Skip to contentSkip to conversation

current git crashes on bootup with pci_iounmap()

DISCUSSION Kay Sievers 3 participants 6 messages Feb 13, 2007 – Feb 24, 2007

lore
Share & tools
lore
#0 lore raw

kernel BUG at lib/iomap.c:254!
invalid opcode: 0000 [#1]
...

The screen picture is here:
http://vrfy.org/pci_iounmap.jpg

It's a Thinkpad T43p.

2.6.20 was working fine.

Commenting out:
IO_COND(addr, /* nothing */, iounmap(addr));
in:
lib/iomap.c:254
makes at least booting up possible.

Thanks,
Kay

#1 lore raw

Hi Kay,

Quoted 9 lines

kernel BUG at lib/iomap.c:254!
invalid opcode: 0000 [#1]
...

The screen picture is here:
http://vrfy.org/pci_iounmap.jpg

It's a Thinkpad T43p.

Quoted 7 lines

2.6.20 was working fine.

Commenting out:
IO_COND(addr, /* nothing */, iounmap(addr));
in:
lib/iomap.c:254
makes at least booting up possible.

I saw a similar one on my X41. Disabling the AHCI driver made the machine booting again.

Regards

Marcel

#2 lore raw
Quoted 11 lines · Marcel Holtmann
Quoted 11 lines

kernel BUG at lib/iomap.c:254!
invalid opcode: 0000 [#1]
...

The screen picture is here:
http://vrfy.org/pci_iounmap.jpg

It's a Thinkpad T43p.

2.6.20 was working fine.

Quoted 8 lines
Quoted 5 lines

Commenting out:
IO_COND(addr, /* nothing */, iounmap(addr));
in:
lib/iomap.c:254
makes at least booting up possible.

I saw a similar one on my X41. Disabling the AHCI driver made the machine booting again.

Hey Marcel,
yeah, that works fine here too.

Thanks,
Kay

#3 lore raw
Quoted 18 lines · Kay Sievers
Quoted 17 lines · Marcel Holtmann
Quoted 16 lines

kernel BUG at lib/iomap.c:254!
invalid opcode: 0000 [#1]
...

The screen picture is here:
http://vrfy.org/pci_iounmap.jpg

It's a Thinkpad T43p.

2.6.20 was working fine.

Commenting out:
IO_COND(addr, /* nothing */, iounmap(addr));
in:
lib/iomap.c:254
makes at least booting up possible.

Quoted 5 lines
Quoted 2 lines

I saw a similar one on my X41. Disabling the AHCI driver made the machine booting again.

Hey Marcel,
yeah, that works fine here too.

Tejun, Jeff, any ideas what's going wrong with the iomap change and the AHCI driver?

Thanks,
Kay

#4 lore raw
Quoted 19 lines · Kay Sievers
Quoted 18 lines · Kay Sievers
Quoted 17 lines · Marcel Holtmann
Quoted 16 lines

kernel BUG at lib/iomap.c:254!
invalid opcode: 0000 [#1]
...

The screen picture is here:
http://vrfy.org/pci_iounmap.jpg

It's a Thinkpad T43p.

2.6.20 was working fine.

Commenting out:
IO_COND(addr, /* nothing */, iounmap(addr));
in:
lib/iomap.c:254
makes at least booting up possible.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

I saw a similar one on my X41. Disabling the AHCI driver made the machine booting again.

Hey Marcel,
yeah, that works fine here too.

Tejun, Jeff, any ideas what's going wrong with the iomap change and the AHCI driver?

The following commit should have fixed the problem, at least the oops. Care to give the current git head a shot?

commit fb4d64e78ceab77cf20f7796f74aa10ebe862032

Author: Frederik Deweerdt <[email protected]>
Date:   Fri Feb 16 01:27:15 2007 -0800

[PATCH] pci_iomap_regions() error handling fix

It appears that the pcim_iomap_regions() function doesn't get the error handling right. It BUGs early at boot with a backtrace along the
lines of:

ahci_init
pci_register_driver
driver_register
[...]
ahci_init_one
pcim_iomap_region
pcim_iounmap

The following patch allows me to boot. Only the if(mask..) continue; part fixes the problem actually, the gotos where changed so that we don't try to unmap something we couldn't map anyway.

Signature · 1 line
tejun
#5 lore raw
Quoted 23 lines · Tejun Heo
Quoted 22 lines · Kay Sievers
Quoted 21 lines · Kay Sievers
Quoted 19 lines · Marcel Holtmann
Quoted 9 lines

kernel BUG at lib/iomap.c:254!
invalid opcode: 0000 [#1]
...

The screen picture is here:
http://vrfy.org/pci_iounmap.jpg

It's a Thinkpad T43p.

Quoted 7 lines

2.6.20 was working fine.

Commenting out:
IO_COND(addr, /* nothing */, iounmap(addr));
in:
lib/iomap.c:254
makes at least booting up possible.

I saw a similar one on my X41. Disabling the AHCI driver made the machine booting again.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

Hey Marcel,
yeah, that works fine here too.

Tejun, Jeff, any ideas what's going wrong with the iomap change and the AHCI driver?

The following commit should have fixed the problem, at least the oops. Care to give the current git head a shot?

Yes, it boots fine now with the AHCI driver compiled in.

Thanks,
Kay