linux-kernel - [PATCH 4.15 100/122] EDAC, sb_edac: Fix out of bound writes during DIMM configuration on KNL

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]

Message-Id: <20180307191743.848319164@linuxfoundation.org>
Date:   Wed,  7 Mar 2018 11:38:32 -0800
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     linux-kernel@...r.kernel.org
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        stable@...r.kernel.org, Dan Carpenter <dan.carpenter@...cle.com>,
        Anna Karbownik <anna.karbownik@...el.com>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Tony Luck <tony.luck@...el.com>, jim.m.snow@...el.com,
        krzysztof.paliswiat@...el.com, lukasz.odzioba@...el.com,
        qiuxu.zhuo@...el.com, linux-edac <linux-edac@...r.kernel.org>,
        Borislav Petkov <bp@...e.de>
Subject: [PATCH 4.15 100/122] EDAC, sb_edac: Fix out of bound writes during DIMM configuration on KNL

4.15-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Anna Karbownik <anna.karbownik@...el.com>

commit bf8486709ac7fad99e4040dea73fe466c57a4ae1 upstream.

Commit

  3286d3eb906c ("EDAC, sb_edac: Drop NUM_CHANNELS from 8 back to 4")

decreased NUM_CHANNELS from 8 to 4, but this is not enough for Knights
Landing which supports up to 6 channels.

This caused out-of-bounds writes to pvt->mirror_mode and pvt->tolm
variables which don't pay critical role on KNL code path, so the memory
corruption wasn't causing any visible driver failures.

The easiest way of fixing it is to change NUM_CHANNELS to 6. Do that.

An alternative solution would be to restructure the KNL part of the
driver to 2MC/3channel representation.

Reported-by: Dan Carpenter <dan.carpenter@...cle.com>
Signed-off-by: Anna Karbownik <anna.karbownik@...el.com>
Cc: Mauro Carvalho Chehab <mchehab@...nel.org>
Cc: Tony Luck <tony.luck@...el.com>
Cc: jim.m.snow@...el.com
Cc: krzysztof.paliswiat@...el.com
Cc: lukasz.odzioba@...el.com
Cc: qiuxu.zhuo@...el.com
Cc: linux-edac <linux-edac@...r.kernel.org>
Cc: <stable@...r.kernel.org>
Fixes: 3286d3eb906c ("EDAC, sb_edac: Drop NUM_CHANNELS from 8 back to 4")
Link: http://lkml.kernel.org/r/1519312693-4789-1-git-send-email-anna.karbownik@intel.com
[ Massage commit message. ]
Signed-off-by: Borislav Petkov <bp@...e.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>

---
 drivers/edac/sb_edac.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/drivers/edac/sb_edac.c
+++ b/drivers/edac/sb_edac.c
@@ -279,7 +279,7 @@ static const u32 correrrthrsld[] = {
  * sbridge structs
  */

-#define NUM_CHANNELS		4	/* Max channels per MC */
+#define NUM_CHANNELS		6	/* Max channels per MC */
 #define MAX_DIMMS		3	/* Max DIMMS per channel */
 #define KNL_MAX_CHAS		38	/* KNL max num. of Cache Home Agents */
 #define KNL_MAX_CHANNELS	6	/* KNL max num. of PCI channels */