irqpoll didn't seem to help either...I've grepped irq and 537 in my dmesg and here is what it is: [root@pb3000 ~]# dmesg | grep irq [ 0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl) [ 0.000000] ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level) [ 0.000000] nr_irqs_gsi: 40 [ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-3.1.6-1.fc16.i686 root=/dev/sda3 ro rd.md=0 rd.lvm=0 rd.dm=0 KEYTABLE=us quiet SYSFONT=latarcyrheb-sun16 rhgb rd.luks=0 LANG=en_US.UTF-8 irqpoll [ 0.000000] CPU 0 irqstacks, hard=f2c20000 soft=f2c22000 [ 0.020998] CPU 1 irqstacks, hard=f2cca000 soft=f2ccc000 [ 0.143946] pnp 00:03: [irq 8] [ 0.144078] pnp 00:05: [irq 13] [ 0.144438] pnp 00:06: [irq 4] [ 0.144770] pnp 00:07: [irq 3] [ 0.145190] pnp 00:08: [irq 7] [ 1.152832] serial8250: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A [ 1.173462] serial8250: ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A [ 1.194533] 00:06: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A [ 1.218635] 00:07: ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A [ 1.226389] ata1: PATA max UDMA/100 cmd 0x1f0 ctl 0x3f6 bmdma 0xf000 irq 14 [ 1.226393] ata2: PATA max UDMA/100 cmd 0x170 ctl 0x376 bmdma 0xf008 irq 15 [ 1.227871] ata3: SATA max UDMA/133 cmd 0xd400 ctl 0xd800 bmdma 0xe400 irq 19 [ 1.227876] ata4: SATA max UDMA/133 cmd 0xdc00 ctl 0xe000 bmdma 0xe408 irq 19 [ 1.232362] ehci_hcd 0000:00:1d.7: irq 23, io mem 0xf20c4000 [ 1.242567] uhci_hcd 0000:00:1d.0: irq 23, io base 0x0000b000 [ 1.242993] uhci_hcd 0000:00:1d.1: irq 19, io base 0x0000b400 [ 1.243452] uhci_hcd 0000:00:1d.2: irq 18, io base 0x0000b800 [ 1.243907] uhci_hcd 0000:00:1d.3: irq 16, io base 0x0000bc00 [ 1.247417] serio: i8042 KBD port at 0x60,0x64 irq 1 [ 1.247426] serio: i8042 AUX port at 0x60,0x64 irq 12 [ 1.248073] rtc0: alarms up to one month, 242 bytes nvram, hpet irqs [ 9.678735] parport0: PC-style at 0x378, irq 7 [PCSPP,TRISTATE] [ 9.699813] snd_hda_intel 0000:00:1b.0: irq 40 for MSI/MSI-X [ 64.179719] WARNING: at kernel/softirq.c:159 local_bh_enable_ip+0x60/0x90() [root@pb3000 ~]# dmesg | grep 537 [ 2.415376] firewire_ohci: Added fw-ohci device 0000:01:07.0, OHCI v1.10, 4 IR + 8 IT contexts, quirks 0x2 [ 14.916794] Intel 537 card found [ 46.119822] 537: Loaded [ 64.179725] Modules linked in: Intel537 8021q garp bridge stp llc it87 hwmon_vid nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack joydev ppdev microcode snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device parport_pc parport snd_pcm netjet isdnhdlc crc_ccitt mISDNipac mISDN_core serio_raw r8169 mii i2c_i801 iTCO_wdt iTCO_vendor_support snd_timer snd soundcore snd_page_alloc uinput usb_storage firewire_ohci firewire_core crc_itu_t i915 drm_kms_helper drm i2c_algo_bit i2c_core video [last unloaded: scsi_wait_scan] [ 64.179942] [<f9de6b12>] linux_MT_ReleaseLock_mt_dsp+0x12/0x20 [Intel537] [ 64.180048] [<f9de8862>] InterruptDSP+0x22/0x30 [Intel537] [ 64.180167] [<fa03e676>] set_register+0x30/0x34 [Intel537] [ 64.180167] [<f9de6c12>] ? linux_MT_AcquireLock_acu_rx_buffer+0x12/0x20 [Intel537] [ 64.180167] [<fa085866>] ResetDspInternal+0x1a/0x6a [Intel537] [ 64.180167] [<f9de55ab>] clm_terminate+0x5b/0x70 [Intel537] [ 64.180167] [<f9de440d>] close+0xd/0x30 [Intel537] [ 64.180167] [<f9de78e7>] softserial_close+0x77/0xb0 [Intel537] [ 64.180167] [<c0537396>] ? do_vfs_ioctl+0x86/0x5a0 for some reason I think the softserial_close line near the end is the culprit here...but then it's just a guess. Patrick On Fri, Jan 06, 2012 at 12:06:50PM +0100, Philippe Vouters wrote: > Patrick, > > I am exploiting all informations you sent me. When your modem board > interrupts at IRQ #19, you have this line your dmesg output: > [ 5123.808126] irq 19: nobody cared (try booting with the "irqpoll" option) > I do not know whether Linux booting with the irqpoll option will > make a real difference for you. > > After you have suppressed "_bh" at the two line lines I mentioned, > rebuilt, uninstalled the old driver, reinstalled the new one with > the two code changes and finally made all the tests I suggest, it > might perhaps be worth giving the irqpoll option a try provided you > keep your Tiger modem board in the same PCI slot. > -- > Philippe Vouters (Fontainebleau/France) > URL: http://vouters.dyndns.org/
