Messages in this thread | | | Date | Thu, 15 Jan 1998 17:49:56 +0100 (MET) | From | Paul Wouters <> | Subject | RE: AIC7xxx.c oops |
| |
On Wed, 7 Jan 1998, Doug Ledford wrote:
Hi, I am back :)
You answered:
> I think your machine is pretty broken. It looks like that HP Lan card and > driver are stomping on the 2742T's I/O address. I say this because the card > is working fine through the SCSI bus scan, then the HP driver starts up, > then we get a BRKADRINT on the card, which means that the card was accessed > when it hadn't been paused, then when we go to look at the card, all three > of our hardware lists on that card have been stomped on. I think we have a > bad configuration here somewhere.
I removed the LAN card, replaced the AMD chip with an Intel DX2. Disabled onboard mouse, two serial ports, 1 par port. Re-ran EISA config manager. So, there isn't really much more that can give interrupts. I still get an oops. This is running pre-2.1.80-2.
>>EIP: c01ac3ca <aic7xxx_match_scb+4e/d0>
Trace: c01acd96 <aic7xxx_reset_device+156/4c4> Trace: c01151bd <printk+155/164> Trace: c01af28a <aic7xxx_isr+3ce/4bc> Trace: c01ce194 <RCSid+27f5/3dd9> Trace: c0112ec9 <timer_bh+dd/394> Trace: c0112f08 <timer_bh+11c/394> Trace: c010ba16 <do_IRQ+b2/158> Trace: c0106000 <init_task_union> Trace: c010a0b0 <ret_from_intr> Trace: c0106000 <init_task_union> Trace: c0100018 <startup_32+18/b8> Trace: c0106000 <init_task_union> Trace: c011a240 <find_gap+64/74> Trace: c0106000 <init_task_union> Trace: c011a301 <release_region+31/3c> Trace: c0106000 <init_task_union> Trace: c0132368 <sys_setup+ac/d8> Trace: c0109fca <system_call+3a/40> Trace: c0121e60 <kswapd> Trace: c0106000 <init_task_union> Trace: c01b0018 <aic7xxx_register+12c/8c8> Trace: c01080bb <init+4f/1b4> Trace: c0107fe4 <this_must_match_init_task+1fe4/2000> Trace: c0106000 <init_task_union> Trace: c01080bb <init+4f/1b4>
Code: c01ac3ca <aic7xxx_match_scb+4e/d0> Code: c01ac3ca <aic7xxx_match_scb+4e/d0> 8b 40 08 movl 0x8(%eax),%eax Code: c01ac3cd <aic7xxx_match_scb+51/d0> 8b 40 0c movl 0xc(%eax),%eax Code: c01ac3d0 <aic7xxx_match_scb+54/d0> 0f b7 40 32 movzwl 0x32(%eax),%eax Code: c01ac3da <aic7xxx_match_scb+5e/d0> 50 pushl %eax Code: c01ac3db <aic7xxx_match_scb+5f/d0> 68 a4 d4 1c c0 pushl $0xc01cd4a4 Code: c01ac3e0 <aic7xxx_match_scb+64/d0> e8 89 8c f6 00 call f68c9e <_EIP+f68c9e> Code: c01ac3eb <aic7xxx_match_scb+6f/d0> 90 nop Code: c01ac3ec <aic7xxx_match_scb+70/d0> 90 nop Code: c01ac3ed <aic7xxx_match_scb+71/d0> 90 nop
I still see some irq problems, and still the "Illegal host access" is there :(
I don't know what else I can do to either debug or solve this problem, so any clue as to what to do (apart from burning this Intel machine) would be appreciated. (It's a really decent machine with the AMD in there)
Paul
| |