On Tue, 13 Nov 2007 18:30:30 +0100 (CET)
Tilman Schmidt <[email protected]> wrote:
> From: Tilman Schmidt <[email protected]>
>
> This patch adds basic suspend/resume support to the bas_gigaset ISDN
> driver for the Siemens Gigaset SX255 series of ISDN DECT bases.
>
> Only the USB aspects are handled so far; the ISDN subsystem is not
> notified in any way, for lack of information about how to do that.
> The driver will refuse to suspend if a connection is active.
>
> ...
>
> + if (atomic_read(&cs->hw.bas->basstate) & BS_SUSPEND) {
that's pretty peculiar. We'd only expect to see atomics being used in
conjunction with atomic_add/sub/inc/etc. Here the driver is using an
atomic_t as a state variable. And here's the magic bit:
spin_lock_irqsave(&ucs->lock, flags);
state = atomic_read(&ucs->basstate);
atomic_set(&ucs->basstate, (state & ~clear) | set);
spin_unlock_irqrestore(&ucs->lock, flags);
I'm suspecting that a plain old `int' would be more appropriate here.
Am 15.11.2007 23:50 schrieb Andrew Morton:
>> ...
>>
>> + if (atomic_read(&cs->hw.bas->basstate) & BS_SUSPEND) {
>
> that's pretty peculiar. We'd only expect to see atomics being used in
> conjunction with atomic_add/sub/inc/etc. Here the driver is using an
> atomic_t as a state variable. And here's the magic bit:
>
> spin_lock_irqsave(&ucs->lock, flags);
> state = atomic_read(&ucs->basstate);
> atomic_set(&ucs->basstate, (state & ~clear) | set);
> spin_unlock_irqrestore(&ucs->lock, flags);
>
> I'm suspecting that a plain old `int' would be more appropriate here.
You're right. That's a prehistoric leftover. That variable was
originally accessed using atomic_set_mask() and atomic_clear_mask()
which are unfortunately x86 platform specific.
I'll prepare a cleanup patch.
Thanks,
Tilman
--
Tilman Schmidt E-Mail: [email protected]
Bonn, Germany
Diese Nachricht besteht zu 100% aus wiederverwerteten Bits.
Unge?ffnet mindestens haltbar bis: (siehe R?ckseite)