linux-kernel - [PATCH] firmware: dmi_scan: Prevent dmi

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

Date:	Fri, 20 Mar 2015 09:59:47 +0100
From:	Jean Delvare <jdelvare@...e.de>
To:	LKML <linux-kernel@...r.kernel.org>
Cc:	Matt Fleming <matt.fleming@...el.com>,
	Ard Biesheuvel <ard.biesheuvel@...aro.org>,
	Ivan Khoronzhuk <ivan.khoronzhuk@...aro.org>
Subject: [PATCH] firmware: dmi_scan: Prevent dmi_num integer overflow

dmi_num is a u16, dmi_len is a u32, so this construct:

	dmi_num = dmi_len / 4;

would result in an integer overflow for a DMI table larger than
256 kB. I've never see such a large table so far, but SMBIOS 3.0
makes it possible so maybe we'll see such tables in the future.

So instead of faking a structure count when the entry point does
not provide it, adjust the loop condition in dmi_table() to properly
deal with the case where dmi_num is not set.

Signed-off-by: Jean Delvare <jdelvare@...e.de>
Cc: Matt Fleming <matt.fleming@...el.com>
Cc: Ard Biesheuvel <ard.biesheuvel@...aro.org>
Cc: Ivan Khoronzhuk <ivan.khoronzhuk@...aro.org>
---
 drivers/firmware/dmi_scan.c |   22 +++++++---------------
 1 file changed, 7 insertions(+), 15 deletions(-)

--- linux-4.0-rc4.orig/drivers/firmware/dmi_scan.c	2015-03-19 14:19:56.384426825 +0100
+++ linux-4.0-rc4/drivers/firmware/dmi_scan.c	2015-03-20 09:23:37.689542485 +0100
@@ -86,10 +86,13 @@ static void dmi_table(u8 *buf, u32 len,
 	int i = 0;
 
 	/*
-	 *	Stop when we see all the items the table claimed to have
-	 *	OR we run off the end of the table (also happens)
+	 * Stop when we have seen all the items the table claimed to have
+	 * (SMBIOS < 3.0 only) OR we reach an end-of-table marker OR we run
+	 * off the end of the table (should never happen but sometimes does
+	 * on bogus implementations.)
 	 */
-	while ((i < num) && (data - buf + sizeof(struct dmi_header)) <= len) {
+	while ((!num || i < num) &&
+	       (data - buf + sizeof(struct dmi_header)) <= len) {
 		const struct dmi_header *dm = (const struct dmi_header *)data;
 
 		/*
@@ -529,21 +532,10 @@ static int __init dmi_smbios3_present(co
 	if (memcmp(buf, "_SM3_", 5) == 0 &&
 	    buf[6] < 32 && dmi_checksum(buf, buf[6])) {
 		dmi_ver = get_unaligned_be16(buf + 7);
+		dmi_num = 0;			/* No longer specified */
 		dmi_len = get_unaligned_le32(buf + 12);
 		dmi_base = get_unaligned_le64(buf + 16);
 
-		/*
-		 * The 64-bit SMBIOS 3.0 entry point no longer has a field
-		 * containing the number of structures present in the table.
-		 * Instead, it defines the table size as a maximum size, and
-		 * relies on the end-of-table structure type (#127) to be used
-		 * to signal the end of the table.
-		 * So let's define dmi_num as an upper bound as well: each
-		 * structure has a 4 byte header, so dmi_len / 4 is an upper
-		 * bound for the number of structures in the table.
-		 */
-		dmi_num = dmi_len / 4;
-
 		if (dmi_walk_early(dmi_decode) == 0) {
 			pr_info("SMBIOS %d.%d present.\n",
 				dmi_ver >> 8, dmi_ver & 0xFF);


-- 
Jean Delvare
SUSE L3 Support
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/