On Sunday 25 June 2006 22:31, Dimitar Georgiev Popov wrote:

Hi, again.

The problem turned out to be a buggy program - namely ktorrent; it worked well until yesterday; but now, it opens several HUNDRED kio_http processes; I updated the system yesterday and probably some of the libs have modified the program's behavior. Anyway, many thanks to you both Grahame and Ralph.

Greetings, Mitko

Hi, Well, the problem is not ktorrent. I don't use it, but the hangs remain. I noticed that the system hangs whenever there is a high network load (e.g. when torrents are running). So I suspected the NIC (Realtec chip). I changed the NIC with a 3Com 3c905C-TX which works just fine on another system. But, in vain: more hangs. Once, I managed to save the syslog before the system hanged, and I found this: 07/08/2006 05:24:47 PM myhost kernel in_atomic():1, irqs_disabled():0 07/08/2006 05:24:47 PM myhost kernel <ffffffff8832543a>{:nvidia:_nv004467rm+38} <ffffffff80209e2e>{acpi_pci_allocate_irq+36} 07/08/2006 05:24:47 PM myhost kernel <ffffffff881b36e7>{:nvidia:_nv002125rm+39} <ffffffff8815304e>{:3c59x:boomerang_interrupt+733} 07/08/2006 05:24:47 PM myhost kernel <ffffffff881517eb>{:3c59x:vortex_up+172} <ffffffff88152c9b>{:3c59x:vortex_error+666} 07/08/2006 05:24:47 PM myhost kernel <ffffffff80404259>{_sinittext+601} 07/08/2006 05:24:47 PM myhost kernel <ffffffff80209f87>{acpi_pci_irq_enable+167} <ffffffff80209e0a>{acpi_pci_allocate_irq+0} 07/08/2006 05:24:47 PM myhost kernel <ffffffff801d38ad>{pci_enable_device_bars+39} <ffffffff801d38cd>{pci_enable_device+19} 07/08/2006 05:24:47 PM myhost kernel <ffffffff8015f4fb>{kmem_cache_free+57} <ffffffff80134976>{__rcu_process_callbacks+295} 07/08/2006 05:24:47 PM myhost kernel <ffffffff801452a4>{handle_IRQ_event+41} <ffffffff80145342>{__do_IRQ+109} 07/08/2006 05:24:47 PM myhost kernel <ffffffff80134a1b>{rcu_process_callbacks+25} <ffffffff8012da5f>{process_timeout+0} 07/08/2006 05:24:47 PM myhost kernel <ffffffff8012a70d>{tasklet_action+61} <ffffffff8012a7e0>{__do_softirq+70} 07/08/2006 05:24:47 PM myhost kernel <ffffffff8011016e>{nommu_map_sg+0} <ffffffff80109619>{default_idle+42} 07/08/2006 05:24:47 PM myhost kernel <ffffffff8010c2a7>{do_IRQ+64} <ffffffff801095ef>{default_idle+0} 07/08/2006 05:24:47 PM myhost kernel <ffffffff8010c2a2>{do_IRQ+59} <ffffffff8010a918>{ret_from_intr+0} 07/08/2006 05:24:47 PM myhost kernel <ffffffff8010b1aa>{call_softirq+30} <ffffffff8010c0a4>{do_softirq+44} 07/08/2006 05:24:47 PM myhost kernel <ffffffff8010a918>{ret_from_intr+0} <EOI> <ffffffff802aa292>{thread_return+0} 07/08/2006 05:24:47 PM myhost kernel <ffffffff8010967e>{cpu_idle+61} <ffffffff8040477c>{start_kernel+456} 07/08/2006 05:24:47 PM myhost kernel eth2: Too much work in interrupt, status e003. 07/08/2006 05:24:47 PM myhost kernel eth2: PCI bus error, bus status 800000a0 07/08/2006 05:24:47 PM myhost kernel eth2: PCI bus error, bus status 80000020 ... 07/08/2006 05:24:47 PM myhost kernel eth2: PCI bus error, bus status 80000020 07/08/2006 05:24:47 PM myhost kernel eth2: Host error, FIFO diagnostic register 8000. 07/08/2006 05:24:47 PM myhost kernel eth2: Host error, FIFO diagnostic register 0000. ... 07/08/2006 05:24:47 PM myhost kernel eth2: Host error, FIFO diagnostic register 0000. 07/08/2006 05:24:47 PM myhost kernel Debug: sleeping function called from invalid context at include/asm/semaphore.h:105 07/08/2006 05:24:47 PM myhost kernel Call Trace: <IRQ> <ffffffff802096f7>{acpi_pci_link_allocate_irq+169} 07/08/2006 05:24:47 PM myhost kernel ACPI: PCI Interrupt 0000:01:06.0[A] -> Link [LNKD] -> GSI 17 (level, high) -> IRQ 193 ... 07/08/2006 05:24:47 PM myhost kernel ACPI: PCI Interrupt 0000:01:06.0[A] -> Link [LNKD] -> GSI 17 (level, high) -> IRQ 193 More info about eth2: # lspci -vv 01:06.0 Ethernet controller: 3Com Corporation 3c905C-TX/TX-M [Tornado] (rev 78) Subsystem: 3Com Corporation 3C905C-TX Fast Etherlink for PC Management NIC Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr- Stepping- SERR+ FastB2B- Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort+ >SERR- <PERR- Latency: 64 (2500ns min, 2500ns max), Cache Line Size 08 Interrupt: pin A routed to IRQ 193 Region 0: I/O ports at dc00 [size=128] Region 1: Memory at fa9ffc00 (32-bit, non-prefetchable) [size=128] Expansion ROM at fa9c0000 [disabled] [size=128K] Capabilities: [dc] Power Management version 2 Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold+) Status: D0 PME-Enable- DSel=0 DScale=2 PME- # uname -a Linux myhost 2.6.16.13-4-default #1 Wed May 3 04:53:23 UTC 2006 x86_64 x86_64 x86_64 GNU/Linux The mainboard is Asus SK8N with nForce3 chipset. I don't know what other information could be helpful, but I'm ready to provide it. Thanks for any help! Greetings, Mitko

On Saturday 08 July 2006 18:56, Bernd Felsche wrote:

Try booting with irqpoll as a boot parameter. ACPI may be setting up the wrong interrupts. This seems related to the nForce3 chipset or ASUS.

It doesn't work still the same massages in the syslog, still the same behavior :( Greetings, Mitko