Hi On Tue, Aug 31, 2021 at 9:17 PM Michael Tokarev <[email protected]> wrote:
> 31.08.2021 01:54, Michael Tokarev wrote: > > Linux kernel can return size of af_unix socket to be > > one byte larger than sockaddr_un structure - adding > > the trailing zero byte. > > > > Signed-off-by: Michael Tokarev <[email protected]> > > Fixes: 4cfd970ec188558daa6214f26203fe553fb1e01f (first in 6.1.0) > > Cc: [email protected] > > > > diff --git a/util/qemu-sockets.c b/util/qemu-sockets.c > > index f2f3676d1f..83926dc2bc 100644 > > --- a/util/qemu-sockets.c > > +++ b/util/qemu-sockets.c > > @@ -1345,8 +1345,9 @@ socket_sockaddr_to_address_unix(struct > sockaddr_storage *sa, > > SocketAddress *addr; > > struct sockaddr_un *su = (struct sockaddr_un *)sa; > > > > + /* kernel might have added \0 terminator to non-abstract socket */ > > assert(salen >= sizeof(su->sun_family) + 1 && > > - salen <= sizeof(struct sockaddr_un)); > > + salen <= sizeof(struct sockaddr_un) + su->sun_path[0] ? 1 : > 0); > > > > addr = g_new0(SocketAddress, 1); > > addr->type = SOCKET_ADDRESS_TYPE_UNIX; > > Actually, this is not sufficient. > > While this change fixes one issue (the famous trailing null byte \0), > the actual assertion failure occurs because salen = 2, ie, too SMALL, > not too large. > > So it looks like libvirt provides an unnamed socket there, -- > maybe from a socketpair(2)? > Yes Ok, I guess it should still check for salen >= sizeof(su->sun_family) and then modify if (salen > sizeof(su->sun_family) && !su->sun_path[0]) { > Hwell.. > hmm, too bad we didn't catch it during RC! (strange that it seems to hit Debian libvirt/virt-manager users and apparently not on Fedora)
