fatal kernel trap

[Nuno Dias]

Ok, thank's anyway.

For now the SCSI card in 64 bit pci slot work, and i'm going to use this
configuration.

ND

On Tue, 2005-03-29 at 23:23 +0200, Bernd Walter wrote:
> On Tue, Mar 29, 2005 at 04:43:54PM +0100, Nuno Dias wrote:
> > On Tue, 2005-03-29 at 17:00 +0200, Bernd Walter wrote:
> > > On Tue, Mar 29, 2005 at 03:47:32PM +0100, Nuno Dias wrote:
> > The crash that i have reported now is from boot, after the install ...
> > This is the crash (IDE Disk)...
> > ------------------
> > Hit [Enter] to boot immediately, or any other key for command prompt.
> > Booting [/boot/kernel/kernel]...
> > Entering /boot/kernel/kernel at 0xfffffc0000345b60...
> > 
> > halted CPU 0
> > 
> > halt code = 2
> > kernel stack not valid halt
> > PC = 200000000
> > ------------------
> 
> This is very early and an interessting PC address.
> I have no idea about this.
> 
> > And yes, i see the same problem as JoséM.Fandiño when i try to install
> > and the scsi card is in 32 bits slot pci.
> > 
> > > 
> > > >  Then i change my approach to a SCSI DISK. 
> > > > 
> > > >  I discovery that the scsi card only work in 64 bits slot pci, if i put
> > > > the card in 32 bits slot pci the install crash.
> > > 
> > > Can you deliver boot messages from detecting the SCSI card in
> > > different slots?
> > > AFAIK the 32 bit slots on Miate are behind a PCI-PCI bridge.
> > 
> > 64 bit slot
> > 
> > isp0: <Qlogic ISP 1020/1040 PCI SCSI Adapter> port 0x8000-0x80ff mem
> > 0x800580000isp0: interrupting at CIA irq 4
> > isp0: [GIANT-LOCKED]
> > ...
> > Waiting 15 seconds for SCSI devices to settle
> > da0 at isp0 bus 0 target 0 lun 0
> > da0: <DEC RZ1CC-BA (C) DEC 883F> Fixed Direct Access SCSI-2 device
> > da0: 40.000MB/s transfers (20.000MHz, offset 8, 16bit), Tagged Queueing
> > Enabled
> > da0: 4091MB (8380080 512 byte sectors: 255H 63S/T 521C)
> > 
> > 32 bit slot
> > 
> > isp0: <Qlogic ISP 1020/1040 PCI SCSI Adapter> port 0x8000-0x80ff mem
> > 0x800100001isp0: interrupting at CIA irq 16
> > isp0: [GIANT-LOCKED]
> > 
> > halted CPU 0
> > 
> > halt code = 5
> > HALT instruction executed
> > PC = 1f00
> > >>>
> 
> That makes sense now.
> If the drivers tries to acces the port range it will trap as it is
> very likely not bridged.
> Havn't thought about this case yet.
> imp told be that we are verifying the path now, but it doesn't seem
> to work :(
> Well such cases are very specific to alpha world.
> 
-- 
Nuno Dias <ndias at lip.pt>
LIP