2940UW & 2.4.0-test-ac7 PCI probs

2940UW & 2.4.0-test-ac7 PCI probs

[Andreas Tobler]

Hi,

I try to get the 2.4.0-test1-ac7 to boot on my 7200. It fails. I have a
2940UW where all my disks are connected to. A 2.3.99p9 from linuxcare
boots, but the ac7 from bk fails with the following message:

Total memory = 80MB; using 512kB for hash table (at c0300000)
Linux version 2.4.0-test1-ac7 (root@pm7200) (gcc version 2.95.3 19991030
(prerelease/franzo)) #4 Mon Jun 5 17:14:33 CEST 2000
PCI bus 0 controlled by bandit at f2000000
[snip]
PCI: Device 00:0e.0 not available because of resource collisions
[snip]

The kernel can't boot since it doesn't find the root. ;-)

I'm aware of problems with the adaptec on 2.4.0-test1 and up. (Alan Cox
writes so) But I'm not sure if that is the same problems, since on lk
people reported about timeouts etc. And on my side I have a general PCI problem.

Are any PCI-fixup patches integrated in bk? Or should I try Michel fixes?

Thanks,

Andreas

Below the lspci -vv from the 2.3.99p9

00:0b.0 Host bridge: Apple Computer Inc. Bandit PowerPC host bridge (rev 03)
	Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr-
Stepping- SERR- FastB2B-
	Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort+ >SERR- <PERR-
	Latency: 32 set, cache line size 08

00:0d.0 Ethernet controller: 3Com Corporation 3c590 10BaseT [Vortex]
	Control: I/O- Mem- BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B-
	Status: Cap- 66Mhz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
	Latency: 3 min, 8 max, 32 set
	Interrupt: pin A routed to IRQ 23
	Region 0: I/O ports at 0480 [disabled]

00:0e.0 SCSI storage controller: Adaptec AIC-7881U
	Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR- FastB2B-
	Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
	Latency: 8 min, 8 max, 32 set, cache line size 08
	Interrupt: pin A routed to IRQ 24
	Region 0: I/O ports at <unassigned>
	Region 1: Memory at 80801000 (32-bit, non-prefetchable)

00:0f.0 Ethernet controller: 3Com Corporation 3c905B 100BaseTX [Cyclone]
(rev 64)
	Subsystem: Unknown device 10b7:9055
	Control: I/O- Mem- BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr-
Stepping- SERR- FastB2B-
	Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
	Latency: 10 min, 10 max, 32 set, cache line size 08
	Interrupt: pin A routed to IRQ 25
	Region 0: I/O ports at 0400 [disabled]
	Region 1: Memory at 80800000 (32-bit, non-prefetchable) [disabled]
	Capabilities: <available only to root>

00:10.0 Class ff00: Apple Computer Inc. Grand Central I/O (rev 02)
	Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr-
Stepping- SERR- FastB2B-
	Status: Cap- 66Mhz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR+
	Latency: 32 set, cache line size 08
	Region 0: Memory at f3000000 (32-bit, non-prefetchable)

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 2940UW & 2.4.0-test-ac7 PCI probs

[Michel Lanners]

Hi Andreas,

On   5 Jun, this message from Andreas Tobler echoed through cyberspace:
> I try to get the 2.4.0-test1-ac7 to boot on my 7200. It fails. I have a
> 2940UW where all my disks are connected to. A 2.3.99p9 from linuxcare
> boots, but the ac7 from bk fails with the following message:
>
> Total memory = 80MB; using 512kB for hash table (at c0300000)
> Linux version 2.4.0-test1-ac7 (root@pm7200) (gcc version 2.95.3 19991030
> (prerelease/franzo)) #4 Mon Jun 5 17:14:33 CEST 2000
> PCI bus 0 controlled by bandit at f2000000
> [snip]
> PCI: Device 00:0e.0 not available because of resource collisions
> [snip]

Get either my PCI patches or Geert's patches posted here a few hours
ago. Then enable hotswap (still needed, right, Geert?), and the resource
code should fix any collisions it finds.

Michel

-------------------------------------------------------------------------
Michel Lanners                 |  " Read Philosophy.  Study Art.
23, Rue Paul Henkes            |    Ask Questions.  Make Mistakes.
L-1710 Luxembourg              |
email   mlan@cpu.lu            |
http://www.cpu.lu/~mlan        |                     Learn Always. "

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 2940UW & 2.4.0-test-ac7 PCI probs

[Geert Uytterhoeven]

On Mon, 5 Jun 2000, Michel Lanners wrote:
> Hi Andreas,
> On   5 Jun, this message from Andreas Tobler echoed through cyberspace:
> > I try to get the 2.4.0-test1-ac7 to boot on my 7200. It fails. I have a
> > 2940UW where all my disks are connected to. A 2.3.99p9 from linuxcare
> > boots, but the ac7 from bk fails with the following message:
> >
> > Total memory = 80MB; using 512kB for hash table (at c0300000)
> > Linux version 2.4.0-test1-ac7 (root@pm7200) (gcc version 2.95.3 19991030
> > (prerelease/franzo)) #4 Mon Jun 5 17:14:33 CEST 2000
> > PCI bus 0 controlled by bandit at f2000000
> > [snip]
> > PCI: Device 00:0e.0 not available because of resource collisions
> > [snip]
>
> Get either my PCI patches or Geert's patches posted here a few hours

Note that my patches need some changes to make them work on PowerMac.

> ago. Then enable hotswap (still needed, right, Geert?), and the resource
> code should fix any collisions it finds.

I don't think you need hotswap to fix collisions.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 2940UW & 2.4.0-test-ac7 PCI probs

[Andreas Tobler]

Hi Michel,

Michel Lanners wrote:
> Get either my PCI patches or Geert's patches posted here a few hours
> ago. Then enable hotswap (still needed, right, Geert?), and the resource
> code should fix any collisions it finds.

Fine, I had to take yours, I wasn't able to apply Geerts one. (Too late/
too tired ???)
I applied the one for 2.3.99p9 to 2.4.0-test1-ac7, success, including
your serial console patch and Takashi's mftb patch for 601 CPU's.
Below the output. I think now I am in the same situation as other
adaptec users on the linux front.
One Q, hotswap == hotswap PCI in 'General Setup'?

Thanks

Andreas

-------

Total memory = 80MB; using 512kB for hash table (at c0300000)
Linux version 2.4.0-test1-ac7 (root@pm7200) (gcc version 2.95.3 19991030
(prerelease/franzo)) #7 Tue Jun 6 00:20:04 CEST 2000
PCI bus 0 controlled by bandit at f2000000
Cache coherency enabled for bandit/PSX at f2000000
[snip]
PCI: Probing PCI hardware (semiautomatic)
Scanning bus 00
Found 00:58 [106b/0001] 000600 00
Found 00:68 [10b7/5900] 000200 00
Found 00:70 [9004/8178] 000100 00
Found 00:78 [10b7/9055] 000200 00
Found 00:80 [106b/0002] 00ff00 00
Fixups for bus 00
Bus scan for 00 returning with max=00
PCI: Fixing device 00:0b.0 (106b:0001)
PCI: Fixing device 00:0d.0 (10b7:5900)
PCI: Setting IRQ 23 on device 00:0d.0.
PCI: Correcting IO address 0 on device 00:0d.0, now f2000480.
PCI: Enabling device 00:0d.0 (0004 -> 0005)
PCI: Fixing device 00:0e.0 (9004:8178)
PCI: Setting IRQ 24 on device 00:0e.0.
PCI: Device 00:0e.0 not available because of resource collisions
PCI: Fixing device 00:0f.0 (10b7:9055)
PCI: Setting IRQ 25 on device 00:0f.0.
PCI: Correcting IO address 0 on device 00:0f.0, now f2000400.
PCI: Enabling device 00:0f.0 (0014 -> 0017)
PCI: Fixing device 00:10.0 (106b:0002)
PCI: Resource f2000480-f200049f (f=101, d=0, p=0)
PCI: Resource 80801000-80801fff (f=200, d=0, p=0)
PCI: Resource f2000400-f200047f (f=101, d=0, p=0)
PCI: Resource 80800000-8080007f (f=200, d=0, p=0)
PCI: Resource f3000000-f301ffff (f=200, d=0, p=0)
  for root[0:ffffffff] min[1000] size[100]
  got res[1000:10ff] for resource 0
[snip]
(scsi0) <Adaptec AHA-294X Ultra SCSI host adapter> found at PCI 0/14/0
(scsi0) Wide Channel, SCSI ID=7, 16/255 SCBs
(scsi0) Warning - detected auto-termination
(scsi0) Please verify driver detected settings are correct.
(scsi0) If not, then please properly set the device termination
(scsi0) in the Adaptec SCSI BIOS by hitting CTRL-A when prompted
(scsi0) during machine bootup.
(scsi0) Cables present (Int-50 NO, Int-68 YES, Ext-68 NO)
(scsi0) Downloading sequencer code... 422 instructions downloaded
scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 5.2.0/3.2.4
       <Adaptec AHA-294X Ultra SCSI host adapter>
scsi1 : 53C94
scsi : 2 hosts.
scsi : aborting command due to timeout : pid 0, scsi0, channel 0, id 0,
lun 0 Inquiry 00 00 00 ff 00
SCSI host 0 abort (pid 0) timed out - resetting
SCSI bus is being reset for host 0 channel 0.
SCSI host 0 channel 0 reset (pid 0) timed out - trying harder
SCSI bus is being reset for host 0 channel 0.
SCSI host 0 abort (pid 0) timed out - resetting
SCSI bus is being reset for host 0 channel 0.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 2940UW & 2.4.0-test-ac7 PCI probs

[Michel Lanners]

Hi there,

On   6 Jun, this message from Andreas Tobler echoed through cyberspace:
>> Get either my PCI patches or Geert's patches posted here a few hours
>> ago. Then enable hotswap (still needed, right, Geert?), and the resource
>> code should fix any collisions it finds.
>
> Fine, I had to take yours, I wasn't able to apply Geerts one. (Too late/
> too tired ???)
> I applied the one for 2.3.99p9 to 2.4.0-test1-ac7, success, including
> your serial console patch and Takashi's mftb patch for 601 CPU's.
> Below the output. I think now I am in the same situation as other
> adaptec users on the linux front.

See below; I've spotted one thing that probably breaks.

> One Q, hotswap == hotswap PCI in 'General Setup'?

Yes, but it is not needed, I think.

>   for root[0:ffffffff] min[1000] size[100]
>   got res[1000:10ff] for resource 0

This must be the conflicting resource, but it gets an equally-wrong new
address :-(( In fact, the PCI resource code still misses a patch to make
the correct resources available to the resource allocation code (so that
IO regions can be taken out of the IO region of the correct PCI bus..)

I'm still working on that part ;-)

So if the SCSI driver actually uses the IO region, it will most
certainly not work....

Michel

-------------------------------------------------------------------------
Michel Lanners                 |  " Read Philosophy.  Study Art.
23, Rue Paul Henkes            |    Ask Questions.  Make Mistakes.
L-1710 Luxembourg              |
email   mlan@cpu.lu            |
http://www.cpu.lu/~mlan        |                     Learn Always. "

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 2940UW & 2.4.0-test-ac7 PCI probs

[Andreas Tobler]

Hi,

Michel Lanners wrote:
> >   for root[0:ffffffff] min[1000] size[100]
> >   got res[1000:10ff] for resource 0
>
> This must be the conflicting resource, but it gets an equally-wrong new
> address :-(( In fact, the PCI resource code still misses a patch to make
> the correct resources available to the resource allocation code (so that
> IO regions can be taken out of the IO region of the correct PCI bus..)
>
> I'm still working on that part ;-)

Thanks!

>
> So if the SCSI driver actually uses the IO region, it will most
> certainly not work....

How do I find that out?

Below it says I/O ports unassigned and it works. (2399p9)

00:0e.0 SCSI storage controller: Adaptec AIC-7881U
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr-
Stepping- SERR- FastB2B-
        Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
        Latency: 8 min, 8 max, 32 set, cache line size 08
        Interrupt: pin A routed to IRQ 24
        Region 0: I/O ports at <unassigned>
        Region 1: Memory at 80801000 (32-bit, non-prefetchable)

Andreas

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 2940UW & 2.4.0-test-ac7 PCI probs

[Michel Lanners]

Hi Andreas,

On   6 Jun, this message from Andreas Tobler echoed through cyberspace:
>> So if the SCSI driver actually uses the IO region, it will most
>> certainly not work....
>
> How do I find that out?

RTFS ;-)

> Below it says I/O ports unassigned and it works. (2399p9)

So it is probably using the memory region. A lot of those chips support
two different views on their register file: via an IO port window and
via a memory(-mapped) region. But the functionality available is the
same on both...

Michel

-------------------------------------------------------------------------
Michel Lanners                 |  " Read Philosophy.  Study Art.
23, Rue Paul Henkes            |    Ask Questions.  Make Mistakes.
L-1710 Luxembourg              |
email   mlan@cpu.lu            |
http://www.cpu.lu/~mlan        |                     Learn Always. "

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/