| To: | Harald Welte <laforge@xxxxxxxxxxxx> |
|---|---|
| Subject: | Re: [BUG] sundance "Something Wicked happened!" (IntrPCIErr) |
| From: | "David S. Miller" <davem@xxxxxxxxxxxxx> |
| Date: | Mon, 31 Jan 2005 11:38:34 -0800 |
| Cc: | netdev@xxxxxxxxxxx |
| In-reply-to: | <20050131165947.GJ6878@sunbeam.de.gnumonks.org> |
| References: | <20050131165947.GJ6878@sunbeam.de.gnumonks.org> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
On Mon, 31 Jan 2005 17:59:48 +0100
Harald Welte <laforge@xxxxxxxxxxxx> wrote:
> Looking at the source code, the error message only happens if IntrPCIErr
> is set.
Try adding a printout here of the PCI status register in config
space when this happens, something like:
{
u16 val;
pci_read_config_word(np->pci_dev, PCI_STATUS, &val);
printk("%s: PCI status is %x\n", dev->name, val);
}
What we'll be looking for are bits set there such as
PCI_STATUS_SIG_TARGET_ABORT, PCI_STATUS_REC_TARGET_ABORT,
PCI_STATUS_REC_MASTER_ABORT, PCI_SIG_SYSTEM_ERROR, or
PCI_STATUS_DETECTED_PARITY.
That will tell us who is seeing the error and thus
signalling the event (either PCI bus or the sundance
chip itself).
I don't know anything specific about the sundance
hardware though, sorry.
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [PATCH] OpenBSD Networking-related randomization port, Valdis . Kletnieks |
|---|---|
| Next by Date: | Re: Fw: [Bugme-new] [Bug 4138] New: ipsec with racoon in transport mode with esp and ah hangs (problem is in xfrm_state_add), David S. Miller |
| Previous by Thread: | [BUG] sundance "Something Wicked happened!" (IntrPCIErr), Harald Welte |
| Next by Thread: | where is the proper place for r8169 bug reports?, Dale E. Martin |
| Indexes: | [Date] [Thread] [Top] [All Lists] |