Re: [PATCH] xfrm: Fix oops in __xfrm_state_delete()

[Antony Antony <antony.antony@secunet.com>]

On Tue, Nov 01, 2022 at 12:39:48 +0800, Herbert Xu wrote:
> On Mon, Oct 31, 2022 at 04:26:12PM +0100, Thomas Jarosch wrote:
> >
> > diff --git a/net/key/af_key.c b/net/key/af_key.c
> > index c85df5b958d2..65a9ede62d65 100644
> > --- a/net/key/af_key.c
> > +++ b/net/key/af_key.c
> > @@ -3382,7 +3382,7 @@ static int pfkey_send_new_mapping(struct xfrm_state *x, xfrm_address_t *ipaddr,
> >  	hdr->sadb_msg_len = size / sizeof(uint64_t);
> >  	hdr->sadb_msg_errno = 0;
> >  	hdr->sadb_msg_reserved = 0;
> > -	hdr->sadb_msg_seq = x->km.seq = get_acqseq();

This line looks very odd.

> > +	hdr->sadb_msg_seq = get_acqseq();
> 
> This looks broken.  x->km.seq is part of the state which you are
> changing.  Shouldn't you do whatever xfrm_user does in the same
> situation?

xfrm_user sets msg_seq to zero in mapping change message. seq is only useful for
acquire message. I think setting to zero would be a better fix.

-	hdr->sadb_msg_seq = x->km.seq = get_acqseq();
+	hdr->sadb_msg_seq = 0;

While increasing x->km.seq in every call to pfkey_send_new_mapping()
could be an issue, would it alone explan the crash?

Tobias would pfkey_send_new_mapping() called in a default setting?