All of lore.kernel.org
 help / color / mirror / Atom feed
From: Kevin Wolf <kwolf@redhat.com>
To: Max Reitz <mreitz@redhat.com>
Cc: qemu-devel@nongnu.org, qemu-block@nongnu.org
Subject: Re: [PATCH 4/4] block: Use blk_make_empty() after commits
Date: Tue, 28 Apr 2020 17:03:11 +0200	[thread overview]
Message-ID: <20200428150311.GK5789@linux.fritz.box> (raw)
In-Reply-To: <20200428132629.796753-5-mreitz@redhat.com>

Am 28.04.2020 um 15:26 hat Max Reitz geschrieben:
> bdrv_commit() already has a BlockBackend pointing to the BDS that we
> want to empty, it just has the wrong permissions.
> 
> qemu-img commit has no BlockBackend pointing to the old backing file
> yet, but introducing one is simple.
> 
> After this commit, bdrv_make_empty() is the only remaining caller of
> BlockDriver.bdrv_make_empty().
> 
> Signed-off-by: Max Reitz <mreitz@redhat.com>
> ---
>  block/commit.c |  8 +++++++-
>  qemu-img.c     | 19 ++++++++++++++-----
>  2 files changed, 21 insertions(+), 6 deletions(-)
> 
> diff --git a/block/commit.c b/block/commit.c
> index 8e672799af..24720ba67d 100644
> --- a/block/commit.c
> +++ b/block/commit.c
> @@ -493,10 +493,16 @@ int bdrv_commit(BlockDriverState *bs)
>      }
>  
>      if (drv->bdrv_make_empty) {
> -        ret = drv->bdrv_make_empty(bs);
> +        ret = blk_set_perm(src, BLK_PERM_WRITE, BLK_PERM_ALL, NULL);

This is very likely to fail because the common case is that the source
node is attached to a guest device that doesn't share writes.
(qemu-iotests 131 and 274 catch this.)

So I think after my theoretical comment in patch 1, this is the
practical reason why we need WRITE_UNCHANGED rather than WRITE.

Also, why don't you take this permission from the start so that we would
error out right away rather than failing after waiting for the all the
data to be copied?

>          if (ret < 0) {
>              goto ro_cleanup;
>          }
> +
> +        ret = blk_make_empty(src, NULL);
> +        if (ret < 0) {
> +            goto ro_cleanup;
> +        }
> +
>          blk_flush(src);
>      }
>  
> diff --git a/qemu-img.c b/qemu-img.c
> index 821cbf610e..a5e8659867 100644
> --- a/qemu-img.c
> +++ b/qemu-img.c
> @@ -1065,11 +1065,20 @@ static int img_commit(int argc, char **argv)
>          goto unref_backing;
>      }
>  
> -    if (!drop && bs->drv->bdrv_make_empty) {
> -        ret = bs->drv->bdrv_make_empty(bs);
> -        if (ret) {
> -            error_setg_errno(&local_err, -ret, "Could not empty %s",
> -                             filename);
> +    if (!drop) {
> +        BlockBackend *old_backing_blk;
> +
> +        old_backing_blk = blk_new_with_bs(bs, BLK_PERM_WRITE, BLK_PERM_ALL,
> +                                          &local_err);

Oh, you actually depend on another series that you didn't mention in
the cover letter.

> +        if (!old_backing_blk) {
> +            goto unref_backing;
> +        }
> +        ret = blk_make_empty(old_backing_blk, &local_err);
> +        blk_unref(old_backing_blk);
> +        if (ret == -ENOTSUP) {
> +            error_free(local_err);
> +            local_err = NULL;
> +        } else if (ret < 0) {
>              goto unref_backing;
>          }
>      }

Kevin



  parent reply	other threads:[~2020-04-28 15:07 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-28 13:26 [PATCH 0/4] block: Do not call BlockDriver.bdrv_make_empty() directly Max Reitz
2020-04-28 13:26 ` [PATCH 1/4] block: Add bdrv_make_empty() Max Reitz
2020-04-28 13:53   ` Eric Blake
2020-04-28 14:01     ` Kevin Wolf
2020-04-28 14:07       ` Eric Blake
2020-04-28 14:16         ` Kevin Wolf
2020-04-28 14:25           ` Eric Blake
2020-04-28 14:21   ` Kevin Wolf
2020-04-29  7:39     ` Max Reitz
2020-04-28 13:26 ` [PATCH 2/4] block: Use bdrv_make_empty() where possible Max Reitz
2020-04-28 13:54   ` Eric Blake
2020-04-28 15:03   ` Kevin Wolf
2020-04-28 13:26 ` [PATCH 3/4] block: Add blk_make_empty() Max Reitz
2020-04-28 13:55   ` Eric Blake
2020-04-28 14:28     ` Eric Blake
2020-04-28 14:47   ` Kevin Wolf
2020-04-29  7:39     ` Max Reitz
2020-04-28 13:26 ` [PATCH 4/4] block: Use blk_make_empty() after commits Max Reitz
2020-04-28 14:07   ` Eric Blake
2020-04-29  7:58     ` Max Reitz
2020-04-28 15:03   ` Kevin Wolf [this message]
2020-04-29  8:01     ` Max Reitz
2020-04-28 13:38 ` [PATCH 0/4] block: Do not call BlockDriver.bdrv_make_empty() directly no-reply
2020-04-28 13:43 ` no-reply
2020-04-28 13:57   ` Eric Blake
2020-04-28 13:48 ` no-reply
2020-04-28 13:49 ` Eric Blake
2020-04-28 14:05   ` Eric Blake
2020-04-28 14:53 ` no-reply
2020-04-28 14:57 ` no-reply
2020-04-28 15:02 ` no-reply

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=20200428150311.GK5789@linux.fritz.box \
    --to=kwolf@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.