module_load_install should check and return results . . .

Makes some sense - however, I would think the return value should be exactly consistent with module_load_include()... In other words, maybe it should just do return module_load_include('install', $module) instead?

missing doxygen for @return

looks good. wait for green before commit.

Er, in the latest version of the patch you left out the most important part - the actual code change :)

I added that back in, plus clarified the @return docs a tiny bit to indicate that the name of the file is what's returned. I also added a blank line before the @return, to comply with coding standards.

I think it's probably still RTBC.

#9: modules-inc-828648-9.patch queued for re-testing.

Although badly needed, this is D8 material according to the rules (I had to learn today). It may be backported at a later point in time (though that's unlikely).

I think that making a function which previously returned nothing now return something (and be consistent with other similar functions) is, although technically an API "change", not one that could ever break anyone's code under any realistic scenario.

Moving back to D7 for consideration.

module.inc_.patch queued for re-testing.

Agreed, this seems to just be fixing an oversight in the existing API.

Committed to HEAD.