All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Xu <peterx@redhat.com>
To: Roman Kiryanov <rkir@google.com>
Cc: farosas@suse.de, qemu-devel@nongnu.org, whollins@google.com,
	jansene@google.com, jpcottin@google.com
Subject: Re: [PATCH v2] vmstate: validate VMStateDescription::fields upon registration
Date: Fri, 13 Mar 2026 10:30:44 -0400	[thread overview]
Message-ID: <abQflF_Ju2VI7pBa@x1.local> (raw)
In-Reply-To: <20260312210802.899803-1-rkir@google.com>

On Thu, Mar 12, 2026 at 09:08:02PM +0000, Roman Kiryanov wrote:
> The vmstate_save_state_v() function does not support
> NULL in VMStateDescription::fields and will crash if
> one is provided.
> 
> This change prevents this situation from happening
> by explicitly crashing earlier.
> 
> Suggested-by: Fabiano Rosas <farosas@suse.de>
> Suggested-by: Peter Xu <peterx@redhat.com>

Thanks for the generosity, all kudos to Fabiano.  Let's replace this line
with:

Reviewed-by: Peter Xu <peterx@redhat.com>

> Signed-off-by: Roman Kiryanov <rkir@google.com>
> ---
> v2:
>  - Moved the assert from vmstate_save_state_v to the registration
>    path (vmstate_register_with_alias_id) and the qtest validation path
>    (vmstate_check) to catch missing fields earlier.
> ---
>  migration/savevm.c | 5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/migration/savevm.c b/migration/savevm.c
> index 197c89e0e6..20c2b99231 100644
> --- a/migration/savevm.c
> +++ b/migration/savevm.c
> @@ -861,6 +861,8 @@ static void vmstate_check(const VMStateDescription *vmsd)
>      const VMStateField *field = vmsd->fields;
>      const VMStateDescription * const *subsection = vmsd->subsections;
>  
> +    assert(field || vmsd->unmigratable);
> +
>      if (field) {
>          while (field->name) {
>              if (field->flags & (VMS_STRUCT | VMS_VSTRUCT)) {
> @@ -900,6 +902,9 @@ int vmstate_register_with_alias_id(VMStateIf *obj, uint32_t instance_id,
>      /* If this triggers, alias support can be dropped for the vmsd. */
>      assert(alias_id == -1 || required_for_version >= vmsd->minimum_version_id);
>  
> +    /* If vmsd is migratable it MUST have valid fields. */
> +    assert(vmsd->fields || vmsd->unmigratable);
> +
>      se = g_new0(SaveStateEntry, 1);
>      se->version_id = vmsd->version_id;
>      se->section_id = savevm_state.global_section_id++;
> -- 
> 2.53.0.851.ga537e3e6e9-goog
> 

-- 
Peter Xu



  reply	other threads:[~2026-03-13 14:31 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-12 21:08 [PATCH v2] vmstate: validate VMStateDescription::fields upon registration Roman Kiryanov
2026-03-13 14:30 ` Peter Xu [this message]
2026-03-13 23:34   ` Roman Kiryanov

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=abQflF_Ju2VI7pBa@x1.local \
    --to=peterx@redhat.com \
    --cc=farosas@suse.de \
    --cc=jansene@google.com \
    --cc=jpcottin@google.com \
    --cc=qemu-devel@nongnu.org \
    --cc=rkir@google.com \
    --cc=whollins@google.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 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.