Re: Regression AR2413 from 9.1 to 10
Thats an old pre 11n nic. That said, nothing should have broken. Are you able to try 10.0 from say, 6 months back and see if it works? Adrian On Sep 22, 2012 11:43 PM, "Sven Hazejager" wrote: > All, > > I am in the process of upgrading my NanoBSD-ALIX kit from 802.11g to > 802.11n. First step is to move from 9.1-PRERELEASE (100% stable) to > 10-CURRENT with current hardware (AR2413). And alas, I am experiencing > stalling wifi on all my clients: after about 30-60 seconds of being > online, the client loses all connectivity (my Mac thinks it's still > connected but nothing gets through anymore). *sigh*. I did not change > my kernel config. MALLOC_PRODUCTION=Y etc. Where shall I start to > debug this? All the ath* tools are already on my image so let's hack > :-) > > Sven > ___ > freebsd-wireless@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-wireless > To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org > " > ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: Regression AR2413 from 9.1 to 10
Hi Adrian, I am unable to build any svn sources from around March and April due to some yacc/parse.y failure that I cannot fix within the time I have. Can I simply restore an older version of sys/modules/ath and sys/dev/ath into an up-to-date /usr/src? Thanks Sven On Sun, Sep 23, 2012 at 10:14 AM, Adrian Chadd wrote: > Are you able to try 10.0 from say, 6 months back and see if it works? ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: Regression AR2413 from 9.1 to 10
On 23 September 2012 10:06, Sven Hazejager wrote: > Hi Adrian, > > I am unable to build any svn sources from around March and April due > to some yacc/parse.y failure that I cannot fix within the time I have. > > Can I simply restore an older version of sys/modules/ath and > sys/dev/ath into an up-to-date /usr/src? Well you can, but you have to likely hack on sys/conf/files to remove the source files I've added since then. Adrian ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
athratestats doesn't support `-a' option :)
Hello, Freebsd-wireless. root@gateway:/usr/home/lev # ~lev/athtools/athratestats -a athratestats: /usr/home/lev/athtools/athratestats: macaddress wasn't supplied and no -a given root@gateway:/usr/home/lev # -- // Black Lion AKA Lev Serebryakov ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: Regression AR2413 from 9.1 to 10
Looks like someone else is reporting a problem. I wonder if I broke beacon handling somehow.. Adrian ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
New ath code on AP: strange client behavioir
Hello, Freebsd-wireless. I've build new image for my router (r240742), and now traffic stuc shortly after client association. After manual re-assotiation (diconnect/connect cycle) traffic stuck after several packets (2-3MiB of traffic) again. According to tcpdump on AP, it sees some traffic from client, and send something back, but client could not resolve any name, etc... Even more, it looks like, tcpdump sees last fiew packets from client multiple times, because at moment when client could not do anything, tcpdump SOMETIMES shows latest MCAST UDP traffic from client (Microsoft network) over and over again! AP is AR9220, 802.11gn mode. Client is Intel Advanced-N, but in 802.11g only mode. -- // Black Lion AKA Lev Serebryakov ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
broken TX in HEAD (AR5212)
Hi .. i have updated to latest -HEAD and found that afters some time or under havy traffic load card stop to trasmit frames... i roll back to ath rev r240677 which works OK .. Beri ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: New ath code on AP: strange client behavioir
On 23 September 2012 11:16, Lev Serebryakov wrote: > Hello, Freebsd-wireless. > > I've build new image for my router (r240742), and now traffic stuc > shortly after client association. > > After manual re-assotiation (diconnect/connect cycle) traffic stuck > after several packets (2-3MiB of traffic) again. > > According to tcpdump on AP, it sees some traffic from client, and > send something back, but client could not resolve any name, etc... Ok ok I've broken something. I'm currently tracking down one bug (which may just turn out to be a hardware quirk..) so I'd really appreciate it if you all could work together to isolate which commit(s) broke things. Thanks so very much and I'm sorry I've broken things! Adrian ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: broken TX in HEAD (AR5212)
On 23 September 2012 11:17, Berislav Purgar wrote: > Hi .. > > i have updated to latest -HEAD and found that afters some time or under > havy traffic load card stop to trasmit frames... i roll back to ath > rev r240677 which works OK .. Lev, can you try r240677 and test? I'd like to establish whether that commit is stable for you (and others.) Thanks, Adrian ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: broken TX in HEAD (AR5212)
Hello, Adrian. You wrote 23 сентября 2012 г., 22:22:11: AC> Lev, can you try r240677 and test? I've rolled back to my previous build, 240324, but I'm building 240677 right now (it takes about hour :( ) -- // Black Lion AKA Lev Serebryakov ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: broken TX in HEAD (AR5212)
Hello, Adrian. You wrote 23 сентября 2012 г., 22:22:11: AC> Lev, can you try r240677 and test? AC> I'd like to establish whether that commit is stable for you (and others.) Yep, it passes smoke test -- copying of 1GiB over WiFi. -- // Black Lion AKA Lev Serebryakov ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: broken TX in HEAD (AR5212)
On 23 September 2012 12:52, Lev Serebryakov wrote: > AC> Lev, can you try r240677 and test? > AC> I'd like to establish whether that commit is stable for you (and others.) > Yep, it passes smoke test -- copying of 1GiB over WiFi. Ok, can you work with Berislav to figure out where I broke things? :) Adrian ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: broken TX in HEAD (AR5212)
Hello, Adrian. You wrote 23 сентября 2012 г., 23:54:19: >> AC> Lev, can you try r240677 and test? >> AC> I'd like to establish whether that commit is stable for you (and others.) >> Yep, it passes smoke test -- copying of 1GiB over WiFi. AC> Ok, can you work with Berislav to figure out where I broke things? :) It looks like here are four commits from you in this time period: # svn log -r 240677:HEAD sys | grep "| adrian |" r240677 | adrian | 2012-09-19 00:33:04 +0400 (Wed, 19 Sep 2012) | 14 lines r240721 | adrian | 2012-09-20 07:03:01 +0400 (Thu, 20 Sep 2012) | 8 lines r240722 | adrian | 2012-09-20 07:04:19 +0400 (Thu, 20 Sep 2012) | 2 lines r240724 | adrian | 2012-09-20 07:13:20 +0400 (Thu, 20 Sep 2012) | 32 lines # It is not too hard to try all of them, as 240677 is known to work. -- // Black Lion AKA Lev Serebryakov ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: broken TX in HEAD (AR5212)
On 23 September 2012 13:02, Lev Serebryakov wrote: > # svn log -r 240677:HEAD sys | grep "| adrian |" > r240677 | adrian | 2012-09-19 00:33:04 +0400 (Wed, 19 Sep 2012) | 14 lines > r240721 | adrian | 2012-09-20 07:03:01 +0400 (Thu, 20 Sep 2012) | 8 lines > r240722 | adrian | 2012-09-20 07:04:19 +0400 (Thu, 20 Sep 2012) | 2 lines > r240724 | adrian | 2012-09-20 07:13:20 +0400 (Thu, 20 Sep 2012) | 32 lines > # > > It is not too hard to try all of them, as 240677 is known to work. Sweet, thanks! Adrian ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: broken TX in HEAD (AR5212)
Hiya, Ok. Please keep me posted. I've found some annoying behaviours this weekend. * The rx proc tasklet gets called slightly too often- if an RX interrupt comes in whilst ath_rx_proc is running, the kernel will preempt that thread, run ath_intr(), which will reschedule ath_rx_proc to run. The next time it runs (which will be immediatley after the previous one finishes), it'll discover there's nothing left in the RX queue. That's not too scary; just slightly wasteful. But the annoying one: * I'm getting TXEOL interrupts appearing (which is by design, so that's OK) - but then when the descriptor is checked, the first descriptor in the list shows up as HAL_EINPROGRESS - and sure enough, the contents of that descriptor are not yet initialised. Now on my MIPS boards the descriptors are in KSEG1 - which is supposed to be unmapped, uncached memory. A subsequent TXEOL (from another descriptor being pushed into the queue) or TXOK (from that descriptor completing TX successfully) will kick the queue along. Now I don't _think_ this is the core problem. I thought it was a problem before but it's very likely just something that I've noticed going slightly "odd" when I began digging into it. (Yes, I may have spent a chunk of yesterday chasing a red herring. Sigh.) What actually does happen is that hostapd actually boots my station off because it fails to re-synchronise the group key. When hostapd decides its time, it regenerates a group key and pushes it out to each STA. It's done STA-at-a-time - ie, the STA has to ACK the new group key. I'm seeing hostapd not "see" an ACK and it's thus getting angry and leaving. I'm just adding some debugging code to my tree now to see if I can pinpoint what's going on - whether the frame is going out, whether it's being successfully transmitted, whether the other end hears it, whether the receive path is working .. heck, since it's all done via raw BPF sockets, it may be some BPF related hilarity. :( Adrian ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: broken TX in HEAD (AR5212)
Hi i have tested r240722 which is OK .. so problem is in r240724. (for me only with AR5212 card .. 11N cards 9220 and 9280 work without problem on r240724) Beri On Sun, Sep 23, 2012 at 10:02 PM, Lev Serebryakov wrote: > Hello, Adrian. > You wrote 23 сентября 2012 г., 23:54:19: > > >> AC> Lev, can you try r240677 and test? > >> AC> I'd like to establish whether that commit is stable for you (and > others.) > >> Yep, it passes smoke test -- copying of 1GiB over WiFi. > AC> Ok, can you work with Berislav to figure out where I broke things? :) > It looks like here are four commits from you in this time period: > > # svn log -r 240677:HEAD sys | grep "| adrian |" > r240677 | adrian | 2012-09-19 00:33:04 +0400 (Wed, 19 Sep 2012) | 14 lines > r240721 | adrian | 2012-09-20 07:03:01 +0400 (Thu, 20 Sep 2012) | 8 lines > r240722 | adrian | 2012-09-20 07:04:19 +0400 (Thu, 20 Sep 2012) | 2 lines > r240724 | adrian | 2012-09-20 07:13:20 +0400 (Thu, 20 Sep 2012) | 32 lines > # > > It is not too hard to try all of them, as 240677 is known to work. > > -- > // Black Lion AKA Lev Serebryakov > > ___ > freebsd-wireless@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-wireless > To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org > " > ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: broken TX in HEAD (AR5212)
Hm, ok. Change this bit in if_ath_tx.c: - flags = HAL_TXDESC_CLRDMASK;/* XXX needed for crypto errs */ + //flags = HAL_TXDESC_CLRDMASK; /* XXX needed for crypto errs */ + flags = 0; to just flags = HAL_TXDESC_CLRDMASK; See if that works. Hmph. I thought I caught all the filtered frames weirdness. Obviously I didn't.. Oh for gods sake. I know what I did wrong. Yes, do the above for now. I totally screwed up; I thought I had set CLRDMASK correctly for non-aggregate traffic. Ok, here's what I did wrong - I didn't set the CLRDMASK unconditionally for non-aggregate traffic. So the non-11n (non-aggregate path) has it selectively set - but there's no completion handler which is putting the traffic on the filtered frames list and setting the CLRDMASK bit again. So do the above and please, please file a bug with what you've gathered. I'll add this to the email and I'll fix it in -HEAD. Thanks for pointing this out! Adrian ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
Re: broken TX in HEAD (AR5212)
hi, I've committed some code to -HEAD that should do the trick. Or, it could set everything on fire. I'm not sure yet. :-) Please try it out and let me know how it goes. Adrian ___ freebsd-wireless@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-wireless To unsubscribe, send any mail to "freebsd-wireless-unsubscr...@freebsd.org"
