netdev - Re: [PATCH bpf-next v10 2/9] bpf: Propagate modified uaddrlen from cgroup sockaddr programs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20231011194520.60480-1-kuniyu@amazon.com>
Date: Wed, 11 Oct 2023 12:45:20 -0700
From: Kuniyuki Iwashima <kuniyu@...zon.com>
To: <daan.j.demeyer@...il.com>
CC: <bpf@...r.kernel.org>, <kernel-team@...a.com>, <kuniyu@...zon.com>,
	<martin.lau@...ux.dev>, <netdev@...r.kernel.org>
Subject: Re: [PATCH bpf-next v10 2/9] bpf: Propagate modified uaddrlen from cgroup sockaddr programs

From: Daan De Meyer <daan.j.demeyer@...il.com>
Date: Wed, 11 Oct 2023 21:09:33 +0200
> > From: Daan De Meyer <daan.j.demeyer@...il.com>
> > Date: Wed, 11 Oct 2023 20:37:49 +0200
> > > > > @@ -1483,11 +1488,18 @@ int __cgroup_bpf_run_filter_sock_addr(struct sock *sk,
> > > > >       if (!ctx.uaddr) {
> > > > >               memset(&unspec, 0, sizeof(unspec));
> > > > >               ctx.uaddr = (struct sockaddr *)&unspec;
> > > > > -     }
> > > > > +             ctx.uaddrlen = 0;
> > > > > +     } else
> > > > > +             ctx.uaddrlen = *uaddrlen;
> > > > >
> > > > >       cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data);
> > > > > -     return bpf_prog_run_array_cg(&cgrp->bpf, atype, &ctx, bpf_prog_run,
> > > > > -                                  0, flags);
> > > > > +     ret = bpf_prog_run_array_cg(&cgrp->bpf, atype, &ctx, bpf_prog_run,
> > > > > +                                 0, flags);
> > > > > +
> > > > > +     if (!ret && uaddrlen)
> > > >
> > > > nit: no need to check uaddrlen here or maybe check ctx.uaddrlen.
> > >
> > > Are you sure? uaddrlen can still be NULL if uaddr is also NULL
> >
> > How?  In the patch 2 and 4, it seems uaddrlen always points to an
> > actual variable.
> 
> Right, I was assuming we don't know for sure how callers are calling
> this function. It is right that right now no caller calls it with uaddrlen set
> to NULL.

We need not to be defensive for future in-kernel users who should take
care of that properly.


> 
> It still seems like a good idea to check for uaddr instead of uaddrlen though,
> to mimic the same check that is done earlier in this function.

Sounds good.