mpi3mr: Check for copyin errors in mpi3mr_map_data_buffer_dma()

[markj]

Summary

A failed copyin will cause the driver to use the contents of
uninitialized buffers instead, which is unlikely to be the behaviour
that we want. Check for errors.

This is in preparation for annotating copyin() and related functions
with __result_use_check.

[imp]

sys/dev/mpi3mr/mpi3mr_app.c

586

This seems kinda noisy. Wouldn't the application just fail the I/O? And isn't that enoguh?

[markj]

sys/dev/mpi3mr/mpi3mr_app.c

586

It'd only be due to a bug in the application, or something malicious happening. I can remove it, I was just following the example of an earlier error on line 551. If you still think it's better to remove the print I'm happy to.

[imp]

sys/dev/mpi3mr/mpi3mr_app.c

586

They both seem kinda dubious. They are both programming errors, and aren't something we generally warn about in drivers... And we know that there's a lot of 'warnings' that just scare the horses (eg, it's an expected failure because, for example, they are probing the maximum size)... but this does seem like an actual problem.... I half suspect we'll need to remove it in the future if we leave it in now...

tl;dr: Looking at the rest of the driver, this is fine, since you can't get to this point w/o permission to open the device which means it can't be used easily to ddos the logs...