Re: Kernel build fails on ARM: Cannot fork: Cannot allocate memory

On Sun, 23 Jun 2013, Ruslan Bukin wrote:

> On Sun, Jun 23, 2013 at 07:50:40PM +0300, Konstantin Belousov wrote:
>> On Sun, Jun 23, 2013 at 08:44:25PM +0400, Ruslan Bukin wrote:
>>> On Sun, Jun 23, 2013 at 07:16:17PM +0300, Konstantin Belousov wrote:
>>>> On Sun, Jun 23, 2013 at 06:43:46PM +0400, Ruslan Bukin wrote:
>>>>>
>>>>> Trying to mount root from ufs:/dev/da0 []...
>>>>> WARNING: / was not properly dismounted
>>>>> warning: no time-of-day clock registered, system time will not be set accurately
>>>>> panic: __rw_wlock_hard: recursing but non-recursive rw pmap pv global _at_ /usr/home/br/dev/head/sys/arm/arm/pmap-v6.c:1289
>>>>>
>>>>> KDB: enter: panic
>>>>> [ thread pid 1 tid 100001 ]
>>>>> Stopped at      kdb_enter+0x48: ldrb    r15, [r15, r15, ror r15]!
>>>>> db> bt
>>>>> Tracing pid 1 tid 100001 td 0xc547f620
>>>>> _end() at 0xde9d0530
>>>>> scp=0xde9d0530 rlv=0xc1211458 (db_trace_thread+0x34)
>>>>>         rsp=0xde9d0514 rfp=0xc12d1b60
>>>>> Bad frame pointer: 0xc12d1b60
>>>>> db>
>>>> This is completely broken.  It seems that witness triggered the panic,
>>>> and ddb is unable to obtain a backtrace from the normal panic(9) call.
>>>>
>>>> Show the output of the 'show alllocks'.
>>>
>>> No such command
>> Do you have witness in the kernel config ? If not, add it to the config
>> and retry.
>
> Trying to mount root from ufs:/dev/da0 []...
> WARNING: / was not properly dismounted
> warning: no time-of-day clock registered, system time will not be set accurately
> panic: __rw_wlock_hard: recursing but non-recursive rw pmap pv global _at_ /usr/home/br/dev/head/sys/arm/arm/pmap-v6.c:1289
>
> KDB: enter: panic
> [ thread pid 1 tid 100001 ]
> Stopped at      kdb_enter+0x48: ldrb    r15, [r15, r15, ror r15]!
> db> show alllocks
> Process 1 (kernel) thread 0xc55fc620 (100001)
> exclusive sleep mutex pmap (pmap) r = 0 (0xc5600590) locked _at_ /usr/home/br/dev/head/sys/arm/arm/pmap-v6.c:729
> exclusive rw pmap pv global (pmap pv global) r = 0 (0xc1479dd0) locked _at_ /usr/home/br/dev/head/sys/arm/arm/pmap-v6.c:728
> shared rw vm object (vm object) r = 0 (0xc1551d4c) locked _at_ /usr/home/br/dev/head/sys/vm/vm_map.c:1809
> exclusive sx vm map (user) (vm map (user)) r = 0 (0xc5600528) locked _at_ /usr/home/br/dev/head/sys/kern/imgact_elf.c:445
> exclusive lockmgr ufs (ufs) r = 0 (0xc56f7914) locked _at_ /usr/home/br/dev/head/sys/kern/imgact_elf.c:821
> exclusive sleep mutex Giant (Giant) r = 0 (0xc147c778) locked _at_ /usr/home/br/dev/head/sys/kern/vfs_mount.c:1093
> db>
>

Would any of the arm users be interested in testing a larger patch that 
changes the way the kernel allocations KVA?  It also has some UMA code 
that lessens kernel memory utilization.

http://people.freebsd.org/~jeff/vmem.diff

Any reports would be helpful.  Is there any ETA on getting stack tracing 
fixed?  I suspect the pmap recursion encountered with Kostik's patch exist 
in the current kernel.  The other changes in this patch my fix that as 
well.

Thanks,
Jeff