Re: Please include user-namespace.git in linux-next

From: Eric W. Biederman
Date: Tue May 22 2012 - 13:26:47 EST


Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx> writes:

> I think something is wrong with commit e1c972b6 "userns: Add negative depends
> on entries to avoid building code that is userns unsafe".

My apologies for the delayed reply.

> With gcc 4.4.6 on CentOS 6.2, "make allnoconfig" where UIDGID_CONVERTED should
> become y is showing
>
> Symbol: USER_NS [=n]
> Type : boolean
> Prompt: User namespace (EXPERIMENTAL)
> Defined at init/Kconfig:880
> Depends on: NAMESPACES [=y] && EXPERIMENTAL [=n] && UIDGID_CONVERTED [=n]
> Location:
> -> General setup
> -> Namespaces support (NAMESPACES [=y])
> Selects: UIDGID_STRICT_TYPE_CHECKS [=n]
>
> .
> I think this commit meant "!FOO" rather than "FOO = n",
> othwewise there is no way for linux-next-20120518 to enable USER_NS.
> ----------------------------------------
> PATCH: user_ns: Fix wrong dependency in UIDGID_CONVERTED.

It turns out to be more subtle than that. The issue is that I have
"depends on USB_DEVICEFS = n" and then USB_DEVICEFS was removed.

"depends on FOO = n" is the only way I found that will succeed when in
verifying that FOO is neither enabled nor modular. "depends on !FOO"
appears to succeed when FOO = m which is not at all what I want.

After both trees are merged I will have to remove that
"depends on USB_DEVICEFS = n" dependency.

I can't see any possible way to do hand this before then.

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