[v3,3/9] qcow2: Implement copy offloading
diff mbox series

Message ID 20180509145815.3330-4-famz@redhat.com
State New
Headers show
Series
  • qemu-img convert with copy offloading
Related show

Commit Message

Fam Zheng May 9, 2018, 2:58 p.m. UTC
The two callbacks are implemented quite similarly to the read/write
functions: bdrv_co_copy_range_from maps for read and calls into bs->file
or bs->backing depending on the allocation status; bdrv_co_copy_range_to
maps for write and calls into bs->file.

Signed-off-by: Fam Zheng <famz@redhat.com>
---
 block/qcow2.c | 228 ++++++++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 198 insertions(+), 30 deletions(-)

Comments

Stefan Hajnoczi May 10, 2018, 9:25 a.m. UTC | #1
On Wed, May 09, 2018 at 10:58:09PM +0800, Fam Zheng wrote:
> +static int qcow2_co_copy_range_from(BlockDriverState *bs,
> +                                    BdrvChild *src, uint64_t src_offset,
> +                                    BdrvChild *dst, uint64_t dst_offset,
> +                                    uint64_t bytes, BdrvRequestFlags flags)
> +{
> +    BDRVQcow2State *s = bs->opaque;
> +    int offset_in_cluster;
> +    int ret;
> +    unsigned int cur_bytes; /* number of bytes in current iteration */
> +    uint64_t cluster_offset = 0;
> +    BdrvChild *child = NULL;
> +
> +    assert(!bs->encrypted);
> +    qemu_co_mutex_lock(&s->lock);
> +
> +    while (bytes != 0) {
> +
> +        /* prepare next request */
> +        cur_bytes = MIN(bytes, INT_MAX);
> +
> +        ret = qcow2_get_cluster_offset(bs, src_offset, &cur_bytes, &cluster_offset);
> +        if (ret < 0) {
> +            goto out;
> +        }
> +
> +        offset_in_cluster = offset_into_cluster(s, src_offset);
> +
> +        switch (ret) {
> +        case QCOW2_CLUSTER_UNALLOCATED:
> +            if (bs->backing) {
> +                child = bs->backing;
> +            } else {
> +                flags |= BDRV_REQ_ZERO_WRITE;
> +            }
> +            break;

Do we need a special case if the backing file is shorter than this
image?
Fam Zheng May 10, 2018, 2:51 p.m. UTC | #2
On Thu, 05/10 10:25, Stefan Hajnoczi wrote:
> On Wed, May 09, 2018 at 10:58:09PM +0800, Fam Zheng wrote:
> > +static int qcow2_co_copy_range_from(BlockDriverState *bs,
> > +                                    BdrvChild *src, uint64_t src_offset,
> > +                                    BdrvChild *dst, uint64_t dst_offset,
> > +                                    uint64_t bytes, BdrvRequestFlags flags)
> > +{
> > +    BDRVQcow2State *s = bs->opaque;
> > +    int offset_in_cluster;
> > +    int ret;
> > +    unsigned int cur_bytes; /* number of bytes in current iteration */
> > +    uint64_t cluster_offset = 0;
> > +    BdrvChild *child = NULL;
> > +
> > +    assert(!bs->encrypted);
> > +    qemu_co_mutex_lock(&s->lock);
> > +
> > +    while (bytes != 0) {
> > +
> > +        /* prepare next request */
> > +        cur_bytes = MIN(bytes, INT_MAX);
> > +
> > +        ret = qcow2_get_cluster_offset(bs, src_offset, &cur_bytes, &cluster_offset);
> > +        if (ret < 0) {
> > +            goto out;
> > +        }
> > +
> > +        offset_in_cluster = offset_into_cluster(s, src_offset);
> > +
> > +        switch (ret) {
> > +        case QCOW2_CLUSTER_UNALLOCATED:
> > +            if (bs->backing) {
> > +                child = bs->backing;
> > +            } else {
> > +                flags |= BDRV_REQ_ZERO_WRITE;
> > +            }
> > +            break;
> 
> Do we need a special case if the backing file is shorter than this
> image?

Yes, we should take the BDRV_REQ_ZERO_WRITE path if the range is fully beyond
EOF, otherwise clip cur_bytes at the backing image offset.

Fam

Patch
diff mbox series

diff --git a/block/qcow2.c b/block/qcow2.c
index 2f36e632f9..07e38b548a 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -1755,6 +1755,31 @@  static int coroutine_fn qcow2_co_block_status(BlockDriverState *bs,
     return status;
 }
 
+static int qcow2_handle_l2meta(BlockDriverState *bs, QCowL2Meta *l2meta)
+{
+    int ret = 0;
+
+    while (l2meta != NULL) {
+        QCowL2Meta *next;
+
+        if (!ret) {
+            ret = qcow2_alloc_cluster_link_l2(bs, l2meta);
+        }
+
+        /* Take the request off the list of running requests */
+        if (l2meta->nb_clusters != 0) {
+            QLIST_REMOVE(l2meta, next_in_flight);
+        }
+
+        qemu_co_queue_restart_all(&l2meta->dependent_requests);
+
+        next = l2meta->next;
+        g_free(l2meta);
+        l2meta = next;
+    }
+    return ret;
+}
+
 static coroutine_fn int qcow2_co_preadv(BlockDriverState *bs, uint64_t offset,
                                         uint64_t bytes, QEMUIOVector *qiov,
                                         int flags)
@@ -2041,24 +2066,10 @@  static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset,
             }
         }
 
-        while (l2meta != NULL) {
-            QCowL2Meta *next;
-
-            ret = qcow2_alloc_cluster_link_l2(bs, l2meta);
-            if (ret < 0) {
-                goto fail;
-            }
-
-            /* Take the request off the list of running requests */
-            if (l2meta->nb_clusters != 0) {
-                QLIST_REMOVE(l2meta, next_in_flight);
-            }
-
-            qemu_co_queue_restart_all(&l2meta->dependent_requests);
-
-            next = l2meta->next;
-            g_free(l2meta);
-            l2meta = next;
+        ret = qcow2_handle_l2meta(bs, l2meta);
+        l2meta = NULL;
+        if (ret) {
+            goto fail;
         }
 
         bytes -= cur_bytes;
@@ -2069,18 +2080,7 @@  static coroutine_fn int qcow2_co_pwritev(BlockDriverState *bs, uint64_t offset,
     ret = 0;
 
 fail:
-    while (l2meta != NULL) {
-        QCowL2Meta *next;
-
-        if (l2meta->nb_clusters != 0) {
-            QLIST_REMOVE(l2meta, next_in_flight);
-        }
-        qemu_co_queue_restart_all(&l2meta->dependent_requests);
-
-        next = l2meta->next;
-        g_free(l2meta);
-        l2meta = next;
-    }
+    qcow2_handle_l2meta(bs, l2meta);
 
     qemu_co_mutex_unlock(&s->lock);
 
@@ -3267,6 +3267,172 @@  static coroutine_fn int qcow2_co_pdiscard(BlockDriverState *bs,
     return ret;
 }
 
+static int qcow2_co_copy_range_from(BlockDriverState *bs,
+                                    BdrvChild *src, uint64_t src_offset,
+                                    BdrvChild *dst, uint64_t dst_offset,
+                                    uint64_t bytes, BdrvRequestFlags flags)
+{
+    BDRVQcow2State *s = bs->opaque;
+    int offset_in_cluster;
+    int ret;
+    unsigned int cur_bytes; /* number of bytes in current iteration */
+    uint64_t cluster_offset = 0;
+    BdrvChild *child = NULL;
+
+    assert(!bs->encrypted);
+    qemu_co_mutex_lock(&s->lock);
+
+    while (bytes != 0) {
+
+        /* prepare next request */
+        cur_bytes = MIN(bytes, INT_MAX);
+
+        ret = qcow2_get_cluster_offset(bs, src_offset, &cur_bytes, &cluster_offset);
+        if (ret < 0) {
+            goto out;
+        }
+
+        offset_in_cluster = offset_into_cluster(s, src_offset);
+
+        switch (ret) {
+        case QCOW2_CLUSTER_UNALLOCATED:
+            if (bs->backing) {
+                child = bs->backing;
+            } else {
+                flags |= BDRV_REQ_ZERO_WRITE;
+            }
+            break;
+
+        case QCOW2_CLUSTER_ZERO_PLAIN:
+        case QCOW2_CLUSTER_ZERO_ALLOC:
+            flags |= BDRV_REQ_ZERO_WRITE;
+            break;
+
+        case QCOW2_CLUSTER_COMPRESSED:
+            ret = -ENOTSUP;
+            goto out;
+            break;
+
+        case QCOW2_CLUSTER_NORMAL:
+            child = bs->file;
+            if ((cluster_offset & 511) != 0) {
+                ret = -EIO;
+                goto out;
+            }
+            break;
+
+        default:
+            abort();
+        }
+        qemu_co_mutex_unlock(&s->lock);
+        ret = bdrv_co_copy_range_from(child,
+                                      cluster_offset + offset_in_cluster,
+                                      dst, dst_offset,
+                                      cur_bytes, flags);
+        qemu_co_mutex_lock(&s->lock);
+        if (ret < 0) {
+            goto out;
+        }
+
+        bytes -= cur_bytes;
+        src_offset += cur_bytes;
+    }
+    ret = 0;
+
+out:
+    qemu_co_mutex_unlock(&s->lock);
+    return ret;
+}
+
+static int qcow2_co_copy_range_to(BlockDriverState *bs,
+                                  BdrvChild *src, uint64_t src_offset,
+                                  BdrvChild *dst, uint64_t dst_offset,
+                                  uint64_t bytes, BdrvRequestFlags flags)
+{
+    BDRVQcow2State *s = bs->opaque;
+    int offset_in_cluster;
+    int ret;
+    unsigned int cur_bytes; /* number of sectors in current iteration */
+    uint64_t cluster_offset;
+    uint8_t *cluster_data = NULL;
+    QCowL2Meta *l2meta = NULL;
+
+    assert(!bs->encrypted);
+    s->cluster_cache_offset = -1; /* disable compressed cache */
+
+    qemu_co_mutex_lock(&s->lock);
+
+    while (bytes != 0) {
+
+        l2meta = NULL;
+
+        offset_in_cluster = offset_into_cluster(s, dst_offset);
+        cur_bytes = MIN(bytes, INT_MAX);
+
+        /* TODO:
+         * If src->bs == dst->bs, we could simply copy by incrementing
+         * the refcnt, without copying user data.
+         * Or if src->bs == dst->bs->backing->bs, we could copy by discarding. */
+        ret = qcow2_alloc_cluster_offset(bs, dst_offset, &cur_bytes,
+                                         &cluster_offset, &l2meta);
+        if (ret < 0) {
+            goto fail;
+        }
+
+        assert((cluster_offset & 511) == 0);
+
+        ret = qcow2_pre_write_overlap_check(bs, 0,
+                cluster_offset + offset_in_cluster, cur_bytes);
+        if (ret < 0) {
+            goto fail;
+        }
+
+        qemu_co_mutex_unlock(&s->lock);
+        ret = bdrv_co_copy_range_to(src, src_offset,
+                                    bs->file,
+                                    cluster_offset + offset_in_cluster,
+                                    cur_bytes, flags);
+        qemu_co_mutex_lock(&s->lock);
+        if (ret < 0) {
+            goto fail;
+        }
+
+        while (l2meta != NULL) {
+            QCowL2Meta *next;
+
+            ret = qcow2_alloc_cluster_link_l2(bs, l2meta);
+            if (ret < 0) {
+                goto fail;
+            }
+
+            /* Take the request off the list of running requests */
+            if (l2meta->nb_clusters != 0) {
+                QLIST_REMOVE(l2meta, next_in_flight);
+            }
+
+            qemu_co_queue_restart_all(&l2meta->dependent_requests);
+
+            next = l2meta->next;
+            g_free(l2meta);
+            l2meta = next;
+        }
+
+        bytes -= cur_bytes;
+        dst_offset += cur_bytes;
+    }
+    ret = 0;
+
+fail:
+    qcow2_handle_l2meta(bs, l2meta);
+
+    qemu_co_mutex_unlock(&s->lock);
+
+    qemu_vfree(cluster_data);
+    trace_qcow2_writev_done_req(qemu_coroutine_self(), ret);
+
+    return ret;
+}
+
 static int qcow2_truncate(BlockDriverState *bs, int64_t offset,
                           PreallocMode prealloc, Error **errp)
 {
@@ -4515,6 +4681,8 @@  BlockDriver bdrv_qcow2 = {
 
     .bdrv_co_pwrite_zeroes  = qcow2_co_pwrite_zeroes,
     .bdrv_co_pdiscard       = qcow2_co_pdiscard,
+    .bdrv_co_copy_range_from = qcow2_co_copy_range_from,
+    .bdrv_co_copy_range_to  = qcow2_co_copy_range_to,
     .bdrv_truncate          = qcow2_truncate,
     .bdrv_co_pwritev_compressed = qcow2_co_pwritev_compressed,
     .bdrv_make_empty        = qcow2_make_empty,