Date: Wed, 25 Mar 2015 05:42:19 +0000 From: Miguel Clara <miguelmclara@gmail.com> To: Adrian Chadd <adrian@freebsd.org> Cc: "freebsd-wireless@freebsd.org" <freebsd-wireless@freebsd.org> Subject: Re: ath0 performence issues "ar9300_Stub_GetCTSTimeout" "ar9300_Stub_GetCTSTimeout Message-ID: <CADGo8CW%2Bfuq5BfzYUcHd9S3GCKXdSvvcLLd1mjz3bTvAVWmGcw@mail.gmail.com> In-Reply-To: <CADGo8CWLTvGcErHzkK49NM3GH%2BizDfCD=a5G3ryOOeHzV2Bn_Q@mail.gmail.com> References: <CADGo8CWitJUhU0%2BjDX59gv%2Bc-4uhpgf58=r_K%2Bfq=4acDTni8A@mail.gmail.com> <CADGo8CUMzY7j6h55dscW5t0jm5Jb_yBRj56UfDei0oY1GpiWww@mail.gmail.com> <CADGo8CWDhwDfiJr=gDDvvpe7pX6ZJgUJhVxtDx6PEavSLzX7Ew@mail.gmail.com> <CADGo8CXeHiWypp3O6LT9x3aBwxHcUFGpfx0GpDJeE-OcG0YjgQ@mail.gmail.com> <CADGo8CWBjVvvAxpXKSrr6BqOE7SBBv%2B2trcGwa6JSf1KPsAB6A@mail.gmail.com> <CAJ-VmonrnATPDX4pFNbw6=ip7XyJ2c7P682K0xciDr8YLn%2BXMw@mail.gmail.com> <CADGo8CVQ633WEKiLz%2BKdJONxPjimpTEPg9wxytgA2qQL-7qZyw@mail.gmail.com> <CAJ-Vmomd0M5h64rH5JxF0rOyiA5Joup2GvxfHuJG_6YSh-00HQ@mail.gmail.com> <CADGo8CUcVoRqTw9cpHgVHcwjXssn_5UwwgmnO9A1hAOQB-1f9w@mail.gmail.com> <CAJ-VmonsV08QMEmQCkoWeWib-2dJG8Z598UTQjPWzBBE9vKRww@mail.gmail.com> <CADGo8CW0E9nLeUanMcUJtNJb4=5trYBJ0FngujsWLkp7G-Fn6A@mail.gmail.com> <CAJ-VmoksmN8vXXeex%2BGBQkWvFUg-pGczBLi1esuMJgmXrs0P6A@mail.gmail.com> <CADGo8CVak7jXQ2FfGeHQPWK6%2B1ojP1bp_SY3X=cW7ofx1QL-TA@mail.gmail.com> <CAJ-Vmo=rQbABVuU1bszKDMMbc8BS01sBwCqg31VLgFECS_Qd3Q@mail.gmail.com> <C50E2C57-7821-46A2-BC86-3ED5DE278E60@gmail.com> <CAJ-VmokCB2ztx8Gw%2B7qAQPUrzgdTB8=s9yvgyU1n%2BH0f1nxFmg@mail.gmail.com> <CADGo8CVuOoJhTLY93oiMeAxRLfT5DrSJ2FHhyo4sTd4GhRBc5Q@mail.gmail.com> <CADGo8CXx8r6NLJWVKzBB-77WeoE7q8vubEOrat-LKfYkdhedXw@mail.gmail.com> <CAJ-VmonDXrLVaMYsJ538_LCqPZLFRyWXkUAQoq7OSNVAwYvsaA@mail.gmail.com> <CADGo8CWLTvGcErHzkK49NM3GH%2BizDfCD=a5G3ryOOeHzV2Bn_Q@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Mar 25, 2015 at 5:26 AM, Miguel Clara <miguelmclara@gmail.com> wrote: > > It seems re-aggregation for 11n is still broken, as I just faced the same > issue... > > I had to connect to the 2.4 network since I have 5g set only to n. > > truning on ath.0.hal.debug i see this: > getchannels: cc 0 regDmn 0x1ff mode 0xffffff ecm > getregstate: cc 0 rd 0x1ff > getchannels: !avail mode 0x1f800d (0x2) flags 0x2150 > getchannels: too many channels for channel table > getchannels: cc 0 regDmn 0x1ff mode 0xffffff ecm > getregstate: cc 0 rd 0x1ff > getchannels: !avail mode 0x1f800d (0x2) flags 0x2150 > getchannels: too many channels for channel table > wlan0: link state changed to UP > wlan0: link state changed to DOWN > getchannels: cc 0 regDmn 0x1ff mode 0xffffff ecm > getregstate: cc 0 rd 0x1ff > getchannels: !avail mode 0x1f800d (0x2) flags 0x2150 > getchannels: too many channels for channel table > > > with dev.ath.0.debug I see a lot of this: > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=86, isfrag=0, iswep=64, > m0=0xfffff800308a2e00 > ath0: ath_tx_dmasetup: m 0xfffff800308a2e00 len 96 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 04 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 3010 0000 0000 > 0401 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=1504, isfrag=0, iswep=64, > m0=0xfffff80009800600 > ath0: ath_tx_dmasetup: m 0xfffff80009800600 len 1514 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 05 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 4010 0000 0000 > 0501 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=1504, isfrag=0, iswep=64, > m0=0xfffff80096b04b00 > ath0: ath_tx_dmasetup: m 0xfffff80096b04b00 len 1514 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 06 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 5010 0000 0000 > 0601 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=1445, isfrag=0, iswep=64, > m0=0xfffff80080644200 > ath0: ath_tx_dmasetup: m 0xfffff80080644200 len 1455 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 07 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 6010 0000 0000 > 0701 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=127, isfrag=0, iswep=64, > m0=0xfffff8006b484400 > ath0: ath_tx_dmasetup: m 0xfffff8006b484400 len 137 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 08 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 7010 0000 0000 > 0801 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=198, isfrag=0, iswep=64, > m0=0xfffff80022d4a700 > ath0: ath_tx_dmasetup: m 0xfffff80022d4a700 len 208 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 09 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 8010 0000 0000 > 0901 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=86, isfrag=0, iswep=64, > m0=0xfffff800099caa00 > > [51/4841] > ath0: ath_tx_dmasetup: m 0xfffff800099caa00 len 96 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 03 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 2010 0000 0000 > 0301 0020 0000 0000 aaaa 0300 0000 0800 > linux: pid 81570 (sublime_text): syscall inotify_init not implemented > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=86, isfrag=0, iswep=64, > m0=0xfffff800308a2e00 > ath0: ath_tx_dmasetup: m 0xfffff800308a2e00 len 96 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 04 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 3010 0000 0000 > 0401 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=1504, isfrag=0, iswep=64, > m0=0xfffff80009800600 > ath0: ath_tx_dmasetup: m 0xfffff80009800600 len 1514 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 05 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 4010 0000 0000 > 0501 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=1504, isfrag=0, iswep=64, > m0=0xfffff80096b04b00 > ath0: ath_tx_dmasetup: m 0xfffff80096b04b00 len 1514 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 06 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 5010 0000 0000 > 0601 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=1445, isfrag=0, iswep=64, > m0=0xfffff80080644200 > ath0: ath_tx_dmasetup: m 0xfffff80080644200 len 1455 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 07 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 6010 0000 0000 > 0701 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=127, isfrag=0, iswep=64, > m0=0xfffff8006b484400 > ath0: ath_tx_dmasetup: m 0xfffff8006b484400 len 137 > TODS e0:00:00:00:00:5b->60:00:00:00:00:c0(60:00:00:00:00:c0) data QoS [TID > 0] WEP [IV 08 01 00 00 00 00 KID 0] 1M > 8841 0000 60a4 4c28 13c0 e006 e6aa bf5b 60a4 4c28 13c0 7010 0000 0000 > 0801 0020 0000 0000 aaaa 0300 0000 0800 > ath0: ath_tx_tag_crypto: hdrlen=26, pktlen=198, isfrag=0, iswep=64, > m0=0xfffff80022d4a700 > ath0: ath_tx_dmasetup: m 0xfffff80022d4a700 len 208 > [......] > > > Might be worth adding that this was after a ~9h uptime And also after wlandebug +11n and ifconfig wlan0 scan I see this is dmesg: error: [drm:pid8:i915_gem_object_unbind] *ERROR* Attempting to unbind pinned buffer wlan0: [60:**:c0] discard MPDU frame, BA win <2338:2401> (0 frames) rxseq 2337 tid 0 (retransmit) wlan0: [60:**:c0] discard MPDU frame, BA win <2662:2725> (0 frames) rxseq 2661 tid 0 (retransmit) wlan0: [60:**:c0] discard MPDU frame, BA win <2762:2825> (0 frames) rxseq 2761 tid 0 (retransmit) wlan0: [60:**:c0] discard MPDU frame, BA win <2818:2881> (0 frames) rxseq 2817 tid 0 (retransmit) wlan0: [60:**:c0] discard MPDU frame, BA win <2833:2896> (0 frames) rxseq 2832 tid 0 (retransmit) The MAC address belongs to the 2.4g and not the 5g though... odd no? while this the scan results are just "hanged" showing nothing until I ^C and do the list scan which shows only the 2.4g networks. > Melhores Cumprimentos // Best Regards > ----------------------------------------------- > *Miguel Clara* > *IT - Sys Admin & Developer* > > On Mon, Mar 23, 2015 at 5:16 PM, Adrian Chadd <adrian@freebsd.org> wrote: > >> A-MPDU support makes 11n surprisingly fast. :) >> >> >> >> -a >> >> >> On 23 March 2015 at 09:48, Miguel Clara <miguelmclara@gmail.com> wrote: >> > Was link aggregation the only fix? >> > >> > ############################## >> > % speedtest >> > Retrieving speedtest.net configuration... >> > Retrieving speedtest.net server list... >> > Testing from Vodafone Portugal (89.114.40.155)... >> > Selecting best server based on latency... >> > Hosted by Vodafone PT (Porto) [16.63 km]: 18.06 ms >> > Testing download speed........................................ >> > Download: 104.07 Mbits/s >> > Testing upload speed.................................................. >> > Upload: 9.06 Mbits/s >> > >> > % speedtest >> > Retrieving speedtest.net configuration... >> > Retrieving speedtest.net server list... >> > Testing from Vodafone Portugal (89.114.40.155)... >> > Selecting best server based on latency... >> > Hosted by Vodafone PT (Porto) [16.63 km]: 9.814 ms >> > Testing download speed........................................ >> > Download: 107.18 Mbits/s >> > Testing upload speed.................................................. >> > Upload: 9.54 Mbits/s >> > #################################### >> > >> > I'm back to awesome performance... it can't get better than that since >> my >> > link is Fibre 100/10 :) >> > >> > >> > Melhores Cumprimentos // Best Regards >> > ----------------------------------------------- >> > Miguel Clara >> > IT - Sys Admin & Developer >> > >> > On Sun, Mar 22, 2015 at 11:48 PM, Miguel Clara <miguelmclara@gmail.com> >> > wrote: >> >> >> >> >> >> On Sun, Mar 22, 2015 at 5:48 PM, Adrian Chadd <adrian@freebsd.org> >> wrote: >> >>> >> >>> Hi, >> >>> >> >>> * use force_full_reset at startup - maybe put it in /etc/sysctl.conf >> >>> * upgrade to what I just committed to net80211 today. >> >>> >> >>> It turns out that after some recent changes in -HEAD, 11n aggregation >> >>> wouldn't happen if you tried using it > 10 minutes after boot. >> >>> >> >>> Grr. >> >> >> >> >> >> Haha!! So that's why it was switching to the 2.4 network and 11g... >> >> >> >> Thanks for the fix, compiling now, and I'll still keep ath_debug stuff >> >> present, might be useful. >> >> I need to read up on athstats and other tools, which seem to be quite >> >> interesting for analyses/debugging... guessing this is the best place >> to >> >> look --> https://wiki.freebsd.org/dev/ath(4)/Debugging ? >> >> >> >>> >> >>> >> >>> (Now i Have to update all the wifi APs in my apartment.) >> >>> >> >>> >> >>> -a >> >> >> >> >> > >> > >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CADGo8CW%2Bfuq5BfzYUcHd9S3GCKXdSvvcLLd1mjz3bTvAVWmGcw>