kumquat-buildroot/package/botan/0001-Avoid-using-short-exponents-with-ElGamal.patch
Peter Korsgaard 31c94080d2 package/botan: add upstream security fix for CVE-2021-40529
Fixes the following security issue:

- CVE-2021-40529: The ElGamal implementation in Botan through 2.18.1, as
  used in Thunderbird and other products, allows plaintext recovery because,
  during interaction between two cryptographic libraries, a certain
  dangerous combination of the prime defined by the receiver's public key,
  the generator defined by the receiver's public key, and the sender's
  ephemeral exponents can lead to a cross-configuration attack against
  OpenPGP

For more details, see the upstream bug and issue writeup:
- https://github.com/randombit/botan/pull/2790
- https://ibm.github.io/system-security-research-updates/2021/07/20/insecurity-elgamal-pt1

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
2021-09-18 19:42:48 +02:00

39 lines
1.4 KiB
Diff

From 9a23e4e3bc3966340531f2ff608fa9d33b5185a2 Mon Sep 17 00:00:00 2001
From: Jack Lloyd <jack@randombit.net>
Date: Tue, 3 Aug 2021 18:20:29 -0400
Subject: [PATCH] Avoid using short exponents with ElGamal
Some off-brand PGP implementation generates keys where p - 1 is
smooth, as a result short exponents can leak enough information about
k to allow decryption.
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
[Peter: Drop tests, CVE-2021-40529]
---
src/lib/pubkey/elgamal/elgamal.cpp | 8 +++-
1 file changed, 1 insertions(+), 1 deletions(-)
diff --git a/src/lib/pubkey/elgamal/elgamal.cpp b/src/lib/pubkey/elgamal/elgamal.cpp
index b3ec6df2c..0e33c2ca5 100644
--- a/src/lib/pubkey/elgamal/elgamal.cpp
+++ b/src/lib/pubkey/elgamal/elgamal.cpp
@@ -113,8 +113,12 @@ ElGamal_Encryption_Operation::raw_encrypt(const uint8_t msg[], size_t msg_len,
if(m >= m_group.get_p())
throw Invalid_Argument("ElGamal encryption: Input is too large");
- const size_t k_bits = m_group.exponent_bits();
- const BigInt k(rng, k_bits);
+ /*
+ Some ElGamal implementations foolishly use prime fields where p - 1 is
+ smooth, as a result it is unsafe to use short exponents.
+ */
+ const size_t k_bits = m_group.p_bits() - 1;
+ const BigInt k(rng, k_bits, false);
const BigInt a = m_group.power_g_p(k, k_bits);
const BigInt b = m_group.multiply_mod_p(m, monty_execute(*m_monty_y_p, k, k_bits));
-
--
2.20.1