From: Michael Roth Date: Tue, 24 Sep 2019 15:35:09 +0000 (-0500) Subject: slirp: Fix heap overflow in ip_reass on big packet input X-Git-Tag: v3.1.1.1~2 X-Git-Url: http://git.ipfire.org/?a=commitdiff_plain;h=28c1dde9aa2a22724f81134035959d1a33a57690;p=thirdparty%2Fqemu.git slirp: Fix heap overflow in ip_reass on big packet input When the first fragment does not fit in the preallocated buffer, q will already be pointing to the ext buffer, so we mustn't try to update it. Signed-off-by: Samuel Thibault (from libslirp.git commit 126c04acbabd7ad32c2b018fe10dfac2a3bc1210) (from libslirp.git commit e0be80430c390bce181ea04dfcdd6ea3dfa97de1) *squash in e0be80 (clarifying comments) Signed-off-by: Michael Roth --- diff --git a/slirp/ip_input.c b/slirp/ip_input.c index 348e1dca5ab..6e291eabf00 100644 --- a/slirp/ip_input.c +++ b/slirp/ip_input.c @@ -334,6 +334,8 @@ insert: q = fp->frag_link.next; m = dtom(slirp, q); + int was_ext = m->m_flags & M_EXT; + q = (struct ipasfrag *) q->ipf_next; while (q != (struct ipasfrag*)&fp->frag_link) { struct mbuf *t = dtom(slirp, q); @@ -350,13 +352,12 @@ insert: q = fp->frag_link.next; /* - * If the fragments concatenated to an mbuf that's - * bigger than the total size of the fragment, then and - * m_ext buffer was alloced. But fp->ipq_next points to - * the old buffer (in the mbuf), so we must point ip - * into the new buffer. + * If the fragments concatenated to an mbuf that's bigger than the total + * size of the fragment and the mbuf was not already using an m_ext buffer, + * then an m_ext buffer was alloced. But fp->ipq_next points to the old + * buffer (in the mbuf), so we must point ip into the new buffer. */ - if (m->m_flags & M_EXT) { + if (!was_ext && m->m_flags & M_EXT) { int delta = (char *)q - m->m_dat; q = (struct ipasfrag *)(m->m_ext + delta); }