Next-in-Thread Next Message

Ok Problem solved; card working! 

Forum: PCMCIA Network Adapter Issues
Re: None "eth0: Tx hung, 9 vs 0." from NETGEAR FA510c (David Brooks)
Re: None Interrupt setup problem (David Hinds)
Date: 2000, Aug 03
From: David Brooks dbrooks

 Thank you so very much for the accurate [and very rapid, as well] advice.

And, while I'm at it, I would be amiss to not especially thank you for the excellent and extensive work you have devoted to making linux on laptops a real world phenomena. I hold you within the same pantheon as Linus!

Sure enough, as soon as I looked in /proc/interrupts (at the same time that the card was reported in the syslog by the pcmcia tools at irq5), I noticed that it was not showing up there. Aha! You pointed me in exactly the right direction, and I proceeded to experiment with all of the interrupt options of the i82365 and pcmcia_core modules. [Although I already tried several permutations of "do_scan", "do_pci_probe" "pci_int", I had not been focusing explicitly on an interrupt problem... finally found that PCIC_OPTS="pci_irq_list=5 " and CORE_OPTS="cb_pci_irq=5 " did the trick quite marvelously. Perhaps you might want to provide a more extensive list of errors and their possible implications in the howto.

Oh, also. Minor bug report, unrelated; perhaps you would find it usefull to know that on my system, pcmcia-cs-3.3.8 had never crashed the system, but when I upgraded to pcmcia-cs-3.3.19, it would oops whenever it was called by rc at system boot up. I disabled it and tried a few things before I realized that it would sometimes start fine if I activated the init.d script by hand after the system was up... Weird! So I put a "sleep 1s" command after each insmod of the core modules, and it has worked perfectly and with no oops since then. [Laptop is a pokey old 80586 'pentium', 133Mhz]. It also did this when I was trying v3.3.18.

--David Brooks

...oh, and I do have the output from dump_pirq if you should want it for anything...

---------------------

This is from before the explicit irq=5 configuration, just a fresh install of pcmcia

Interrupt routing table found at address 0xfd3d0:
  Version 1.0, size 0x0060
  Interrupt router is device 00:01.0
  PCI exclusive interrupt mask: 0x0000
  Compatible router: vendor 0x8086 device 0x122e

Device 00:00.0 (slot 0):

Device 00:02.0 (slot 0):
  INTA: link 0x62, irq mask 0x8eb8

Device 00:01.0 (slot 0):

Device 00:13.0 (slot 0):
  INTA: link 0x60, irq mask 0x8eb8
  INTB: link 0x61, irq mask 0x8eb8

Interrupt router at 00:01.0: Intel 82371AB PIIX4/PIIX4E PCI-to-ISA bridge
  PIRQ1 (link 0x60): unrouted
  PIRQ2 (link 0x61): unrouted
  PIRQ3 (link 0x62): unrouted
  PIRQ4 (link 0x63): unrouted
  Serial IRQ: [disabled] [quiet] [frame=21] [pulse=4]

---------

this is from the working card, with the explicit force to irq 5

---------

Interrupt routing table found at address 0xfd3d0:
  Version 1.0, size 0x0060
  Interrupt router is device 00:01.0
  PCI exclusive interrupt mask: 0x0000
  Compatible router: vendor 0x8086 device 0x122e

Device 00:00.0 (slot 0):

Device 00:02.0 (slot 0):
  INTA: link 0x62, irq mask 0x8eb8

Device 00:01.0 (slot 0):

Device 00:13.0 (slot 0):
  INTA: link 0x60, irq mask 0x8eb8
  INTB: link 0x61, irq mask 0x8eb8

Interrupt router at 00:01.0: Intel 82371AB PIIX4/PIIX4E PCI-to-ISA bridge
  PIRQ1 (link 0x60): irq 5
  PIRQ2 (link 0x61): irq 5
  PIRQ3 (link 0x62): unrouted
  PIRQ4 (link 0x63): unrouted
  Serial IRQ: [disabled] [quiet] [frame=21] [pulse=4]

Next-in-Thread Next Message

Add Message to: "Problem solved; card working!"

Members Subscribe Admin Mode Show Frames Help for HyperNews at pcmcia-cs.sourceforge.net 1.10