* [PATCH] sctp: use GFP_USER for user-controlled kmalloc [not found] <CACT4Y+a_V5WQZNEnYkuA3Xc5qCWmLV3oScNeNiATZm-wW5eg3Q@mail.gmail.comc> @ 2015-11-30 16:32 ` Marcelo Ricardo Leitner 2015-12-01 10:46 ` David Laight 2015-12-03 4:40 ` David Miller 0 siblings, 2 replies; 4+ messages in thread From: Marcelo Ricardo Leitner @ 2015-11-30 16:32 UTC (permalink / raw) To: netdev Cc: linux-sctp, Vlad Yasevich, Neil Horman, daniel, linux-kernel, davem, syzkaller, dvyukov, kcc, glider, sasha.levin, edumazet Dmitry Vyukov reported that the user could trigger a kernel warning by using a large len value for getsockopt SCTP_GET_LOCAL_ADDRS, as that value directly affects the value used as a kmalloc() parameter. This patch thus switches the allocation flags from all user-controllable kmalloc size to GFP_USER to put some more restrictions on it and also disables the warn, as they are not necessary. Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Acked-by: Daniel Borkmann <daniel@iogearbox.net> --- net/sctp/socket.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 897c01c029cab3d5805cc56b0964c70e06f4143a..676b3bb092e16848fd1c822e1c999af4a2ef198d 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -972,7 +972,7 @@ static int sctp_setsockopt_bindx(struct sock *sk, return -EFAULT; /* Alloc space for the address array in kernel memory. */ - kaddrs = kmalloc(addrs_size, GFP_KERNEL); + kaddrs = kmalloc(addrs_size, GFP_USER | __GFP_NOWARN); if (unlikely(!kaddrs)) return -ENOMEM; @@ -4928,7 +4928,7 @@ static int sctp_getsockopt_local_addrs(struct sock *sk, int len, to = optval + offsetof(struct sctp_getaddrs, addrs); space_left = len - offsetof(struct sctp_getaddrs, addrs); - addrs = kmalloc(space_left, GFP_KERNEL); + addrs = kmalloc(space_left, GFP_USER | __GFP_NOWARN); if (!addrs) return -ENOMEM; -- 2.5.0 ^ permalink raw reply related [flat|nested] 4+ messages in thread
* RE: [PATCH] sctp: use GFP_USER for user-controlled kmalloc 2015-11-30 16:32 ` [PATCH] sctp: use GFP_USER for user-controlled kmalloc Marcelo Ricardo Leitner @ 2015-12-01 10:46 ` David Laight 2015-12-01 11:29 ` Daniel Borkmann 2015-12-03 4:40 ` David Miller 1 sibling, 1 reply; 4+ messages in thread From: David Laight @ 2015-12-01 10:46 UTC (permalink / raw) To: 'Marcelo Ricardo Leitner', netdev@vger.kernel.org Cc: linux-sctp@vger.kernel.org, Vlad Yasevich, Neil Horman, daniel@iogearbox.net, linux-kernel@vger.kernel.org, davem@davemloft.net, syzkaller@googlegroups.com, dvyukov@google.com, kcc@google.com, glider@google.com, sasha.levin@oracle.com, edumazet@google.com From: Marcelo Ricardo Leitner > Sent: 30 November 2015 16:33 > Dmitry Vyukov reported that the user could trigger a kernel warning by > using a large len value for getsockopt SCTP_GET_LOCAL_ADDRS, as that > value directly affects the value used as a kmalloc() parameter. > > This patch thus switches the allocation flags from all user-controllable > kmalloc size to GFP_USER to put some more restrictions on it and also > disables the warn, as they are not necessary. ISTM that the code should put some 'sanity limit' on that size before allocating the kernel buffer. David ^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] sctp: use GFP_USER for user-controlled kmalloc 2015-12-01 10:46 ` David Laight @ 2015-12-01 11:29 ` Daniel Borkmann 0 siblings, 0 replies; 4+ messages in thread From: Daniel Borkmann @ 2015-12-01 11:29 UTC (permalink / raw) To: David Laight, 'Marcelo Ricardo Leitner', netdev@vger.kernel.org Cc: linux-sctp@vger.kernel.org, Vlad Yasevich, Neil Horman, linux-kernel@vger.kernel.org, davem@davemloft.net, syzkaller@googlegroups.com, dvyukov@google.com, kcc@google.com, glider@google.com, sasha.levin@oracle.com, edumazet@google.com On 12/01/2015 11:46 AM, David Laight wrote: > From: Marcelo Ricardo Leitner >> Sent: 30 November 2015 16:33 >> Dmitry Vyukov reported that the user could trigger a kernel warning by >> using a large len value for getsockopt SCTP_GET_LOCAL_ADDRS, as that >> value directly affects the value used as a kmalloc() parameter. >> >> This patch thus switches the allocation flags from all user-controllable >> kmalloc size to GFP_USER to put some more restrictions on it and also >> disables the warn, as they are not necessary. > > ISTM that the code should put some 'sanity limit' on that > size before allocating the kernel buffer. One could do that in addition, but this buffer has just a short lifetime and by using GFP_USER hardwall restrictions apply already. ^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] sctp: use GFP_USER for user-controlled kmalloc 2015-11-30 16:32 ` [PATCH] sctp: use GFP_USER for user-controlled kmalloc Marcelo Ricardo Leitner 2015-12-01 10:46 ` David Laight @ 2015-12-03 4:40 ` David Miller 1 sibling, 0 replies; 4+ messages in thread From: David Miller @ 2015-12-03 4:40 UTC (permalink / raw) To: marcelo.leitner Cc: netdev, linux-sctp, vyasevich, nhorman, daniel, linux-kernel, syzkaller, dvyukov, kcc, glider, sasha.levin, edumazet From: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> Date: Mon, 30 Nov 2015 14:32:54 -0200 > Dmitry Vyukov reported that the user could trigger a kernel warning by > using a large len value for getsockopt SCTP_GET_LOCAL_ADDRS, as that > value directly affects the value used as a kmalloc() parameter. > > This patch thus switches the allocation flags from all user-controllable > kmalloc size to GFP_USER to put some more restrictions on it and also > disables the warn, as they are not necessary. > > Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com> > Acked-by: Daniel Borkmann <daniel@iogearbox.net> Applied. ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2015-12-03 4:40 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <CACT4Y+a_V5WQZNEnYkuA3Xc5qCWmLV3oScNeNiATZm-wW5eg3Q@mail.gmail.comc>
2015-11-30 16:32 ` [PATCH] sctp: use GFP_USER for user-controlled kmalloc Marcelo Ricardo Leitner
2015-12-01 10:46 ` David Laight
2015-12-01 11:29 ` Daniel Borkmann
2015-12-03 4:40 ` David Miller
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).