Re: Kernel Panic on H3 OrnagePi One

[wilde%sha-bang.de@localhost]

Jared McNeill <jmcneill%invisible.ca@localhost> wrote:
> Hi --
>
> I checked in a fix that makes this problem go away for me. Can you try it?

Thanks for working on this.  I will update my kernel, but so far I wasn't
able to reproduce the problem.  The one incident I reported was the only
one for me so far...

> On Fri, 7 Jul 2017, wilde%sha-bang.de@localhost wrote:
>
>> FWIW, during checking out the NetBSD src tree via cvs (on the MMC) I got
>> an kernel panic.  No Idea if this related to the new h3 port, but I
>> think reporting doesn't hurt.  The relevant dmesg snippet follows:
>>
>> WARNING: mclpool limit reached; increase kern.mbuf.nmbclusters
>> panic: kernel diagnostic assertion "offset < map->dm_mapsize" failed:
>> file
>> "/home/wilde/data/qemu/NetBSD-arm/usr/src/sys/arch/arm/arm32/bus_dma.c",
>> line 1056 offset 0 mapsize 0
>> cpu0: Begin traceback...
>> 0x9ba35bdc: netbsd:db_panic+0xc
>> 0x9ba35bf4: netbsd:vpanic+0x1b4
>> 0x9ba35c0c: netbsd:__udivmoddi4
>> 0x9ba35c6c: netbsd:_bus_dmamap_sync+0x188
>> 0x9ba35cbc: netbsd:sunxi_emac_intr+0x2c8
>> 0x9ba35cd4: netbsd:gic_fdt_intr+0x28
>> 0x9ba35cfc: netbsd:pic_dispatch+0x58
>> 0x9ba35d54: netbsd:armgic_irq_handler+0xd4
>> 0x9ba35dd4: netbsd:irq_entry+0x68
>> 0x9ba35df4: netbsd:uvm_obj_destroy+0x38
>> 0x9ba35e14: netbsd:vcache_free+0x8c
>> 0x9ba35e4c: netbsd:vrelel+0x468
>> 0x9ba35e84: netbsd:ufs_remove+0xb4
>> 0x9ba35eac: netbsd:VOP_REMOVE+0x44
>> 0x9ba35f1c: netbsd:do_sys_unlinkat+0x1f8
>> 0x9ba35f34: netbsd:sys_unlink+0x28
>> 0x9ba35fac: netbsd:syscall+0x108
>> cpu0: End traceback...
>>
>> dump to dev 92,1 not possible
>> rebooting...
>>
>>