All of lore.kernel.org
 help / color / mirror / Atom feed
From: Fabiano Rosas <farosas@suse.de>
To: Zhu Wei <zhuwei@sangfor.com.cn>, qemu-devel@nongnu.org
Cc: Paolo Bonzini <pbonzini@redhat.com>, Peter Xu <peterx@redhat.com>,
	zhuwei <zhuwei@sangfor.com.cn>,
	"zachwade . k" <zachwade.k@gmail.com.cn>
Subject: Re: [PATCH 1/1] migration: free migration blocker error in migrate_del_blocker() stub
Date: Mon, 16 Nov 2026 15:19:30 -0300	[thread overview]
Message-ID: <87se10skp9.fsf@suse.de> (raw)
In-Reply-To: <20261111122711.3167511-2-zhuwei@sangfor.com.cn>

Zhu Wei <zhuwei@sangfor.com.cn> writes:

> From: niming <niming@sangfor.com.cn>
>
> The migrate_del_blocker() implementation in stubs/migr-blocker.c was an
> empty no-op, while the real one in migration/migration.c takes ownership
> of the Error from the migration blocker, frees it and then clears the
> pointer:
>
>     void migrate_del_blocker(Error **reasonp)
>     {
>         if (*reasonp) {
>             ...
>             error_free(*reasonp);
>             *reasonp = NULL;
>         }
>     }
>
> Command line tools such as qemu-img link against the stub API. Block
> drivers like vmdk allocate a migration blocker Error during open() (see
> error_setg() in vmdk_open(), "does not support live migration") and pass
> it to migrate_add_blocker_normal(). Because the stub migrate_add_blocker*()
> is a no-op it does not take ownership of the Error, so on close the
> stub migrate_del_blocker() has to free it itself; leaving the body empty
> leaks the error message.
>
> This becomes visible when running under ASan: "qemu-img info foo.vmdk"
> leaks the migration-blocker error and hangs / fails to exit when the
> image is closed.
>
> Free the Error in the stub migrate_del_blocker() and set the pointer to
> NULL to match the semantics of the real implementation. This also covers
> every other driver that uses a migration blocker through the stub.
>
> Signed-off-by: niming <niming@sangfor.com.cn>
> Signed-off-by: zhuwei <zhuwei@sangfor.com.cn>
> ---
>  stubs/migr-blocker.c | 7 +++++++
>  1 file changed, 7 insertions(+)
>
> diff --git a/stubs/migr-blocker.c b/stubs/migr-blocker.c
> index e54c7160d3..575b65f183 100644
> --- a/stubs/migr-blocker.c
> +++ b/stubs/migr-blocker.c
> @@ -1,5 +1,6 @@
>  #include "qemu/osdep.h"
>  #include "migration/blocker.h"
> +#include "qapi/error.h"
>  
>  int migrate_add_blocker(Error **reasonp, Error **errp)
>  {
> @@ -18,4 +19,12 @@ int migrate_add_blocker_modes(Error **reasonp, unsigned modes, Error **errp)
>  
>  void migrate_del_blocker(Error **reasonp)
>  {
> +    /*
> +     * migrate_add_blocker*() is a no-op here, so nobody owns the Error;
> +     * free it to avoid a leak (e.g. "qemu-img info foo.vmdk").
> +     */
> +    if (reasonp && *reasonp) {
> +        error_free(*reasonp);
> +        *reasonp = NULL;
> +    }
>  }

Reviewed-by: Fabiano Rosas <farosas@suse.de>


  reply	other threads:[~2026-11-16 18:20 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-11 12:27 [PATCH 0/1] migration: free migration blocker error in migrate_del_blocker() stub Zhu Wei
2026-11-11 12:27 ` [PATCH 1/1] " Zhu Wei
2026-11-16 18:19   ` Fabiano Rosas [this message]
  -- strict thread matches above, loose matches on Subject: below --
2026-11-13 15:11 [PATCH 0/1] " Zhu Wei
2026-11-13 15:11 ` [PATCH 1/1] " Zhu Wei

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=87se10skp9.fsf@suse.de \
    --to=farosas@suse.de \
    --cc=pbonzini@redhat.com \
    --cc=peterx@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=zachwade.k@gmail.com.cn \
    --cc=zhuwei@sangfor.com.cn \
    /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.