[wip] ar9300 hostap support

[Joshua Isom]

Another panic, but a different cause.  I think I'd seen it once before 
but I'm not sure.

Unread portion of the kernel message buffer:
ath0: ath_edma_recv_proc_queue: handled npkts 0
ath0: ath_edma_tx_processq: Q1: empty?
panic: _mtx_lock_sleep: recursed on non-recursive mutex ath0_txq1 @ 
/root/ATH/head/sys/modules/ath/../../dev/ath/if_ath_tx_edma.c:667

cpuid = 1
KDB: enter: panic

[...]

(kgdb) bt
#0  doadump (textdump=0) at pcpu.h:229
#1  0xffffffff802fb52e in db_dump (dummy=<value optimized out>, 
dummy2=0, dummy3=0, dummy4=0x0)
     at /root/ATH/head/sys/ddb/db_command.c:543
#2  0xffffffff802fafda in db_command (last_cmdp=<value optimized out>, 
cmd_table=<value optimized out>, dopager=1)
     at /root/ATH/head/sys/ddb/db_command.c:449
#3  0xffffffff802fad92 in db_command_loop () at 
/root/ATH/head/sys/ddb/db_command.c:502
#4  0xffffffff802fd740 in db_trap (type=<value optimized out>, code=0) 
at /root/ATH/head/sys/ddb/db_main.c:231
#5  0xffffffff8056e4e3 in kdb_trap (type=3, code=0, tf=<value optimized 
out>) at /root/ATH/head/sys/kern/subr_kdb.c:654
#6  0xffffffff807d5368 in trap (frame=0xffffff8020bdd7d0) at 
/root/ATH/head/sys/amd64/amd64/trap.c:579
#7  0xffffffff807be232 in calltrap () at exception.S:228
#8  0xffffffff8056dcce in kdb_enter (why=0xffffffff808abd45 "panic", 
msg=<value optimized out>) at cpufunc.h:63
#9  0xffffffff805390f7 in vpanic (fmt=<value optimized out>, ap=<value 
optimized out>)
     at /root/ATH/head/sys/kern/kern_shutdown.c:747
#10 0xffffffff80538fa6 in kassert_panic (fmt=<value optimized out>) at 
/root/ATH/head/sys/kern/kern_shutdown.c:642
#11 0xffffffff80524cfa in __mtx_lock_sleep (c=0xffffff800090dea8, 
tid=18446741874793984000, opts=<value optimized out>,
     file=<value optimized out>, line=549311568) at 
/root/ATH/head/sys/kern/kern_mutex.c:394
#12 0xffffffff8052491a in __mtx_lock_flags (c=<value optimized out>, opts=0,
     file=0xffffffff813f2dcf 
"/root/ATH/head/sys/modules/ath/../../dev/ath/if_ath_tx_edma.c", line=667)
     at /root/ATH/head/sys/kern/kern_mutex.c:224
#13 0xffffffff81335467 in ath_edma_tx_processq (sc=0xffffff800090d000, 
dosched=1)
     at /root/ATH/head/sys/modules/ath/../../dev/ath/if_ath_tx_edma.c:667
#14 0xffffffff8057cea0 in taskqueue_run_locked 
(queue=0xfffffe0006488500) at /root/ATH/head/sys/kern/subr_taskqueue.c:333
#15 0xffffffff8057d67b in taskqueue_thread_loop (arg=<value optimized out>)
     at /root/ATH/head/sys/kern/subr_taskqueue.c:535
#16 0xffffffff80508a14 in fork_exit (callout=0xffffffff8057d5e0 
<taskqueue_thread_loop>, arg=0xffffff800090d810,
     frame=0xffffff8020bddc00) at /root/ATH/head/sys/kern/kern_fork.c:991
#17 0xffffffff807be76e in fork_trampoline () at exception.S:602
#18 0x0000000000000000 in ?? ()


On 3/26/2013 7:11 PM, Adrian Chadd wrote:
>
> Hah, that's nice to know.
>
> Please let me know if you see any further issues here.
>
> I have another tweak to add in soon in the TX completion code; I just
> noticed a very subtle difference between what my code does and what
> the reference code does.
>
> Thanks,
>
>
>
> Adrian
>