From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: Juan Quintela <quintela@redhat.com>
Cc: qemu-devel@nongnu.org, Peter Xu <peterx@redhat.com>
Subject: Re: [PATCH] postocpy: Check that postocpy fd's are not NULL
Date: Thu, 4 Nov 2021 12:27:11 +0000 [thread overview]
Message-ID: <YYPRn8geZ21UKsA/@work-vm> (raw)
In-Reply-To: <87r1bxgfvk.fsf@secure.mitica>
* Juan Quintela (quintela@redhat.com) wrote:
> "Dr. David Alan Gilbert" <dgilbert@redhat.com> wrote:
> > * Juan Quintela (quintela@redhat.com) wrote:
> >> If postcopy has finished, it frees the array.
> >> But vhost-user unregister it at cleanup time.
> >
> > I must admit to being confused as the double migrate case vs migrate
> > once and shutdown. It might just be an ordering thing?
> >
> > I notice that 'vhost_user_backend_cleanup' does:
> > if (u->postcopy_fd.handler) {
> > postcopy_unregister_shared_ufd(&u->postcopy_fd);
> > close(u->postcopy_fd.fd);
> > u->postcopy_fd.handler = NULL;
> > }
> >
> > where as the other caller, 'vhost_user_postcopy_end'
> > does:
> > postcopy_unregister_shared_ufd(&u->postcopy_fd);
> > close(u->postcopy_fd.fd);
> > u->postcopy_fd.handler = NULL;
>
> I think that we want ta make here the check to see if it has already
> been freed.
>
> > maybe it would be better to fix them to do the same if check?
>
> But even there, I think that it is more robust that we don't try to
> access a NULL pointer.
>
> I.e. there are two things that we can fix here:
> - postcopy unregister
> - vhost use of postcopy unregister
True we could indeed fix both as a belt-and-braces.
> > (Also note 'post*o*cpy' typo in title, and probably worth a
> > Fixes: c4f7538 ?)
>
> Sure.
>
> What do you think?
So yeh,
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Please do the other if as a separate patch sometime.
Dave
> Later, Juan.
>
> > Dave
> >
> >
> >> Signed-off-by: Juan Quintela <quintela@redhat.com>
> >> ---
> >> migration/postcopy-ram.c | 4 ++++
> >> 1 file changed, 4 insertions(+)
> >>
> >> diff --git a/migration/postcopy-ram.c b/migration/postcopy-ram.c
> >> index e721f69d0f..d18b5d05b2 100644
> >> --- a/migration/postcopy-ram.c
> >> +++ b/migration/postcopy-ram.c
> >> @@ -1457,6 +1457,10 @@ void postcopy_unregister_shared_ufd(struct PostCopyFD *pcfd)
> >> MigrationIncomingState *mis = migration_incoming_get_current();
> >> GArray *pcrfds = mis->postcopy_remote_fds;
> >>
> >> + if (!pcrfds) {
> >> + /* migration has already finished and freed the array */
> >> + return;
> >> + }
> >> for (i = 0; i < pcrfds->len; i++) {
> >> struct PostCopyFD *cur = &g_array_index(pcrfds, struct PostCopyFD, i);
> >> if (cur->fd == pcfd->fd) {
> >> --
> >> 2.33.1
> >>
>
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
prev parent reply other threads:[~2021-11-04 12:28 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-03 19:31 [PATCH] postocpy: Check that postocpy fd's are not NULL Juan Quintela
2021-11-03 20:00 ` Dr. David Alan Gilbert
2021-11-03 20:10 ` Juan Quintela
2021-11-04 12:27 ` Dr. David Alan Gilbert [this message]
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=YYPRn8geZ21UKsA/@work-vm \
--to=dgilbert@redhat.com \
--cc=peterx@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=quintela@redhat.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;
as well as URLs for NNTP newsgroup(s).