Hi Quentin, On Wed, 19 Nov 2025 at 06:46, Quentin Schulz <[email protected]> wrote: > > Hi Simon, > > On 11/19/25 2:16 PM, Simon Glass wrote: > > Hi Quentin, > > > > On Mon, 17 Nov 2025 at 12:43, Quentin Schulz <[email protected]> wrote: > >> > >> From: Quentin Schulz <[email protected]> > >> > >> mkimage has support for OpenSSL engines but binman currently doesn't for > >> direct callers of mkimage (e.g. the fit etype). This prepares for adding > >> support for OpenSSL engines for signing elements of a FIT image, which > >> will done in the next commit. > >> > >> Signed-off-by: Quentin Schulz <[email protected]> > >> --- > >> tools/binman/btool/mkimage.py | 5 ++++- > >> 1 file changed, 4 insertions(+), 1 deletion(-) > > > > patchwork seems to be done, so I cannot try the test coverage > > > > b4 shazam > https://lore.kernel.org/u-boot/[email protected]/ > > should do the trick to fetch the code easily and run whatever you want > to run locally. > > I only tested 4/4 tests individually (you cannot actually pass multiple > even if the help text claims the opposite). The pip dependencies aren't > installable on Fedora 43 (or Debian Trixie AFAICT) because the libgit2 > dependency for building the wheels is too recent compared to what's > expected. Anyway, building in the same container as the CI and repeating > the same steps, it seems that my tests are actually breaking other tests > now when running them all at once, I'll check. I suspect something > related to the environment variables persisting across tests. > > I'll run the coverage test as well, thanks for the reminder, to see what > I can do, we're most likely missing the check for fit,engine + > fit,encrypt raised Error.
OK thanks, that worked. It looks quite good, but you have 4 lines missing coverage in fit.py, from what I can see. Regards, Simon
