[systemd-devel] [PATCH 03/10] rules: load sg module from 80-drivers.rules

[Martin Pitt]

Kay Sievers [2014-07-16 12:53 +0200]:
> On Wed, Jul 16, 2014 at 12:09 PM, Jon Severinsson <jon at severinsson.net> wrote:
> > From: Martin Pitt <martin.pitt at ubuntu.com>
> >
> > Taken from the Debian specific rules, this is the remaining difference over the
> > upstream 80-drivers.rules.
> >
> > Bug-Debian: http://bugs.debian.org/657948
> > ---
> >  rules/80-drivers.rules | 1 +
> >  1 file changed, 1 insertion(+)
> >
> > diff --git a/rules/80-drivers.rules b/rules/80-drivers.rules
> > index 8551f47..f764075 100644
> > --- a/rules/80-drivers.rules
> > +++ b/rules/80-drivers.rules
> > @@ -9,5 +9,6 @@ SUBSYSTEM=="memstick", RUN{builtin}+="kmod load ms_block mspro_block"
> >  SUBSYSTEM=="i2o", RUN{builtin}+="kmod load i2o_block"
> >  SUBSYSTEM=="module", KERNEL=="parport_pc", RUN{builtin}+="kmod load ppdev"
> >  KERNEL=="mtd*ro", ENV{MTD_FTL}=="smartmedia", RUN{builtin}+="kmod load sm_ftl"
> > +SUBSYSTEM=="scsi", ENV{DEVTYPE}=="scsi_device", TEST!="[module/sg]", RUN{builtin}+="kmod load sg"
> 
> We do not want to force-load the sg driver. Why would that be needed?

Most things don't need it. This was re-added because of
https://bugs.debian.org/657948 as a user complained that without it
some tape devices don't work any more with mtx.

Again, I deliberately didn't forward that upstream as this really
should be fixed in mtx. It's just a rule which the Debian package has
had for many years.

Martin

-- 
Martin Pitt                        | http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)