Manuel Pégourié-Gonnard

The CI came back with a number of failures. You can ignore all failures in components with `armcc` in their name, as they are due to some temporary licensing issues,...

Thanks for the update! However our DCO checker points out the second commit is missing the Signed-Off-By line. Also, we're having trouble with our CI at the moment. I'll kick...

It looks like something went wrong when fixing the commit message, now the history still has the original non-signed-off commit, plus the same with a sign-off, and a merge commit....

> What do you want me to do in detail? We require every single commit to include a signed-off-by line in its message. Right now your PR has 4 commits,...

Thanks for your report! This is an API break, and we're two days away from code freeze for the next major version, so I'm not sure if your report comes...

I'm afraid so. There was so much that we _had_ to include in this major version (related to the transition to PSA crypto and the repo split, and to remove...

> Unfortunately `mbedtls_ssl_write` is not included in that list so it's still not conclusive about how to handle it in this case. I think this error can be ignored from...

For the record, here's my reasoning about which public functions can return `MBEDTLS_ERR_SSL_RECEIVED_NEW_SESSION_TICKET`. Starting from the bottom up: - `MBEDTLS_ERR_SSL_RECEIVED_NEW_SESSION_TICKET` can only be returned by `ssl_tls13_process_new_session_ticket()`, - which is only...

> you'll need to call `mbedtls_ssl_conf_tls13_enable_signal_new_session_tickets(1)` to enable the pre-3.6.0 behavior ITYM pre-3.6.1 here.

This is fixed in [Mbed TLS 3.6.1](https://github.com/Mbed-TLS/mbedtls/releases/tag/mbedtls-3.6.1) which has now been released. However I'm not closing this ticket as the following work remains: - [x] Raise a follow-up PR addressing...