[LinuxPPS] Kernel Oops on 2.26.21

[Don Weeks]

I found it. It was a system call change in device_create from 27 to 21. I
corrected that and the drivers are registering with each other.
         Don

On Fri, Dec 11, 2009 at 1:41 PM, Don Weeks <don.l.weeks.jr at gmail.com> wrote:

> I am doing some embedded work with Cavium Octeon (MIPS) which supplies a
> 2.26.21 kernel. I grabbed the lastest (might be the problem) kernel patches
> for Linux PPS and added them to the .21 kernel. Everything compiles fine and
> the pps driver comes up:
>
> 2009-12-10,21:29:28.617,kern,info,localhost,kernel,LinuxPPS API ver. 1
> registered
> 2009-12-10,21:29:28.617,kern,info,localhost,kernel,Software ver. 5.3.6 -
> Copyright 2005-2007 Rodolfo Giometti <giometti at linux.it>
>
>
> However, on loading the ktimer driver, I get the following oops:
>
> Code: 144000b1 00000000 dfa60000 <80c30000> 1060017b 02e0802d 24020025
> 10620021 0212102b
> CPU 9 Unable to handle kernel paging request at virtual address
> 0000000000000000, epc == ffffffff802e05d8, ra == ffffffff80382ef4
> Oops[#2]:
> Cpu 9
> $ 0 : 0000000000000000 0000000000000000 0000000000000000 ffffffff80382bf8
> $ 4 : a80000006c3e4558 0000000000000014 0000000000000000 a80000005ecb7c70
> $ 8 : c000000000208bd0 0000000000000001 a80000041c18a6d8 0000000000000005
> $12 : 0000000000000000 ffffffff802e1eec 0000000000210000 ffffffff807aa5b8
> $16 : a80000006c3e4400 a800000069834a00 a80000006c3e456c 0000000000000000
> $20 : a80000005ecb7c70 000000000fb00001 0000000000000026 a80000006c3e4558
> $24 : 0000000000000000 0000000000000020
> $28 : a80000005ecb4000 a80000005ecb7bc0 0000000000000014 ffffffff80382ef4
> Hi : 000000000000007d
> Lo : fffffffffdfdfe7b
> epc : ffffffff802e05d8 vsnprintf+0x58/0x798 Tainted: P
> ra : ffffffff80382ef4 device_create+0xc4/0x160
> Status: 1000cce3 KX SX UX KERNEL EXL IE
> Cause : 00000008
> BadVA : 0000000000000000
> PrId : 000d0409
> Modules linked in: ktimer cavgpio(P) pps_core ipip xt_dscp nfsd exportfs
> xt_TCPMSS xt_tcpmss xt_string xt_statistic xt_state xt_sctp xt_realm
> xt_quota xt_policy xt_pkttype xt_physdev xt_NFQUEUE xt_NFLOG xt_multiport
> xt_MARK xt_mark xt_mac xt_limit xt_length xt_helper xt_hashlimit xt_esp
> xt_dccp xt_conntrack xt_connmark xt_connbytes xt_comment tun sch_tbf
> sch_prio sch_ingress sch_htb sch_dsmark qosmPktMonitor_driver(P) nf_nat_sip
> nf_conntrack_sip ipv6 ipt_ULOG ipt_TTL ipt_ttl ipt_TOS ipt_tos ipt_SAME
> ipt_REJECT ipt_REDIRECT ipt_recent ipt_owner ipt_NETMAP ipt_MASQUERADE
> ipt_LOG ipt_iprange ipt_ECN ipt_ecn ipt_CLUSTERIP ipt_ah ipt_addrtype
> iptable_nat nf_nat nf_conntrack_ipv4 nf_conntrack nfnetlink iptable_mangle
> iptable_filter ip_tables tunnel4 ip6_tables cls_u32 cls_tcindex act_police
> Process modprobe (pid: 30207, threadinfo=a80000005ecb4000,
> task=a80000006695c1c0)
> Stack : 0000000000000000 c000000000235c00 a80000006c3e4400 a800000069834a00
> fffffffffffffff4 0000000000000000 0000000000000000 000000000fb00001
> 0000000000000026 0000000000000026 ffffffff801981d0 ffffffff80382ef4
> a80000005ecb7c70 ffffffff803893a4 a80000041c18a600 a80000041c18a6d8
> 0000000000000000 c000000000210000 c000000000210000 ffffffff80570000
> c0000000002072f8 ffffffff80570000 c000000000208bd0 0000000000000001
> a80000041c18a6d8 0000000000000005 0000000000000000 c000000000235b30
> a80000041c18a600 c000000000210000 c0000000002083ec c0000000002082a4
> 0000000000000000 0000000100100100 c000000000235c00 c000000000236688
> c000000000230000 c000000000235c00 c000000000217eb0 c00000000020f000
> ...
> Call Trace:
> [<ffffffff802e05d8>] vsnprintf+0x58/0x798
> [<ffffffff80382ef4>] device_create+0xc4/0x160
> [<c0000000002072f8>] pps_register_cdev+0x180/0x1e0 [pps_core]
> [<c0000000002083ec>] pps_register_source+0x414/0x530 [pps_core]
> [<c00000000010c078>] pps_ktimer_init+0x78/0x1a8 [ktimer]
> [<ffffffff80199940>] sys_init_module+0x1c8/0x1cd8
> [<ffffffff80154494>] handle_sys+0x154/0x174
>
>
> Code: 144000b1 00000000 dfa60000 <80c30000> 1060017b 02e0802d 24020025
> 10620021 0212102b
>
> I looked at the archive from the wiki and the 21 patches are vastly
> different than the code I picked up. Any help would be appreciated.
>        Don
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://ml.enneenne.com/pipermail/linuxpps/attachments/20091211/d2198e52/attachment.htm