Hello all,
2.5.69-mm3 works fine, mm4 wasn't tested. Kernel oopses right after attempting to
initialize agpgart. I've managed to copy only the little data from the oops
that is shown below, enough to locate it (oops happened in the swapper task):
agp_via_probe+0x106/0x180
sysfs_create_dir+0x81/0xb0
pci_device_probe+0x5e/0x70
bus_match+0x45/0xb0
driver_attach+0x5c/0x60
bus_add_driver+0xa8/0xc0
driver_register+0x2f/0x40
pci_register_driver+0x41/0x60
agp_via_init+0x13/0x40
do_initcalls+0x2c/0xa0
init_workqueues+0xf/0x30
init+0x33/0x1b0
init+0x0/0x1b0
kernel_thread_helper+0x5/0x18
Code: 5c 24 0c 85 74 24 10 8b 5c 24 18 8b 43 18 85 c0 0f 84 55 01
00 00 a1 e4 5d 44 c0 85 c0 0f 85 32 01 00 00 8b 43 04 be 01
00 00 00 <8b> 10 85 d2 74 2f b8 00 e0 ff ff 21 e0 ff 40 14
85 3a 02 0f 84
Machine config attached,
regards,
marek
On Sat, May 17, 2003 at 01:05:26AM +0200, Marek Habersack wrote:
> Hello all,
>
> 2.5.69-mm3 works fine, mm4 wasn't tested. Kernel oopses right after attempting to
> initialize agpgart. I've managed to copy only the little data from the oops
> that is shown below, enough to locate it (oops happened in the swapper task):
patch from Christoph Hellwig attached.
Still waiting for Linus to pull this (and other) agp bits from bkbits.
Dave
--- 1.39/drivers/char/agp/via-agp.c Mon Apr 28 03:32:35 2003
+++ edited/drivers/char/agp/via-agp.c Tue May 13 10:51:00 2003
@@ -402,6 +402,7 @@
bridge->dev = pdev;
bridge->capndx = cap_ptr;
+ bridge->driver = &via_driver; /* might be overriden later */
switch (pdev->device) {
case PCI_DEVICE_ID_VIA_8367_0:
@@ -427,7 +428,6 @@
}
/*FALLTHROUGH*/
default:
- bridge->driver = &via_driver;
break;
}