Regards, Nishanth Menon > -----Original Message----- > From: Pandita, Vikram > Sent: Friday, June 12, 2009 9:50 AM > To: Menon, Nishanth; felipe.balbi@xxxxxxxxx > Cc: linux-omap@xxxxxxxxxxxxxxx > Subject: RE: [PATCH 1/2] Serial: Define IRQ flags for 8250 driver > > > > >-----Original Message----- > >From: Menon, Nishanth > >Sent: Friday, June 12, 2009 9:46 AM > >To: felipe.balbi@xxxxxxxxx > >Cc: Pandita, Vikram; linux-omap@xxxxxxxxxxxxxxx > >Subject: RE: [PATCH 1/2] Serial: Define IRQ flags for 8250 driver > > > >> -----Original Message----- > >> From: Felipe Balbi [mailto:felipe.balbi@xxxxxxxxx] > >> Sent: Friday, June 12, 2009 1:38 AM > >> To: Menon, Nishanth > >> Cc: Pandita, Vikram; linux-omap@xxxxxxxxxxxxxxx > >> Subject: Re: [PATCH 1/2] Serial: Define IRQ flags for 8250 driver > >> > >> On Fri, Jun 12, 2009 at 03:46:37AM +0200, ext Menon, Nishanth wrote: > >> > > >> > > -----Original Message----- > >> > > From: linux-omap-owner@xxxxxxxxxxxxxxx [mailto:linux-omap- > >> > > owner@xxxxxxxxxxxxxxx] On Behalf Of Pandita, Vikram > >> > > Sent: Thursday, June 11, 2009 7:50 PM > >> > > To: linux-omap@xxxxxxxxxxxxxxx > >> > > Cc: Pandita, Vikram > >> > > Subject: [PATCH 1/2] Serial: Define IRQ flags for 8250 driver > >> > > > >> > > > >> > > + /* Get IRQ Trigger Flag */ > >> > > + if (up->port.flags & UPF_IRQ_TRIG_RISING) > >> > > + irq_flags |= IRQF_TRIGGER_RISING; > >> > > + else if (up->port.flags & UPF_IRQ_TRIG_FALLING) > >> > > + irq_flags |= IRQF_TRIGGER_FALLING; > >> > > + else if (up->port.flags & UPF_IRQ_TRIG_HIGH) > >> > > + irq_flags |= IRQF_TRIGGER_HIGH; > >> > > + else if (up->port.flags & UPF_IRQ_TRIG_LOW) > >> > > + irq_flags |= IRQF_TRIGGER_LOW; > >> > > + > >> > Blame it on my dislike for nested if elseif, but... > >> > irq_flags |= > >> > (up->port.flags & UPF_IRQ_TRIG_RISING)? IRQF_TRIGGER_RISING : > >> > (up->port.flags & UPF_IRQ_TRIG_FALLING)? IRQF_TRIGGER_FALLING : > >> > (up->port.flags & IRQF_TRIGGER_HIGH)? IRQF_TRIGGER_HIGH : > >> > (up->port.flags & UPF_IRQ_TRIG_LOW)? IRQF_TRIGGER_LOW : 0; > >> > > >> > Makes sense? > >> > >> switch (up->port.flags & UPF_IRQ_FLAGS_MASK) { > >> case UPF_IRQ_TRIG_RISING: > >> irq_flags |= IRQF_TRIGGER_RISING; > >> break; > >> case UPF_IRQ_TRIG_FALLING: > >> irq_flags |= IRQF_TRIGGER_FALLING; > >> break; > >> case UPF_IRQ_TRIG_HIGH: > >> irq_flags |= IRQF_TRIGGER_HIGH; > >> break; > >> case UPF_IRQ_TRIG_LOW: > >> irq_flags |= IRQF_TRIGGER_LOW; > >> break; > >> default: > >> printk(KERN_ERR "Unsupported flag\n"); > >> return -EINVAL; > >> } > >> > >Better :) infact, if the port.flags = UPF_IRQ_TRIG_LOW | > UPF_IRQ_TRIG_HIGH it might be better to > >return -EINVAL, which Felipe's change does :). > > Needs more investigation as to how current boards using 8250 driver do not > pass any flag (maybe just SHARED flag) and they still work. Maybe some > default taken by request_irq() > > In short NO_ need to return failure as no flag is a valid use case. Replacing switch (up->port.flags & UPF_IRQ_FLAGS_MASK) With #define UPF_IRQ_LEVEL_FLAGS_MASK \ (UPF_IRQ_TRIG_RISING | UPF_IRQ_TRIG_FALLING |\ UPF_IRQ_TRIG_HIGH | UPF_IRQ_TRIG_LOW) switch (up->port.flags & UPF_IRQ_LEVEL_FLAGS_MASK) { will allow for valid -EINVAL return. Regards, Nishanth Menon -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
