From 85c877d5072866aadbe8ed0c3e0590fbb5e16788 Mon Sep 17 00:00:00 2001
From: Fabian Greffrath <fabian@greffrath.com>
Date: Thu, 28 Sep 2017 12:15:04 +0200
Subject: [PATCH] double64_init: Check psf->sf.channels against upper bound

This prevents division by zero later in the code.

While the trivial case to catch this (i.e. sf.channels < 1) has already
been covered, a crafted file may report a number of channels that is
so high (i.e. > INT_MAX/sizeof(double)) that it "somehow" gets
miscalculated to zero (if this makes sense) in the determination of the
blockwidth. Since we only support a limited number of channels anyway,
make sure to check here as well.

CVE-2017-14634

Closes: https://github.com/erikd/libsndfile/issues/318
Signed-off-by: Erik de Castro Lopo <erikd@mega-nerd.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
---
 src/double64.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/double64.c b/src/double64.c
index b318ea86..78dfef7f 100644
--- a/src/double64.c
+++ b/src/double64.c
@@ -91,7 +91,7 @@ int
 double64_init	(SF_PRIVATE *psf)
 {	static int double64_caps ;
 
-	if (psf->sf.channels < 1)
+	if (psf->sf.channels < 1 || psf->sf.channels > SF_MAX_CHANNELS)
 	{	psf_log_printf (psf, "double64_init : internal error : channels = %d\n", psf->sf.channels) ;
 		return SFE_INTERNAL ;
 		} ;
-- 
2.11.0