[Btrfs-users] [bugreport] btrfs 0.11

[Chris mason]

On Monday 21 January 2008, Jeff Mahoney wrote:
> - gpg control packet
>
> Chris mason wrote:
> > On Saturday 19 January 2008, Jeff Mahoney wrote:
> >> - gpg control packet
> >>
> >> Gabor MICSKO wrote:
> >>> On Sat, 2008-01-19 at 16:16 -0500, Jeff Mahoney wrote:
> >>>> Yeah, that'd do it. I include the following patch in openSUSE's
> >>>> btrfs package.
> >>>>
> >>>> - -Jeff
> >>>>
> >>>> diff -rup a/source/file.c b/source/file.c
> >>>> - --- a/file.c	2007-06-12 11:36:58.000000000 -0400
> >>>> +++ b/file.c	2007-06-12 14:06:53.000000000 -0400
> >>>> @@ -482,7 +482,7 @@ static ssize_t btrfs_file_write(struct f
> >>>>  		goto out;
> >>>>  	if (count == 0)
> >>>>  		goto out;
> >>>> - -	err = remove_suid(file->f_path.dentry);
> >>>> +	err = remove_suid(&file->f_path);
> >>>>  	if (err)
> >>>>  		goto out;
> >>>>  	file_update_time(file);
> >>>
> >>> Is this patch against btrfs-0.11? Because it fails to apply.
> >>
> >> Oh. No. You're right. That's against 0.9. Here's the 0.11 patch.
> >
> > Thanks Jeff.  Is there a #ifdef that I might be able to wrap around this?
>
> Nothing you'd like. I don't have a solution short of an ugly
> compatibility patch that depends on CONFIG_SUSE_KERNEL (and whatever the
> debian/ubuntu equivalent is) or a dependency on autotools. For now, I'm
> fine assigning the responsibility to the distributions. The patch was
> trivial to update for openSUSE.

For people that like to recompile their own modules on the distro kernels, I'm 
fine with including a one liner around this.  It would be nice though if I 
could #ifdef APPARMOUR_PATCHED or something.

#ifdef ILOVESUSE is fine too

-chris