Re: [patch 03/15] cgroup: dont call vfs_mkdir

[Miklos Szeredi]

> > Well, I'd hardly call this whole patchset messing up the kernel, and
> > you seemed to agree with most of the patches.
> 
> Of course.  Good patches will always again, but preparing for apparmor
> is not an excuse for bad patches.

It's not an excuse.  It's one of *several* reasons for why the patches
are good.  And no, "help *cleanly* integrate AA" is not a reason
against patches.

Please try to keep your arguments technical (like code duplication in
reiserfs, etc), which are perfectly valid.  I don't like the reiserfs
duplication, but as I've said, it's needed anyway for some other
reason.  I can dig out the original patch and include the description
in this one if that helps.

And yes, all these are cleanups.  Recursing into the vfs_ functions
from inside the filesystems is plain wrong most of the time.

For example fsnotify_change() call in fat_generic_ioctl() would
actually *break* after the IS_IMMUTABLE() cleanups, because it's
actually calling this function before changing the "SYS" attribute
(which is interpreted as immutable) on the inode.  So without this
cleanup that notify_change() would always fail.

Miklos
--
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/