Re: Atheros AR9380 Panic
On 26 September 2014 05:28, Marc Suttle wrote: > Mark, > > What card do you plan on using? A Ubiquiti SR71-E, with AR9280 chipset. I should reiterate that I'm not yet in possession of the card, but am hopeful that it will do the job :)
Re: Atheros AR9380 Panic
On Thu, Sep 25, 2014 at 02:28:47PM -0500, Marc Suttle wrote: > I understand we can always use alternatives to this setup. In an > enterprise env. you would probably never use one of these anyway. It just > seems that there is quite a bit of development on the FreeBSD side and > why duplicate efforts if an NDA would not be needed for a stack port? This > would allow OpenBSD to support more wireless chipsets especially on laptop > and home/smb firewalls. A port of the FreeBSD driver won't magically be free of bugs so why would replacing the existing driver be any easier than hunting down remaining bugs in the existing driver code?
Re: Atheros AR9380 Panic
Mark, What card do you plan on using? Also here is some more information from an interview with Adrian Chadd on WLAN networking and BSD. Would there need to be any NDA signed if we just ported over the FreeBSD Atheros stack? Interview - http://www.bsdnow.tv/episodes/2014_09_17-the_promised_wlan I understand we can always use alternatives to this setup. In an enterprise env. you would probably never use one of these anyway. It just seems that there is quite a bit of development on the FreeBSD side and why duplicate efforts if an NDA would not be needed for a stack port? This would allow OpenBSD to support more wireless chipsets especially on laptop and home/smb firewalls. Thanks- Marc On Sun, Sep 21, 2014 at 9:18 PM, mark hellewell wrote: > On 21 September 2014 19:49, Stefan Sperling wrote: > > Hunt down an older athn card that works. > > This is what I ended up doing. I've ordered an older Ubiquiti card which > I'll hopefully have more luck with. > > > Send the one you've got to a developer who's interested in fixing > > support for it. > > I'm more than happy to send the card which was the subject of my thread to > somebody. Just let me know who :) > > Cheers, > Mark
Re: Atheros AR9380 Panic
On 21 September 2014 19:49, Stefan Sperling wrote: > Hunt down an older athn card that works. This is what I ended up doing. I've ordered an older Ubiquiti card which I'll hopefully have more luck with. > Send the one you've got to a developer who's interested in fixing > support for it. I'm more than happy to send the card which was the subject of my thread to somebody. Just let me know who :) Cheers, Mark
Re: Atheros AR9380 Panic
On Sat, Sep 20, 2014 at 10:23:17PM -0500, Marc Suttle wrote: > Hello openbsd-misc, > > Long time OpenBSD user here. I was getting sick of using an external AP > for my pf based firewall and decided to try and get an athn device working > in AP mode. I am getting a panic on my Soekris 6501 with an Atheros > AR9380. Model - TP-LINK TL-WDN4800 Dual Band Wireless N900 PCI Express > Adapter,2.4GHz 450Mbps/5Ghz 450Mbps. > > I know FreeBSD has made some progress with this chipset in 10 and even are > now supporting N. It would be great to see this working under OpenBSD. > Please let me know how I can assist or test with efforts on my part. Porting the FreeBSD driver over is not trivial. FreeBSD uses a single ath(4) driver with a different code base from Atheros (this used to be closed sourced but is now open source but access to docs requires signing an NDA with the vendor which we won't do) while OpenBSD uses ath(4) with the reverse-engineed ar5xxx hal for old devices and otherwise athn(4) written by Damien as far as I can tell based on information from early versions of Linux' ath9k driver. The other option is to fix the existing drivers in OpenBSD. Both ath(4) and athn(4) have issues which I'll gladly rant about if you want to help fixing them. Neither option is very pleasant for someone who wants to write a drive-by fix. Having looked into both options I found this task requires a degree of devotion that I personally find hardly sustainable. > This looks very similar to the thread started by Mark Hellewell: > http://marc.info/?t=14093762312&r=1&w=2 > > ddb> show panic > kernel diagnostic assertion "pin < sc->ngpiopins" failed: file > "../../../../dev/ic/ar9003.c", line 515 And I'll say it again: These devices attach but they *don't work*. I think it was a mistake to make athn(4) attach to this device. The code is untested and has never been working. The initial commit of this code from 2010 contains this comment in the log message: Tested for non-regression on various AR5416 (sparc64+i386), AR9281 and AR9285 (i386 only) adapters. AR9380 part is not tested (hardware is not available to the general public yet). 4 years later the hardware has become fairly widespread and people are starting to test it :) Some of the issues have since been fixed but many remain. Hunt down an older athn card that works. Send the one you've got to a developer who's interested in fixing support for it. Or even better try to fix the code yourself. We don't have anyone fully devoted to hacking the wireless stack at the moment.
Atheros AR9380 Panic
Hello openbsd-misc, Long time OpenBSD user here. I was getting sick of using an external AP for my pf based firewall and decided to try and get an athn device working in AP mode. I am getting a panic on my Soekris 6501 with an Atheros AR9380. Model - TP-LINK TL-WDN4800 Dual Band Wireless N900 PCI Express Adapter,2.4GHz 450Mbps/5Ghz 450Mbps. I know FreeBSD has made some progress with this chipset in 10 and even are now supporting N. It would be great to see this working under OpenBSD. Please let me know how I can assist or test with efforts on my part. This looks very similar to the thread started by Mark Hellewell: http://marc.info/?t=14093762312&r=1&w=2 ddb> show panic kernel diagnostic assertion "pin < sc->ngpiopins" failed: file "../../../../dev/ic/ar9003.c", line 515 Trace: Debugger() at Debugger+0x5 panic() at panic+0xe4 __assert() at __assert+0x21 ar9003_gpio_write() at ar9003_gpio_write+0x9d athn_init() at athn_init+0xfa athn_ioctl() at athn_ioctl+0x1e6 in6_ifinit() at in6_ifinit+0xc1 in6_update_ifa() at in6_update_ifa+0x1dd in6_ifattach_linklocal() at in6_ifattach_linklocal+0x113 in6_ifattach() at in6_ifattach+0x104 in6_if_up() at in6_if_up+0x15 if_up() at if_up+0x78 ifioctl() at ifioctl+0xb91 sys_ioctl() at sys_ioctl+0x150 syscall() at syscall+0x162 --- syscall (number 54) --- end of kernel end trace frame: 0x7f7b49a0, count: -15 acpi_pdirpa+0x408b4a: ps: PID PPID PGRPUID S FLAGS WAIT COMMAND *29030 27343 29030 0 7 0ifconfig 27343 1 27343 1000 30x88 pause ksh 27641 1 27641 0 30x80 selectcron 5171 1 5171 0 30x80 kqreadapmd 15129 1 15129601 30x80 kqreadunbound 19154 1 19154 99 30x80 poll sndiod 19605 1 19605 71 30x80 kqreadftp-proxy 15238 1 15238 0 30x80 selectsendmail 12228 1 12228 77 30x80 poll dhcpd 3638 1 3638 0 30x80 selectsshd 3487 31462 26491 83 30x80 poll ntpd 31462 26491 26491 83 30x80 poll ntpd 26491 1 26491 0 30x80 poll ntpd 11316 16 16 74 30x80 bpf pflogd 16 1 16 0 30x80 netio pflogd 12721 5068 5068 73 2 0syslogd 5068 1 5068 0 30x80 netio syslogd 22255 1 22255 77 20x80dhclient 15940 1 15940 0 30x80 poll dhclient 14 0 0 0 30x100200 aiodoned aiodoned 13 0 0 0 30x100200 syncerupdate 12 0 0 0 30x100200 cleaner cleaner 11 0 0 0 30x100200 reaperreaper 10 0 0 0 30x100200 pgdaemon pagedaemon 9 0 0 0 30x100200 bored crypto 8 0 0 0 30x100200 pftm pfpurge 7 0 0 0 30x100200 mmctsksdmmc1 6 0 0 0 30x100200 mmctsksdmmc0 5 0 0 0 30x100200 usbtskusbtask 4 0 0 0 30x100200 usbatsk usbatsk 3 0 0 0 30x100200 bored syswq 2 0 0 0 3 0x40100200idle0 1 0 1 0 30x80 wait init 0 -1 0 0 3 0x200 scheduler swapper show registers: ds0xb03cacpi_pdirpa+0x6adc es 0xd fs0x9770acpi_pdirpa+0x5210 gs0xe4d0acpi_pdirpa+0x9f70 rdi 0x1 rsi 0x5 rbp 0x800021339760 rbx 0x8175e4d0addrmask+0x2f30 rdx 0x7 rcx0x292 rax 0x1 r80x800021339680 r9 0 r10 0x800021339638 r11 0x8 r120x100 r13 0x800021339770 r14 0x8015e562 r15 0x7 rip 0x813a0515Debugger+0x5 cs 0x8 rflags 0x202 rsp 0x800021339760 ss 0x10 Debugger+0x5: leave dmesg: OpenBSD 5.4 (GENERIC) #37: Tue Jul 30 15:24:05 MDT 2013 dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC real mem = 2130575360 (2031MB) avail mem = 2066227200 (1970MB) mainbus0 at root bios0 at mainbus0 acpi at bios0 not configured mpbios0 at bios0: Intel M
