Re: [ath5k-devel] RFC phy timing register bug

2009-05-13 Thread Forrest Zhang
Hi all, The patch is created by me (forr...@hifulltech.com) or (forr...@263.net). When I test ath5k of linux2.6.30-rc1, I found it cannot communicate with other device on 5G band. After comparing the register settings of ath5k with madwifi and atheros driver, I found two major differences:

Re: [ath5k-devel] RFC phy timing register bug

2009-05-13 Thread Bob Copeland
On Wed, May 13, 2009 at 01:32:47PM +0800, Forrest Zhang wrote: > Hi all, > > The patch is created by me (forr...@hifulltech.com) or (forr...@263.net). Ok thanks, Pavel Roskin already sent a patch for AR5K_PHY_PLL, I'll send this one upstream with your 'From'. -- Bob Copeland %% www.bobcopeland.

[ath5k-devel] RFC phy timing register bug

2009-05-12 Thread Bob Copeland
Hi all, Someone pointed out to me a change in the code with the reset rework, due to my suggestion to use get_bitmask_order() which actually returns highest bit set+1. Nick, can you comment whether the patch below does the right thing and that the original code was correct? [Forrest, is this ide