X.400 address type confusion in X.509 GeneralName (CVE-2023-0286)
Use-after-free following BIO_new_NDEF (CVE-2023-0215)
Double free after calling PEM_read_bio_ex (CVE-2022-4450)
After discussing with jsing, we decided that it's best to neuter the "fix" for
Timing Oracle in RSA Decryption (CVE-2022-4304)
for the time being. The upstream patch is ~700 lines of strange code without
license and with barely decipherable comments - initial versions sported
several variants of "mongomery". Testers found issues on aarch64... Even by
OpenSSL standards this is badly written and manifestly poorly tested code.
It is irresponsible to subject users to this in a security update before it
has seen more testing.
X.400 address type confusion in X.509 GeneralName (CVE-2023-0286)
X.509 Name Constraints Read Buffer Overflow (CVE-2022-4203) - Punycode again!
Use-after-free following BIO_new_NDEF (CVE-2023-0215)
Double free after calling PEM_read_bio_ex (CVE-2022-4450)
Invalid pointer dereference in d2i_PKCS7 functions (CVE-2023-0216)
NULL dereference validating DSA public key (CVE-2023-0217)
NULL dereference during PKCS7 data verification (CVE-2023-0401)
After discussing with jsing, we decided that it's best to neuter the "fix" for
Timing Oracle in RSA Decryption (CVE-2022-4304)
for the time being. The upstream patch is ~700 lines of strange code without
license and with barely decipherable comments - initial versions sported
several variants of "mongomery". Testers found issues on aarch64... Even by
OpenSSL standards this is badly written and manifestly poorly tested code.
It is irresponsible to subject users to this in a security update before it
has seen more testing.
The update also contains a fix for this which we had already patched locally:
Fix for X.509 Policy Constraints Double Locking (CVE-2022-3996)
- For non-interactive usage (CI/CD, etc), encryption of private keys can
be disabled with the -W switch.
- A new command, -C was added to change passwords, or remove them (when
combined with -W).