[RFC/RFT] ssb: resolve alternate SPROM offset for 14e4:4315

[Michael Buesch]

On Thursday 06 May 2010 11:53:01 Michael Buesch wrote:
> On Thursday 06 May 2010 07:00:13 Larry Finger wrote:
> > This patch and the patch by Gabor entitled "[PATCH] ssb: Implement
> > fast powerup delay calculation" are enough to allow the netbook from
> > John to work with ssb/b43. As this patch tampers with the SPROM offset for
> > 14e4:4315 devices, it should be tested by anyone with an LP PHY that works
> > to ensure that it is not killed.
> > 
> > The essential elements of this patch will also be tested as part of kernel Bug
> > #15825.
> > 
> > Larry
> >  
> > Index: wireless-testing/drivers/ssb/pci.c
> > ===================================================================
> > --- wireless-testing.orig/drivers/ssb/pci.c
> > +++ wireless-testing/drivers/ssb/pci.c
> > @@ -631,8 +631,17 @@ static int ssb_pci_sprom_get(struct ssb_
> >  		return -ENODEV;
> >  	}
> >  
> > -	bus->sprom_offset = (bus->chipco.dev->id.revision < 31) ?
> > -		SSB_SPROM_BASE1 : SSB_SPROM_BASE31;
> > +	/* get SPROM offset: SSB_SPROM_BASE1 except for chipcommon rev >= 31
> > +	 * or chip ID is 0x4312 and bit 0x2 is set in chipcommon status
> > +	 */
> > +	if (bus->chipco.dev->id.revision >= 31)
> > +		bus->sprom_offset = SSB_SPROM_BASE31;
> > +	else if (bus->chip_id == 0x4312 && (bus->chipco.status & 0x02))
> > +		bus->sprom_offset = SSB_SPROM_BASE31;
> > +	else
> > +		bus->sprom_offset = SSB_SPROM_BASE1;
> > +
> > +	ssb_dprintk(KERN_INFO PFX "SPROM offset is 0x%x\n", bus->sprom_offset);
> >  
> >  	buf = kcalloc(SSB_SPROMSIZE_WORDS_R123, sizeof(u16), GFP_KERNEL);
> >  	if (!buf)
> > Index: wireless-testing/drivers/ssb/scan.c
> > ===================================================================
> > --- wireless-testing.orig/drivers/ssb/scan.c
> > +++ wireless-testing/drivers/ssb/scan.c
> > @@ -306,6 +306,11 @@ int ssb_bus_scan(struct ssb_bus *bus,
> >  		}
> >  		tmp = scan_read32(bus, 0, SSB_CHIPCO_CAP);
> >  		bus->chipco.capabilities = tmp;
> > +		if (bus->chip_rev >= 11)
> 
> This still is wrong. the chip_rev is not the chipcommon core revision.

Although technically in most (all?) cases it might be the same number.
I'm not sure on that one.
But it is read from an entirely different register and I think it's plain
wrong to assume that chip_rev equals the chipcommon core rev.

> We already went through this and we decided to read the chipstat later.
> Read it in chipcommon init. The bus scan is _way_ too early to read this.
> 
> > +			bus->chipco.status = scan_read32(bus, 0,
> > +					     SSB_CHIPCO_CHIPSTAT);
> > +		else
> > +			bus->chipco.status = 0;
> >  	} else {
> >  		if (bus->bustype == SSB_BUSTYPE_PCI) {
> >  			bus->chip_id = pcidev_to_chipid(bus->host_pci);
> > 
> > 
> 
> 

-- 
Greetings, Michael.