The basis for the Lucky 13 family of attacks is for an attacker to be able to
distinguish between (long) valid TLS-CBC padding and invalid TLS-CBC padding.
Since our code sets padlen = 0 for invalid padding, the length of the input to
the HMAC function, and the location where we read the MAC, give information
about that.
A local attacker could gain information about that by observing via a
cache attack whether the bytes at the end of the record (at the location of
would-be padding) have been read during MAC verification (computation +
comparison).
Let's make sure they're always read.
The basis for the Lucky 13 family of attacks is for an attacker to be able to
distinguish between (long) valid TLS-CBC padding and invalid TLS-CBC padding.
Since our code sets padlen = 0 for invalid padding, the length of the input to
the HMAC function gives information about that.
Information about this length (modulo the MD/SHA block size) can be deduced
from how much MD/SHA padding (this is distinct from TLS-CBC padding) is used.
If MD/SHA padding is read from a (static) buffer, a local attacker could get
information about how much is used via a cache attack targeting that buffer.
Let's get rid of this buffer. Now the only buffer used is the internal MD/SHA
one, which is always read fully by the process() function.
The CONTRIBUTING.md referred to 'legacy' branches instead of LTS branches, and
also referenced mbedtls-1.3 which is no longer maintained, and omitted
mbedtls-2.7 which is.
Fix Documentation error in `mbedtls_ssl_get_session`.
This function supports deep copying of the session,
and the peer certificate is not lost anymore, Resolves#926
Move definition of `MBEDTLS_CIPHER_MODE_STREAM` to header file
(`mbedtls_cipher_internal.h`), because it is used by more than
one file. Raised by TrinityTonic in #1719
Functions time with TIME_AND_TSC() didn't have their return values checked.
I'm not sure whether Coverity complained about existing uses, but it did about
new ones, since we consistently check their return values everywhere but here,
which it rightfully finds suspicious.
So, let's check return values. This probably adds a few cycles to existing
loop overhead, but on my machine (x86_64) the added overhead is less than the
random-looking variation between various runs, so it's acceptable.
Some calls had their own particular error checking; remove that in favour of
the new general solution.
