Re: [PATCH] bcm43xx-mac80211: Fix specs typo for baseband attenuation

["John W. Linville" <linville@tuxdriver.com>]

On Thu, Jul 26, 2007 at 12:56:20PM -0500, Larry Finger wrote:
> John W. Linville wrote:
> >On Thu, Jul 26, 2007 at 11:33:01AM -0500, Larry Finger wrote:
> >>A typo in the specs interchanges the branches in an if statement, which
> >>breaks operations for a BCM4306/rev 2 that has phy->analog == 1.
> >
> >>@@ -1895,7 +1895,7 @@ void bcm43xx_phy_set_baseband_attenuatio
> >> 		bcm43xx_write16(dev, BCM43xx_MMIO_PHY0,
> >> 				(bcm43xx_read16(dev, BCM43xx_MMIO_PHY0)
> >> 				 & 0xFFF0) | baseband_attenuation);
> >>-	} else if (phy->analog == 1) {
> >>+	} else if (phy->analog != 1) {
> >> 		bcm43xx_phy_write(dev, BCM43xx_PHY_DACCTL,
> >> 				  (bcm43xx_phy_read(dev, BCM43xx_PHY_DACCTL)
> >> 				   & 0xFFC3) | (baseband_attenuation << 2));
> >
> >Larry,
> >
> >How does this relate to the bcm43xx patch you asked me to revert
> >(and has been reverted in F-7)?  That one change "==" to ">", while
> >this one changes "==" to "!=".  Instead of reverting the other,
> >should it do the same thing as this?
> 
> It really doesn't matter whether one uses ">" or "!=" here. The number in 
> question is >= 0 and the test for for zero occurs earlier in the routine 
> and ends with a return. Now that you mention it, it would be best to make 
> bcm43xx nad bcm43xx-mac80211 look the same. I'll modify and resubmit the 
> patch.

Hmmm...well, if you asked to revert it on bcm43xx, why is it appropriate here?

John
-- 
John W. Linville
linville@tuxdriver.com