Keith Packard

> I'm late to the game, but why adding a new keyword argument instead of allowing path as `output: 'subdir/foo.txt'`, as suggested in [#2320 (comment)](https://github.com/mesonbuild/meson/issues/2320#issuecomment-370418738). That seems more flexible and...

> What I'm a bit worried is the case of custom_target() where each output could be in a different subdir. If we allow `custom_target(..., output: ['subdir/out.txt', 'subdir2/out.txt'])` as requested in...

I'm currently using a hack where I add '-lgcc' to link_args as that wraps the list of static libraries, the --defsym values and then -lgcc in `--start-group`, `--end-group`; the linker...

> Since the feature being added seems generic enough that its effective functionality isn't actually about defsym at all, I wonder if it's worth being a bit more direct about...

> ... or even just passing them all early by default? Are there any (ignoring -l which should be handled via link_with / dependencies anyway) that specifically need to be...

> ... or even just passing them all early by default? Are there any (ignoring -l which should be handled via link_with / dependencies anyway) that specifically need to be...

These are (finally) passing Zephyr tests using both SDK 0.17 and SDK 0.18-alpha4 (although each requires some additional changes which are wending their way upstream). I think they're ready for...

> Had a chat with upstream about these (picolibc) changes. Let's keep them only in this fork until they're not needed anymore (TF-M is moving away from C library usage),...

Sounds like any libc-specific changes will be transient then, which will be nice.

> NACKing until the issues with the Nordic devices are solved How can we fix this? We're running out of time to get SDK 1.0 ready for Zephyr 4.3, and...