Re: efivarfs fixes without the commit being fixed in 6.1 and 6.6 (resending without html)

Re: efivarfs fixes without the commit being fixed in 6.1 and 6.6 (resending without html)

[Ted Brandston]

Hi, this is my first time posting to a kernel list (third try, finally
figured out the html-free -- sorry for the noise).

I noticed that in the 6.6 kernel there's a fix commit from Ard [1] but
not the commit it's fixing ("efivarfs: Add uid/gid mount options").
Same thing in 6.1 [2]. The commit being fixed doesn't appear until 6.7
[3].

I'm not familiar with this code so it's unclear to me if this might
cause problems, but I figured I should point it out.

[1]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/fs/efivarfs/super.c?h=linux-6.6.y&id=48be1364dd387e375e1274b76af986cb8747be2c
[2]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/log/fs/efivarfs/super.c?h=linux-6.1.y
[3]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/log/fs/efivarfs/super.c?h=linux-6.7.y

Thanks!

Re: efivarfs fixes without the commit being fixed in 6.1 and 6.6 (resending without html)

[Greg KH]

On Thu, Mar 21, 2024 at 10:43:05AM -0400, Ted Brandston wrote:
> Hi, this is my first time posting to a kernel list (third try, finally
> figured out the html-free -- sorry for the noise).
> 
> I noticed that in the 6.6 kernel there's a fix commit from Ard [1] but
> not the commit it's fixing ("efivarfs: Add uid/gid mount options").
> Same thing in 6.1 [2]. The commit being fixed doesn't appear until 6.7
> [3].
> 
> I'm not familiar with this code so it's unclear to me if this might
> cause problems, but I figured I should point it out.
> 
> [1]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/fs/efivarfs/super.c?h=linux-6.6.y&id=48be1364dd387e375e1274b76af986cb8747be2c
> [2]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/log/fs/efivarfs/super.c?h=linux-6.1.y
> [3]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/log/fs/efivarfs/super.c?h=linux-6.7.y

Good catch.  Ard, should this be reverted?

thanks,

greg k-h

Re: efivarfs fixes without the commit being fixed in 6.1 and 6.6 (resending without html)

[Ard Biesheuvel]

On Thu, 21 Mar 2024 at 15:59, Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Thu, Mar 21, 2024 at 10:43:05AM -0400, Ted Brandston wrote:
> > Hi, this is my first time posting to a kernel list (third try, finally
> > figured out the html-free -- sorry for the noise).
> >
> > I noticed that in the 6.6 kernel there's a fix commit from Ard [1] but
> > not the commit it's fixing ("efivarfs: Add uid/gid mount options").
> > Same thing in 6.1 [2]. The commit being fixed doesn't appear until 6.7
> > [3].
> >
> > I'm not familiar with this code so it's unclear to me if this might
> > cause problems, but I figured I should point it out.
> >
> > [1]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/fs/efivarfs/super.c?h=linux-6.6.y&id=48be1364dd387e375e1274b76af986cb8747be2c
> > [2]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/log/fs/efivarfs/super.c?h=linux-6.1.y
> > [3]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/log/fs/efivarfs/super.c?h=linux-6.7.y
>
> Good catch.

Indeed. Thanks for reporting this.

>  Ard, should this be reverted?
>

With this fix applied, we'll end up kfree()'ing a pointer that is
guaranteed to be NULL, on a code path that typically executes once per
boot, if at all.

So in practical terms, there is really no difference, and this is the
only thing I personally care about.

So I wouldn't mind if we just left them, unless there are other
concerns wrt to maintenance, tidiness etc.

Re: efivarfs fixes without the commit being fixed in 6.1 and 6.6 (resending without html)

[Greg KH]

On Thu, Mar 21, 2024 at 04:28:56PM +0100, Ard Biesheuvel wrote:
> On Thu, 21 Mar 2024 at 15:59, Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Thu, Mar 21, 2024 at 10:43:05AM -0400, Ted Brandston wrote:
> > > Hi, this is my first time posting to a kernel list (third try, finally
> > > figured out the html-free -- sorry for the noise).
> > >
> > > I noticed that in the 6.6 kernel there's a fix commit from Ard [1] but
> > > not the commit it's fixing ("efivarfs: Add uid/gid mount options").
> > > Same thing in 6.1 [2]. The commit being fixed doesn't appear until 6.7
> > > [3].
> > >
> > > I'm not familiar with this code so it's unclear to me if this might
> > > cause problems, but I figured I should point it out.
> > >
> > > [1]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/fs/efivarfs/super.c?h=linux-6.6.y&id=48be1364dd387e375e1274b76af986cb8747be2c
> > > [2]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/log/fs/efivarfs/super.c?h=linux-6.1.y
> > > [3]: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/log/fs/efivarfs/super.c?h=linux-6.7.y
> >
> > Good catch.
> 
> Indeed. Thanks for reporting this.
> 
> >  Ard, should this be reverted?
> >
> 
> With this fix applied, we'll end up kfree()'ing a pointer that is
> guaranteed to be NULL, on a code path that typically executes once per
> boot, if at all.
> 
> So in practical terms, there is really no difference, and this is the
> only thing I personally care about.
> 
> So I wouldn't mind if we just left them, unless there are other
> concerns wrt to maintenance, tidiness etc.
> 

Ok, let's leave it, as long as there's no bad side affects.

greg k-h