From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dan Carpenter Date: Wed, 29 Jul 2015 21:36:05 +0000 Subject: [patch] [SCSI] atp870u: 64 bit bug in probe() Message-Id: <20150729213605.GD21784@mwanda> List-Id: References: <20130904095002.GC13892@elgon.mountain> In-Reply-To: <20130904095002.GC13892@elgon.mountain> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: "James E.J. Bottomley" Cc: linux-scsi@vger.kernel.org, kernel-janitors@vger.kernel.org On 64 bit CPUs there is a memory corruption bug on probe(). It should be a u32 pointer instead of an unsigned long pointer or we write past the end of the setupdata[] array. Signed-off-by: Dan Carpenter --- Someone reported in 2003 that probe has a NULL deref so maybe it's related to this memory corruption? https://bugzilla.kernel.org/show_bug.cgi?id18 If only we had applied this patch when I originally sent it two years ago, then it would only be 10 years too late instead of 12! :P diff --git a/drivers/scsi/atp870u.c b/drivers/scsi/atp870u.c index 05301bc..62acabd 100644 --- a/drivers/scsi/atp870u.c +++ b/drivers/scsi/atp870u.c @@ -2791,11 +2791,11 @@ next_fblk_885: p->global_map[m]= 0; for (k=0; k < 4; k++) { outw(n++,base_io + 0x3c); - ((unsigned long *)&setupdata[m][0])[k]=inl(base_io + 0x38); + ((u32 *)&setupdata[m][0])[k]=inl(base_io + 0x38); } for (k=0; k < 4; k++) { outw(n++,base_io + 0x3c); - ((unsigned long *)&p->sp[m][0])[k]=inl(base_io + 0x38); + ((u32 *)&p->sp[m][0])[k]=inl(base_io + 0x38); } n += 8; }