Commit Graph

1131 Commits

Author SHA1 Message Date
Hanno Becker
07d420d6ad Remove unnecessary backup of explicit IV in AEAD record decryption
There is no need to hold back the explicit IV for AEAD ciphers.
2019-08-01 09:50:27 +02:00
Hanno Becker
8244cfa8bc Remove redundant minimum length check
Availability of sufficient incoming data should be checked when
it is needed, which is in mbedtls_ssl_fetch_input(), and this
function has the necessary bounds checks in place.
2019-08-01 09:50:27 +02:00
Hanno Becker
6d3db0fa25 Improve documentation of mbedtls_ssl_decrypt_buf() 2019-08-01 09:50:26 +02:00
Hanno Becker
9520b31860 Remove misleading comment in mbedtls_ssl_decrypt_buf()
The comment doesn't seem to relate to the code that follows.
2019-08-01 09:50:26 +02:00
Hanno Becker
b603bd34bc Remove assertion in mbedtls_ssl_decrypt_buf()
mbedtls_ssl_decrypt_buf() asserts that the passed transform is not NULL,
but the function is only invoked in a single place, and this invocation
is clearly visible to be within a branch ensuring that the incoming
transform isn't NULL. Remove the assertion for the benefit of code-size.
2019-08-01 09:50:26 +02:00
Hanno Becker
f024285034 Check architectural bound for max record payload len in one place
The previous code performed architectural maximum record length checks
both before and after record decryption. Since MBEDTLS_SSL_IN_CONTENT_LEN
bounds the maximum length of the record plaintext, it suffices to check
only once after (potential) decryption.

This must not be confused with the internal check that the record
length is small enough to make the record fit into the internal input
buffer; this is done in mbedtls_ssl_fetch_input().
2019-08-01 09:50:26 +02:00
Hanno Becker
408a2742b3 Remove redundant length-0 checks for incoming unprotected records 2019-08-01 09:50:26 +02:00
Hanno Becker
1c26845777 Remove redundant length check during record header parsing
The check is in terms of the internal input buffer length and is
hence likely to be originally intended to protect against overflow
of the input buffer when fetching data from the underlying
transport in mbedtls_ssl_fetch_input(). For locality of reasoning,
it's better to perform such a check close to where it's needed,
and in fact, mbedtls_ssl_fetch_input() _does_ contain an equivalent
bounds check, too, rendering the bounds check in question redundant.
2019-08-01 09:50:26 +02:00
Hanno Becker
02f2609551 Introduce configuration option and API for SSL record checking 2019-07-30 15:38:40 +03:00
Manuel Pégourié-Gonnard
cdb83e7c88
Merge pull request #616 from mpg/context-s11n
[baremetal] Implement context serialization
2019-07-30 00:07:23 +02:00
Manuel Pégourié-Gonnard
69a3e417d8 Improve reability and debugability of large if
Breaking into a series of statements makes things easier when stepping through
the code in a debugger.

Previous comments we stating the opposite or what the code tested for (what we
want vs what we're erroring out on) which was confusing.

Also expand a bit on the reasons for these restrictions.
2019-07-29 12:32:02 +02:00
Manuel Pégourié-Gonnard
18332c5c6c Improve getter for renegotiation enabled 2019-07-29 12:17:52 +02:00
Manuel Pégourié-Gonnard
b3bb31bd90 Introduce getter function for disable_renego 2019-07-26 16:37:45 +02:00
Manuel Pégourié-Gonnard
14e2a8ac06 Fix a typo in a comment 2019-07-26 16:31:53 +02:00
Hanno Becker
42a6b04c4a Don't forget about pending alerts after ssl_get_next_record()
ssl_get_next_record() may pend fatal alerts in response to receiving
invalid records. Previously, however, those were never actually sent
because there was no code-path checking for pending alerts.

This commit adds a call to ssl_send_pending_fatal_alert() after
the invocation of ssl_get_next_record() to fix this.
2019-07-26 07:25:20 +01:00
Hanno Becker
b82350b25f Introduce helper function to send pending fatal alerts 2019-07-26 07:25:02 +01:00
Hanno Becker
c8f529995f Rename pend_alert_msg -> pending_fatal_alert_msg 2019-07-25 12:59:24 +01:00
Hanno Becker
2e8d133ebf Reintroduce return code checking when sending NoRenego alert 2019-07-25 12:58:48 +01:00
Hanno Becker
3caf7189f9 Remove field to store level of pending alert
Pending alerts is so far only used for fatal alerts.
2019-07-25 12:58:44 +01:00
Hanno Becker
de62da9d3c Use separate functions to pend fatal and non-fatal alerts 2019-07-24 13:45:35 +01:00
Hanno Becker
1facd552fc Replace xxx_send_alert by xxx_pend_alert to save code 2019-07-24 13:20:27 +01:00
Hanno Becker
f46e1ce812 Introduce SSL helper function to mark pending alerts 2019-07-24 13:20:27 +01:00
Manuel Pégourié-Gonnard
7af7375473 Fix MSVC warning
We know the length of the ALPN string is always less than 255, so the cast to
uint8_t is safe.
2019-07-24 00:58:27 +02:00
Manuel Pégourié-Gonnard
2cc9223a3b Fix compile error in reduced configurations
Found by running scripts/baremetal.h --rom --gcc --check after adding
MBEDTLS_SSL_CONTEXT_SERIALIZATION to baremetal.h
2019-07-23 17:22:39 +02:00
Simon Butcher
3b014fc23a Merge remote-tracking branch 'origin/pr/604' into baremetal 2019-07-23 16:16:24 +01:00
Manuel Pégourié-Gonnard
7ce9446e4c Avoid duplication of session format header 2019-07-23 17:02:11 +02:00
Manuel Pégourié-Gonnard
a7cd4830ee Implement config-checking header to context s11n
Modelled after the config-checking header from session s11n.

The list of relevant config flags was established by manually checking the
fields serialized in the format, and which config.h flags they depend on.
This probably deserves double-checking by reviewers.
2019-07-23 17:02:10 +02:00
Manuel Pégourié-Gonnard
4c1d06e429 Provide serialisation API only if it's enabled 2019-07-23 17:02:10 +02:00
Manuel Pégourié-Gonnard
73a4636ca4 Adapt to hardcoded single version 2019-07-23 17:02:10 +02:00
Manuel Pégourié-Gonnard
2f3fa62a0a Fix compiler warning: comparing signed to unsigned
Since the type of cid_len is unsigned but shorter than int, it gets
"promoted" to int (which is also the type of the result), unless we make the
other operand an unsigned int which then forces the expression to unsigned int
as well.
2019-07-23 17:02:10 +02:00
Manuel Pégourié-Gonnard
bc847caa33 Actually reset the context on save as advertised
Also fix some wording in the documentation while at it.
2019-07-23 17:02:10 +02:00
Manuel Pégourié-Gonnard
ff22200fab Re-use buffer allocated by handshake_init()
This fixes a memory leak as well (found by running ssl-opt.sh in an Asan
build).
2019-07-23 17:02:10 +02:00
Manuel Pégourié-Gonnard
138079d7d6 Add setting of forced fields when deserializing 2019-07-23 17:02:10 +02:00
Manuel Pégourié-Gonnard
16d1485a3d Add saved fields from top-level structure 2019-07-23 17:02:10 +02:00
Manuel Pégourié-Gonnard
322f3c7377 Add transform (de)serialization 2019-07-23 17:02:10 +02:00
Manuel Pégourié-Gonnard
8175816200 Fix English in comments 2019-07-23 17:02:10 +02:00
Manuel Pégourié-Gonnard
f1f3e529a5 Add session saving/loading
For now, the header (version+format bytes) is duplicated. This might be
optimized later.
2019-07-23 17:02:09 +02:00
Manuel Pégourié-Gonnard
d0dd10469b Add (stub) header writing and checking
The number of meaning of the flags will be determined later, when handling the
relevant struct members. For now three bytes are reserved as an example, but
this number may change later.
2019-07-23 17:02:09 +02:00
Manuel Pégourié-Gonnard
5e534baaec Add usage checks in context_load() 2019-07-23 17:02:09 +02:00
Manuel Pégourié-Gonnard
b6163ef175 Document internal serialisation format
This mainly follows the design document (saving all fields marked "saved" in
the main structure and the transform sub-structure) with two exceptions:

- things related to renegotiation are excluded here (there weren't quite in
  the design document as the possibility of allowing renegotiation was still
on the table, which is no longer is) - also, ssl.secure_renegotiation (which
is not guarded by MBEDTLS_SSL_RENEGOTIATION because it's used in initial
handshakes even with renegotiation disabled) is still excluded, as we don't
need it after the handshake.

- things related to Connection ID are added, as they weren't present at the
  time the design document was written.

The exact format of the header (value of the bitflag indicating compile-time
options, whether and how to merge it with the serialized session header) will
be determined later.
2019-07-23 17:02:09 +02:00
Manuel Pégourié-Gonnard
569ed6ba56 Implement usage checks in context_save()
Enforce restrictions indicated in the documentation.

This allows to make some simplifying assumptions (no need to worry about
saving IVs for CBC in TLS < 1.1, nor about saving handshake data) and
guarantees that all values marked as "forced" in the design document have the
intended values and can be skipped when serialising.

Some of the "forced" values are not checked because their value is a
consequence of other checks (for example, session_negotiated == NULL outside
handshakes). We do however check that session and transform are not NULL (even
if that's also a consequence of the initial handshake being over) as we're
going to dereference them and static analyzers may appreciate the info.
2019-07-23 17:02:09 +02:00
Manuel Pégourié-Gonnard
a3024eef7b Save Hello random bytes for later use 2019-07-23 17:02:09 +02:00
Hanno Becker
95d1b93c69 Don't reset timer during mbedtls_ssl_setup()
At that point, the timer might not yet be configured.

The timer is reset at the following occasions:

- when it is initially configured through
    mbedtls_ssl_set_timer_cb() or
    mbedtls_ssl_set_timer_cb_cx()
- when a session is reset in mbedtls_ssl_session_reset()
- when a handshake finishes via mbedtls_ssl_handshake_wrap()
2019-07-22 11:15:28 +01:00
Hanno Becker
56595f4f7b Allow hardcoding single signature hash at compile-time
This commit introduces the option MBEDTLS_SSL_CONF_SINGLE_HASH
which can be used to register a single supported signature hash
algorithm at compile time. It replaces the runtime configuration
API mbedtls_ssl_conf_sig_hashes() which allows to register a _list_
of supported signature hash algorithms.

In contrast to other options used to hardcode configuration options,
MBEDTLS_SSL_CONF_SINGLE_HASH isn't a numeric option, but instead it's
only relevant if it's defined or not. To actually set the single
supported hash algorithm that should be supported, numeric options

MBEDTLS_SSL_CONF_SINGLE_HASH_TLS_ID
MBEDTLS_SSL_CONF_SINGLE_HASH_MD_ID

must both be defined and provide the TLS ID and the Mbed TLS internal
ID and the chosen hash algorithm, respectively.
2019-07-17 10:19:27 +01:00
Hanno Becker
f1bc9e1c69 Introduce helper functions to traverse signature hashes 2019-07-17 10:19:27 +01:00
Hanno Becker
627fbee41a Don't offer SHA-1 in CertificateRequest message in TLS 1.2
mbedtls_ssL_set_calc_verify_md() is used to select valid hashes when
writing the server's CertificateRequest message, as well as to verify
and act on the client's choice when reading its CertificateVerify
message.

If enabled at compile-time and configured via mbedtls_ssl_conf_sig_hashes()
the current code also offers SHA-1 in TLS 1.2. However, the SHA-1-based
handshake transcript in TLS 1.2 is different from the SHA-1 handshake
transcript used in TLS < 1.2, and we only maintain the latter
(through ssl_update_checksum_md5sha1()), but not the former.
Concretely, this will lead to CertificateVerify verification failure
if the client picks SHA-1 for the CertificateVerify message in a TLS 1.2
handshake.

This commit removes SHA-1 from the list of supported hashes in
the CertificateRequest message, and adapts two tests in ssl-opt.sh
which expect SHA-1 to be listed in the CertificateRequest message.
2019-07-17 10:19:27 +01:00
Hanno Becker
0a6417041e Remove redundant check in mbedtls_ssl_set_calc_verify_md()
mbedtls_ssl_set_calc_verify_md() is only called from places
where it has been checked that TLS 1.2 is being used. The
corresponding compile-time and runtime guards checking the
version in mbedtls_ssl_set_calc_verify_md() are therefore
redundant and can be removed.
2019-07-17 10:19:25 +01:00
Simon Butcher
feb1cee36e Merge remote-tracking branch 'origin/pr/602' into baremetal 2019-07-15 19:24:11 +01:00
Hanno Becker
c1096e7514 Allow hardcoding single supported elliptic curve
This commit introduces the option MBEDTLS_SSL_CONF_SINGLE_EC
which can be used to register a single supported elliptic curve
at compile time. It replaces the runtime configuration API
mbedtls_ssl_conf_curves() which allows to register a _list_
of supported elliptic curves.

In contrast to other options used to hardcode configuration options,
MBEDTLS_SSL_CONF_SINGLE_EC isn't a numeric option, but instead it's
only relevant if it's defined or not. To actually set the single
elliptic curve that should be supported, numeric options

MBEDTLS_SSL_CONF_SINGLE_EC_TLS_ID
MBEDTLS_SSL_CONF_SINGLE_EC_GRP_ID

must both be defined and provide the TLS ID and the Mbed TLS internal
ID and the chosen curve, respectively.
2019-07-12 15:25:03 +01:00
Hanno Becker
ee24f8cecb Remove unnecessary check for presence of supported EC list
For both client/server the EC curve list is assumed not to be NULL:

- On the client-side, it's assumed when writing the
  supported elliptic curve extension:

    c54ee936d7/library/ssl_cli.c (L316)

- On the server, it is assumed when searching for a
  suitable curve for the ECDHE exchange:

    c54ee936d7/library/ssl_srv.c (L3200)

It is therefore not necessary to check this in mbedtls_ssl_check_curve().
2019-07-12 15:25:03 +01:00