Re: [Bugme-new] [Bug 10586] New: High Number of Segfaults onMacBookStartingWith 2.6.24.5 Kernel

From: Ryan Roth
Date: Thu May 01 2008 - 19:57:51 EST

Randy Dunlap wrote:
On Thu, 01 May 2008 16:29:58 -0700 Ryan Roth wrote:

-----Original Message-----
From: Andrew Morton [mailto:akpm@xxxxxxxxxxxxxxxxxxxx]
Sent: Thu 5/1/2008 1:30 PM
To: Roth, Ryan/RDD
Cc: bugme-daemon@xxxxxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx
Subject: Re: [Bugme-new] [Bug 10586] New: High Number of Segfaults onMacBookStarting With 2.6.24.5 Kernel

On Thu, 1 May 2008 14:09:00 -0600
<Ryan.Roth@xxxxxxxx> wrote:
-----Original Message-----
From: Andrew Morton [mailto:akpm@xxxxxxxxxxxxxxxxxxxx]
Sent: Thursday, May 01, 2008 1:05 PM
To: Roth, Ryan/RDD
Cc: bugme-daemon@xxxxxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx
Subject: Re: [Bugme-new] [Bug 10586] New: High Number of Segfaults
onMacBook Starting With 2.6.24.5 Kernel

(switched to email. Please respond via emailed reply-to-all, not via
the bugzilla web interface).

On Thu, 1 May 2008 12:30:18 -0700 (PDT)
bugme-daemon@xxxxxxxxxxxxxxxxxxx wrote:

http://bugzilla.kernel.org/show_bug.cgi?id=10586

Summary: High Number of Segfaults on MacBook Starting
With
2.6.24.5 Kernel
Product: Memory Management
Version: 2.5
KernelVersion: 2.6.24.5
Platform: All
OS/Version: Linux
Tree: Mainline
Status: NEW
Severity: normal
Priority: P1
Component: Other
AssignedTo: akpm@xxxxxxxx
ReportedBy: ryan.roth@xxxxxxxx


Latest working kernel version: 2.6.23?
Earliest failing kernel version: 2.6.24.5
Distribution: Fedora 8 x86_64
Hardware Environment: MacBook
Problem Description: Extremely high number of segfaults. Suspect
that this may have to do with some sort of change in memory
remapping
changes in the kernel.
This report is a bit mysterious. I assume you're referring to x86's

"%s%s[%d]: segfault at %lx ip %08lx sp %08lx error %lx"

printk in the kernel logs?

Can you please send some examples of the log output?

Is there actually any observeable problem apart from the logs? Are
applications dying where they previously did not?

Yes there is a issue with a lot of apps dying, especially the
gnome-power-* apps which means my power management fails to work
frequently. I'm on Git-10 currently and I'm going to patch it up to
date right now because it looks like there have been some memory remap
patches lately. I will update the list after I do so.

OK, worried. This is a huge bug. But afaik you're the only person who has
hit it, so there must be something special in your setup which is
triggering it. This could be hard.

OK here is what I get with the git16 kernel. On boot I get this error:

PCI: Not using MMCONFIG
PCI: BIOS Bug: MCFG area at f0000000 is not reserved in ACPI motherboard resources
PCI: Not using MMCONFIG

Attached is a complete log after this boot

Where are the segfaults?

Do you mean these reports? If so, then this is something
other than segfaults. And not quite as critical.
Or are things not working for you with this kernel?


May 1 16:15:01 kernel: proc_dir_entry 'info' already registered
May 1 16:15:01 kernel: Pid: 1099, comm: modprobe Not tainted 2.6.25-git16 #3
May 1 16:15:01 kernel: May 1 16:15:01 kernel: Call Trace:
May 1 16:15:01 kernel: [<ffffffff81118269>] idr_get_new+0xb/0x28
May 1 16:15:01 kernel: [<ffffffff810d5e92>] proc_register+0xcb/0x14b
May 1 16:15:01 kernel: [<ffffffff810d5ec5>] proc_register+0xfe/0x14b
May 1 16:15:01 kernel: [<ffffffff810d5f96>] proc_create_data+0x84/0x9e
May 1 16:15:01 kernel: [<ffffffffa013dd37>]
:video:acpi_video_bus_add+0x98f/0xc6d
May 1 16:15:01 kernel: [<ffffffff8115d4d9>] acpi_device_probe+0x43/0x91
May 1 16:15:01 kernel: [<ffffffff81192564>] driver_probe_device+0xd0/0x14d
May 1 16:15:01 kernel: [<ffffffff81192627>] __driver_attach+0x46/0x6d
May 1 16:15:01 kernel: [<ffffffff811925e1>] __driver_attach+0x0/0x6d
May 1 16:15:01 kernel: [<ffffffff81191f09>] bus_for_each_dev+0x44/0x6f
May 1 16:15:01 kernel: [<ffffffff8119186e>] bus_add_driver+0xb2/0x1ff
May 1 16:15:01 kernel: [<ffffffff81192846>] driver_register+0x59/0xcd
May 1 16:15:01 kernel: [<ffffffffa00d803c>] :video:acpi_video_init+0x3c/0x5e
May 1 16:15:01 kernel: [<ffffffff81052db2>] sys_init_module+0x171a/0x1896
May 1 16:15:01 kernel: [<ffffffff8115d812>] acpi_bus_register_driver+0x0/0x3e
May 1 16:15:01 kernel: [<ffffffff81098831>] vfs_read+0xa6/0xfe
May 1 16:15:01 kernel: [<ffffffff8100bf3b>]
system_call_after_swapgs+0x7b/0x80
---
~Randy

The segfaults are still present causing apps to not work. I attached a boot log in case it was needed for more details. Should I not worry about the following message?

May 1 16:15:01 kernel: PCI: MCFG configuration 0: base f0000000 segment 0 buses 0 - 255
May 1 16:15:01 kernel: PCI: BIOS Bug: MCFG area at f0000000 is not reserved in ACPI motherboard resources
May 1 16:15:01 kernel: PCI: Not using MMCONFIG.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/