public inbox for pve-devel@lists.proxmox.com
 help / color / mirror / Atom feed
From: Fiona Ebner <f.ebner@proxmox.com>
To: pve-devel@lists.proxmox.com
Subject: [pve-devel] [HACK qemu 07/13] block/block-copy: always consider source cluster size too
Date: Thu, 25 Jan 2024 15:41:43 +0100	[thread overview]
Message-ID: <20240125144149.216064-8-f.ebner@proxmox.com> (raw)
In-Reply-To: <20240125144149.216064-1-f.ebner@proxmox.com>

With backup fleecing, it might be necessary to discard the source.
There will be an assertion failure if bitmaps on the source side have
a bigger granularity than the block copy's cluster size, so just
consider the source side too.

This also supersedes the hunk in block/backup.c added by
"PVE-Backup: add backup-dump block driver".

Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
---

Still need to wait on a response from upstream. For now this hack, so
that the RFC as a whole doesn't have to wait.

 block/backup.c     |  8 --------
 block/block-copy.c | 22 ++++++++++++++--------
 2 files changed, 14 insertions(+), 16 deletions(-)

diff --git a/block/backup.c b/block/backup.c
index 51f9d28115..7950bff27e 100644
--- a/block/backup.c
+++ b/block/backup.c
@@ -435,14 +435,6 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
     }
 
     cluster_size = block_copy_cluster_size(bcs);
-    if (cluster_size < 0) {
-        goto error;
-    }
-
-    BlockDriverInfo bdi;
-    if (bdrv_get_info(bs, &bdi) == 0) {
-        cluster_size = MAX(cluster_size, bdi.cluster_size);
-    }
 
     if (perf->max_chunk && perf->max_chunk < cluster_size) {
         error_setg(errp, "Required max-chunk (%" PRIi64 ") is less than backup "
diff --git a/block/block-copy.c b/block/block-copy.c
index 05cfccfda8..cf750ef670 100644
--- a/block/block-copy.c
+++ b/block/block-copy.c
@@ -310,13 +310,19 @@ void block_copy_set_copy_opts(BlockCopyState *s, bool use_copy_range,
     }
 }
 
-static int64_t block_copy_calculate_cluster_size(BlockDriverState *target,
+static int64_t block_copy_calculate_cluster_size(BlockDriverState *source,
+                                                 BlockDriverState *target,
                                                  Error **errp)
 {
     int ret;
+    int64_t source_cluster_size = BLOCK_COPY_CLUSTER_SIZE_DEFAULT;
     BlockDriverInfo bdi;
     bool target_does_cow = bdrv_backing_chain_next(target);
 
+    if (bdrv_get_info(source, &bdi) == 0) {
+        source_cluster_size = MAX(source_cluster_size, bdi.cluster_size);
+    }
+
     /*
      * If there is no backing file on the target, we cannot rely on COW if our
      * backup cluster size is smaller than the target cluster size. Even for
@@ -327,11 +333,11 @@ static int64_t block_copy_calculate_cluster_size(BlockDriverState *target,
         /* Cluster size is not defined */
         warn_report("The target block device doesn't provide "
                     "information about the block size and it doesn't have a "
-                    "backing file. The default block size of %u bytes is "
+                    "backing file. The source's or default block size of %ld bytes is "
                     "used. If the actual block size of the target exceeds "
-                    "this default, the backup may be unusable",
-                    BLOCK_COPY_CLUSTER_SIZE_DEFAULT);
-        return BLOCK_COPY_CLUSTER_SIZE_DEFAULT;
+                    "this value, the backup may be unusable",
+                    source_cluster_size);
+        return source_cluster_size;
     } else if (ret < 0 && !target_does_cow) {
         error_setg_errno(errp, -ret,
             "Couldn't determine the cluster size of the target image, "
@@ -341,10 +347,10 @@ static int64_t block_copy_calculate_cluster_size(BlockDriverState *target,
         return ret;
     } else if (ret < 0 && target_does_cow) {
         /* Not fatal; just trudge on ahead. */
-        return BLOCK_COPY_CLUSTER_SIZE_DEFAULT;
+        return source_cluster_size;
     }
 
-    return MAX(BLOCK_COPY_CLUSTER_SIZE_DEFAULT, bdi.cluster_size);
+    return MAX(source_cluster_size, bdi.cluster_size);
 }
 
 BlockCopyState *block_copy_state_new(BdrvChild *source, BdrvChild *target,
@@ -358,7 +364,7 @@ BlockCopyState *block_copy_state_new(BdrvChild *source, BdrvChild *target,
     BdrvDirtyBitmap *copy_bitmap;
     bool is_fleecing;
 
-    cluster_size = block_copy_calculate_cluster_size(target->bs, errp);
+    cluster_size = block_copy_calculate_cluster_size(source->bs, target->bs, errp);
     if (cluster_size < 0) {
         return NULL;
     }
-- 
2.39.2





  parent reply	other threads:[~2024-01-25 14:41 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-25 14:41 [pve-devel] [RFC qemu/guest-common/manager/qemu-server/docs 00/13] fix #4136: implement backup fleecing Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [PATCH qemu 01/13] backup: factor out gathering device info into helper Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [PATCH qemu 02/13] backup: get device info: code cleanup Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [PATCH qemu 03/13] block/io: clear BDRV_BLOCK_RECURSE flag after recursing in bdrv_co_block_status Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [RFC qemu 04/13] block/copy-before-write: create block_copy bitmap in filter node Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [RFC qemu 05/13] qapi: blockdev-backup: add discard-source parameter Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [HACK qemu 06/13] block/{copy-before-write, snapshot-access}: implement bdrv_co_get_info driver callback Fiona Ebner
2024-01-29 14:35   ` Fiona Ebner
2024-01-25 14:41 ` Fiona Ebner [this message]
2024-01-25 14:41 ` [pve-devel] [RFC qemu 08/13] PVE backup: add fleecing option Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [RFC guest-common 09/13] vzdump: schema: add fleecing property string Fiona Ebner
2024-01-29 15:41   ` Fiona Ebner
2024-01-30 14:03     ` DERUMIER, Alexandre
2024-02-01  8:28       ` Fiona Ebner
2024-02-01 12:39     ` DERUMIER, Alexandre
2024-02-01 13:11       ` Fiona Ebner
2024-02-01 13:20         ` DERUMIER, Alexandre
2024-02-01 13:27           ` Fiona Ebner
2024-02-01 21:33             ` DERUMIER, Alexandre
2024-02-02  8:30               ` Fiona Ebner
2024-02-01 13:30           ` Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [RFC manager 10/13] vzdump: handle new 'fleecing' " Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [RFC qemu-server 11/13] backup: disk info: also keep track of size Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [RFC qemu-server 12/13] backup: implement fleecing option Fiona Ebner
2024-01-29 15:28   ` Fiona Ebner
2024-01-25 14:41 ` [pve-devel] [RFC docs 13/13] vzdump: add section about backup fleecing Fiona Ebner
2024-01-25 16:13   ` Dietmar Maurer
2024-01-25 16:41     ` DERUMIER, Alexandre
2024-01-25 18:18       ` Dietmar Maurer
2024-01-26  8:39         ` Fiona Ebner
2024-01-25 16:02 ` [pve-devel] [RFC qemu/guest-common/manager/qemu-server/docs 00/13] fix #4136: implement " DERUMIER, Alexandre

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240125144149.216064-8-f.ebner@proxmox.com \
    --to=f.ebner@proxmox.com \
    --cc=pve-devel@lists.proxmox.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Service provided by Proxmox Server Solutions GmbH | Privacy | Legal