Am 30.10.2007 22:01 schrieb Andrew Morton:
quoted text
> On Mon, 29 Oct 2007 22:41:30 +0100 (CET)
> Tilman Schmidt <tilman@imap.cc> wrote:
>=20
>> From: Tilman Schmidt <tilman@imap.cc>
>>
>> Add basic suspend/resume support to the usb_gigaset driver.
>> (Corrected version.)
>>
>=20
> You're not a big fan of checkpatch, I see.

Sorry. I promise to mend my ways.

quoted text
>> +static int gigaset_suspend(struct usb_interface *intf, pm_message_t m=
essage)
quoted text
>> +{
>> +	struct cardstate *cs;
>> +	struct usb_cardstate *ucs;
>> +
>> +	if ((cs =3D usb_get_intfdata(intf)) =3D=3D NULL ||
>> +	    (ucs =3D cs->hw.usb) =3D=3D NULL) {
>> +		err("%s: no cardstate", __func__);
>> +		return -EFAULT;
>> +	}
>=20
> Is the above reeeeeely needed?  I bet it never happens.

I'm a great believer in defensive programming. :-)

Anyway, to be sure these checks aren't needed, I would need the
assurance that the suspend and resume methods are serialized with
the probe and disconnect methods. Are they?

Thanks,
Tilman

--=20
Tilman Schmidt                          E-Mail: tilman@imap.cc
Bonn, Germany
Diese Nachricht besteht zu 100% aus wiederverwerteten Bits.
Unge=F6ffnet mindestens haltbar bis: (siehe R=FCckseite)