Tigran Aivazian wrote:
>
> Hi Martin,
>
> I just found out that my earlier statement "2.2.x is okay" should be
> changed to "win98 is okay" so there are definitely problems with sharing
> PCI irqs between eepro100/3c59x/(rtl)8139(too) in both 2.2.x and 2.4.x. I
> utterly don't care about 2.2.x but I am in the mood to find out what's
> wrong with 2.4.x.
>
Perhaps that patch was applied in test9-pre2?
Because Cardbus on this recent Dell laptop has
suddenly stopped working:
Sep 11 22:52:52 dell kernel: Yenta IRQ list 0498, PCI irq11
Sep 11 22:52:52 dell kernel: Socket status: 30000006
Sep 11 22:52:52 dell kernel: cs: cb_alloc(bus 2): vendor 0x10b7, device 0x0000
Sep 11 22:52:52 dell kernel: PCI: Failed to allocate resource 3 for PCI device 10b7:0000
Sep 11 22:52:52 dell kernel: PCI: Failed to allocate resource 4 for PCI device 10b7:0000
Sep 11 22:52:52 dell kernel: PCI: Failed to allocate resource 5 for PCI device 10b7:0000
Sep 11 22:52:52 dell kernel: PCI: Enabling device 02:00.0 (0000 -> 0003)
I don't see where we _ever_ scan behind a Cardbus bridge???
As a datapoint, this hack makes Cardbus work again:
--- linux-2.4.0-test9-pre2/drivers/pci/pci.c Mon Sep 18 20:31:49 2000
+++ linux-akpm/drivers/pci/pci.c Mon Sep 18 22:17:39 2000
@@ -714,7 +714,7 @@
* We need to blast all three values with a single write.
*/
pci_write_config_dword(dev, PCI_PRIMARY_BUS, buses);
- if (!is_cardbus) {
+ if (1 || !is_cardbus) {
/* Now we can scan all subordinate buses... */
max = pci_do_scan_bus(child);
} else {
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Sep 23 2000 - 21:00:17 EST