Date: Sat, 18 Sep 2010 23:38:08 +0100 From: Rui Paulo <rpaulo@FreeBSD.org> To: Adrian Chadd <adrian.chadd@gmail.com> Cc: freebsd-current <freebsd-current@freebsd.org>, freebsd-mobile@freebsd.org Subject: Re: RFT: if_ath HAL refactoring Message-ID: <813E7FEA-2BA9-450C-9826-CE3D799BBDED@FreeBSD.org> In-Reply-To: <AANLkTikZUZ3%2BW%2Bikyiiw-L-X5v4t4EgTNF4vFf4w=oyK@mail.gmail.com> References: <AANLkTikZUZ3%2BW%2Bikyiiw-L-X5v4t4EgTNF4vFf4w=oyK@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 18 Sep 2010, at 17:09, Adrian Chadd wrote: > Hi all, > > I've uploaded a snapshot of the if_ath HAL which i've been working on. > I've been refactoring out various bits of the AR5416 HAL into > something that resembles the ath9k hardware MAC/PHY operations to make > it easier to port further ath9k updates over. It also includes the > AR9100 support (but it's missing a couple bits of glue needed to use > it outside of my GIT tree.) Finally, it includes the probe/attach > operations for the AR2427, but I haven't at all tested it yet (and > i've explained why it isn't working in a previous email.) > > It's available for download at http://people.freebsd.org/~adrian/ath/ > . There's a diff against src/sys/files/conf and a tarball that just > replaces the ath device/module directory. > > Note you'll need to add "device if_ath_pci" to your kernel > configuration file as the PCI bus glue is now not built by default in > a static kernel in this HAL. (It's included in the module Makefile by > default.) This was done to allow multiple backend bus types - now > being PCI and "AHB" for the AR9100 SoC. > > I'd appreciate testing by AR5416/AR9160/AR9280/AR9285 users. I only > currently have easy access to AR5416/AR9160. Please let me know > immediately if something doesn't work with this which does work in > -head. > > If you're an AR2427 user, I'd appreciate some brief testing with > HAL_DEBUG_ATTACH/HAL_DEBUG_EEPROM enabled (sysctl > hw.ath.hal.debug=0x8002.) I doubt it'll work but it should attach and > then spit out some computetxtime errors. Let me know if that happens > and I'll see about trying to fix that. Can you also provide a diff against HEAD please? Regards, -- Rui Paulo
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?813E7FEA-2BA9-450C-9826-CE3D799BBDED>