_bytes, false,
-                               mode, errp);
+                               mode, 0, errp);
         if (ret < 0) {
             return ret;
         }
@@ -3511,7 +3512,7 @@ qcow2_co_create(BlockdevCreateOptions *create_options, Error **errp)
 
     /* Okay, now that we have a valid image, let's give it the right size */
     ret = blk_truncate(blk, qcow2_opts->size, false, qcow2_opts->preallocation,
-                       errp);
+                       0, errp);
     if (ret < 0) {
         error_prepend(errp, "Could not resize image: ");
         goto out;
@@ -3964,7 +3965,7 @@ fail:
 
 static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset,
                                           bool exact, PreallocMode prealloc,
-                                          Error **errp)
+                                          BdrvRequestFlags flags, Error **errp)
 {
     BDRVQcow2State *s = bs->opaque;
     uint64_t old_length;
@@ -4061,7 +4062,7 @@ static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset,
              * always fulfilled, so there is no need to pass it on.)
              */
             bdrv_co_truncate(bs->file, (last_cluster + 1) * s->cluster_size,
-                             false, PREALLOC_MODE_OFF, &local_err);
+                             false, PREALLOC_MODE_OFF, 0, &local_err);
             if (local_err) {
                 warn_reportf_err(local_err,
                                  "Failed to truncate the tail of the image: ");
@@ -4083,7 +4084,8 @@ static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset,
              * file should be resized to the exact target size, too,
              * so we pass @exact here.
              */
-            ret = bdrv_co_truncate(s->data_file, offset, exact, prealloc, errp);
+            ret = bdrv_co_truncate(s->data_file, offset, exact, prealloc, 0,
+                                   errp);
             if (ret < 0) {
                 goto fail;
             }
@@ -4168,8 +4170,25 @@ static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset,
         /* Allocate the data area */
         new_file_size = allocation_start +
                         nb_new_data_clusters * s->cluster_size;
-        /* Image file grows, so @exact does not matter */
+        /*
-        ret = bdrv_co_truncate(bs->file, new_file_size, false, prealloc, errp);
+         * Image file grows, so @exact does not matter.
+         *
+         * If we need to zero out the new area, try first whether the protocol
+         * driver can already take care of this.
+         */
+        if (flags & BDRV_REQ_ZERO_WRITE) {
+            ret = bdrv_co_truncate(bs->file, new_file_size, false, prealloc,
+                                   BDRV_REQ_ZERO_WRITE, NULL);
+            if (ret >= 0) {
+                flags &= ~BDRV_REQ_ZERO_WRITE;
+            }
+        } else {
+            ret = -1;
+        }
+        if (ret < 0) {
+            ret = bdrv_co_truncate(bs->file, new_file_size, false, prealloc, 0,
+                                   errp);
+        }
         if (ret < 0) {
             error_prepend(errp, "Failed to resize underlying file: ");
             qcow2_free_clusters(bs, allocation_start,
@@ -4212,6 +4231,39 @@ static int coroutine_fn qcow2_co_truncate(BlockDriverState *bs, int64_t offset,
         g_assert_not_reached();
     }
 
+    if ((flags & BDRV_REQ_ZERO_WRITE) && offset > old_length) {
+        uint64_t zero_start = QEMU_ALIGN_UP(old_length, s->cluster_size);
+
+        /*
+         * Use zero clusters as much as we can. qcow2_cluster_zeroize()
+         * requires a cluster-aligned start. The end may be unaligned if it is
+         * at the end of the image (which it is here).
+         */
+        ret = qcow2_cluster_zeroize(bs, zero_start, offset - zero_start, 0);
+        if (ret < 0) {
+            error_setg_errno(errp, -ret, "Failed to zero out new clusters");
+            goto fail;
+        }
+
+        /* Write explicit zeros for the unaligned head */
+        if (zero_start > old_length) {
+            uint64_t len = zero_start - old_length;
+            uint8_t *buf = qemu_blockalign0(bs, len);
+            QEMUIOVector qiov;
+            qemu_iovec_init_buf(&qiov, buf, len);
+
+            qemu_co_mutex_unlock(&s->lock);
+            ret = qcow2_co_pwritev_part(bs, old_length, len, &qiov, 0, 0);
+            qemu_co_mutex_lock(&s->lock);
+
+            qemu_vfree(buf);
+            if (ret < 0) {
+                error_setg_errno(errp, -ret, "Failed to zero out the new area");
+                goto fail;
+            }
+        }
+    }
+
     if (prealloc != PREALLOC_MODE_OFF) {
         /* Flush metadata before actually changing the image size */
         ret = qcow2_write_caches(bs);
@@ -4348,7 +4400,8 @@ qcow2_co_pwritev_compressed_part(BlockDriverState *bs,
         if (len < 0) {
             return len;
         }
-        return bdrv_co_truncate(bs->file, len, false, PREALLOC_MODE_OFF, NULL);
+        return bdrv_co_truncate(bs->file, len, false, PREALLOC_MODE_OFF, 0,
+                                NULL);
     }
 
     if (offset_into_cluster(s, offset)) {
@@ -4563,7 +4616,7 @@ static int make_completely_empty(BlockDriverState *bs)
     }
 
     ret = bdrv_truncate(bs->file, (3 + l1_clusters) * s->cluster_size, false,
-                        PREALLOC_MODE_OFF, &local_err);
+                        PREALLOC_MODE_OFF, 0, &local_err);
     if (ret < 0) {
         error_report_err(local_err);
         goto fail;
@@ -5371,7 +5424,7 @@ static int qcow2_amend_options(BlockDriverState *bs, QemuOpts *opts,
          * Amending image options should ensure that the image has
          * exactly the given new values, so pass exact=true here.
          */
-        ret = blk_truncate(blk, new_size, true, PREALLOC_MODE_OFF, errp);
+        ret = blk_truncate(blk, new_size, true, PREALLOC_MODE_OFF, 0, errp);
         blk_unref(blk);
         if (ret < 0) {
             return ret;

block/qed.c

@@ -677,7 +677,7 @@ static int coroutine_fn bdrv_qed_co_create(BlockdevCreateOptions *opts,
      * The QED format associates file length with allocation status,
      * so a new file (which is empty) must have a length of 0.
      */
-    ret = blk_truncate(blk, 0, true, PREALLOC_MODE_OFF, errp);
+    ret = blk_truncate(blk, 0, true, PREALLOC_MODE_OFF, 0, errp);
     if (ret < 0) {
         goto out;
     }
@@ -1467,6 +1467,7 @@ static int coroutine_fn bdrv_qed_co_truncate(BlockDriverState *bs,
                                              int64_t offset,
                                              bool exact,
                                              PreallocMode prealloc,
+                                             BdrvRequestFlags flags,
                                              Error **errp)
 {
     BDRVQEDState *s = bs->opaque;

block/raw-format.c

@@ -371,7 +371,7 @@ static void raw_refresh_limits(BlockDriverState *bs, Error **errp)
 
 static int coroutine_fn raw_co_truncate(BlockDriverState *bs, int64_t offset,
                                         bool exact, PreallocMode prealloc,
-                                        Error **errp)
+                                        BdrvRequestFlags flags, Error **errp)
 {
     BDRVRawState *s = bs->opaque;
 
@@ -387,7 +387,7 @@ static int coroutine_fn raw_co_truncate(BlockDriverState *bs, int64_t offset,
 
     s->size = offset;
     offset += s->offset;
-    return bdrv_co_truncate(bs->file, offset, exact, prealloc, errp);
+    return bdrv_co_truncate(bs->file, offset, exact, prealloc, flags, errp);
 }
 
 static void raw_eject(BlockDriverState *bs, bool eject_flag)
@@ -445,6 +445,8 @@ static int raw_open(BlockDriverState *bs, QDict *options, int flags,
     bs->supported_zero_flags = BDRV_REQ_WRITE_UNCHANGED |
         ((BDRV_REQ_FUA | BDRV_REQ_MAY_UNMAP | BDRV_REQ_NO_FALLBACK) &
             bs->file->bs->supported_zero_flags);
+    bs->supported_truncate_flags = bs->file->bs->supported_truncate_flags &
+                                   BDRV_REQ_ZERO_WRITE;
 
     if (bs->probed && !bdrv_is_read_only(bs)) {
         bdrv_refresh_filename(bs->file->bs);

block/rbd.c

@@ -1108,6 +1108,7 @@ static int coroutine_fn qemu_rbd_co_truncate(BlockDriverState *bs,
                                              int64_t offset,
                                              bool exact,
                                              PreallocMode prealloc,
+                                             BdrvRequestFlags flags,
                                              Error **errp)
 {
     int r;

block/sheepdog.c

@@ -2281,7 +2281,7 @@ static int64_t sd_getlength(BlockDriverState *bs)
 
 static int coroutine_fn sd_co_truncate(BlockDriverState *bs, int64_t offset,
                                        bool exact, PreallocMode prealloc,
-                                       Error **errp)
+                                       BdrvRequestFlags flags, Error **errp)
 {
     BDRVSheepdogState *s = bs->opaque;
     int ret, fd;
@@ -2597,7 +2597,7 @@ static coroutine_fn int sd_co_writev(BlockDriverState *bs, int64_t sector_num,
 
     assert(!flags);
     if (offset > s->inode.vdi_size) {
-        ret = sd_co_truncate(bs, offset, false, PREALLOC_MODE_OFF, NULL);
+        ret = sd_co_truncate(bs, offset, false, PREALLOC_MODE_OFF, 0, NULL);
         if (ret < 0) {
             return ret;
         }

block/ssh.c

@@ -1298,7 +1298,7 @@ static int64_t ssh_getlength(BlockDriverState *bs)
 
 static int coroutine_fn ssh_co_truncate(BlockDriverState *bs, int64_t offset,
                                         bool exact, PreallocMode prealloc,
-                                        Error **errp)
+                                        BdrvRequestFlags flags, Error **errp)
 {
     BDRVSSHState *s = bs->opaque;
 

block/vdi.c

@@ -875,7 +875,7 @@ static int coroutine_fn vdi_co_do_create(BlockdevCreateOptions *create_options,
 
     if (image_type == VDI_TYPE_STATIC) {
         ret = blk_truncate(blk, offset + blocks * block_size, false,
-                           PREALLOC_MODE_OFF, errp);
+                           PREALLOC_MODE_OFF, 0, errp);
         if (ret < 0) {
             error_prepend(errp, "Failed to statically allocate file");
             goto exit;

block/vhdx-log.c

@@ -558,7 +558,7 @@ static int vhdx_log_flush(BlockDriverState *bs, BDRVVHDXState *s,
                     goto exit;
                 }
                 ret = bdrv_truncate(bs->file, new_file_size, false,
-                                    PREALLOC_MODE_OFF, NULL);
+                                    PREALLOC_MODE_OFF, 0, NULL);
                 if (ret < 0) {
                     goto exit;
                 }

block/vhdx.c

@@ -1264,7 +1264,7 @@ static int vhdx_allocate_block(BlockDriverState *bs, BDRVVHDXState *s,
     }
 
     return bdrv_truncate(bs->file, *new_offset + s->block_size, false,
-                         PREALLOC_MODE_OFF, NULL);
+                         PREALLOC_MODE_OFF, 0, NULL);
 }
 
 /*
@@ -1703,13 +1703,13 @@ static int vhdx_create_bat(BlockBackend *blk, BDRVVHDXState *s,
         /* All zeroes, so we can just extend the file - the end of the BAT
          * is the furthest thing we have written yet */
         ret = blk_truncate(blk, data_file_offset, false, PREALLOC_MODE_OFF,
-                           errp);
+                           0, errp);
         if (ret < 0) {
             goto exit;
         }
     } else if (type == VHDX_TYPE_FIXED) {
         ret = blk_truncate(blk, data_file_offset + image_size, false,
-                           PREALLOC_MODE_OFF, errp);
+                           PREALLOC_MODE_OFF, 0, errp);
         if (ret < 0) {
             goto exit;
         }

block/vmdk.c

@@ -2077,7 +2077,7 @@ vmdk_co_pwritev_compressed(BlockDriverState *bs, uint64_t offset,
             }
             length = QEMU_ALIGN_UP(length, BDRV_SECTOR_SIZE);
             ret = bdrv_truncate(s->extents[i].file, length, false,
-                                PREALLOC_MODE_OFF, NULL);
+                                PREALLOC_MODE_OFF, 0, NULL);
             if (ret < 0) {
                 return ret;
             }
@@ -2118,7 +2118,7 @@ static int vmdk_init_extent(BlockBackend *blk,
     int gd_buf_size;
 
     if (flat) {
-        ret = blk_truncate(blk, filesize, false, PREALLOC_MODE_OFF, errp);
+        ret = blk_truncate(blk, filesize, false, PREALLOC_MODE_OFF, 0, errp);
         goto exit;
     }
     magic = cpu_to_be32(VMDK4_MAGIC);
@@ -2182,7 +2182,7 @@ static int vmdk_init_extent(BlockBackend *blk,
     }
 
     ret = blk_truncate(blk, le64_to_cpu(header.grain_offset) << 9, false,
-                       PREALLOC_MODE_OFF, errp);
+                       PREALLOC_MODE_OFF, 0, errp);
     if (ret < 0) {
         goto exit;
     }
@@ -2523,7 +2523,7 @@ static int coroutine_fn vmdk_co_do_create(int64_t size,
     /* bdrv_pwrite write padding zeros to align to sector, we don't need that
      * for description file */
     if (desc_offset == 0) {
-        ret = blk_truncate(blk, desc_len, false, PREALLOC_MODE_OFF, errp);
+        ret = blk_truncate(blk, desc_len, false, PREALLOC_MODE_OFF, 0, errp);
         if (ret < 0) {
             goto exit;
         }

block/vpc.c

@@ -898,7 +898,7 @@ static int create_fixed_disk(BlockBackend *blk, uint8_t *buf,
     /* Add footer to total size */
     total_size += HEADER_SIZE;
 
-    ret = blk_truncate(blk, total_size, false, PREALLOC_MODE_OFF, errp);
+    ret = blk_truncate(blk, total_size, false, PREALLOC_MODE_OFF, 0, errp);
     if (ret < 0) {
         return ret;
     }

blockdev.c

@@ -2741,7 +2741,7 @@ void qmp_block_resize(bool has_device, const char *device,
     }
 
     bdrv_drained_begin(bs);
-    ret = blk_truncate(blk, size, false, PREALLOC_MODE_OFF, errp);
+    ret = blk_truncate(blk, size, false, PREALLOC_MODE_OFF, 0, errp);
     bdrv_drained_end(bs);
 
 out:

docs/interop/qcow2.txt

@@ -25,6 +25,9 @@ The first cluster of a qcow2 image contains the file header:
                     is stored (NB: The string is not null terminated). 0 if the
                     image doesn't have a backing file.
 
+                    Note: backing files are incompatible with raw external data
+                    files (auto-clear feature bit 1).
+
          16 - 19:   backing_file_size
                     Length of the backing file name in bytes. Must not be
                     longer than 1023 bytes. Undefined if the image doesn't have

hw/block/Makefile.objs

@@ -7,12 +7,12 @@ common-obj-$(CONFIG_PFLASH_CFI02) += pflash_cfi02.o
 common-obj-$(CONFIG_XEN) += xen-block.o
 common-obj-$(CONFIG_ECC) += ecc.o
 common-obj-$(CONFIG_ONENAND) += onenand.o
-common-obj-$(CONFIG_NVME_PCI) += nvme.o
 common-obj-$(CONFIG_SWIM) += swim.o
 
 common-obj-$(CONFIG_SH4) += tc58128.o
 
 obj-$(CONFIG_VIRTIO_BLK) += virtio-blk.o
 obj-$(CONFIG_VHOST_USER_BLK) += vhost-user-blk.o
+obj-$(CONFIG_NVME_PCI) += nvme.o
 
 obj-y += dataplane/

hw/block/nvme.c

@@ -19,10 +19,19 @@
  *      -drive file=<file>,if=none,id=<drive_id>
  *      -device nvme,drive=<drive_id>,serial=<serial>,id=<id[optional]>, \
  *              cmb_size_mb=<cmb_size_mb[optional]>, \
+ *              [pmrdev=<mem_backend_file_id>,] \
  *              num_queues=<N[optional]>
  *
  * Note cmb_size_mb denotes size of CMB in MB. CMB is assumed to be at
  * offset 0 in BAR2 and supports only WDS, RDS and SQS for now.
+ *
+ * cmb_size_mb= and pmrdev= options are mutually exclusive due to limitation
+ * in available BAR's. cmb_size_mb= will take precedence over pmrdev= when
+ * both provided.
+ * Enabling pmr emulation can be achieved by pointing to memory-backend-file.
+ * For example:
+ * -object memory-backend-file,id=<mem_id>,share=on,mem-path=<file_path>, \
+ *  size=<size> .... -device nvme,...,pmrdev=<mem_id>
  */
 
 #include "qemu/osdep.h"
@@ -35,7 +44,9 @@
 #include "sysemu/sysemu.h"
 #include "qapi/error.h"
 #include "qapi/visitor.h"
+#include "sysemu/hostmem.h"
 #include "sysemu/block-backend.h"
+#include "exec/ram_addr.h"
 
 #include "qemu/log.h"
 #include "qemu/module.h"
@@ -1141,6 +1152,26 @@ static void nvme_write_bar(NvmeCtrl *n, hwaddr offset, uint64_t data,
         NVME_GUEST_ERR(nvme_ub_mmiowr_cmbsz_readonly,
                        "invalid write to read only CMBSZ, ignored");
         return;
+    case 0xE00: /* PMRCAP */
+        NVME_GUEST_ERR(nvme_ub_mmiowr_pmrcap_readonly,
+                       "invalid write to PMRCAP register, ignored");
+        return;
+    case 0xE04: /* TODO PMRCTL */
+        break;
+    case 0xE08: /* PMRSTS */
+        NVME_GUEST_ERR(nvme_ub_mmiowr_pmrsts_readonly,
+                       "invalid write to PMRSTS register, ignored");
+        return;
+    case 0xE0C: /* PMREBS */
+        NVME_GUEST_ERR(nvme_ub_mmiowr_pmrebs_readonly,
+                       "invalid write to PMREBS register, ignored");
+        return;
+    case 0xE10: /* PMRSWTP */
+        NVME_GUEST_ERR(nvme_ub_mmiowr_pmrswtp_readonly,
+                       "invalid write to PMRSWTP register, ignored");
+        return;
+    case 0xE14: /* TODO PMRMSC */
+         break;
     default:
         NVME_GUEST_ERR(nvme_ub_mmiowr_invalid,
                        "invalid MMIO write,"
@@ -1169,6 +1200,16 @@ static uint64_t nvme_mmio_read(void *opaque, hwaddr addr, unsigned size)
     }
 
     if (addr < sizeof(n->bar)) {
+        /*
+         * When PMRWBM bit 1 is set then read from
+         * from PMRSTS should ensure prior writes
+         * made it to persistent media
+         */
+        if (addr == 0xE08 &&
+            (NVME_PMRCAP_PMRWBM(n->bar.pmrcap) & 0x02)) {
+            qemu_ram_writeback(n->pmrdev->mr.ram_block,
+                               0, n->pmrdev->size);
+        }
         memcpy(&val, ptr + addr, size);
     } else {
         NVME_GUEST_ERR(nvme_ub_mmiord_invalid_ofs,
@@ -1332,6 +1373,23 @@ static void nvme_realize(PCIDevice *pci_dev, Error **errp)
         error_setg(errp, "serial property not set");
         return;
     }
+
+    if (!n->cmb_size_mb && n->pmrdev) {
+        if (host_memory_backend_is_mapped(n->pmrdev)) {
+            char *path = object_get_canonical_path_component(OBJECT(n->pmrdev));
+            error_setg(errp, "can't use already busy memdev: %s", path);
+            g_free(path);
+            return;
+        }
+
+        if (!is_power_of_2(n->pmrdev->size)) {
+            error_setg(errp, "pmr backend size needs to be power of 2 in size");
+            return;
+        }
+
+        host_memory_backend_set_mapped(n->pmrdev, true);
+    }
+
     blkconf_blocksizes(&n->conf);
     if (!blkconf_apply_backend_options(&n->conf, blk_is_read_only(n->conf.blk),
                                        false, errp)) {
@@ -1415,6 +1473,51 @@ static void nvme_realize(PCIDevice *pci_dev, Error **errp)
             PCI_BASE_ADDRESS_SPACE_MEMORY | PCI_BASE_ADDRESS_MEM_TYPE_64 |
             PCI_BASE_ADDRESS_MEM_PREFETCH, &n->ctrl_mem);
 
+    } else if (n->pmrdev) {
+        /* Controller Capabilities register */
+        NVME_CAP_SET_PMRS(n->bar.cap, 1);
+
+        /* PMR Capabities register */
+        n->bar.pmrcap = 0;
+        NVME_PMRCAP_SET_RDS(n->bar.pmrcap, 0);
+        NVME_PMRCAP_SET_WDS(n->bar.pmrcap, 0);
+        NVME_PMRCAP_SET_BIR(n->bar.pmrcap, 2);
+        NVME_PMRCAP_SET_PMRTU(n->bar.pmrcap, 0);
+        /* Turn on bit 1 support */
+        NVME_PMRCAP_SET_PMRWBM(n->bar.pmrcap, 0x02);
+        NVME_PMRCAP_SET_PMRTO(n->bar.pmrcap, 0);
+        NVME_PMRCAP_SET_CMSS(n->bar.pmrcap, 0);
+
+        /* PMR Control register */
+        n->bar.pmrctl = 0;
+        NVME_PMRCTL_SET_EN(n->bar.pmrctl, 0);
+
+        /* PMR Status register */
+        n->bar.pmrsts = 0;
+        NVME_PMRSTS_SET_ERR(n->bar.pmrsts, 0);
+        NVME_PMRSTS_SET_NRDY(n->bar.pmrsts, 0);
+        NVME_PMRSTS_SET_HSTS(n->bar.pmrsts, 0);
+        NVME_PMRSTS_SET_CBAI(n->bar.pmrsts, 0);
+
+        /* PMR Elasticity Buffer Size register */
+        n->bar.pmrebs = 0;
+        NVME_PMREBS_SET_PMRSZU(n->bar.pmrebs, 0);
+        NVME_PMREBS_SET_RBB(n->bar.pmrebs, 0);
+        NVME_PMREBS_SET_PMRWBZ(n->bar.pmrebs, 0);
+
+        /* PMR Sustained Write Throughput register */
+        n->bar.pmrswtp = 0;
+        NVME_PMRSWTP_SET_PMRSWTU(n->bar.pmrswtp, 0);
+        NVME_PMRSWTP_SET_PMRSWTV(n->bar.pmrswtp, 0);
+
+        /* PMR Memory Space Control register */
+        n->bar.pmrmsc = 0;
+        NVME_PMRMSC_SET_CMSE(n->bar.pmrmsc, 0);
+        NVME_PMRMSC_SET_CBA(n->bar.pmrmsc, 0);
+
+        pci_register_bar(pci_dev, NVME_PMRCAP_BIR(n->bar.pmrcap),
+            PCI_BASE_ADDRESS_SPACE_MEMORY | PCI_BASE_ADDRESS_MEM_TYPE_64 |
+            PCI_BASE_ADDRESS_MEM_PREFETCH, &n->pmrdev->mr);
     }
 
     for (i = 0; i < n->num_namespaces; i++) {
@@ -1445,11 +1548,17 @@ static void nvme_exit(PCIDevice *pci_dev)
     if (n->cmb_size_mb) {
         g_free(n->cmbuf);
     }
+
+    if (n->pmrdev) {
+        host_memory_backend_set_mapped(n->pmrdev, false);
+    }
     msix_uninit_exclusive_bar(pci_dev);
 }
 
 static Property nvme_props[] = {
     DEFINE_BLOCK_PROPERTIES(NvmeCtrl, conf),
+    DEFINE_PROP_LINK("pmrdev", NvmeCtrl, pmrdev, TYPE_MEMORY_BACKEND,
+                     HostMemoryBackend *),
     DEFINE_PROP_STRING("serial", NvmeCtrl, serial),
     DEFINE_PROP_UINT32("cmb_size_mb", NvmeCtrl, cmb_size_mb, 0),
     DEFINE_PROP_UINT32("num_queues", NvmeCtrl, num_queues, 64),

hw/block/nvme.h

@@ -83,6 +83,8 @@ typedef struct NvmeCtrl {
     uint64_t    timestamp_set_qemu_clock_ms;    /* QEMU clock time */
 
     char            *serial;
+    HostMemoryBackend *pmrdev;
+
     NvmeNamespace   *namespaces;
     NvmeSQueue      **sq;
     NvmeCQueue      **cq;

hw/block/trace-events

@@ -110,6 +110,10 @@ nvme_ub_mmiowr_ssreset_w1c_unsupported(void) "attempted to W1C CSTS.NSSRO but CA
 nvme_ub_mmiowr_ssreset_unsupported(void) "attempted NVM subsystem reset but CAP.NSSRS is zero (not supported)"
 nvme_ub_mmiowr_cmbloc_reserved(void) "invalid write to reserved CMBLOC when CMBSZ is zero, ignored"
 nvme_ub_mmiowr_cmbsz_readonly(void) "invalid write to read only CMBSZ, ignored"
+nvme_ub_mmiowr_pmrcap_readonly(void) "invalid write to read only PMRCAP, ignored"
+nvme_ub_mmiowr_pmrsts_readonly(void) "invalid write to read only PMRSTS, ignored"
+nvme_ub_mmiowr_pmrebs_readonly(void) "invalid write to read only PMREBS, ignored"
+nvme_ub_mmiowr_pmrswtp_readonly(void) "invalid write to read only PMRSWTP, ignored"
 nvme_ub_mmiowr_invalid(uint64_t offset, uint64_t data) "invalid MMIO write, offset=0x%"PRIx64", data=0x%"PRIx64""
 nvme_ub_mmiord_misaligned32(uint64_t offset) "MMIO read not 32-bit aligned, offset=0x%"PRIx64""
 nvme_ub_mmiord_toosmall(uint64_t offset) "MMIO read smaller than 32-bits, offset=0x%"PRIx64""

include/block/block.h

@@ -339,9 +339,10 @@ BlockDriverState *bdrv_find_backing_image(BlockDriverState *bs,
 void bdrv_refresh_filename(BlockDriverState *bs);
 
 int coroutine_fn bdrv_co_truncate(BdrvChild *child, int64_t offset, bool exact,
-                                  PreallocMode prealloc, Error **errp);
+                                  PreallocMode prealloc, BdrvRequestFlags flags,
+                                  Error **errp);
 int bdrv_truncate(BdrvChild *child, int64_t offset, bool exact,
-                  PreallocMode prealloc, Error **errp);
+                  PreallocMode prealloc, BdrvRequestFlags flags, Error **errp);
 
 int64_t bdrv_nb_sectors(BlockDriverState *bs);
 int64_t bdrv_getlength(BlockDriverState *bs);

include/block/block_int.h

@@ -355,7 +355,7 @@ struct BlockDriver {
      */
     int coroutine_fn (*bdrv_co_truncate)(BlockDriverState *bs, int64_t offset,
                                          bool exact, PreallocMode prealloc,
-                                         Error **errp);
+                                         BdrvRequestFlags flags, Error **errp);
 
     int64_t (*bdrv_getlength)(BlockDriverState *bs);
     bool has_variable_length;
@@ -847,6 +847,14 @@ struct BlockDriverState {
     /* Flags honored during pwrite_zeroes (so far: BDRV_REQ_FUA,
      * BDRV_REQ_MAY_UNMAP, BDRV_REQ_WRITE_UNCHANGED) */
     unsigned int supported_zero_flags;
+    /*
+     * Flags honoured during truncate (so far: BDRV_REQ_ZERO_WRITE).
+     *
+     * If BDRV_REQ_ZERO_WRITE is given, the truncate operation must make sure
+     * that any added space reads as all zeros. If this can't be guaranteed,
+     * the operation must fail.
+     */
+    unsigned int supported_truncate_flags;
 
     /* the following member gives a name to every node on the bs graph. */
     char node_name[32];

include/block/nvme.h

@@ -15,6 +15,13 @@ typedef struct NvmeBar {
     uint64_t    acq;
     uint32_t    cmbloc;
     uint32_t    cmbsz;
+    uint8_t     padding[3520]; /* not used by QEMU */
+    uint32_t    pmrcap;
+    uint32_t    pmrctl;
+    uint32_t    pmrsts;
+    uint32_t    pmrebs;
+    uint32_t    pmrswtp;
+    uint32_t    pmrmsc;
 } NvmeBar;
 
 enum NvmeCapShift {
@@ -27,6 +34,7 @@ enum NvmeCapShift {
     CAP_CSS_SHIFT      = 37,
     CAP_MPSMIN_SHIFT   = 48,
     CAP_MPSMAX_SHIFT   = 52,
+    CAP_PMR_SHIFT      = 56,
 };
 
 enum NvmeCapMask {
@@ -39,6 +47,7 @@ enum NvmeCapMask {
     CAP_CSS_MASK       = 0xff,
     CAP_MPSMIN_MASK    = 0xf,
     CAP_MPSMAX_MASK    = 0xf,
+    CAP_PMR_MASK       = 0x1,
 };
 
 #define NVME_CAP_MQES(cap)  (((cap) >> CAP_MQES_SHIFT)   & CAP_MQES_MASK)
@@ -69,6 +78,8 @@ enum NvmeCapMask {
                                                            << CAP_MPSMIN_SHIFT)
 #define NVME_CAP_SET_MPSMAX(cap, val) (cap |= (uint64_t)(val & CAP_MPSMAX_MASK)\
                                                             << CAP_MPSMAX_SHIFT)
+#define NVME_CAP_SET_PMRS(cap, val) (cap |= (uint64_t)(val & CAP_PMR_MASK)\
+                                                            << CAP_PMR_SHIFT)
 
 enum NvmeCcShift {
     CC_EN_SHIFT     = 0,
@@ -205,6 +216,167 @@ enum NvmeCmbszMask {
 #define NVME_CMBSZ_GETSIZE(cmbsz) \
     (NVME_CMBSZ_SZ(cmbsz) * (1 << (12 + 4 * NVME_CMBSZ_SZU(cmbsz))))
 
+enum NvmePmrcapShift {
+    PMRCAP_RDS_SHIFT      = 3,
+    PMRCAP_WDS_SHIFT      = 4,
+    PMRCAP_BIR_SHIFT      = 5,
+    PMRCAP_PMRTU_SHIFT    = 8,
+    PMRCAP_PMRWBM_SHIFT   = 10,
+    PMRCAP_PMRTO_SHIFT    = 16,
+    PMRCAP_CMSS_SHIFT     = 24,
+};
+
+enum NvmePmrcapMask {
+    PMRCAP_RDS_MASK      = 0x1,
+    PMRCAP_WDS_MASK      = 0x1,
+    PMRCAP_BIR_MASK      = 0x7,
+    PMRCAP_PMRTU_MASK    = 0x3,
+    PMRCAP_PMRWBM_MASK   = 0xf,
+    PMRCAP_PMRTO_MASK    = 0xff,
+    PMRCAP_CMSS_MASK     = 0x1,
+};
+
+#define NVME_PMRCAP_RDS(pmrcap)    \
+    ((pmrcap >> PMRCAP_RDS_SHIFT)   & PMRCAP_RDS_MASK)
+#define NVME_PMRCAP_WDS(pmrcap)    \
+    ((pmrcap >> PMRCAP_WDS_SHIFT)   & PMRCAP_WDS_MASK)
+#define NVME_PMRCAP_BIR(pmrcap)    \
+    ((pmrcap >> PMRCAP_BIR_SHIFT)   & PMRCAP_BIR_MASK)
+#define NVME_PMRCAP_PMRTU(pmrcap)    \
+    ((pmrcap >> PMRCAP_PMRTU_SHIFT)   & PMRCAP_PMRTU_MASK)
+#define NVME_PMRCAP_PMRWBM(pmrcap)    \
+    ((pmrcap >> PMRCAP_PMRWBM_SHIFT)   & PMRCAP_PMRWBM_MASK)
+#define NVME_PMRCAP_PMRTO(pmrcap)    \
+    ((pmrcap >> PMRCAP_PMRTO_SHIFT)   & PMRCAP_PMRTO_MASK)
+#define NVME_PMRCAP_CMSS(pmrcap)    \
+    ((pmrcap >> PMRCAP_CMSS_SHIFT)   & PMRCAP_CMSS_MASK)
+
+#define NVME_PMRCAP_SET_RDS(pmrcap, val)   \
+    (pmrcap |= (uint64_t)(val & PMRCAP_RDS_MASK) << PMRCAP_RDS_SHIFT)
+#define NVME_PMRCAP_SET_WDS(pmrcap, val)   \
+    (pmrcap |= (uint64_t)(val & PMRCAP_WDS_MASK) << PMRCAP_WDS_SHIFT)
+#define NVME_PMRCAP_SET_BIR(pmrcap, val)   \
+    (pmrcap |= (uint64_t)(val & PMRCAP_BIR_MASK) << PMRCAP_BIR_SHIFT)
+#define NVME_PMRCAP_SET_PMRTU(pmrcap, val)   \
+    (pmrcap |= (uint64_t)(val & PMRCAP_PMRTU_MASK) << PMRCAP_PMRTU_SHIFT)
+#define NVME_PMRCAP_SET_PMRWBM(pmrcap, val)   \
+    (pmrcap |= (uint64_t)(val & PMRCAP_PMRWBM_MASK) << PMRCAP_PMRWBM_SHIFT)
+#define NVME_PMRCAP_SET_PMRTO(pmrcap, val)   \
+    (pmrcap |= (uint64_t)(val & PMRCAP_PMRTO_MASK) << PMRCAP_PMRTO_SHIFT)
+#define NVME_PMRCAP_SET_CMSS(pmrcap, val)   \
+    (pmrcap |= (uint64_t)(val & PMRCAP_CMSS_MASK) << PMRCAP_CMSS_SHIFT)
+
+enum NvmePmrctlShift {
+    PMRCTL_EN_SHIFT   = 0,
+};
+
+enum NvmePmrctlMask {
+    PMRCTL_EN_MASK   = 0x1,
+};
+
+#define NVME_PMRCTL_EN(pmrctl)  ((pmrctl >> PMRCTL_EN_SHIFT)   & PMRCTL_EN_MASK)
+
+#define NVME_PMRCTL_SET_EN(pmrctl, val)   \
+    (pmrctl |= (uint64_t)(val & PMRCTL_EN_MASK) << PMRCTL_EN_SHIFT)
+
+enum NvmePmrstsShift {
+    PMRSTS_ERR_SHIFT    = 0,
+    PMRSTS_NRDY_SHIFT   = 8,
+    PMRSTS_HSTS_SHIFT   = 9,
+    PMRSTS_CBAI_SHIFT   = 12,
+};
+
+enum NvmePmrstsMask {
+    PMRSTS_ERR_MASK    = 0xff,
+    PMRSTS_NRDY_MASK   = 0x1,
+    PMRSTS_HSTS_MASK   = 0x7,
+    PMRSTS_CBAI_MASK   = 0x1,
+};
+
+#define NVME_PMRSTS_ERR(pmrsts)     \
+    ((pmrsts >> PMRSTS_ERR_SHIFT)   & PMRSTS_ERR_MASK)
+#define NVME_PMRSTS_NRDY(pmrsts)    \
+    ((pmrsts >> PMRSTS_NRDY_SHIFT)   & PMRSTS_NRDY_MASK)
+#define NVME_PMRSTS_HSTS(pmrsts)    \
+    ((pmrsts >> PMRSTS_HSTS_SHIFT)   & PMRSTS_HSTS_MASK)
+#define NVME_PMRSTS_CBAI(pmrsts)    \
+    ((pmrsts >> PMRSTS_CBAI_SHIFT)   & PMRSTS_CBAI_MASK)
+
+#define NVME_PMRSTS_SET_ERR(pmrsts, val)   \
+    (pmrsts |= (uint64_t)(val & PMRSTS_ERR_MASK) << PMRSTS_ERR_SHIFT)
+#define NVME_PMRSTS_SET_NRDY(pmrsts, val)   \
+    (pmrsts |= (uint64_t)(val & PMRSTS_NRDY_MASK) << PMRSTS_NRDY_SHIFT)
+#define NVME_PMRSTS_SET_HSTS(pmrsts, val)   \
+    (pmrsts |= (uint64_t)(val & PMRSTS_HSTS_MASK) << PMRSTS_HSTS_SHIFT)
+#define NVME_PMRSTS_SET_CBAI(pmrsts, val)   \
+    (pmrsts |= (uint64_t)(val & PMRSTS_CBAI_MASK) << PMRSTS_CBAI_SHIFT)
+
+enum NvmePmrebsShift {
+    PMREBS_PMRSZU_SHIFT   = 0,
+    PMREBS_RBB_SHIFT      = 4,
+    PMREBS_PMRWBZ_SHIFT   = 8,
+};
+
+enum NvmePmrebsMask {
+    PMREBS_PMRSZU_MASK   = 0xf,
+    PMREBS_RBB_MASK      = 0x1,
+    PMREBS_PMRWBZ_MASK   = 0xffffff,
+};
+
+#define NVME_PMREBS_PMRSZU(pmrebs)  \
+    ((pmrebs >> PMREBS_PMRSZU_SHIFT)   & PMREBS_PMRSZU_MASK)
+#define NVME_PMREBS_RBB(pmrebs)     \
+    ((pmrebs >> PMREBS_RBB_SHIFT)   & PMREBS_RBB_MASK)
+#define NVME_PMREBS_PMRWBZ(pmrebs)  \
+    ((pmrebs >> PMREBS_PMRWBZ_SHIFT)   & PMREBS_PMRWBZ_MASK)
+
+#define NVME_PMREBS_SET_PMRSZU(pmrebs, val)   \
+    (pmrebs |= (uint64_t)(val & PMREBS_PMRSZU_MASK) << PMREBS_PMRSZU_SHIFT)
+#define NVME_PMREBS_SET_RBB(pmrebs, val)   \
+    (pmrebs |= (uint64_t)(val & PMREBS_RBB_MASK) << PMREBS_RBB_SHIFT)
+#define NVME_PMREBS_SET_PMRWBZ(pmrebs, val)   \
+    (pmrebs |= (uint64_t)(val & PMREBS_PMRWBZ_MASK) << PMREBS_PMRWBZ_SHIFT)
+
+enum NvmePmrswtpShift {
+    PMRSWTP_PMRSWTU_SHIFT   = 0,
+    PMRSWTP_PMRSWTV_SHIFT   = 8,
+};
+
+enum NvmePmrswtpMask {
+    PMRSWTP_PMRSWTU_MASK   = 0xf,
+    PMRSWTP_PMRSWTV_MASK   = 0xffffff,
+};
+
+#define NVME_PMRSWTP_PMRSWTU(pmrswtp)   \
+    ((pmrswtp >> PMRSWTP_PMRSWTU_SHIFT)   & PMRSWTP_PMRSWTU_MASK)
+#define NVME_PMRSWTP_PMRSWTV(pmrswtp)   \
+    ((pmrswtp >> PMRSWTP_PMRSWTV_SHIFT)   & PMRSWTP_PMRSWTV_MASK)
+
+#define NVME_PMRSWTP_SET_PMRSWTU(pmrswtp, val)   \
+    (pmrswtp |= (uint64_t)(val & PMRSWTP_PMRSWTU_MASK) << PMRSWTP_PMRSWTU_SHIFT)
+#define NVME_PMRSWTP_SET_PMRSWTV(pmrswtp, val)   \
+    (pmrswtp |= (uint64_t)(val & PMRSWTP_PMRSWTV_MASK) << PMRSWTP_PMRSWTV_SHIFT)
+
+enum NvmePmrmscShift {
+    PMRMSC_CMSE_SHIFT   = 1,
+    PMRMSC_CBA_SHIFT    = 12,
+};
+
+enum NvmePmrmscMask {
+    PMRMSC_CMSE_MASK   = 0x1,
+    PMRMSC_CBA_MASK    = 0xfffffffffffff,
+};
+
+#define NVME_PMRMSC_CMSE(pmrmsc)    \
+    ((pmrmsc >> PMRMSC_CMSE_SHIFT)   & PMRMSC_CMSE_MASK)
+#define NVME_PMRMSC_CBA(pmrmsc)     \
+    ((pmrmsc >> PMRMSC_CBA_SHIFT)   & PMRMSC_CBA_MASK)
+
+#define NVME_PMRMSC_SET_CMSE(pmrmsc, val)   \
+    (pmrmsc |= (uint64_t)(val & PMRMSC_CMSE_MASK) << PMRMSC_CMSE_SHIFT)
+#define NVME_PMRMSC_SET_CBA(pmrmsc, val)   \
+    (pmrmsc |= (uint64_t)(val & PMRMSC_CBA_MASK) << PMRMSC_CBA_SHIFT)
+
 typedef struct NvmeCmd {
     uint8_t     opcode;
     uint8_t     fuse;

include/qom/object_interfaces.h

@@ -87,6 +87,22 @@ Object *user_creatable_add_type(const char *type, const char *id,
                                 const QDict *qdict,
                                 Visitor *v, Error **errp);
 
+/**
+ * user_creatable_add_dict:
+ * @qdict: the object definition
+ * @keyval: if true, use a keyval visitor for processing @qdict (i.e.
+ *          assume that all @qdict values are strings); otherwise, use
+ *          the normal QObject visitor (i.e. assume all @qdict values
+ *          have the QType expected by the QOM object type)
+ * @errp: if an error occurs, a pointer to an area to store the error
+ *
+ * Create an instance of the user creatable object that is defined by
+ * @qdict.  The object type is taken from the QDict key 'qom-type', its
+ * ID from the key 'id'. The remaining entries in @qdict are used to
+ * initialize the object properties.
+ */
+void user_creatable_add_dict(QDict *qdict, bool keyval, Error **errp);
+
 /**
  * user_creatable_add_opts:
  * @opts: the object definition

include/sysemu/block-backend.h

@@ -237,7 +237,7 @@ int coroutine_fn blk_co_pwrite_zeroes(BlockBackend *blk, int64_t offset,
 int blk_pwrite_compressed(BlockBackend *blk, int64_t offset, const void *buf,
                           int bytes);
 int blk_truncate(BlockBackend *blk, int64_t offset, bool exact,
-                 PreallocMode prealloc, Error **errp);
+                 PreallocMode prealloc, BdrvRequestFlags flags, Error **errp);
 int blk_pdiscard(BlockBackend *blk, int64_t offset, int bytes);
 int blk_save_vmstate(BlockBackend *blk, const uint8_t *buf,
                      int64_t pos, int size);

qemu-img.c

@@ -3897,7 +3897,7 @@ static int img_resize(int argc, char **argv)
      * resizing, so pass @exact=true.  It is of no use to report
      * success when the image has not actually been resized.
      */
-    ret = blk_truncate(blk, total_size, true, prealloc, &err);
+    ret = blk_truncate(blk, total_size, true, prealloc, 0, &err);
     if (!ret) {
         qprintf(quiet, "Image resized.\n");
     } else {

qemu-io-cmds.c

@@ -1715,7 +1715,7 @@ static int truncate_f(BlockBackend *blk, int argc, char **argv)
      * exact=true.  It is better to err on the "emit more errors" side
      * than to be overly permissive.
      */
-    ret = blk_truncate(blk, offset, true, PREALLOC_MODE_OFF, &local_err);
+    ret = blk_truncate(blk, offset, true, PREALLOC_MODE_OFF, 0, &local_err);
     if (ret < 0) {
         error_report_err(local_err);
         return ret;

qemu-storage-daemon.c

@@ -278,7 +278,6 @@ static void process_options(int argc, char *argv[])
                 QemuOpts *opts;
                 const char *type;
                 QDict *args;
-                QObject *ret_data = NULL;
 
                 /* FIXME The keyval parser rejects 'help' arguments, so we must
                  * unconditionall try QemuOpts first. */
@@ -291,9 +290,8 @@ static void process_options(int argc, char *argv[])
                 qemu_opts_del(opts);
 
                 args = keyval_parse(optarg, "qom-type", &error_fatal);
-                qmp_object_add(args, &ret_data, &error_fatal);
+                user_creatable_add_dict(args, true, &error_fatal);
                 qobject_unref(args);
-                qobject_unref(ret_data);
                 break;
             }
         default:

qom/object_interfaces.c

@@ -6,6 +6,7 @@
 #include "qapi/qmp/qerror.h"
 #include "qapi/qmp/qjson.h"
 #include "qapi/qmp/qstring.h"
+#include "qapi/qobject-input-visitor.h"
 #include "qom/object_interfaces.h"
 #include "qemu/help_option.h"
 #include "qemu/module.h"
@@ -105,6 +106,36 @@ out:
     return obj;
 }
 
+void user_creatable_add_dict(QDict *qdict, bool keyval, Error **errp)
+{
+    Visitor *v;
+    Object *obj;
+    g_autofree char *type = NULL;
+    g_autofree char *id = NULL;
+
+    type = g_strdup(qdict_get_try_str(qdict, "qom-type"));
+    if (!type) {
+        error_setg(errp, QERR_MISSING_PARAMETER, "qom-type");
+        return;
+    }
+    qdict_del(qdict, "qom-type");
+
+    id = g_strdup(qdict_get_try_str(qdict, "id"));
+    if (!id) {
+        error_setg(errp, QERR_MISSING_PARAMETER, "id");
+        return;
+    }
+    qdict_del(qdict, "id");
+
+    if (keyval) {
+        v = qobject_input_visitor_new_keyval(QOBJECT(qdict));
+    } else {
+        v = qobject_input_visitor_new(QOBJECT(qdict));
+    }
+    obj = user_creatable_add_type(type, id, qdict, v, errp);
+    visit_free(v);
+    object_unref(obj);
+}
 
 Object *user_creatable_add_opts(QemuOpts *opts, Error **errp)
 {

qom/qom-qmp-cmds.c

@@ -21,7 +21,6 @@
 #include "qapi/qapi-commands-qom.h"
 #include "qapi/qmp/qdict.h"
 #include "qapi/qmp/qerror.h"
-#include "qapi/qobject-input-visitor.h"
 #include "qemu/cutils.h"
 #include "qom/object_interfaces.h"
 #include "qom/qom-qobject.h"
@@ -245,24 +244,6 @@ void qmp_object_add(QDict *qdict, QObject **ret_data, Error **errp)
 {
     QObject *props;
     QDict *pdict;
-    Visitor *v;
-    Object *obj;
-    g_autofree char *type = NULL;
-    g_autofree char *id = NULL;
-
-    type = g_strdup(qdict_get_try_str(qdict, "qom-type"));
-    if (!type) {
-        error_setg(errp, QERR_MISSING_PARAMETER, "qom-type");
-        return;
-    }
-    qdict_del(qdict, "qom-type");
-
-    id = g_strdup(qdict_get_try_str(qdict, "id"));
-    if (!id) {
-        error_setg(errp, QERR_MISSING_PARAMETER, "id");
-        return;
-    }
-    qdict_del(qdict, "id");
 
     props = qdict_get(qdict, "props");
     if (props) {
@@ -282,10 +263,7 @@ void qmp_object_add(QDict *qdict, QObject **ret_data, Error **errp)
         qobject_unref(pdict);
     }
 
-    v = qobject_input_visitor_new(QOBJECT(qdict));
+    user_creatable_add_dict(qdict, false, errp);
-    obj = user_creatable_add_type(type, id, qdict, v, errp);
-    visit_free(v);
-    object_unref(obj);
 }
 
 void qmp_object_del(const char *id, Error **errp)

tests/qemu-iotests/244

@@ -143,7 +143,6 @@ $QEMU_IO -c 'read -P 0 0 1M' \
 echo
 $QEMU_IO -c 'read -P 0 0 1M' \
          -c 'read -P 0x11 1M 1M' \
-         -c 'read -P 0 2M 2M' \
          -c 'read -P 0x11 4M 1M' \
          -c 'read -P 0 5M 1M' \
          -f raw "$TEST_IMG.data" |
@@ -180,8 +179,15 @@ $QEMU_IO -c 'read -P 0 0 1M' \
          -f $IMGFMT "$TEST_IMG" |
          _filter_qemu_io
 
+# Discarded clusters are only marked as such in the qcow2 metadata, but
+# they can contain stale data in the external data file.  Instead, zero
+# clusters must be zeroed in the external data file too.
 echo
-$QEMU_IMG compare "$TEST_IMG" "$TEST_IMG.data"
+$QEMU_IO -c 'read -P 0 0 1M' \
+         -c 'read -P 0x11 1M 1M' \
+         -c 'read -P 0 3M 3M' \
+         -f raw "$TEST_IMG".data |
+         _filter_qemu_io
 
 echo -n "qcow2 file size after I/O: "
 du -b $TEST_IMG | cut -f1

tests/qemu-iotests/244.out

@@ -74,8 +74,6 @@ read 1048576/1048576 bytes at offset 0
 1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 read 1048576/1048576 bytes at offset 1048576
 1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
-read 2097152/2097152 bytes at offset 2097152
-2 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 read 1048576/1048576 bytes at offset 4194304
 1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 read 1048576/1048576 bytes at offset 5242880
@@ -108,7 +106,12 @@ read 1048576/1048576 bytes at offset 1048576
 read 4194304/4194304 bytes at offset 2097152
 4 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 
-Images are identical.
+read 1048576/1048576 bytes at offset 0
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 1048576/1048576 bytes at offset 1048576
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+read 3145728/3145728 bytes at offset 3145728
+3 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
 qcow2 file size after I/O: 327680
 
 === bdrv_co_block_status test for file and offset=0 ===

tests/qemu-iotests/274

@@ -0,0 +1,155 @@
+#!/usr/bin/env python3
+#
+# Copyright (C) 2019 Red Hat, Inc.
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+#
+# Creator/Owner: Kevin Wolf <kwolf@redhat.com>
+#
+# Some tests for short backing files and short overlays
+
+import iotests
+
+iotests.verify_image_format(supported_fmts=['qcow2'])
+iotests.verify_platform(['linux'])
+
+size_short = 1 * 1024 * 1024
+size_long = 2 * 1024 * 1024
+size_diff = size_long - size_short
+
+def create_chain() -> None:
+    iotests.qemu_img_log('create', '-f', iotests.imgfmt, base,
+                         str(size_long))
+    iotests.qemu_img_log('create', '-f', iotests.imgfmt, '-b', base, mid,
+                         str(size_short))
+    iotests.qemu_img_log('create', '-f', iotests.imgfmt, '-b', mid, top,
+                         str(size_long))
+
+    iotests.qemu_io_log('-c', 'write -P 1 0 %d' % size_long, base)
+
+def create_vm() -> iotests.VM:
+    vm = iotests.VM()
+    vm.add_blockdev('file,filename=%s,node-name=base-file' % base)
+    vm.add_blockdev('%s,file=base-file,node-name=base' % iotests.imgfmt)
+    vm.add_blockdev('file,filename=%s,node-name=mid-file' % mid)
+    vm.add_blockdev('%s,file=mid-file,node-name=mid,backing=base'
+                    % iotests.imgfmt)
+    vm.add_drive(top, 'backing=mid,node-name=top')
+    return vm
+
+with iotests.FilePath('base') as base, \
+     iotests.FilePath('mid') as mid, \
+     iotests.FilePath('top') as top:
+
+    iotests.log('== Commit tests ==')
+
+    create_chain()
+
+    iotests.log('=== Check visible data ===')
+
+    iotests.qemu_io_log('-c', 'read -P 1 0 %d' % size_short, top)
+    iotests.qemu_io_log('-c', 'read -P 0 %d %d' % (size_short, size_diff), top)
+
+    iotests.log('=== Checking allocation status ===')
+
+    iotests.qemu_io_log('-c', 'alloc 0 %d' % size_short,
+                        '-c', 'alloc %d %d' % (size_short, size_diff),
+                        base)
+
+    iotests.qemu_io_log('-c', 'alloc 0 %d' % size_short,
+                        '-c', 'alloc %d %d' % (size_short, size_diff),
+                        mid)
+
+    iotests.qemu_io_log('-c', 'alloc 0 %d' % size_short,
+                        '-c', 'alloc %d %d' % (size_short, size_diff),
+                        top)
+
+    iotests.log('=== Checking map ===')
+
+    iotests.qemu_img_log('map', '--output=json', base)
+    iotests.qemu_img_log('map', '--output=human', base)
+    iotests.qemu_img_log('map', '--output=json', mid)
+    iotests.qemu_img_log('map', '--output=human', mid)
+    iotests.qemu_img_log('map', '--output=json', top)
+    iotests.qemu_img_log('map', '--output=human', top)
+
+    iotests.log('=== Testing qemu-img commit (top -> mid) ===')
+
+    iotests.qemu_img_log('commit', top)
+    iotests.img_info_log(mid)
+    iotests.qemu_io_log('-c', 'read -P 1 0 %d' % size_short, mid)
+    iotests.qemu_io_log('-c', 'read -P 0 %d %d' % (size_short, size_diff), mid)
+
+    iotests.log('=== Testing HMP commit (top -> mid) ===')
+
+    create_chain()
+    with create_vm() as vm:
+        vm.launch()
+        vm.qmp_log('human-monitor-command', command_line='commit drive0')
+
+    iotests.img_info_log(mid)
+    iotests.qemu_io_log('-c', 'read -P 1 0 %d' % size_short, mid)
+    iotests.qemu_io_log('-c', 'read -P 0 %d %d' % (size_short, size_diff), mid)
+
+    iotests.log('=== Testing QMP active commit (top -> mid) ===')
+
+    create_chain()
+    with create_vm() as vm:
+        vm.launch()
+        vm.qmp_log('block-commit', device='top', base_node='mid',
+                   job_id='job0', auto_dismiss=False)
+        vm.run_job('job0', wait=5)
+
+    iotests.img_info_log(mid)
+    iotests.qemu_io_log('-c', 'read -P 1 0 %d' % size_short, mid)
+    iotests.qemu_io_log('-c', 'read -P 0 %d %d' % (size_short, size_diff), mid)
+
+
+    iotests.log('== Resize tests ==')
+
+    # Use different sizes for different allocation modes:
+    #
+    # We want to have at least one test where 32 bit truncation in the size of
+    # the overlapping area becomes visible. This is covered by the
+    # prealloc='off' case (1G to 6G is an overlap of 5G).
+    #
+    # However, we can only do this for modes that don't preallocate data
+    # because otherwise we might run out of space on the test host.
+    #
+    # We also want to test some unaligned combinations.
+    for (prealloc, base_size, top_size_old, top_size_new, off) in [
+            ('off',       '6G',    '1G',   '8G',   '5G'),
+            ('metadata', '32G',   '30G',  '33G',  '31G'),
+            ('falloc',   '10M',    '5M',  '15M',   '9M'),
+            ('full',     '16M',    '8M',  '12M',  '11M'),
+            ('off',      '384k', '253k', '512k', '253k'),
+            ('off',      '400k', '256k', '512k', '336k'),
+            ('off',      '512k', '256k', '500k', '436k')]:
+
+        iotests.log('=== preallocation=%s ===' % prealloc)
+        iotests.qemu_img_log('create', '-f', iotests.imgfmt, base, base_size)
+        iotests.qemu_img_log('create', '-f', iotests.imgfmt, '-b', base, top,
+                             top_size_old)
+        iotests.qemu_io_log('-c', 'write -P 1 %s 64k' % off, base)
+
+        # After this, top_size_old to base_size should be allocated/zeroed.
+        #
+        # In theory, leaving base_size to top_size_new unallocated would be
+        # correct, but in practice, if we zero out anything, we zero out
+        # everything up to top_size_new.
+        iotests.qemu_img_log('resize', '-f', iotests.imgfmt,
+                             '--preallocation', prealloc, top, top_size_new)
+        iotests.qemu_io_log('-c', 'read -P 0 %s 64k' % off, top)
+        iotests.qemu_io_log('-c', 'map', top)
+        iotests.qemu_img_log('map', '--output=json', top)

tests/qemu-iotests/274.out

@@ -0,0 +1,268 @@
+== Commit tests ==
+Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=2097152 cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-mid', fmt=qcow2 size=1048576 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=2097152 backing_file=TEST_DIR/PID-mid cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+wrote 2097152/2097152 bytes at offset 0
+2 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+=== Check visible data ===
+read 1048576/1048576 bytes at offset 0
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+read 1048576/1048576 bytes at offset 1048576
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+=== Checking allocation status ===
+1048576/1048576 bytes allocated at offset 0 bytes
+1048576/1048576 bytes allocated at offset 1 MiB
+
+0/1048576 bytes allocated at offset 0 bytes
+0/0 bytes allocated at offset 1 MiB
+
+0/1048576 bytes allocated at offset 0 bytes
+0/1048576 bytes allocated at offset 1 MiB
+
+=== Checking map ===
+[{ "start": 0, "length": 2097152, "depth": 0, "zero": false, "data": true, "offset": 327680}]
+
+Offset          Length          Mapped to       File
+0               0x200000        0x50000         TEST_DIR/PID-base
+
+[{ "start": 0, "length": 1048576, "depth": 1, "zero": false, "data": true, "offset": 327680}]
+
+Offset          Length          Mapped to       File
+0               0x100000        0x50000         TEST_DIR/PID-base
+
+[{ "start": 0, "length": 1048576, "depth": 2, "zero": false, "data": true, "offset": 327680},
+{ "start": 1048576, "length": 1048576, "depth": 0, "zero": true, "data": false}]
+
+Offset          Length          Mapped to       File
+0               0x100000        0x50000         TEST_DIR/PID-base
+
+=== Testing qemu-img commit (top -> mid) ===
+Image committed.
+
+image: TEST_IMG
+file format: IMGFMT
+virtual size: 2 MiB (2097152 bytes)
+cluster_size: 65536
+backing file: TEST_DIR/PID-base
+Format specific information:
+    compat: 1.1
+    lazy refcounts: false
+    refcount bits: 16
+    corrupt: false
+
+read 1048576/1048576 bytes at offset 0
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+read 1048576/1048576 bytes at offset 1048576
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+=== Testing HMP commit (top -> mid) ===
+Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=2097152 cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-mid', fmt=qcow2 size=1048576 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=2097152 backing_file=TEST_DIR/PID-mid cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+wrote 2097152/2097152 bytes at offset 0
+2 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+{"execute": "human-monitor-command", "arguments": {"command-line": "commit drive0"}}
+{"return": ""}
+image: TEST_IMG
+file format: IMGFMT
+virtual size: 2 MiB (2097152 bytes)
+cluster_size: 65536
+backing file: TEST_DIR/PID-base
+Format specific information:
+    compat: 1.1
+    lazy refcounts: false
+    refcount bits: 16
+    corrupt: false
+
+read 1048576/1048576 bytes at offset 0
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+read 1048576/1048576 bytes at offset 1048576
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+=== Testing QMP active commit (top -> mid) ===
+Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=2097152 cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-mid', fmt=qcow2 size=1048576 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=2097152 backing_file=TEST_DIR/PID-mid cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+wrote 2097152/2097152 bytes at offset 0
+2 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+{"execute": "block-commit", "arguments": {"auto-dismiss": false, "base-node": "mid", "device": "top", "job-id": "job0"}}
+{"return": {}}
+{"execute": "job-complete", "arguments": {"id": "job0"}}
+{"return": {}}
+{"data": {"device": "job0", "len": 0, "offset": 0, "speed": 0, "type": "commit"}, "event": "BLOCK_JOB_READY", "timestamp": {"microseconds": "USECS", "seconds": "SECS"}}
+{"data": {"device": "job0", "len": 0, "offset": 0, "speed": 0, "type": "commit"}, "event": "BLOCK_JOB_COMPLETED", "timestamp": {"microseconds": "USECS", "seconds": "SECS"}}
+{"execute": "job-dismiss", "arguments": {"id": "job0"}}
+{"return": {}}
+image: TEST_IMG
+file format: IMGFMT
+virtual size: 2 MiB (2097152 bytes)
+cluster_size: 65536
+backing file: TEST_DIR/PID-base
+Format specific information:
+    compat: 1.1
+    lazy refcounts: false
+    refcount bits: 16
+    corrupt: false
+
+read 1048576/1048576 bytes at offset 0
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+read 1048576/1048576 bytes at offset 1048576
+1 MiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+== Resize tests ==
+=== preallocation=off ===
+Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=6442450944 cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=1073741824 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+wrote 65536/65536 bytes at offset 5368709120
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+Image resized.
+
+read 65536/65536 bytes at offset 5368709120
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+1 GiB (0x40000000) bytes not allocated at offset 0 bytes (0x0)
+7 GiB (0x1c0000000) bytes     allocated at offset 1 GiB (0x40000000)
+
+[{ "start": 0, "length": 1073741824, "depth": 1, "zero": true, "data": false},
+{ "start": 1073741824, "length": 7516192768, "depth": 0, "zero": true, "data": false}]
+
+=== preallocation=metadata ===
+Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=34359738368 cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=32212254720 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+wrote 65536/65536 bytes at offset 33285996544
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+Image resized.
+
+read 65536/65536 bytes at offset 33285996544
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+30 GiB (0x780000000) bytes not allocated at offset 0 bytes (0x0)
+3 GiB (0xc0000000) bytes     allocated at offset 30 GiB (0x780000000)
+
+[{ "start": 0, "length": 32212254720, "depth": 1, "zero": true, "data": false},
+{ "start": 32212254720, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 327680},
+{ "start": 32749125632, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 537264128},
+{ "start": 33285996544, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 1074200576},
+{ "start": 33822867456, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 1611137024},
+{ "start": 34359738368, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 2148139008},
+{ "start": 34896609280, "length": 536870912, "depth": 0, "zero": true, "data": false, "offset": 2685075456}]
+
+=== preallocation=falloc ===
+Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=10485760 cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=5242880 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+wrote 65536/65536 bytes at offset 9437184
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+Image resized.
+
+read 65536/65536 bytes at offset 9437184
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+5 MiB (0x500000) bytes not allocated at offset 0 bytes (0x0)
+10 MiB (0xa00000) bytes     allocated at offset 5 MiB (0x500000)
+
+[{ "start": 0, "length": 5242880, "depth": 1, "zero": true, "data": false},
+{ "start": 5242880, "length": 10485760, "depth": 0, "zero": false, "data": true, "offset": 327680}]
+
+=== preallocation=full ===
+Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=16777216 cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=8388608 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+wrote 65536/65536 bytes at offset 11534336
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+Image resized.
+
+read 65536/65536 bytes at offset 11534336
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+8 MiB (0x800000) bytes not allocated at offset 0 bytes (0x0)
+4 MiB (0x400000) bytes     allocated at offset 8 MiB (0x800000)
+
+[{ "start": 0, "length": 8388608, "depth": 1, "zero": true, "data": false},
+{ "start": 8388608, "length": 4194304, "depth": 0, "zero": false, "data": true, "offset": 327680}]
+
+=== preallocation=off ===
+Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=393216 cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=259072 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+wrote 65536/65536 bytes at offset 259072
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+Image resized.
+
+read 65536/65536 bytes at offset 259072
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+192 KiB (0x30000) bytes not allocated at offset 0 bytes (0x0)
+320 KiB (0x50000) bytes     allocated at offset 192 KiB (0x30000)
+
+[{ "start": 0, "length": 196608, "depth": 1, "zero": true, "data": false},
+{ "start": 196608, "length": 65536, "depth": 0, "zero": false, "data": true, "offset": 327680},
+{ "start": 262144, "length": 262144, "depth": 0, "zero": true, "data": false}]
+
+=== preallocation=off ===
+Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=409600 cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=262144 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+wrote 65536/65536 bytes at offset 344064
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+Image resized.
+
+read 65536/65536 bytes at offset 344064
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+256 KiB (0x40000) bytes not allocated at offset 0 bytes (0x0)
+256 KiB (0x40000) bytes     allocated at offset 256 KiB (0x40000)
+
+[{ "start": 0, "length": 262144, "depth": 1, "zero": true, "data": false},
+{ "start": 262144, "length": 262144, "depth": 0, "zero": true, "data": false}]
+
+=== preallocation=off ===
+Formatting 'TEST_DIR/PID-base', fmt=qcow2 size=524288 cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+Formatting 'TEST_DIR/PID-top', fmt=qcow2 size=262144 backing_file=TEST_DIR/PID-base cluster_size=65536 lazy_refcounts=off refcount_bits=16
+
+wrote 65536/65536 bytes at offset 446464
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+Image resized.
+
+read 65536/65536 bytes at offset 446464
+64 KiB, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec)
+
+256 KiB (0x40000) bytes not allocated at offset 0 bytes (0x0)
+244 KiB (0x3d000) bytes     allocated at offset 256 KiB (0x40000)
+
+[{ "start": 0, "length": 262144, "depth": 1, "zero": true, "data": false},
+{ "start": 262144, "length": 249856, "depth": 0, "zero": true, "data": false}]
+

tests/qemu-iotests/group

@@ -286,6 +286,7 @@
 270 rw backing quick
 272 rw
 273 backing quick
+274 rw backing
 277 rw quick
 279 rw backing quick
 280 rw migration quick

tests/qemu-iotests/iotests.py

@@ -338,8 +338,9 @@ def filter_img_info(output, filename):
     for line in output.split('\n'):
         if 'disk size' in line or 'actual-size' in line:
             continue
-        line = line.replace(filename, 'TEST_IMG') \
+        line = line.replace(filename, 'TEST_IMG')
-                   .replace(imgfmt, 'IMGFMT')
+        line = filter_testfiles(line)
+        line = line.replace(imgfmt, 'IMGFMT')
         line = re.sub('iters: [0-9]+', 'iters: XXX', line)
         line = re.sub('uuid: [-a-f0-9]+', 'uuid: XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX', line)
         line = re.sub('cid: [0-9]+', 'cid: XXXXXXXXXX', line)

tests/test-block-iothread.c

@@ -46,7 +46,8 @@ static int coroutine_fn bdrv_test_co_pdiscard(BlockDriverState *bs,
 
 static int coroutine_fn
 bdrv_test_co_truncate(BlockDriverState *bs, int64_t offset, bool exact,
-                      PreallocMode prealloc, Error **errp)
+                      PreallocMode prealloc, BdrvRequestFlags flags,
+                      Error **errp)
 {
     return 0;
 }
@@ -185,18 +186,18 @@ static void test_sync_op_truncate(BdrvChild *c)
     int ret;
 
     /* Normal success path */
-    ret = bdrv_truncate(c, 65536, false, PREALLOC_MODE_OFF, NULL);
+    ret = bdrv_truncate(c, 65536, false, PREALLOC_MODE_OFF, 0, NULL);
     g_assert_cmpint(ret, ==, 0);
 
     /* Early error: Negative offset */
-    ret = bdrv_truncate(c, -2, false, PREALLOC_MODE_OFF, NULL);
+    ret = bdrv_truncate(c, -2, false, PREALLOC_MODE_OFF, 0, NULL);
     g_assert_cmpint(ret, ==, -EINVAL);
 
     /* Error: Read-only image */
     c->bs->read_only = true;
     c->bs->open_flags &= ~BDRV_O_RDWR;
 
-    ret = bdrv_truncate(c, 65536, false, PREALLOC_MODE_OFF, NULL);
+    ret = bdrv_truncate(c, 65536, false, PREALLOC_MODE_OFF, 0, NULL);
     g_assert_cmpint(ret, ==, -EACCES);
 
     c->bs->read_only = false;