Date: Fri, 14 Oct 2011 09:38:30 +0800 From: Adrian Chadd <adrian@freebsd.org> To: Sceak <sceakie@gmail.com> Cc: wireless@freebsd.org Subject: Re: AR9285 RX sensitivity (was Re: Scanning with ath) Message-ID: <CAJ-Vmom5LMvZV4yrGitqKAJnpMVwkDkCTTy90j=%2B3mhjNnOuZA@mail.gmail.com>
index | next in thread | raw e-mail
Hi, Right. Please enable hal/ath debugging: options ATH_DEBUG options AH_DEBUG options ATH_DIAGAPI Then enable diversity debugging: sysctl dev.ath.0.hal.debug=0x00100000 The relevant code is in sys/dev/ath/ath_hal/ar9002/ar9285_diversity.c . I "stole" the tx antenna switch hooks to implement the ar9285 RX antenna diversity. It's a hack which I'll make go away at some point (or if someone would like a mini-project, please step up and let me know.) sysctl dev.ath.0.txantenna thus: 0 = variable RX diversity 1 = RX prefer antenna A 2 = RX prefer antenna B You could try setting it to 1 or 2 and doing some scans. "Variable" allows the hardware to choose which antenna to use. There's more complicated software-driven AR9285 diversity which I've included the code for, but it's disabled at the moment due to a lack of time to thoroughly debug issues (much like what you were seeing, but when doing live traffic.) I'm happy to help out anyone who would like to toy around with AR9285 software driven diversity, I have plenty of documentation and notes which I can share. Adrianhelp
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAJ-Vmom5LMvZV4yrGitqKAJnpMVwkDkCTTy90j=%2B3mhjNnOuZA>