Hi! > -----Original Message----- > From: freeman groups [mailto:[EMAIL PROTECTED] > Sent: Friday, June 18, 2004 3:13 AM > To: LEAF > Subject: Re: [leaf-user] uClibc 2.2.0_b4 - Caution using > via-rhine - "kernel BUG at slab.c:1130!"/"In interrupt > handler - not syncing"/"Kernel panic" > > Luis, thanks for your thoughts. > > Luis.F.Correia wrote: > > >Hi! > > > >Without any cable attached to the network cards, try to cat > /proc/interrupts > >and check if there is any interrupt sharing... > > > > > I didn't do that particular check but I know who was getting what > interrupts and no-one was sharing. >
Beware, most boards have a quirk that makes the leftmost PCI slot to share the same int of the closest ISA slot, which means if you have a PCI right next to an ISA card, chances are that thay might get the same one... > ... In all fairness though I should check specifically what > you took the > time to recommend, so here's the cat (after unplugging the > cable which I > forgot to do in my first retest)... > > [BTW, can I ask why you suggested I have no network cable > plugged in - > this shouldn't make a difference, right?] It does not make any difference to the interrupt, but it won't bring your system down while we are gathering info :) > > > [EMAIL PROTECTED] /root> # cat /proc/interrupts > > CPU0 > > 0: 4547 XT-PIC timer > > 1: 2 XT-PIC keyboard > > 2: 0 XT-PIC cascade > > 4: 650 XT-PIC serial > > 8: 0 XT-PIC rtc > > 10: 0 XT-PIC eth0 > > 14: 134 XT-PIC ide0 > > NMI: 0 > > ERR: 0 > > Yup, got the problem (error msg et al) again. > > Odd though, that COM2 isn't showing up with an IRQ... ??? ... It's > active in BIOS and shows up in the boot-up messages. At this point, which network card drivers are loaded? Interrupts only show on this list after the driver is loaded AND the card is properly initialized (ifup). I've seen the COM2 behaviour as well, but since noone is using int 3, it does not really matter. One thing though, you can disable COM2 in BIOS, so that you have another available int. > > >Maybe if you swap that card to another PCI slot, it would > get a different > >interrupt and then > >stop behaving like that. > > > > > For one test I had reserved the normally-assigned IRQ 10 for ISA (via > BIOS), which forced the card to get IRQ 11, but the problem > remained. As > well in another test I recall moving this net card to the > neighbouring > slot and it still got IRQ 10 (i.e. mobo-based IRQ assignment > algorithm). > > FWIW I had also removed all cards except the VGA and the > D-Link and the > problem persisted (and the VGA was, via a jumper on the card, > told not > to get an interrupt and indeed it was not ... but would when > I changed > the jumper setting ... I mention this only because my > experience is that > IRQ assignment for VGA cards is usually done via BIOS, but not the > crappy Phoenix BIOS on my mobo ... instead it was done via > the jumper on > the VGA card itself - and via watching the POST screen I saw that the > jumper on/off did what was expected in terms of assigning, or not, an > interrupt). > > >Did you disable the USB controller on the motherboard? > >You would gain an extra interrupt. > > > > > Never enabled it at all, since I have no use for USB on the mobo. > > Here's my IRQ usage (all cards inserted): > 3 - COM2 > 4 - COM1 > 5 - COM 3 (ISA modem, IRQ reserved for ISA in BIOS) > 7 - LPT - mobo > 9 - ETH0 (ISA, IRQ reserved for ISA in BIOS) > 10 - ETH1 (PCI) > 11 - ETH2 (PCI) > 12 - available (since PS/2 mouse is disabled via mobo jumper) > > I do have a second LPT port (ISA card) in the machine but the > jumper on > it - IRQ5 or IRQ 7 or <none> (=unjumpered, and the /only/ officially > recommended setting, per the doc'n for the card) seemed to have no > effect from what I could tell in terms of running an old DOS > util called > IRQStat - also confirmed via checking /proc/interrupts with various > jumper configs tested). The DMESG info from boot-up indicates > that IRQ 7 > is assigned for LPT1 but makes no mention of an IRQ for LPT2, > though it > recognizes both LPT's and does polling on both of them. I > only have IRQ > 7 assigned because the BIOS gives no choice of "no-IRQ" for > the built-in > LPT port. > > It (/net/via-rhine) not working sure confounded me, since I > was pretty > darned fastidious about my IRQ assignments, given how full my > slots are. Well, if nothing else works, try with another motherboard or even invest on a four-port ethernet card. Ther are some cheap ones around. > > scott; canada > > >>----Original Message----- > >>From: freeman groups [mailto:[EMAIL PROTECTED] > >>Sent: Wednesday, June 16, 2004 9:50 PM > >>To: LEAF > >>Subject: [leaf-user] uClibc 2.2.0_b4 - Caution using > >>via-rhine - "kernel BUG at slab.c:1130!"/"In interrupt > >>handler - not syncing"/"Kernel panic" > >> > >> My net card in question is a D-Link DFE-530TX Rev A1. I > have been > >>placing the modules into the modules package (/lib/modules > >>folder) and > >>backing up. > >> > > > ------------------------------------------------------- > This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference > Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer > Conference, June 28 - July 1 at the Moscone Center in San > Francisco, CA > REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority > Code NWMGYKND > -------------------------------------------------------------- > ---------- > leaf-user mailing list: [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/leaf-user > SR FAQ: http://leaf-project.org/pub/doc/docmanager/docid_1891.html > ------------------------------------------------------- This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer Conference, June 28 - July 1 at the Moscone Center in San Francisco, CA REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code NWMGYKND ------------------------------------------------------------------------ leaf-user mailing list: [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-user SR FAQ: http://leaf-project.org/pub/doc/docmanager/docid_1891.html
