On Fri, 20 Jan 2023 at 17:04, Johan Hovold <johan@xxxxxxxxxx> wrote:
>
> On Fri, Jan 20, 2023 at 10:23:18AM +0100, Ard Biesheuvel wrote:
> > (cc Peter, Heinrich)
> >
> > On Thu, 19 Jan 2023 at 17:45, Johan Hovold <johan+linaro@xxxxxxxxxx> wrote:
> > >
> > > The efivar ops are typically registered at subsys init time so that
> > > they are available when efivarfs is registered at module init time.
> > >
> > > Other efivars implementations, such as Google SMI, exists and can
> > > currently be build as modules which means that efivar may not be
> > > available when efivarfs is initialised.
> > >
> > > Move the efivar availability check from module init to when the
> > > filesystem is mounted to allow late registration of efivars.
> > >
> > > Signed-off-by: Johan Hovold <johan+linaro@xxxxxxxxxx>
> >
> > I think this change is fine in principle, but I 'm not sure if there
> > is user space code that the distros are carrying that might get
> > confused by this: beforehand, efivarfs would not exist in
> > /proc/filesystems and now, it will but trying to mount it might fail.
>
> User space must already handle mount failing since commit 483028edacab
> ("efivars: respect EFI_UNSUPPORTED return from firmware") so that should
> not be an issue.
>
Fair enough
