Re: [PATCH] 2.5.41 s390 (8/8): 16 bit uid/gids.

From: Tim Hockin (thockin@hockin.org)
Date: Wed Oct 09 2002 - 17:43:35 EST


> > Because Sparc64/s390x/? still need to tell highuid.h to do macro magic for
> > NEW_TO_OLD_UID() and friends in some places and not others. A CONFIG_XXX
> > applies all the time to all files.
>
> If __UID16 works, then renaming it to CONFIG_UID16_ONLY _must_ also work.
>
> I don't understand your argument about other architectures. I'm claiming
> that __UID16 is a config option, and that it must be renamed to _show_
> that it is a config option.

Renaming it, if that is all you want, is fine by everyone, I'm sure. I was
trying to make the point that it is NOT a config option, because for some
architectures, you only want it defined for SOME FILES.

architectures that define CONFIG_UID16 get high2lowuid() all the time
        - no changes needed on any of them

architectures that do not define CONFIG_UID16 and don't want it ever get
straight uids, no high2lowuid() cruft
        - no changes needed on any of them

architectures that can use uid16.c code, but do not want high2lowuid()
everywhere need to change
        - define CONFIG_UID16_SYSCALLS (build uid16.o)
        - explicitly call out where it wants high2lowuid()
                - this was called __UID16, but can be called
                CONFIG_UID16_HERE or REALLY_DO_HIGHUID_CRAP or FRED for all
                I care.

I just want to make sure that you're telling me to name something CONFIG_FOO
when, in fact, it is defined by various bits of code that want to change the
behavior of a particular header.

In that case we can just have them define CONFIG_UID16 and not add an OR in
there at all. I just thought you'd find a .c file defining CONFIG_FOO to be
grosser, yet. But on further argument, I guess it's not.

I'll consider it a done deal, unless you want to stop me :) Martin, do you
want to just change the __UID16 defines to CONFIG_UID16 and nix the || in
highuid.h and resubmit?

I have the same patch in my local tree.

Tim
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Tue Oct 15 2002 - 22:00:33 EST