Message-ID: <20120210225123.GA5752@black.transpect.com>
Date:	Fri, 10 Feb 2012 17:51:24 -0500
From:	Whit Blauvelt <whit@...nspect.com>
To:	linux-kernel@...r.kernel.org
Subject: Ignorant question: Is via rhine dangerous when on but not networked?

The system that the reports below were from halted again, even after
swapping out the tulip nic for an e1000 (along with the e1000e that's long
been there). In checking through the BIOS settings after the most recent
halt I turned off the via rhine onboard NIC that I stopped using long ago
(wasn't dependable). Then I ran Google's stress test extensively on the
system (http://code.google.com/p/stressapptest/) and Red Hat's
(http://people.redhat.com/dledford/memtest.shtml) plus memtest86, and
everything was happy. 

There's been a lot of recent traffic here about workarounds for via rhine
being a crappy NIC, causing system lockups under load. Could merely having
the via rhine NIC enabled in the BIOS (and probably the module loaded for it
on boot) have made the system vulnerable even though the via rhine NIC
wasn't networked?

This is on a Shuttle SK22V10 with an Athlon 2 core CPU, and a 3.2.2 vanilla
kernel.

Whit



On Thu, Feb 09, 2012 at 09:43:50AM -0500, Whit Blauvelt wrote:
> Another instance, in this case with the card becoming non-operational. There
> have been several other instances of the kernel halting possibly related. In
> any case, stress testing CPU and RAM doesn't reproduce the halt.
> 
> Feb  8 19:38:08 whit kernel: [11045.627031] net eth0: (99048) System Error occurred (0)
> Feb  8 19:38:08 whit kernel: [11045.627088] net eth0: (99048) System Error occurred (0)
> Feb  8 19:38:19 whit kernel: [11057.050016] ------------[ cut here ]------------
> Feb  8 19:38:19 whit kernel: [11057.050035] WARNING: at net/sched/sch_generic.c:255 dev_watchdog+0x262/0x270()
> Feb  8 19:38:19 whit kernel: [11057.050041] Hardware name: SK22V10
> Feb  8 19:38:19 whit kernel: [11057.050046] NETDEV WATCHDOG: eth0 (tulip): transmit queue 0 timed out
> Feb  8 19:38:19 whit kernel: [11057.050050] Modules linked in: Modules linked in: binfmt_misc nfsd lockd nfs_acl auth_rpcgss sunrpc
>   exportfs dm_crypt xt_limit autofs4 ipt_REJECT xt_multiport ipt_LOG xt_tcpudp xt_state iptable_filter ip_tables
>   nf_conntrack_ipv4 nf_defrag_ipv4 nf_conntrack_proto_udplite nf_conntrack_proto_dccp nf_conntrack_tftp speedstep_lib
>   nf_conntrack_netbios_ns nf_conntrack_snmp nf_conntrack_broadcast nf_conntrack_proto_sctp nf_conntrack_irc          
>   nf_conntrack_sane nf_conntrack_h323 nf_conntrack_pptp kvm_amd nf_conntrack_proto_gre kvm ts_kmp nf_conntrack_netlink
>   nfnetlink nf_conntrack_amanda xt_conntrack x_tables nf_conntrack_sip snd_via82xx gameport snd_ac97_codec ac97_bus
>   snd_pcm snd_page_alloc snd_mpu401_uart snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq nf_conntrack_ftp
>   nf_conntrack snd_timer snd_seq_device firewire_sbp2 ppdev shpchp lp parport_pc k8temp bridge stp snd soundcore
>   i2c_viapro parport edac_core edac_mce_amd e1000e floppy firewire_ohci sata_via via_rhine pata_via firewire_core
>   crc_itu_t tulip
> Feb  8 19:38:19 whit kernel: [11057.050182] Pid: 0, comm: swapper/0 Not tainted 3.2.2-black01 #1
> Feb  8 19:38:19 whit kernel: [11057.050186] Call Trace:
> Feb  8 19:38:19 whit kernel: [11057.050191]  <IRQ>  [<ffffffff81063d8f>] warn_slowpath_common+0x7f/0xc0
> Feb  8 19:38:19 whit kernel: [11057.050211]  [<ffffffff81063e86>] warn_slowpath_fmt+0x46/0x50
> Feb  8 19:38:19 whit kernel: [11057.050220]  [<ffffffff8108d638>] ? sched_clock_cpu+0xa8/0x110
> Feb  8 19:38:19 whit kernel: [11057.050233]  [<ffffffff814f0f32>] dev_watchdog+0x262/0x270
> Feb  8 19:38:19 whit kernel: [11057.050240]  [<ffffffff8105db02>] ? scheduler_tick+0x122/0x280
> Feb  8 19:38:19 whit kernel: [11057.050249]  [<ffffffff810740a4>] run_timer_softirq+0x134/0x360
> Feb  8 19:38:19 whit kernel: [11057.050256]  [<ffffffff814f0cd0>] ? netif_carrier_off+0x30/0x30
> Feb  8 19:38:19 whit kernel: [11057.050265]  [<ffffffff8106b388>] __do_softirq+0xa8/0x1c0
> Feb  8 19:38:19 whit kernel: [11057.050274]  [<ffffffff810981f4>] ? tick_program_event+0x24/0x30
> Feb  8 19:38:19 whit kernel: [11057.050282]  [<ffffffff815e916c>] call_softirq+0x1c/0x30
> Feb  8 19:38:19 whit kernel: [11057.050292]  [<ffffffff810152b5>] do_softirq+0x65/0xa0
> Feb  8 19:38:19 whit kernel: [11057.050299]  [<ffffffff8106b71e>] irq_exit+0x8e/0xb0
> Feb  8 19:38:19 whit kernel: [11057.050307]  [<ffffffff815e9b0e>] smp_apic_timer_interrupt+0x6e/0x99
> Feb  8 19:38:19 whit kernel: [11057.050316]  [<ffffffff815e79de>] apic_timer_interrupt+0x6e/0x80
> Feb  8 19:38:19 whit kernel: [11057.050321]  <EOI>  [<ffffffff8103b21b>] ? native_safe_halt+0xb/0x10
> Feb  8 19:38:19 whit kernel: [11057.050337]  [<ffffffff8101b9c1>] default_idle+0x41/0xe0
> Feb  8 19:38:19 whit kernel: [11057.050344]  [<ffffffff8101babd>] amd_e400_idle+0x5d/0x120
> Feb  8 19:38:19 whit kernel: [11057.050351]  [<ffffffff81012276>] cpu_idle+0xd6/0x110
> Feb  8 19:38:19 whit kernel: [11057.050362]  [<ffffffff815c1c45>] rest_init+0x75/0x80
> Feb  8 19:38:19 whit kernel: [11057.050371]  [<ffffffff81ad7c0a>] start_kernel+0x3b3/0x3be
> Feb  8 19:38:19 whit kernel: [11057.050379]  [<ffffffff81ad7347>] x86_64_start_reservations+0x132/0x136
> Feb  8 19:38:19 whit kernel: [11057.050386]  [<ffffffff81ad744c>] x86_64_start_kernel+0x101/0x110
> Feb  8 19:38:19 whit kernel: [11057.050393] ---[ end trace ff676ffe84ac588c ]---
> Feb  8 19:38:19 whit kernel: [11057.050406] net eth0: PNIC2 transmit timed out, status ec69000d, CSR6/7 e7fe42c5 / e7fe42cd CSR12 000000c4, resetting...
> Feb  8 19:38:27 whit kernel: [11065.050029] net eth0: PNIC2 transmit timed out, status ec69010f, CSR6/7 e7fe42c5 / e7fe42cd CSR12 000000c4, resetting...
> 
> Needless to say, I've swapped out the old NIC now.
> 
> Whit
> 
> 
> On Sun, Jan 29, 2012 at 01:44:37PM -0500, Whit Blauvelt wrote:
> > 3.2.2 kernel. Just in case it's of interest:
> > 
> > Jan 29 11:15:31 whit kernel: [50599.304905] ------------[ cut here ]------------
> > Jan 29 11:15:31 whit kernel: [50599.304924] WARNING: at kernel/timer.c:1011 del_timer_sync+0x56/0x60()
> > Jan 29 11:15:31 whit kernel: [50599.304929] Hardware name: SK22V10
> > Jan 29 11:15:31 whit kernel: [50599.304933] Modules linked in: joydev st lp parport_pc binfmt_misc nfsd lockd nfs_acl auth_rpcgss sunrpc exportfs xt_limit ipt_REJECT xt_multiport ipt_LOG xt_tcpudp dm_crypt bridge xt_state stp iptable_filter ip_
> > Jan 29 11:15:31 whit kernel: crc_itu_t sata_via [last unloaded: ppa]
> > Jan 29 11:15:31 whit kernel: [50599.305073] Pid: 0, comm: swapper/1 Not tainted 3.2.2-black01 #1
> > Jan 29 11:15:31 whit kernel: [50599.305078] Call Trace:
> > Jan 29 11:15:31 whit kernel: [50599.305083]  <IRQ>  [<ffffffff81063d8f>] warn_slowpath_common+0x7f/0xc0
> > Jan 29 11:15:31 whit kernel: [50599.305105]  [<ffffffff81063dea>] warn_slowpath_null+0x1a/0x20
> > Jan 29 11:15:31 whit kernel: [50599.305111]  [<ffffffff81074406>] del_timer_sync+0x56/0x60
> > Jan 29 11:15:31 whit kernel: [50599.305130]  [<ffffffffa002ee79>] pnic2_lnk_change+0xe9/0x698 [tulip]
> > Jan 29 11:15:31 whit kernel: [50599.305145]  [<ffffffffa00283a9>] tulip_interrupt+0x1e9/0x1010 [tulip]
> > Jan 29 11:15:31 whit kernel: [50599.305159]  [<ffffffff810d4d04>] handle_irq_event_percpu+0x54/0x200  
> > Jan 29 11:15:31 whit kernel: [50599.305169]  [<ffffffff810d4f04>] handle_irq_event+0x54/0x80
> > Jan 29 11:15:31 whit kernel: [50599.305178]  [<ffffffff810d7eca>] handle_fasteoi_irq+0x6a/0x100
> > Jan 29 11:15:31 whit kernel: [50599.305190]  [<ffffffff81015232>] handle_irq+0x22/0x40
> > Jan 29 11:15:31 whit kernel: [50599.305199]  [<ffffffff815e9a1d>] do_IRQ+0x5d/0xe0
> > Jan 29 11:15:31 whit kernel: [50599.305210]  [<ffffffff815df0ee>] common_interrupt+0x6e/0x6e
> > Jan 29 11:15:31 whit kernel: [50599.305215]  <EOI>  [<ffffffff8103b21b>] ? native_safe_halt+0xb/0x10  
> > Jan 29 11:15:31 whit kernel: [50599.305233]  [<ffffffff8101b9c1>] default_idle+0x41/0xe0
> > Jan 29 11:15:31 whit kernel: [50599.305241]  [<ffffffff8101babd>] amd_e400_idle+0x5d/0x120
> > Jan 29 11:15:31 whit kernel: [50599.305250]  [<ffffffff81012276>] cpu_idle+0xd6/0x110     
> > Jan 29 11:15:31 whit kernel: [50599.305259]  [<ffffffff815d70e8>] start_secondary+0x1e0/0x1e7
> > Jan 29 11:15:31 whit kernel: [50599.305267] ---[ end trace 525e0a5d40ea44de ]---
> > 
> > Seeing "tulip" there, I'm guessing this relates to this old NIC:
> > 
> >   LNE100TX [Linksys EtherFast 10/100]
> > 
> > Whit
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists