[3.0-git16] Oops at driver_uevent_store().

[3.0-git16] Oops at driver_uevent_store().

[Tetsuo Handa]

I'm getting below oops on CentOS 6.0 (x86_32) since at least 3.0-git11.
3.0 works OK.

Config is at http://I-love.SAKURA.ne.jp/tmp/config-3.0-git16 .

[    5.434284] NET: Registered protocol family 17
[    5.439180] Using IPI No-Shortcut mode
[    5.441357] input: ImPS/2 Generic Wheel Mouse as /devices/platform/i8042/serio1/input/input2
[    5.451427] debug: unmapping init memory c1583000..c173d000
[    5.453428] Write protecting the kernel text: 3848k
[    5.454413] Testing CPA: Reverting c1000000-c13c2000
[    5.456225] Testing CPA: write protecting again
[    5.457817] Write protecting the kernel read-only data: 1488k
[    5.459136] Testing CPA: undo c13c2000-c1536000
[    5.460662] Testing CPA: write protecting again
[    5.474749] dracut: dracut-004-33.2.el6_0
[    5.484396] udev: starting version 147
[    5.486118] ------------[ cut here ]------------
[    5.487813] WARNING: at fs/proc/base.c:1123 oom_adjust_write+0x266/0x280()
[    5.490401] Hardware name: VMware Virtual Platform
[    5.491594] udevd (1193): /proc/1193/oom_adj is deprecated, please use /proc/1193/oom_score_adj instead.
[    5.494400] Modules linked in:
[    5.495771] Pid: 1193, comm: udevd Tainted: G        W   3.0.0-git16 #1
[    5.497469] Call Trace:
[    5.498398]  [<c103a2bd>] warn_slowpath_common+0x6d/0xa0
[    5.499915]  [<c1113046>] ? oom_adjust_write+0x266/0x280
[    5.502065]  [<c1113046>] ? oom_adjust_write+0x266/0x280
[    5.503676]  [<c103a36e>] warn_slowpath_fmt+0x2e/0x30
[    5.504858]  [<c1113046>] oom_adjust_write+0x266/0x280
[    5.507912]  [<c10c6b7a>] vfs_write+0x9a/0x140
[    5.509834]  [<c1112de0>] ? oom_score_adj_write+0x250/0x250
[    5.511514]  [<c10c6ced>] sys_write+0x3d/0x70
[    5.512328]  [<c13c0d0c>] sysenter_do_call+0x12/0x32
[    5.515909] ---[ end trace 93d72a36b9146f24 ]---
[    5.819969] dracut: Starting plymouth daemon
[    5.887203] ata_id used greatest stack depth: 5840 bytes left
[    6.224024] EXT4-fs (sda1): mounted filesystem with ordered data mode. Opts: (null)
[    6.280249] dracut: Mounted root filesystem /dev/sda1
[    6.355938] dracut: Loading SELinux policy
[    6.626868] load_policy used greatest stack depth: 5696 bytes left
[    6.629000] dracut: /sbin/load_policy: Can't load policy: No such device
[    6.692244] dracut: Switching root
[    7.657652] udev: starting version 147
[    7.669087] BUG: unable to handle kernel paging request at c15e2744
[    7.669648] IP: [<c1242f20>] driver_uevent_store+0x20/0x40
[    7.669648] *pde = 01d65067 *pte = 015e2162 
[    7.671783] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC
[    7.673560] Modules linked in:
[    7.673647] 
[    7.673647] Pid: 2874, comm: udevadm Tainted: G        W   3.0.0-git16 #1 VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform
[    7.673648] EIP: 0060:[<c1242f20>] EFLAGS: 00010246 CPU: 1
[    7.675163] EIP is at driver_uevent_store+0x20/0x40
[    7.676075] EAX: 00000000 EBX: 00000003 ECX: ffffffff EDX: f67ce000
[    7.677645] ESI: c15e2710 EDI: 00000003 EBP: f1b13f30 ESP: f1b13f24
[    7.677646]  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
[    7.677646] Process udevadm (pid: 2874, ti=f1b12000 task=f138c3a0 task.ti=f1b12000)
[    7.679454] Stack:
[    7.680206]  00000000 c1242f00 f4a7df70 f1b13f40 c1242c67 c13e24a0 f6662d28 f1b13f64
[    7.682162]  c111ce8c 00000003 c13e24a0 f106bf84 f4a7df70 f411ef20 00000003 bfe2df1e
[    7.686981]  f1b13f8c c10c6b7a f1b13f98 f411ef28 00000020 c111cdf0 f1b13fa0 f411ef20
[    7.689530] Call Trace:
[    7.689816]  [<c1242f00>] ? bus_uevent_store+0x40/0x40
[    7.691158]  [<c1242c67>] drv_attr_store+0x27/0x40
[    7.692418]  [<c111ce8c>] sysfs_write_file+0x9c/0xf0
[    7.695157]  [<c10c6b7a>] vfs_write+0x9a/0x140
[    7.696104]  [<c111cdf0>] ? sysfs_open_file+0x1e0/0x1e0
[    7.697762]  [<c10c6ced>] sys_write+0x3d/0x70
[    7.699155]  [<c13c0d0c>] sysenter_do_call+0x12/0x32
[    7.700128] Code: f8 89 ec 5d c3 8d b6 00 00 00 00 55 89 e5 83 ec 0c 89 5d f8 89 cb 89 75 fc 8d 4d f4 89 c6 89 d0 89 da e8 34 ae f7 ff 85 c0 75 0b <8b> 46 34 8b 55 f4 e8 b5 b2 f7 ff 89 d8 8b 75 fc 8b 5d f8 89 ec 
[    7.711151] EIP: [<c1242f20>] driver_uevent_store+0x20/0x40 SS:ESP 0068:f1b13f24
[    7.713324] CR2: 00000000c15e2744
[    7.714362] ---[ end trace 93d72a36b9146f25 ]---
[    8.165941] pcnet32: pcnet32.c:v1.35 21.Apr.2008 tsbogend@alpha.franken.de
[    8.168320] pcnet32 0000:02:01.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
[    8.170581] pcnet32: PCnet/PCI II 79C970A at 0x2000, 00:0c:29:6c:1a:76 assigned IRQ 19
[    8.175791] pcnet32: eth0: registered as PCnet/PCI II 79C970A
[    8.177491] pcnet32: 1 cards_found
[    8.281700] ip used greatest stack depth: 5588 bytes left
[    9.131734] EXT4-fs (sda1): re-mounted. Opts: (null)
[    9.687123] NET: Registered protocol family 10

Re: [3.0-git16] Oops at driver_uevent_store().

[Arnaud Lacombe]

Hi,

On Thu, Aug 4, 2011 at 3:41 AM, Linus Torvalds
<torvalds@linux-foundation.org> wrote:
> 2011/8/3 Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>:
>> Arnaud Lacombe wrote:
>>> Marking `pci_eisa_driver' as __refdata fix the warning and leave the
>>> struct untouched:
>>>
>>> diff --git a/drivers/eisa/pci_eisa.c b/drivers/eisa/pci_eisa.c
>>> index 30da70d..cdae207 100644
>>> --- a/drivers/eisa/pci_eisa.c
>>> +++ b/drivers/eisa/pci_eisa.c
>>
>> This patch solved the oops. Thank you.
>
> Arnaud, thanks. Sign-off and slightly cleaned-up changelog?
>
Sent in a separate patch; sorry for the breakage.

 - Arnaud

Re: [3.0-git16] Oops at driver_uevent_store().

[Arnaud Lacombe]

Hi,

On Thu, Aug 4, 2011 at 4:40 AM, Adrian Bunk <bunk@kernel.org> wrote:
> On Thu, Aug 04, 2011 at 01:47:11AM -0400, Arnaud Lacombe wrote:
>> Hi,
>>
>> [Added Adrian Bunk to the Cc: list]
>
> Thanks.
>
>> 2011/8/3 Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>:
>>...
>> commit 74b9a297866d0416edd0be5014cb0810de049c6a
>> Author: Adrian Bunk <bunk@stusta.de>
>> Date:   Mon Mar 26 21:32:27 2007 -0800
>>
>>     [PATCH] drivers/eisa/pci_eisa.c:pci_eisa_init() should be init
>>
>>       WARNING: drivers/built-in.o - Section mismatch: reference to
>> .init.text:eisa_root_register from .text between 'pci_eisa_init' (at
>> offset 0xabf670) and 'virtual_eisa_release'
>>
>>     AFAIK a PCI to EISA bridge isn't anything hotpluggable, so
>>     pci_eisa_init() can become __init.
>>
>>     Signed-off-by: Adrian Bunk <bunk@stusta.de>
>>     Cc: Greg KH <greg@kroah.com>
>>     Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
>>     Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
>>
>>
>> To gain a few bytes, 'pci_eisa_init' is freed, but its reference is
>> kept in `pci_eisa_driver' which triggers the warning. The commit log
>> points out that PCI to EISA should not be hotpluggable, so it is not
>> expected to be ever executed again (which would trigger a crash).
>
> This was not for gaining a few bytes, and I'm not exactly seeing why I'm
> to blame here, since my patch is invariant to what your patch does:
>
> Calling pci_eisa_init outside of __init:
> - might have crashed before my patch and
> - after my patch and
> - after your patch
> since eisa_root_register() is __init.
>
>> Marking `pci_eisa_driver' as __refdata fix the warning and leave the
>> struct untouched:
>>
>> diff --git a/drivers/eisa/pci_eisa.c b/drivers/eisa/pci_eisa.c
>> index 30da70d..cdae207 100644
>> --- a/drivers/eisa/pci_eisa.c
>> +++ b/drivers/eisa/pci_eisa.c
>> @@ -45,13 +45,13 @@ static int __init pci_eisa_init(struct pci_dev *pdev,
>>         return 0;
>>  }
>>
>> -static struct pci_device_id __initdata pci_eisa_pci_tbl[] = {
>> +static struct pci_device_id pci_eisa_pci_tbl[] = {
>>         { PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
>>           PCI_CLASS_BRIDGE_EISA << 8, 0xffff00, 0 },
>>         { 0, }
>>  };
>>
>> -static struct pci_driver __initdata pci_eisa_driver = {
>> +static struct pci_driver __refdata pci_eisa_driver = {
>>         .name           = "pci_eisa",
>>         .id_table       = pci_eisa_pci_tbl,
>>         .probe          = pci_eisa_init,
>
> So the commit that broke it was not mine, but
>
> commit 005bdad7b80ac017ca21d795639d4214b9844a84
> Author: Arnaud Lacombe <lacombar@gmail.com>
> Date:   Mon Jul 25 17:13:04 2011 -0700
>
>    eisa/pci_eisa.c: fix section mismatch
>
>    Fixes
>
>      WARNING: vmlinux.o(.data+0x15d3ac): Section mismatch in reference from the
>      The variable pci_eisa_driver references the function __init pci_eisa_init(
>      If the reference is valid then annotate the variable with __init* or __ref
>      *_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console
>
>    Signed-off-by: Arnaud Lacombe <lacombar@gmail.com>
>    Cc: Greg KH <greg@kroah.com>
>    Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
>    Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
>
>
>
> Please don't blame me for bugs you introduced.
>
I was merely pointing out the modpost warning, but yes, I broke the stuff.

 - Arnaud

>
>>  - Arnaud
>
> cu
> Adrian
>
> --
>
>       "Is there not promise of rain?" Ling Tan asked suddenly out
>        of the darkness. There had been need of rain for many days.
>       "Only a promise," Lao Er said.
>                                       Pearl S. Buck - Dragon Seed
>
>

Re: [3.0-git16] Oops at driver_uevent_store().

[Adrian Bunk]

On Thu, Aug 04, 2011 at 01:47:11AM -0400, Arnaud Lacombe wrote:
> Hi,
> 
> [Added Adrian Bunk to the Cc: list]

Thanks.

> 2011/8/3 Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>:
>...
> commit 74b9a297866d0416edd0be5014cb0810de049c6a
> Author: Adrian Bunk <bunk@stusta.de>
> Date:   Mon Mar 26 21:32:27 2007 -0800
> 
>     [PATCH] drivers/eisa/pci_eisa.c:pci_eisa_init() should be init
> 
>       WARNING: drivers/built-in.o - Section mismatch: reference to
> .init.text:eisa_root_register from .text between 'pci_eisa_init' (at
> offset 0xabf670) and 'virtual_eisa_release'
> 
>     AFAIK a PCI to EISA bridge isn't anything hotpluggable, so
>     pci_eisa_init() can become __init.
> 
>     Signed-off-by: Adrian Bunk <bunk@stusta.de>
>     Cc: Greg KH <greg@kroah.com>
>     Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
>     Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
> 
> 
> To gain a few bytes, 'pci_eisa_init' is freed, but its reference is
> kept in `pci_eisa_driver' which triggers the warning. The commit log
> points out that PCI to EISA should not be hotpluggable, so it is not
> expected to be ever executed again (which would trigger a crash).

This was not for gaining a few bytes, and I'm not exactly seeing why I'm 
to blame here, since my patch is invariant to what your patch does:

Calling pci_eisa_init outside of __init:
- might have crashed before my patch and
- after my patch and
- after your patch
since eisa_root_register() is __init.

> Marking `pci_eisa_driver' as __refdata fix the warning and leave the
> struct untouched:
> 
> diff --git a/drivers/eisa/pci_eisa.c b/drivers/eisa/pci_eisa.c
> index 30da70d..cdae207 100644
> --- a/drivers/eisa/pci_eisa.c
> +++ b/drivers/eisa/pci_eisa.c
> @@ -45,13 +45,13 @@ static int __init pci_eisa_init(struct pci_dev *pdev,
>         return 0;
>  }
> 
> -static struct pci_device_id __initdata pci_eisa_pci_tbl[] = {
> +static struct pci_device_id pci_eisa_pci_tbl[] = {
>         { PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
>           PCI_CLASS_BRIDGE_EISA << 8, 0xffff00, 0 },
>         { 0, }
>  };
> 
> -static struct pci_driver __initdata pci_eisa_driver = {
> +static struct pci_driver __refdata pci_eisa_driver = {
>         .name           = "pci_eisa",
>         .id_table       = pci_eisa_pci_tbl,
>         .probe          = pci_eisa_init,

So the commit that broke it was not mine, but

commit 005bdad7b80ac017ca21d795639d4214b9844a84
Author: Arnaud Lacombe <lacombar@gmail.com>
Date:   Mon Jul 25 17:13:04 2011 -0700

    eisa/pci_eisa.c: fix section mismatch
    
    Fixes
    
      WARNING: vmlinux.o(.data+0x15d3ac): Section mismatch in reference from the
      The variable pci_eisa_driver references the function __init pci_eisa_init(
      If the reference is valid then annotate the variable with __init* or __ref
      *_template, *_timer, *_sht, *_ops, *_probe, *_probe_one, *_console
    
    Signed-off-by: Arnaud Lacombe <lacombar@gmail.com>
    Cc: Greg KH <greg@kroah.com>
    Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
    Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>



Please don't blame me for bugs you introduced.


>  - Arnaud

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: [3.0-git16] Oops at driver_uevent_store().

[Linus Torvalds]

2011/8/3 Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>:
> Arnaud Lacombe wrote:
>> Marking `pci_eisa_driver' as __refdata fix the warning and leave the
>> struct untouched:
>>
>> diff --git a/drivers/eisa/pci_eisa.c b/drivers/eisa/pci_eisa.c
>> index 30da70d..cdae207 100644
>> --- a/drivers/eisa/pci_eisa.c
>> +++ b/drivers/eisa/pci_eisa.c
>
> This patch solved the oops. Thank you.

Arnaud, thanks. Sign-off and slightly cleaned-up changelog?

          Linus

Re: [3.0-git16] Oops at driver_uevent_store().

[Tetsuo Handa]

Arnaud Lacombe wrote:
> Marking `pci_eisa_driver' as __refdata fix the warning and leave the
> struct untouched:
> 
> diff --git a/drivers/eisa/pci_eisa.c b/drivers/eisa/pci_eisa.c
> index 30da70d..cdae207 100644
> --- a/drivers/eisa/pci_eisa.c
> +++ b/drivers/eisa/pci_eisa.c

This patch solved the oops. Thank you.

Re: [3.0-git16] Oops at driver_uevent_store().

[Arnaud Lacombe]

Hi,

[Added Adrian Bunk to the Cc: list]

2011/8/3 Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>:
> Arnaud Lacombe wrote:
>> Looking more closely, I'd say I have been too eager in the marking.
>> The warning concern `pci_eisa_driver', not `pci_eisa_pci_tbl' which is
>> also referenced by MODULE_DEVICE_TABLE().
>>
>> Does removing the annotation from `pci_eisa_pci_tbl' fix the BUG ?
> No.
>
ok, I had to enable DEBUG_PAGEALLOC to trigger the crash.

> -static struct pci_device_id __initdata pci_eisa_pci_tbl[] = {
> +static struct pci_device_id pci_eisa_pci_tbl[] = {
>
> did not help.
>
> Printing dmesg in verbose showed below three lines.
>
> [    0.262828] EISA bus registered
> [    6.291117] EISA: Probing bus 0 at eisa.0
> [    6.295445] EISA: Cannot allocate resource for mainboard
>
> Manual printk() tracing showed below callchain.
>
> virtual_eisa_root_init() calls eisa_root_register().
>  eisa_root_register() calls eisa_probe().
>    eisa_probe() returns -EBUSY since root->force_probe == 0.
>  eisa_root_register() calls release_resource() and returns -EBUSY.
> virtual_eisa_root_init() calls platform_device_unregister() and returns -1.
>
> Also, eisa_root_register() in pci_eisa_init() is not called.
>
I'd guess you just have no PCI-EISA bridge in your VM. The trace you
point out should only indicate that you have no EISA device either, as
the comment a bit before point out:

        /* First try to get hold of slot 0. If there is no device
         * here, simply fail, unless root->force_probe is set. */


> So, it seems that something is wrong with cleanup.
> pci_unregister_driver() is missing?
>
The crash happens when sysfs tries to access the device_driver inside
'pci_eisa_driver'. I'd argue that the root of the problem is commit
74b9a297866d which created the inconsistency at the first place:

commit 74b9a297866d0416edd0be5014cb0810de049c6a
Author: Adrian Bunk <bunk@stusta.de>
Date:   Mon Mar 26 21:32:27 2007 -0800

    [PATCH] drivers/eisa/pci_eisa.c:pci_eisa_init() should be init

      WARNING: drivers/built-in.o - Section mismatch: reference to
.init.text:eisa_root_register from .text between 'pci_eisa_init' (at
offset 0xabf670) and 'virtual_eisa_release'

    AFAIK a PCI to EISA bridge isn't anything hotpluggable, so
    pci_eisa_init() can become __init.

    Signed-off-by: Adrian Bunk <bunk@stusta.de>
    Cc: Greg KH <greg@kroah.com>
    Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
    Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>


To gain a few bytes, 'pci_eisa_init' is freed, but its reference is
kept in `pci_eisa_driver' which triggers the warning. The commit log
points out that PCI to EISA should not be hotpluggable, so it is not
expected to be ever executed again (which would trigger a crash).

Marking `pci_eisa_driver' as __refdata fix the warning and leave the
struct untouched:

diff --git a/drivers/eisa/pci_eisa.c b/drivers/eisa/pci_eisa.c
index 30da70d..cdae207 100644
--- a/drivers/eisa/pci_eisa.c
+++ b/drivers/eisa/pci_eisa.c
@@ -45,13 +45,13 @@ static int __init pci_eisa_init(struct pci_dev *pdev,
        return 0;
 }

-static struct pci_device_id __initdata pci_eisa_pci_tbl[] = {
+static struct pci_device_id pci_eisa_pci_tbl[] = {
        { PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
          PCI_CLASS_BRIDGE_EISA << 8, 0xffff00, 0 },
        { 0, }
 };

-static struct pci_driver __initdata pci_eisa_driver = {
+static struct pci_driver __refdata pci_eisa_driver = {
        .name           = "pci_eisa",
        .id_table       = pci_eisa_pci_tbl,
        .probe          = pci_eisa_init,


 - Arnaud

Re: [3.0-git16] Oops at driver_uevent_store().

[Tetsuo Handa]

Arnaud Lacombe wrote:
> Looking more closely, I'd say I have been too eager in the marking.
> The warning concern `pci_eisa_driver', not `pci_eisa_pci_tbl' which is
> also referenced by MODULE_DEVICE_TABLE().
> 
> Does removing the annotation from `pci_eisa_pci_tbl' fix the BUG ?
No.

-static struct pci_device_id __initdata pci_eisa_pci_tbl[] = {
+static struct pci_device_id pci_eisa_pci_tbl[] = {

did not help.

Printing dmesg in verbose showed below three lines.

[    0.262828] EISA bus registered
[    6.291117] EISA: Probing bus 0 at eisa.0
[    6.295445] EISA: Cannot allocate resource for mainboard

Manual printk() tracing showed below callchain.

virtual_eisa_root_init() calls eisa_root_register().
  eisa_root_register() calls eisa_probe().
    eisa_probe() returns -EBUSY since root->force_probe == 0.
  eisa_root_register() calls release_resource() and returns -EBUSY.
virtual_eisa_root_init() calls platform_device_unregister() and returns -1.

Also, eisa_root_register() in pci_eisa_init() is not called.

So, it seems that something is wrong with cleanup.
pci_unregister_driver() is missing?

Re: [3.0-git16] Oops at driver_uevent_store().

[Arnaud Lacombe]

2011/8/3 Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>:
> Commit 005bdad7 "eisa/pci_eisa.c: fix section mismatch" triggers oops.
> Reverting it solved the oops. Is this rather a sysfs bug?
>
> [    7.460292] dracut: Mounted root filesystem /dev/sda1
> [    7.603017] dracut: Loading SELinux policy
> [    7.976804] dracut: /sbin/load_policy: Can't load policy: No such device
> [    8.095959] dracut: Switching root
> [    9.452334] udev: starting version 147
> [    9.456695] BUG: unable to handle kernel paging request at c15dd684
> [    9.457860] IP: [<c12408a0>] driver_uevent_store+0x20/0x40
> [    9.457860] *pde = 01d60067 *pte = 015dd162
> [    9.457860] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC
> [    9.457860] Modules linked in:
> [    9.457860]
> [    9.457860] Pid: 2914, comm: udevadm Tainted: G        W   3.0.0-03636-g005bdad #12 VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform
> [    9.457860] EIP: 0060:[<c12408a0>] EFLAGS: 00010246 CPU: 1
> [    9.457861] EIP is at driver_uevent_store+0x20/0x40
> [    9.457861] EAX: 00000000 EBX: 00000003 ECX: ffffffff EDX: f4422000
> [    9.457861] ESI: c15dd650 EDI: 00000003 EBP: f4b63f30 ESP: f4b63f24
> [    9.457861]  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
> [    9.459164] Process udevadm (pid: 2914, ti=f4b62000 task=f49f87e0 task.ti=f4b62000)
> [    9.463514] Stack:
> [    9.464278]  00000000 c1240880 f4d61f70 f4b63f40 c12405e7 c13e0240 f5c1bea8 f4b63f64
> [    9.465858]  c111cf6c 00000003 c13e0240 f4f82f84 f4d61f70 f18cdf20 00000003 bfe20f1e
> [    9.465858]  f4b63f8c c10c6bba f4b63f98 f18cdf28 00000020 c111ced0 f4b63fa0 f18cdf20
> [    9.467123] Call Trace:
> [    9.468030]  [<c1240880>] ? bus_uevent_store+0x40/0x40
> [    9.469764]  [<c12405e7>] drv_attr_store+0x27/0x40
> [    9.469856]  [<c111cf6c>] sysfs_write_file+0x9c/0xf0
> [    9.469856]  [<c10c6bba>] vfs_write+0x9a/0x140
> [    9.469857]  [<c111ced0>] ? sysfs_open_file+0x1e0/0x1e0
> [    9.469857]  [<c10c6d2d>] sys_write+0x3d/0x70
> [    9.469857]  [<c13be60c>] sysenter_do_call+0x12/0x32
> [    9.469857] Code: f8 89 ec 5d c3 8d b6 00 00 00 00 55 89 e5 83 ec 0c 89 5d f8 89 cb 89 75 fc 8d 4d f4 89 c6 89 d0 89 da e8 34 b5 f7 ff 85 c0 75 0b <8b> 46 34 8b 55 f4 e8 b5 b9 f7 ff 89 d8 8b 75 fc 8b 5d f8 89 ec
> [    9.469964] EIP: [<c12408a0>] driver_uevent_store+0x20/0x40 SS:ESP 0068:f4b63f24
> [    9.471828] CR2: 00000000c15dd684
> [    9.472987] ---[ end trace 93d72a36b9146f25 ]---
> [    9.959305] pcnet32: pcnet32.c:v1.35 21.Apr.2008 tsbogend@alpha.franken.de
> [    9.961686] pcnet32 0000:02:01.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
> [    9.963766] pcnet32: PCnet/PCI II 79C970A at 0x2000, 00:0c:29:6c:1a:76 assigned IRQ 19
>
> Config is at http://I-love.SAKURA.ne.jp/tmp/config-3.0-git16 .
>
Looking more closely, I'd say I have been too eager in the marking.
The warning concern `pci_eisa_driver', not `pci_eisa_pci_tbl' which is
also referenced by MODULE_DEVICE_TABLE().

Does removing the annotation from `pci_eisa_pci_tbl' fix the BUG ?

Thanks,
 - Arnaud

> Regards.
>

Re: [3.0-git16] Oops at driver_uevent_store().

[Tetsuo Handa]

Commit 005bdad7 "eisa/pci_eisa.c: fix section mismatch" triggers oops.
Reverting it solved the oops. Is this rather a sysfs bug?

[    7.460292] dracut: Mounted root filesystem /dev/sda1
[    7.603017] dracut: Loading SELinux policy
[    7.976804] dracut: /sbin/load_policy: Can't load policy: No such device
[    8.095959] dracut: Switching root
[    9.452334] udev: starting version 147
[    9.456695] BUG: unable to handle kernel paging request at c15dd684
[    9.457860] IP: [<c12408a0>] driver_uevent_store+0x20/0x40
[    9.457860] *pde = 01d60067 *pte = 015dd162 
[    9.457860] Oops: 0000 [#1] SMP DEBUG_PAGEALLOC
[    9.457860] Modules linked in:
[    9.457860] 
[    9.457860] Pid: 2914, comm: udevadm Tainted: G        W   3.0.0-03636-g005bdad #12 VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform
[    9.457860] EIP: 0060:[<c12408a0>] EFLAGS: 00010246 CPU: 1
[    9.457861] EIP is at driver_uevent_store+0x20/0x40
[    9.457861] EAX: 00000000 EBX: 00000003 ECX: ffffffff EDX: f4422000
[    9.457861] ESI: c15dd650 EDI: 00000003 EBP: f4b63f30 ESP: f4b63f24
[    9.457861]  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
[    9.459164] Process udevadm (pid: 2914, ti=f4b62000 task=f49f87e0 task.ti=f4b62000)
[    9.463514] Stack:
[    9.464278]  00000000 c1240880 f4d61f70 f4b63f40 c12405e7 c13e0240 f5c1bea8 f4b63f64
[    9.465858]  c111cf6c 00000003 c13e0240 f4f82f84 f4d61f70 f18cdf20 00000003 bfe20f1e
[    9.465858]  f4b63f8c c10c6bba f4b63f98 f18cdf28 00000020 c111ced0 f4b63fa0 f18cdf20
[    9.467123] Call Trace:
[    9.468030]  [<c1240880>] ? bus_uevent_store+0x40/0x40
[    9.469764]  [<c12405e7>] drv_attr_store+0x27/0x40
[    9.469856]  [<c111cf6c>] sysfs_write_file+0x9c/0xf0
[    9.469856]  [<c10c6bba>] vfs_write+0x9a/0x140
[    9.469857]  [<c111ced0>] ? sysfs_open_file+0x1e0/0x1e0
[    9.469857]  [<c10c6d2d>] sys_write+0x3d/0x70
[    9.469857]  [<c13be60c>] sysenter_do_call+0x12/0x32
[    9.469857] Code: f8 89 ec 5d c3 8d b6 00 00 00 00 55 89 e5 83 ec 0c 89 5d f8 89 cb 89 75 fc 8d 4d f4 89 c6 89 d0 89 da e8 34 b5 f7 ff 85 c0 75 0b <8b> 46 34 8b 55 f4 e8 b5 b9 f7 ff 89 d8 8b 75 fc 8b 5d f8 89 ec 
[    9.469964] EIP: [<c12408a0>] driver_uevent_store+0x20/0x40 SS:ESP 0068:f4b63f24
[    9.471828] CR2: 00000000c15dd684
[    9.472987] ---[ end trace 93d72a36b9146f25 ]---
[    9.959305] pcnet32: pcnet32.c:v1.35 21.Apr.2008 tsbogend@alpha.franken.de
[    9.961686] pcnet32 0000:02:01.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19
[    9.963766] pcnet32: PCnet/PCI II 79C970A at 0x2000, 00:0c:29:6c:1a:76 assigned IRQ 19

Config is at http://I-love.SAKURA.ne.jp/tmp/config-3.0-git16 .

Regards.