Date: Thu, 24 Oct 2013 20:49:56 -0400 From: Sean Bruno <sean_bruno@yahoo.com> To: Adrian Chadd <adrian.chadd@gmail.com> Cc: "freebsd-wireless@freebsd.org" <freebsd-wireless@freebsd.org> Subject: Re: [iwn] request to test: linkq, plcp, txrate fixes Message-ID: <1382662196.2498.1.camel@localhost> In-Reply-To: <CAJ-VmokLJX7Mjt=7VrPrecS8odycGPOzz=knMXVckip_TTHdqw@mail.gmail.com> References: <CAJ-VmokLJX7Mjt=7VrPrecS8odycGPOzz=knMXVckip_TTHdqw@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
--=-kutRUkuTFt8Sr5xjNGCZ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable On Thu, 2013-10-24 at 13:04 -0700, Adrian Chadd wrote: > Hi! >=20 > I have a patch against iwn(4): >=20 > http://people.freebsd.org/~adrian/ath/20131024-iwn-1-fix-xmit-rate-plcp-l= inkq.diff >=20 > This does a few things: >=20 > * The multi-rate retry ("link quality" stuff) on iwn is a table, and I > think I broke it slightly when I implemented 11n rate control. This > disables it for now, at least until I / someone implements a replacement > table setup and linkq decision. >=20 > * get rid of the rate index stuff - it was used to decide a few things an= d > it was doing legacy-only rate index lookups. I've just removed it; it's n= ot > needed. >=20 > * Teach the raw transmit path to use the new plcp lookup routine that set= s > things up. >=20 > * Correctly implement 11g protection. Right now it was a bit of a hack an= d > it was not correct for 11a or 11n operation. >=20 > I still get transmit hangs on my 4965 but the "invalid rate" failures hav= e > disappeared. But for some odd reason, the transmits completed but weren't > being marked as failed. I don't yet know why. >=20 > I'm also not sure why transmit is hanging; sometimes it looks like DMA > stalls somehow. >=20 > So - maybe this'll help with various 11n issues on iwn(4), maybe it won't= . > I'm not sure yet. I'm trying to avoid taking ownership of iwn(4) but I > rather dislike the occasional 5100/4965 hangs.. >=20 > Thanks, >=20 >=20 >=20 > -adrian running this now against 11-CURRENT. No issues noted on my adapter (Centrino Advanced-N 6205 [Taylor Peak]). All seems very good. I haven't tried it the office though, and that's the one that's the most interesting I fear. sean --=-kutRUkuTFt8Sr5xjNGCZ Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.15 (FreeBSD) iQEcBAABAgAGBQJSacAxAAoJEBkJRdwI6BaHrQIH/RlMGJQi5cCnvKcJz6UdWgj7 rruLond4psLWy2exENV2Yj8+rI8GpFSxfQq2vBKhQHfBwIIPZQ9KcA7opiaIrYR+ oD8cKcAJuBQpB1K/rtzKzu6DHoHevZ/toUkguD46oPsh2cBKaCy4voOIRQm9Lxw8 2b0QeUBe6GhZVy3670aaIKcfuGPsJiQNiZNiBJSUDTjoEEpe5M00104q82pnhqtN iVA6f06smwFSCZxrj1CVxI4fTGN3w3CL7n26UA6qmJZM+9wbZ49Pg9BomRZv8yDl CGaoAox/c5b+S4bOTYICUTVDLNZm51IXIN0cxszN9jnwAx4g1piHjkW4Rhhy0Fo= =neCW -----END PGP SIGNATURE----- --=-kutRUkuTFt8Sr5xjNGCZ--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1382662196.2498.1.camel>