Re: [PATCH v4 2/2] efi: an sysfs interface for user to update efi firmware

[Greg KH <gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org>]

On Wed, Apr 15, 2015 at 05:19:09PM -0700, Roy Franz wrote:
> On Wed, Apr 15, 2015 at 8:45 AM, Andy Lutomirski <luto-kltTT9wpgjJwATOyAt5JVQ@public.gmane.org> wrote:
> > On Apr 15, 2015 6:20 AM, "Greg Kroah-Hartman"
> > <gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org> wrote:
> >>
> >> On Tue, Apr 14, 2015 at 11:52:48AM -0400, Andy Lutomirski wrote:
> >> > On Tue, Apr 14, 2015 at 10:09 AM, Greg Kroah-Hartman
> >> > <gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r@public.gmane.org> wrote:
> >> > > On Tue, Apr 14, 2015 at 05:44:56PM +0800, Kweh, Hock Leong wrote:
> >> > >> From: "Kweh, Hock Leong" <hock.leong.kweh-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> >> > >>
> >> > >> Introducing a kernel module to expose capsule loader interface
> >> > >> for user to upload capsule binaries. This module leverage the
> >> > >> request_firmware_direct_full_path() to obtain the binary at a
> >> > >> specific path input by user.
> >> > >>
> >> > >> Example method to load the capsule binary:
> >> > >> echo -n "/path/to/capsule/binary" > /sys/devices/platform/efi_capsule_loader/capsule_loader
> >> > >
> >> > > Ick, why not just have the firmware file location present, and copy it
> >> > > to the sysfs file directly from userspace, instead of this two-step
> >> > > process?
> >> >
> >> > Because it's not at all obvious how error handling should work in that case.
> >>
> >> I don't understand how the error handling is any different.  The kernel
> >> ends up copying the data in through the firmware interface both ways, we
> >> just aren't creating yet-another-firmware-path in the system.
> >
> > If I run uefi-update-capsule foo.bin, I want it to complain if the
> > UEFI call fails.  In contrast, if I boot and my ath10k firmware is
> > bad, there's no explicit user interaction that should fail; instead I
> > have no network device and I get to read the logs and figure out why.
> > IOW bad volatile device firmware is just like a bad device driver,
> > whereas bad UEFI capsules are problems that should be reported to
> > whatever tried to send them to UEFI.
> >
> > --Andy
> >
> There are several types of errors that can be returned by
> UpdateCapsule(), allowing
> us to distinguish between capsules that are not supported by the
> platform, capsules
> that must be updated at boot time, and capsule updates that failed due
> to a device error.
> I think we really do want this type of information returned to capsule loader.

Ok, all of that sounds like you really want a character device, with an
ioctl, to do this.  Just use a misc device and your infrastructure will
be handled for you (sysfs, character device, etc.) and away you go.

thanks,

greg k-h