freebsd32_setcred: Copy all of the setcred fields individually

[jhb]

Summary

This is the more typical style used in compat syscalls. Modern
compilers are smart enough to coalesce multiple member assignments
into a bulk copy.

Obtained from: CheriBSD

[jhb]

This is the important fix. The two prior changes are cleanups, but are pretty small. If you wanted to do an EN for 15.0, you could fix the size in the existing memcpy() to use __rangeof(), but this particular sort of bug is why we generally don't use memcpy() for this sort of thing.

[olce]

Hadn't noticed this new revision (Phabricator mail sending seems to be delayed), so reposting the relevant part of the comment I added on D47878 in the meantime:

Indeed. I should have cast the pointers to uintptr_t or ptrdiff_t to avoid that. I don't think the memcpy() is particularly clever, but these structures are not supposed to change often, so explicitly listing all fields another time is not horrible (avoiding this is the real point for using memcpy(), not performance optimization). The really "clever" fix would be to use X macros, but given I'd like to ask for inclusion of a fix in 15.0, let's forget about it for the time being.

Thanks for noticing and the fix! Testing it now.

sys/compat/freebsd32/freebsd32_misc.c

4258

Not necessary, but OK to keep it, which probably even has advantages (removing it might disturb compiler optimizations, and some later modification re-purposing sc_pad would trigger a compile error here, reminding the author that he has to do something about it).

[jhb]

sys/compat/freebsd32/freebsd32_misc.c

4258

Yes, I actually only included it to keep the one memcpy(). I would normally do a memset() of the structure to zero and then avoid copying the padding field. The compiler will know to only zero the padding in practice. If you would prefer that I'm happy to adjust it.

BTW, the reason I found this bug is that for COMPAT_FREEBSD64 on CheriBSD, the subtraction was between different pointers types as &sc_supp_groups was a uint64_t *.

[olce]

sys/compat/freebsd32/freebsd32_misc.c

4258

If you would prefer that I'm happy to adjust it.

I'd prefer that, but the current code is also fine so really I'm not insisting.

BTW, the reason I found this bug is that for COMPAT_FREEBSD64 on CheriBSD, the subtraction was between different pointers types as &sc_supp_groups was a uint64_t *.

I see. Too bad it happens that, for 32-bit, sc_uid and sc_supp_groups have compatible types, else I would have got this error as well.

[olce]

Planning to commit the minimal fix at D53767, is that OK for you?

[jhb]

Rebase