Skip to contentSkip to conversation

package/mbedtls: security bump to version 2.16.3

PATCH New Fabrice Fontaine 2 participants 3 messages Sep 18, 2019 – Sep 26, 2019

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 package/mbedtls: security bump to version 2.16.3

- Remove second patch (already in version)
- Fix a missing error detection in ECJPAKE. This could have caused a predictable shared secret if a hardware accelerator failed and the other side of the key exchange had a similar bug.
- When writing a private EC key, use a constant size for the private value, as specified in RFC 5915. Previously, the value was written as an ASN.1 INTEGER, which caused the size of the key to leak about 1 bit of information on average and could cause the value to be 1 byte too large for the output buffer.
- The deterministic ECDSA calculation reused the scheme's HMAC-DRBG to implement blinding. Because of this for the same key and message the same blinding value was generated. This reduced the effectiveness of the countermeasure and leaked information about the private key through side channels. Reported by Jack Lloyd.

---
 ...-at-least-ARMv6-to-enable-the-ARM-DS.patch | 38 -------------------
 package/mbedtls/mbedtls.hash                  |  6 +--
 package/mbedtls/mbedtls.mk                    |  2 +-
 3 files changed, 4 insertions(+), 42 deletions(-)
 delete mode 100644 package/mbedtls/0002-bn_mul.h-require-at-least-ARMv6-to-enable-the-ARM-DS.patch
package/mbedtls/0002-bn_mul.h-require-at-least-ARMv6-to-enable-the-ARM-DS.patch deleted +0 -38
deleted file mode 100644
index ee66be4b21..0000000000
@@ -1,38 +0,0 @@
-From 63f5b85e700677270197d176475be609187e9cdb Mon Sep 17 00:00:00 2001
-From: Aurelien Jarno <[email protected]>
-Date: Sat, 3 Nov 2018 00:46:06 +0100
-Subject: [PATCH] bn_mul.h: require at least ARMv6 to enable the ARM DSP code
-
-Commit 16b1bd89326e "bn_mul.h: add ARM DSP optimized MULADDC code"
-added some ARM DSP instructions that was assumed to always be available
-when __ARM_FEATURE_DSP is defined to 1. Unfortunately it appears that
-the ARMv5TE architecture (GCC flag -march=armv5te) supports the DSP
-instructions, but only in Thumb mode and not in ARM mode, despite
-defining __ARM_FEATURE_DSP in both cases.
-
-This patch fixes the build issue by requiring at least ARMv6 in addition
-to the DSP feature.
-
-[Upstream status: https://github.com/ARMmbed/mbedtls/pull/2169]
-Signed-off-by: Pierre-Jean Texier <[email protected]>
----
- include/mbedtls/bn_mul.h | 3 ++-
- 1 file changed, 2 insertions(+), 1 deletion(-)
-
-diff --git a/include/mbedtls/bn_mul.h b/include/mbedtls/bn_mul.h
-index 0af694c..565615f 100644
---- a/include/mbedtls/bn_mul.h
-+++ b/include/mbedtls/bn_mul.h
-@@ -636,7 +636,8 @@
-            "r6", "r7", "r8", "r9", "cc"         \
-          );
- 
--#elif defined (__ARM_FEATURE_DSP) && (__ARM_FEATURE_DSP == 1)
-+#elif (__ARM_ARCH >= 6) && \
-+    defined (__ARM_FEATURE_DSP) && (__ARM_FEATURE_DSP == 1)
- 
- #define MULADDC_INIT                            \
-     asm(
--- 
-2.7.4
-
package/mbedtls/mbedtls.hash +3 -3
index 1e65f8224a..db136c17d9 100644
@@ -1,5 +1,5 @@
-# From https://tls.mbed.org/tech-updates/releases/mbedtls-2.16.2-and-2.7.11-released
-sha1	ba809acfd4b41b86895b92e98d936695b5b62b73	mbedtls-2.16.2-apache.tgz
-sha256	a6834fcd7b7e64b83dfaaa6ee695198cb5019a929b2806cb0162e049f98206a4	mbedtls-2.16.2-apache.tgz
+# From https://tls.mbed.org/tech-updates/releases/mbedtls-2.16.3-and-2.7.12-released
+sha1	dce8550f8f9465f3aea44cb7d0f9d0ba8140034a	mbedtls-2.16.3-apache.tgz
+sha256	ec1bee6d82090ed6ea2690784ea4b294ab576a65d428da9fe8750f932d2da661	mbedtls-2.16.3-apache.tgz
 # Locally calculated
 sha256	cfc7749b96f63bd31c3c42b5c471bf756814053e847c10f3eb003417bc523d30	apache-2.0.txt
package/mbedtls/mbedtls.mk +1 -1
index fdf1d33b2e..f58aad4bca 100644
@@ -5,7 +5,7 @@
 ################################################################################
 
 MBEDTLS_SITE = https://tls.mbed.org/code/releases
-MBEDTLS_VERSION = 2.16.2
+MBEDTLS_VERSION = 2.16.3
 MBEDTLS_SOURCE = mbedtls-$(MBEDTLS_VERSION)-apache.tgz
 MBEDTLS_CONF_OPTS = \
 	-DENABLE_PROGRAMS=$(if $(BR2_PACKAGE_MBEDTLS_PROGRAMS),ON,OFF) \
Signature · 1 line
2.23.0
#1 lore raw

PATCH 1/1 package/mbedtls: security bump to version 2.16.3

Quoted 17 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Fabrice" == Fabrice Fontaine <[email protected]> writes:

- Remove second patch (already in version)
- Fix a missing error detection in ECJPAKE. This could have caused a predictable shared secret if a hardware accelerator failed and the other side of the key exchange had a similar bug.
- When writing a private EC key, use a constant size for the private value, as specified in RFC 5915. Previously, the value was written as an ASN.1 INTEGER, which caused the size of the key to leak about 1 bit of information on average and could cause the value to be 1 byte too large for the output buffer.
- The deterministic ECDSA calculation reused the scheme's HMAC-DRBG to implement blinding. Because of this for the same key and message the same blinding value was generated. This reduced the effectiveness of the countermeasure and leaked information about the private key through side channels. Reported by Jack Lloyd.

Quoted 1 line

Committed, thanks.

Signature · 1 line
Bye, Peter Korsgaard
#2 lore raw

PATCH 1/1 package/mbedtls: security bump to version 2.16.3

In reply to Fabrice Fontaine
Quoted 17 lines
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

"Fabrice" == Fabrice Fontaine <[email protected]> writes:

- Remove second patch (already in version)
- Fix a missing error detection in ECJPAKE. This could have caused a predictable shared secret if a hardware accelerator failed and the other side of the key exchange had a similar bug.
- When writing a private EC key, use a constant size for the private value, as specified in RFC 5915. Previously, the value was written as an ASN.1 INTEGER, which caused the size of the key to leak about 1 bit of information on average and could cause the value to be 1 byte too large for the output buffer.
- The deterministic ECDSA calculation reused the scheme's HMAC-DRBG to implement blinding. Because of this for the same key and message the same blinding value was generated. This reduced the effectiveness of the countermeasure and leaked information about the private key through side channels. Reported by Jack Lloyd.

Quoted 1 line

Committed to 2019.08.x, thanks. For 2019.02.x and 2019.05.x I will instead bump to 2.7.12.

Signature · 1 line
Bye, Peter Korsgaard