netdev - Re: [PATCH] Prevent reading uninitialized memory with socket filters

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

Date:	Wed, 10 Nov 2010 08:22:51 +0100
From:	Eric Dumazet <eric.dumazet@...il.com>
To:	David Miller <davem@...emloft.net>
Cc:	drosenberg@...curity.com, netdev@...r.kernel.org,
	stable@...nel.org, security@...nel.org
Subject: Re: [PATCH] Prevent reading uninitialized memory with socket
 filters

Le mercredi 10 novembre 2010 à 06:53 +0100, Eric Dumazet a écrit :
> Le mardi 09 novembre 2010 à 21:28 -0800, David Miller a écrit :
> > From: Dan Rosenberg <drosenberg@...curity.com>
> > Date: Tue, 09 Nov 2010 17:28:44 -0500
> > 
> > > The "mem" array used as scratch space for socket filters is not
> > > initialized, allowing unprivileged users to leak kernel stack bytes.
> > > 
> > > Signed-off-by: Dan Rosenberg <drosenberg@...curity.com>
> > 
> > Prove it.
> 
> And once done, add the checks in sk_chk_filter() ?
> 
> Allow a load of mem[X] only if a prior store of mem[X] is proven.
> 
> 

This seems complex, and might fail on some valid filters.

What about the following patch then ?

[PATCH] filter: make sure filters dont read uninitialized memory

There is a possibility malicious users can get limited information about
uninitialized stack mem array. Even if sk_run_filter() result is bound
to packet length (0 .. 65535), we could imagine this can be used by
hostile user.

Initializing mem[] array, like Dan Rosenberg suggested in his patch is
expensive since most filters dont even use this array.

Its hard to make the filter validation in sk_chk_filter(), because of
the jumps. This might be done later.

In this patch, I use a bitmap (a single long var) so that only filters
using mem[] loads/stores pay the price of added security checks.

For other filters, additional cost is a single instruction.

Reported-by: Dan Rosenberg <drosenberg@...curity.com>
Signed-off-by: Eric Dumazet <eric.dumazet@...il.com>
---
 net/core/filter.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/net/core/filter.c b/net/core/filter.c
index 7beaec3..4d84dc2 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -117,10 +117,12 @@ unsigned int sk_run_filter(struct sk_buff *skb, struct sock_filter *filter, int
 	u32 A = 0;			/* Accumulator */
 	u32 X = 0;			/* Index Register */
 	u32 mem[BPF_MEMWORDS];		/* Scratch Memory Store */
+	unsigned long memvalid = 0;
 	u32 tmp;
 	int k;
 	int pc;
 
+	BUILD_BUG_ON(BPF_MEMWORDS > BITS_PER_LONG);
 	/*
 	 * Process array of filter instructions.
 	 */
@@ -264,10 +266,12 @@ load_b:
 			X = fentry->k;
 			continue;
 		case BPF_S_LD_MEM:
-			A = mem[fentry->k];
+			A = (memvalid & (1UL << fentry->k)) ?
+				mem[fentry->k] : 0;
 			continue;
 		case BPF_S_LDX_MEM:
-			X = mem[fentry->k];
+			X = (memvalid & (1UL << fentry->k)) ?
+				mem[fentry->k] : 0;
 			continue;
 		case BPF_S_MISC_TAX:
 			X = A;
@@ -280,9 +284,11 @@ load_b:
 		case BPF_S_RET_A:
 			return A;
 		case BPF_S_ST:
+			memvalid |= 1UL << fentry->k;
 			mem[fentry->k] = A;
 			continue;
 		case BPF_S_STX:
+			memvalid |= 1UL << fentry->k;
 			mem[fentry->k] = X;
 			continue;
 		default:


--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html