As reported by Alicja Kario, we ignored excess bytes after the
signature payload in TLS CertificateVerify Messages. These
should not be present.
Fixes: #25298
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/25302)
(cherry picked from commit b4e4bf29ba)
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit ca1d2db291)
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/25082)
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit b1f4aebb74)
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/25082)
Fixes#25054
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(cherry picked from commit b7f93c7fcb)
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/25082)
The ALPN protocol selected by the server must be one that we originally
advertised. We should verify that it is.
Follow on from CVE-2024-5535
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24717)
Return EXT_RETURN_NOT_SENT in the event that we don't send the extension,
rather than EXT_RETURN_SENT. This actually makes no difference at all to
the current control flow since this return value is ignored in this case
anyway. But lets make it correct anyway.
Follow on from CVE-2024-5535
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24717)
The QUIC test server was using incorrectly formatted ALPN data. With the
previous implementation of SSL_select_next_proto this went unnoticed. With
the new stricter implemenation it was failing.
Follow on from CVE-2024-5535
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24717)
In the case where the NPN callback returns with SSL_TLEXT_ERR_OK, but
the selected_len is 0 we should fail. Previously this would fail with an
internal_error alert because calling OPENSSL_malloc(selected_len) will
return NULL when selected_len is 0. We make this error detection more
explicit and return a handshake failure alert.
Follow on from CVE-2024-5535
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24717)
Ensure that the provided client list is non-NULL and starts with a valid
entry. When called from the ALPN callback the client list should already
have been validated by OpenSSL so this should not cause a problem. When
called from the NPN callback the client list is locally configured and
will not have already been validated. Therefore SSL_select_next_proto
should not assume that it is correctly formatted.
We implement stricter checking of the client protocol list. We also do the
same for the server list while we are about it.
CVE-2024-5535
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24717)
Fixes#4545
If free is called for an SSL BIO that is in initialization phase,
the `SSL_shutdown` call is omitted.
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Paul Dale <ppzgs1@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24705)
(cherry picked from commit 57b83edc46)
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24673)
(cherry picked from commit af82623d32)
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24673)
(cherry picked from commit 8d934a7592)
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24673)
(cherry picked from commit 79886c85b3)
A psk session was assumed to be a resumption which failed a check
when parsing the max_fragment_length extension hello from the client.
Relevant code from PR#18130 which was a suggested fix to the issue
was cherry-picked.
Fixes#18121
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24513)
(cherry picked from commit fa49560451)
In order to ensure we do not have a UAF we reset the rl->packet pointer
to NULL after we free it.
Follow on from CVE-2024-4741
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24395)
(cherry picked from commit bfb8128190)
If we're part way through processing a record, or the application has
not released all the records then we should not free our buffer because
they are still needed.
CVE-2024-4741
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24395)
(cherry picked from commit 38690cab18)
Fixes#24300. The current values of SSL_R_NO_APPLICATION_PROTOCOL and
SSL_R_PSK_IDENTITY_NOT_FOUND don't allow for a correct lookup of the
corresponding reason strings.
CLA: trivial
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24351)
(cherry picked from commit a401aaf9ed)
Don't attempt to memcpy a NULL pointer if the length is 0.
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24309)
(cherry picked from commit 97c6489b39)
Setting the server sig algs sets up the certificate "s3->tmp.valid_flags".
These are needed when calling ssl3_choose_cipher() which can happen
immediately after calling the session_secret_cb
Fixes#24213
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24309)
(cherry picked from commit 91c7ab27ce)
Make sure we can't inadvertently use a not_resumable session
Related to CVE-2024-2511
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24043)
In TLSv1.3 we create a new session object for each ticket that we send.
We do this by duplicating the original session. If SSL_OP_NO_TICKET is in
use then the new session will be added to the session cache. However, if
early data is not in use (and therefore anti-replay protection is being
used), then multiple threads could be resuming from the same session
simultaneously. If this happens and a problem occurs on one of the threads,
then the original session object could be marked as not_resumable. When we
duplicate the session object this not_resumable status gets copied into the
new session object. The new session object is then added to the session
cache even though it is not_resumable.
Subsequently, another bug means that the session_id_length is set to 0 for
sessions that are marked as not_resumable - even though that session is
still in the cache. Once this happens the session can never be removed from
the cache. When that object gets to be the session cache tail object the
cache never shrinks again and grows indefinitely.
CVE-2024-2511
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/24043)
Though support for provider-based signature algorithms was added in
ee58915 this functionality did not work with the SignatureAlgorithms
configuration command. If SignatureAlgorithms is set then the provider
sigalgs are not used and instead it used the default value.
This PR adds a check against the provider-base sigalg list when parsing
the SignatureAlgorithms value.
Based-on-patch-by: Martin Schmatz <mrt@zurich.ibm.com>
Fixes#22761
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/22779)
(cherry picked from commit 4169d58c85)
Fixes#23624
The calculation of the size for gid_arr reallocation was wrong.
A multiplication by gid_arr array item size was missing.
Testcase is added.
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23625)
(cherry picked from commit f4ed6eed2c)
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23581)
(cherry picked from commit 5c16e9d384)
There was really no need for this to be void and it made bugs very easy
to introduce accidentally, especially given that the free functions
needed to be passed a pointer to the pointer.
Also fix some bugs in the QUIC code detected immediately by this change.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23519)
(cherry picked from commit 62cb7c810e)
Freeing the QUIC TLS object may make calls back into QUIC so we should
free it first.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23256)
(cherry picked from commit f7f2b665cf)
The TLS object may make callbacks into QUIC during cleanup so we must
free it first.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23256)
(cherry picked from commit fa4b1151c8)
We had two functions which were very similarly named, that did almost the
same thing, but not quite. We bring the two together. Doing this also fixes
a possible bug where some data may not be correctly freed when the
RECORD_LAYER_clear() version was used.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23256)
(cherry picked from commit 4a0e4849af)
Using the client one when the server once should be used could cause a
later call to SSL_set_accept_state() to unexpectedly fail.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23256)
(cherry picked from commit a86714041d)
When we clear the record layer, we better make sure we clear all relevant
fields, otherwise we can get ourselves into an unexpected state.
Fixes#23255
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23256)
(cherry picked from commit 5fb065589d)
Don't check the Max Fragment Length if the it hasn't been negotiated. We
were checking it anyway, and using the default value
(SSL3_RT_MAX_PLAIN_LENGTH). This works in most cases but KTLS can cause the
record length to actually exceed this in some cases.
Fixes#23169
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Neil Horman <nhorman@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23182)
(cherry picked from commit c1decd6246)
Also document that it is ok to use this for control flow decisions.
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23304)
(cherry picked from commit ead44e19fa)
Signed-off-by: lan1120 <lanming@huawei.com>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23045)
(cherry picked from commit e8e95f20a9)
Even if some configuration entry is incorrect, do not
skip the remaining ones.
Fixes#20789
Reviewed-by: Neil Horman <nhorman@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/23048)
(cherry picked from commit 69c067ffbc)
TLSv1.3 includes an extra byte after the payload for the content type.
We should incorporate that in the calculation of the default buffer length.
Fixes#23015
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/23021)
(cherry picked from commit e07b5e1a0a)
We remove a function that was left behind and is no longer called after the
record layer refactor
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/22864)
(cherry picked from commit e46a6b1a5d)
If an ENGINE has been loaded after the SSL_CTX has been created then
the cipher we have cached might be provider based, but the cipher we
actually end up using might not be. Don't try to set provider params on
a cipher that is actually ENGINE based.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/22864)
(cherry picked from commit afcc12c41a)
We don't need the decoded X.509 Full(0) certificate for the EE usages 1 and 3,
because the leaf certificate is always part of the presented chain, so the
certificate is only validated as well-formed, and then discarded, but the
TLSA record is of course still used after the validation step.
Added DANE test cases for: 3 0 0, 3 1 0, 1 0 0, and 1 1 0
Reported by Claus Assmann.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/22821)
(cherry picked from commit f636e7e6bd)
Instead of trying to move the doomed sct back
to the src stack, which may fail as well, simply
free the sct object, as the src list will be
deleted anyway.
Reviewed-by: Paul Yang <kaishen.yy@antfin.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/22762)
(cherry picked from commit a435d78604)
This may happen when ssl_cert_dup calls custom_exts_copy, where
a possible memory allocation error causes custom_exts_free
to be called twice: once in the error handling of custom_exts_copy
and a second time in the error handling of ssl_cert_dup.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/22772)
(cherry picked from commit bc0773bbbd)
Tomas Mraz
Viktor Dukhovni
Matt Caswell
erbsland-dev
Robert Schulze
Frederik Wedel-Heinen
Amir Mohammadi
irosay
Jacob Champion
Hugo Landau
Alex Bozarth
Michael Baentsch
lan1120
Bernd Edlinger