Manuel Pégourié-Gonnard

Thanks for addressing my feedback. I've been thinking a bit more about this "working directory" issue, and the way the functions are written, the constraint is not really that scripts...

> I am curious as to why that necessitates a change to the way the file is read etc. Is this approach more robust? > I don't think it _requires_...

> Changing to another directory can simplify a script, but you have to be careful. I don't disagree, what I'm saying is that changing to another directory is something our...

> * **tests** not required - Sufficient tests already in place Changes to `all.sh` are somewhat tested by the PR job; I've checked that the number of never executed tests...

Looks unrelated indeed, probably wrong version of something. Actually we can test in on the CI using [the release job](https://mbedtls.trustedfirmware.org/job/mbedtls-release-ci-testing/) (need to be logged in to access that page) -...

I am. Let's see if @gabor-mezei-arm agrees now :)

I've just been thinking: since the keys are held as PK contexts in X.509 and TLS so far, and those modules always use PK functions to operate on them (even...

Alternatively, as a completely different strategy, we can leave PK alone and instead change the TLS and X.509 code to: - create a PSA key from the PK context they...

I've not yet used this script myself, but looking at its result there seems to be extra commits that I wouldn't expect to be present in the history, see https://github.com/Mbed-TLS/mbedtls-framework/pull/55#issuecomment-2404287672

> I'm marking this as priority-high because we need this to move files to the framework with the history preserved. Arguably this should be in the framework repository as an...