qemu-block
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-block] [PATCH] resize qcow2 with snapshot


From: zhangzhiming
Subject: Re: [Qemu-block] [PATCH] resize qcow2 with snapshot
Date: Tue, 2 Aug 2016 21:43:43 +0800

ping

zhangzhiming



On Jun 24, 2016, at 7:10 PM, zhangzhiming <address@hidden> wrote:

qcow2 can’t be resized while there is a snapshot in qcow2 image but in version 3 image of qcow2,
each disk size of snapshot is stored in the image, so we can resize image even through this are some snapshots 
since version 3 image. the function of resize nearly finished by others,  i just do very little job.


Signed-off-by: zhangzhiming <address@hidden>

---
 block.c                  |   19 +++++++++++++++++++
 block/qcow2-cluster.c    |   29 +++++++++++++++++++++++++++++
 block/qcow2-snapshot.c   |   37 +++++++++++++++++++++++++------------
 block/qcow2.c            |   29 ++++++++++++++++++++---------
 block/qcow2.h            |    1 +
 include/block/snapshot.h |    1 +
 6 files changed, 95 insertions(+), 21 deletions(-)

diff --git a/block.c b/block.c
index f54bc25..0de7b2d 100644
--- a/block.c
+++ b/block.c
@@ -2586,6 +2586,25 @@ int bdrv_truncate(BlockDriverState *bs, int64_t offset)
     return ret;
 }
 
+int bdrv_apply_snapshot(BlockDriverState *bs, const char *snapshot_id,
+                        uint64_t snapshot_size)
+{
+    int ret = bdrv_snapshot_goto(bs, snapshot_id);
+    if (ret < 0) {
+        return ret;
+    }
+
+    ret = refresh_total_sectors(bs, snapshot_size >> BDRV_SECTOR_BITS);
+    if (ret < 0) {
+        return ret;
+    }
+    bdrv_dirty_bitmap_truncate(bs);
+    if (bs->blk) {
+        blk_dev_resize_cb(bs->blk);
+    }
+    return ret;
+}
+
 /**
  * Length of a allocated file in bytes. Sparse files are counted by actual
  * allocated space. Return < 0 if error or unknown.
diff --git a/block/qcow2-cluster.c b/block/qcow2-cluster.c
index b04bfaf..ebadddf 100644
--- a/block/qcow2-cluster.c
+++ b/block/qcow2-cluster.c
@@ -32,6 +32,35 @@
 #include "qemu/bswap.h"
 #include "trace.h"
 
+int shrink_l1_table(BlockDriverState *bs, int64_t new_l1_size)
+{
+    BDRVQcow2State *s = bs->opaque;
+    int64_t old_l1_size = s->l1_size;
+    s->l1_size = new_l1_size;
+    int ret = qcow2_update_snapshot_refcount(bs, s->l1_table_offset,
+                                             s->l1_size, 1);
+    if (ret < 0) {
+        return ret;
+    }
+
+    s->l1_size = old_l1_size;
+    ret = qcow2_update_snapshot_refcount(bs, s->l1_table_offset,
+                                         s->l1_size, -1);
+    if (ret < 0) {
+        return ret;
+    }
+    s->l1_size = new_l1_size;
+
+    uint32_t be_l1_size = cpu_to_be32(s->l1_size);
+    ret = bdrv_pwrite_sync(bs->file->bs, offsetof(QCowHeader, l1_size),
+                           &be_l1_size, sizeof(be_l1_size));
+    if (ret < 0) {
+        return ret;
+    }
+
+    return 0;
+}
+
 int qcow2_grow_l1_table(BlockDriverState *bs, uint64_t min_size,
                         bool exact_size)
 {
diff --git a/block/qcow2-snapshot.c b/block/qcow2-snapshot.c
index 242fb21..6dd6769 100644
--- a/block/qcow2-snapshot.c
+++ b/block/qcow2-snapshot.c
@@ -478,13 +478,6 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
     }
     sn = &s->snapshots[snapshot_index];
 
-    if (sn->disk_size != bs->total_sectors * BDRV_SECTOR_SIZE) {
-        error_report("qcow2: Loading snapshots with different disk "
-            "size is not implemented");
-        ret = -ENOTSUP;
-        goto fail;
-    }
-
     /*
      * Make sure that the current L1 table is big enough to contain the whole
      * L1 table of the snapshot. If the snapshot L1 table is smaller, the
@@ -506,8 +499,8 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
      * Decrease the refcount referenced by the old one only when the L1
      * table is overwritten.
      */
-    sn_l1_table = g_try_malloc0(cur_l1_bytes);
-    if (cur_l1_bytes && sn_l1_table == NULL) {
+    sn_l1_table = g_try_malloc0(sn_l1_bytes);
+    if (sn_l1_bytes && sn_l1_table == NULL) {
         ret = -ENOMEM;
         goto fail;
     }
@@ -525,17 +518,34 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
     }
 
     ret = qcow2_pre_write_overlap_check(bs, QCOW2_OL_ACTIVE_L1,
-                                        s->l1_table_offset, cur_l1_bytes);
+                                        s->l1_table_offset, sn_l1_bytes);
     if (ret < 0) {
         goto fail;
     }
 
     ret = bdrv_pwrite_sync(bs->file->bs, s->l1_table_offset, sn_l1_table,
-                           cur_l1_bytes);
+                           sn_l1_bytes);
+    if (ret < 0) {
+        goto fail;
+    }
+
+    /* write updated header.size */
+    uint64_t be_disk_size = cpu_to_be64(sn->disk_size);
+    ret = bdrv_pwrite_sync(bs->file->bs, offsetof(QCowHeader, size),
+                           &be_disk_size, sizeof(uint64_t));
     if (ret < 0) {
         goto fail;
     }
 
+    uint32_t be_l1_size = cpu_to_be32(sn->l1_size);
+    ret = bdrv_pwrite_sync(bs->file->bs, offsetof(QCowHeader, l1_size),
+                           &be_l1_size, sizeof(be_l1_size));
+    if (ret < 0) {
+        goto fail;
+    }
+
+    s->l1_vm_state_index = sn->l1_size;
+
     /*
      * Decrease refcount of clusters of current L1 table.
      *
@@ -553,7 +563,8 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
      * Now update the in-memory L1 table to be in sync with the on-disk one. We
      * need to do this even if updating refcounts failed.
      */
-    for(i = 0;i < s->l1_size; i++) {
+    memset(s->l1_table, 0, s->l1_size * sizeof(uint64_t));
+    for (i = 0; i < sn->l1_size; i++) {
         s->l1_table[i] = be64_to_cpu(sn_l1_table[i]);
     }
 
@@ -564,6 +575,7 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
     g_free(sn_l1_table);
     sn_l1_table = NULL;
 
+    s->l1_size = sn->l1_size;
     /*
      * Update QCOW_OFLAG_COPIED in the active L1 table (it may have changed
      * when we decreased the refcount of the old snapshot.
@@ -676,6 +688,7 @@ int qcow2_snapshot_list(BlockDriverState *bs, QEMUSnapshotInfo **psn_tab)
         sn_info->date_sec = sn->date_sec;
         sn_info->date_nsec = sn->date_nsec;
         sn_info->vm_clock_nsec = sn->vm_clock_nsec;
+        sn_info->disk_size = sn->disk_size;
     }
     *psn_tab = sn_tab;
     return s->nb_snapshots;
diff --git a/block/qcow2.c b/block/qcow2.c
index 6f5fb81..612a534 100644
--- a/block/qcow2.c
+++ b/block/qcow2.c
@@ -2495,22 +2495,33 @@ static int qcow2_truncate(BlockDriverState *bs, int64_t offset)
         return -EINVAL;
     }
 
-    /* cannot proceed if image has snapshots */
-    if (s->nb_snapshots) {
-        error_report("Can't resize an image which has snapshots");
+    bool v3_truncate = (s->qcow_version == 3);
+
+    /* cannot proceed if image has snapshots and qcow_version is not 3 */
+    if (!v3_truncate && s->nb_snapshots) {
+        error_report("Can't resize an image which has snapshots and "
+                     "qcow_version is not 3");
         return -ENOTSUP;
     }
 
-    /* shrinking is currently not supported */
-    if (offset < bs->total_sectors * 512) {
-        error_report("qcow2 doesn't support shrinking images yet");
+    /* shrinking is supported from version 3 */
+    if (!v3_truncate && offset < bs->total_sectors * 512) {
+        error_report("qcow2 doesn't support shrinking images yet while"
+                     " qcow_version is not 3");
         return -ENOTSUP;
     }
 
     new_l1_size = size_to_l1(s, offset);
-    ret = qcow2_grow_l1_table(bs, new_l1_size, true);
-    if (ret < 0) {
-        return ret;
+    if (offset < bs->total_sectors * 512) {
+        ret = shrink_l1_table(bs, new_l1_size);
+        if (ret < 0) {
+            return ret;
+        }
+    } else {
+        ret = qcow2_grow_l1_table(bs, new_l1_size, true);
+        if (ret < 0) {
+            return ret;
+        }
     }
 
     /* write updated header.size */
diff --git a/block/qcow2.h b/block/qcow2.h
index 7db9795..8efa735 100644
--- a/block/qcow2.h
+++ b/block/qcow2.h
@@ -534,6 +534,7 @@ int qcow2_change_refcount_order(BlockDriverState *bs, int refcount_order,
                                 void *cb_opaque, Error **errp);
 
 /* qcow2-cluster.c functions */
+int shrink_l1_table(BlockDriverState *bs, int64_t new_l1_size);
 int qcow2_grow_l1_table(BlockDriverState *bs, uint64_t min_size,
                         bool exact_size);
 int qcow2_write_l1_entry(BlockDriverState *bs, int l1_index);
diff --git a/include/block/snapshot.h b/include/block/snapshot.h
index e5c0553..7279c12 100644
--- a/include/block/snapshot.h
+++ b/include/block/snapshot.h
@@ -44,6 +44,7 @@ typedef struct QEMUSnapshotInfo {
     uint32_t date_sec; /* UTC date of the snapshot */
     uint32_t date_nsec;
     uint64_t vm_clock_nsec; /* VM clock relative to boot */
+    uint64_t disk_size;
 } QEMUSnapshotInfo;
 
 int bdrv_snapshot_find(BlockDriverState *bs, QEMUSnapshotInfo *sn_info,
-- 
1.7.1




reply via email to

[Prev in Thread] Current Thread [Next in Thread]