acpi-20040715+D600+Suspend

acpi-20040715+D600+Suspend

[Ow Mun Heng]

Hi guys,
	
	Lo and Behold, I managed to get S3 to work on my laptop by following
instructions on this website.

http://www.loria.fr/~thome/d600/

However, I've not tested it further(as in using the normal kernel boot
up line in grub.conf) to verify if it is the same whether or not I use
video=radeonfb.

I'm getting these errors during Suspend to disk.

Warning: CPU frequency out of sync: cpufreq and timingcore thinks of 600000, is 1400000 kHz.
Debug: sleeping function called from invalid context at mm/slab.c:1989
in_atomic():1, irqs_disabled():0
 [<c01156ac>] __might_sleep+0xa5/0xaf
 [<c0135d3f>] __kmalloc+0x40/0x76
 [<c023bec9>] acpi_os_allocate+0xa/0xb
 [<c024fa1f>] acpi_ut_allocate+0x34/0x57
 [<c024f9b0>] acpi_ut_initialize_buffer+0x42/0x7d
 [<c024c85e>] acpi_rs_create_byte_stream+0x23/0x39
 [<c024dc1e>] acpi_rs_set_srs_method_data+0x1b/0x9d
 [<c011ecd4>] __group_send_sig_info+0x36/0x9d
 [<c011ee0a>] group_send_sig_info+0x58/0x61
 [<c02556fd>] acpi_pci_link_set+0xde/0x155
 [<c0255b0f>] irqrouter_resume+0x1c/0x30
 [<c027bfff>] sysdev_resume+0x3e/0xc7
 [<c027ea69>] device_power_up+0x5/0xa
 [<c012b332>] do_magic_resume_2+0x63/0xb9
 [<c02db0d9>] copy_done+0x49/0x4c
 [<c012b536>] software_suspend+0x98/0xae
 [<c0252322>] acpi_system_write_sleep+0x6a/0x85
 [<c014650c>] vfs_write+0xb8/0xe4
 [<c01465d6>] sys_write+0x3c/0x62
 [<c01059fd>] sysenter_past_esp+0x52/0x71
ACPI: PCI interrupt 0000:00:1d.0[A] -> GSI 11 (level, low) -> IRQ 11
PCI: Setting latency timer of device 0000:00:1d.0 to 64
ACPI: PCI interrupt 0000:00:1d.1[B] -> GSI 11 (level, low) -> IRQ 11
PCI: Setting latency timer of device 0000:00:1d.1 to 64
ACPI: PCI interrupt 0000:00:1d.2[C] -> GSI 11 (level, low) -> IRQ 11
PCI: Setting latency timer of device 0000:00:1d.2 to 64
ACPI: PCI interrupt 0000:00:1d.7[D] -> GSI 11 (level, low) -> IRQ 11
PCI: Setting latency timer of device 0000:00:1d.7 to 64
ACPI: PCI interrupt 0000:00:1f.1[A] -> GSI 11 (level, low) -> IRQ 11
ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 7 (level, low) -> IRQ 7
PCI: Setting latency timer of device 0000:00:1f.5 to 64
radeonfb: resumed !
ACPI: PCI interrupt 0000:02:00.0[A] -> GSI 11 (level, low) -> IRQ 11
Fixing swap signatures... ok


I also noticed that now, suspend does not work as it should/or was when
I was using 2.6.7 kernel. 

Right now, it will go to the Freeing memory section.. then the screen
goes blank. (blacklight still on) and then stays that way for quite
sometime before I see the words "radeonfb: resumed" and when that
happens, it's as though time stood still. It won't start writing to the
disk until I press a key. (I usually just press the spacebar) and then
it will start writing and works OK from that. (I am writing on my laptop
which is been suspended(hibernated) and resumed a few times already.

Anything changed or broke??

-- 
Ow Mun Heng
Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel 2.6.8.1.Aug.29 
Neuromancer 01:06:38 up 1 day, 3:25, 1 user, load average: 0.73, 0.88,
0.93 


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

RE: acpi-20040715+D600+Suspend

[Li, Shaohua]

Thanks you report this. Looks like the Link device resume issue. Link
device is a sysdev, so its resume is in irq disabled. But ACPI method
executed by Link device resume can't be in irq disabled. Most code in
ACPI CA is designed to execute not in IRQ disabled. Looks like we
ignored this.

Thanks,
Shaohua

>-----Original Message-----
>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:acpi-devel-
>admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Ow Mun Heng
>Sent: Wednesday, September 01, 2004 9:17 AM
>To: Acpi-List
>Subject: [ACPI] acpi-20040715+D600+Suspend
>
>Hi guys,
>
>	Lo and Behold, I managed to get S3 to work on my laptop by
following
>instructions on this website.
>
>http://www.loria.fr/~thome/d600/
>
>However, I've not tested it further(as in using the normal kernel boot
>up line in grub.conf) to verify if it is the same whether or not I use
>video=radeonfb.
>
>I'm getting these errors during Suspend to disk.
>
>Warning: CPU frequency out of sync: cpufreq and timingcore thinks of
600000,
>is 1400000 kHz.
>Debug: sleeping function called from invalid context at mm/slab.c:1989
>in_atomic():1, irqs_disabled():0
> [<c01156ac>] __might_sleep+0xa5/0xaf
> [<c0135d3f>] __kmalloc+0x40/0x76
> [<c023bec9>] acpi_os_allocate+0xa/0xb
> [<c024fa1f>] acpi_ut_allocate+0x34/0x57
> [<c024f9b0>] acpi_ut_initialize_buffer+0x42/0x7d
> [<c024c85e>] acpi_rs_create_byte_stream+0x23/0x39
> [<c024dc1e>] acpi_rs_set_srs_method_data+0x1b/0x9d
> [<c011ecd4>] __group_send_sig_info+0x36/0x9d
> [<c011ee0a>] group_send_sig_info+0x58/0x61
> [<c02556fd>] acpi_pci_link_set+0xde/0x155
> [<c0255b0f>] irqrouter_resume+0x1c/0x30
> [<c027bfff>] sysdev_resume+0x3e/0xc7
> [<c027ea69>] device_power_up+0x5/0xa
> [<c012b332>] do_magic_resume_2+0x63/0xb9
> [<c02db0d9>] copy_done+0x49/0x4c
> [<c012b536>] software_suspend+0x98/0xae
> [<c0252322>] acpi_system_write_sleep+0x6a/0x85
> [<c014650c>] vfs_write+0xb8/0xe4
> [<c01465d6>] sys_write+0x3c/0x62
> [<c01059fd>] sysenter_past_esp+0x52/0x71
>ACPI: PCI interrupt 0000:00:1d.0[A] -> GSI 11 (level, low) -> IRQ 11
>PCI: Setting latency timer of device 0000:00:1d.0 to 64
>ACPI: PCI interrupt 0000:00:1d.1[B] -> GSI 11 (level, low) -> IRQ 11
>PCI: Setting latency timer of device 0000:00:1d.1 to 64
>ACPI: PCI interrupt 0000:00:1d.2[C] -> GSI 11 (level, low) -> IRQ 11
>PCI: Setting latency timer of device 0000:00:1d.2 to 64
>ACPI: PCI interrupt 0000:00:1d.7[D] -> GSI 11 (level, low) -> IRQ 11
>PCI: Setting latency timer of device 0000:00:1d.7 to 64
>ACPI: PCI interrupt 0000:00:1f.1[A] -> GSI 11 (level, low) -> IRQ 11
>ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 7 (level, low) -> IRQ 7
>PCI: Setting latency timer of device 0000:00:1f.5 to 64
>radeonfb: resumed !
>ACPI: PCI interrupt 0000:02:00.0[A] -> GSI 11 (level, low) -> IRQ 11
>Fixing swap signatures... ok
>
>
>I also noticed that now, suspend does not work as it should/or was when
>I was using 2.6.7 kernel.
>
>Right now, it will go to the Freeing memory section.. then the screen
>goes blank. (blacklight still on) and then stays that way for quite
>sometime before I see the words "radeonfb: resumed" and when that
>happens, it's as though time stood still. It won't start writing to the
>disk until I press a key. (I usually just press the spacebar) and then
>it will start writing and works OK from that. (I am writing on my
laptop
>which is been suspended(hibernated) and resumed a few times already.
>
>Anything changed or broke??
>
>--
>Ow Mun Heng
>Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel 2.6.8.1.Aug.29
>Neuromancer 01:06:38 up 1 day, 3:25, 1 user, load average: 0.73, 0.88,
>0.93
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by BEA Weblogic Workshop
>FREE Java Enterprise J2EE developer tools!
>Get your free copy of BEA WebLogic Workshop 8.1 today.
>http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
>_______________________________________________
>Acpi-devel mailing list
>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>https://lists.sourceforge.net/lists/listinfo/acpi-devel


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_idP47&alloc_id\x10808&op=click

Re: acpi-20040715+D600+Suspend

[Nathan Bryant]

Oops, that's my bad. Thanks for the quick explanation, Shaohua, looks 
like we have to move the resume to a special case that runs before 
sysdevs. If that's ok with everyone, I'll try to make a patch.

Nathan

Li, Shaohua wrote:

>Thanks you report this. Looks like the Link device resume issue. Link
>device is a sysdev, so its resume is in irq disabled. But ACPI method
>executed by Link device resume can't be in irq disabled. Most code in
>ACPI CA is designed to execute not in IRQ disabled. Looks like we
>ignored this.
>
>Thanks,
>Shaohua
>
>  
>
>>-----Original Message-----
>>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:acpi-devel-
>>admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Ow Mun Heng
>>Sent: Wednesday, September 01, 2004 9:17 AM
>>To: Acpi-List
>>Subject: [ACPI] acpi-20040715+D600+Suspend
>>
>>Hi guys,
>>
>>	Lo and Behold, I managed to get S3 to work on my laptop by
>>    
>>
>following
>  
>
>>instructions on this website.
>>
>>http://www.loria.fr/~thome/d600/
>>
>>However, I've not tested it further(as in using the normal kernel boot
>>up line in grub.conf) to verify if it is the same whether or not I use
>>video=radeonfb.
>>
>>I'm getting these errors during Suspend to disk.
>>
>>Warning: CPU frequency out of sync: cpufreq and timingcore thinks of
>>    
>>
>600000,
>  
>
>>is 1400000 kHz.
>>Debug: sleeping function called from invalid context at mm/slab.c:1989
>>in_atomic():1, irqs_disabled():0
>>[<c01156ac>] __might_sleep+0xa5/0xaf
>>[<c0135d3f>] __kmalloc+0x40/0x76
>>[<c023bec9>] acpi_os_allocate+0xa/0xb
>>[<c024fa1f>] acpi_ut_allocate+0x34/0x57
>>[<c024f9b0>] acpi_ut_initialize_buffer+0x42/0x7d
>>[<c024c85e>] acpi_rs_create_byte_stream+0x23/0x39
>>[<c024dc1e>] acpi_rs_set_srs_method_data+0x1b/0x9d
>>[<c011ecd4>] __group_send_sig_info+0x36/0x9d
>>[<c011ee0a>] group_send_sig_info+0x58/0x61
>>[<c02556fd>] acpi_pci_link_set+0xde/0x155
>>[<c0255b0f>] irqrouter_resume+0x1c/0x30
>>[<c027bfff>] sysdev_resume+0x3e/0xc7
>>[<c027ea69>] device_power_up+0x5/0xa
>>[<c012b332>] do_magic_resume_2+0x63/0xb9
>>[<c02db0d9>] copy_done+0x49/0x4c
>>[<c012b536>] software_suspend+0x98/0xae
>>[<c0252322>] acpi_system_write_sleep+0x6a/0x85
>>[<c014650c>] vfs_write+0xb8/0xe4
>>[<c01465d6>] sys_write+0x3c/0x62
>>[<c01059fd>] sysenter_past_esp+0x52/0x71
>>ACPI: PCI interrupt 0000:00:1d.0[A] -> GSI 11 (level, low) -> IRQ 11
>>PCI: Setting latency timer of device 0000:00:1d.0 to 64
>>ACPI: PCI interrupt 0000:00:1d.1[B] -> GSI 11 (level, low) -> IRQ 11
>>PCI: Setting latency timer of device 0000:00:1d.1 to 64
>>ACPI: PCI interrupt 0000:00:1d.2[C] -> GSI 11 (level, low) -> IRQ 11
>>PCI: Setting latency timer of device 0000:00:1d.2 to 64
>>ACPI: PCI interrupt 0000:00:1d.7[D] -> GSI 11 (level, low) -> IRQ 11
>>PCI: Setting latency timer of device 0000:00:1d.7 to 64
>>ACPI: PCI interrupt 0000:00:1f.1[A] -> GSI 11 (level, low) -> IRQ 11
>>ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 7 (level, low) -> IRQ 7
>>PCI: Setting latency timer of device 0000:00:1f.5 to 64
>>radeonfb: resumed !
>>ACPI: PCI interrupt 0000:02:00.0[A] -> GSI 11 (level, low) -> IRQ 11
>>Fixing swap signatures... ok
>>
>>
>>I also noticed that now, suspend does not work as it should/or was when
>>I was using 2.6.7 kernel.
>>
>>Right now, it will go to the Freeing memory section.. then the screen
>>goes blank. (blacklight still on) and then stays that way for quite
>>sometime before I see the words "radeonfb: resumed" and when that
>>happens, it's as though time stood still. It won't start writing to the
>>disk until I press a key. (I usually just press the spacebar) and then
>>it will start writing and works OK from that. (I am writing on my
>>    
>>
>laptop
>  
>
>>which is been suspended(hibernated) and resumed a few times already.
>>
>>Anything changed or broke??
>>
>>--
>>Ow Mun Heng
>>Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel 2.6.8.1.Aug.29
>>Neuromancer 01:06:38 up 1 day, 3:25, 1 user, load average: 0.73, 0.88,
>>0.93
>>
>>
>>-------------------------------------------------------
>>This SF.Net email is sponsored by BEA Weblogic Workshop
>>FREE Java Enterprise J2EE developer tools!
>>Get your free copy of BEA WebLogic Workshop 8.1 today.
>>http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
>>_______________________________________________
>>Acpi-devel mailing list
>>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>>https://lists.sourceforge.net/lists/listinfo/acpi-devel
>>    
>>
>
>  
>




-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

RE: acpi-20040715+D600+Suspend

[Li, Shaohua]

Wednesday, September 01, 2004, Nathan Bryant wrote:
>Oops, that's my bad. Thanks for the quick explanation, Shaohua, looks
>like we have to move the resume to a special case that runs before
>sysdevs. If that's ok with everyone, I'll try to make a patch.
The limitation of ACPI CA code always confuses people. Run it before
sysdev means we must enable IRQ. It sounds like risky before we resume
LAPIC and PIC/IOAPIC. I'm considering resume it after sysdev but before
any device driver. But this way it still breaks some systems (In
previous email, we even propose to resume irqrouter before PIC). 

Thanks,
Shaohua
>
>>>-----Original Message-----
>>>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:acpi-devel-
>>>admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Ow Mun Heng
>>>Sent: Wednesday, September 01, 2004 9:17 AM
>>>To: Acpi-List
>>>Subject: [ACPI] acpi-20040715+D600+Suspend
>>>
>>>Hi guys,
>>>
>>>	Lo and Behold, I managed to get S3 to work on my laptop by
>>>
>>>
>>following
>>
>>
>>>instructions on this website.
>>>
>>>http://www.loria.fr/~thome/d600/
>>>
>>>However, I've not tested it further(as in using the normal kernel
boot
>>>up line in grub.conf) to verify if it is the same whether or not I
use
>>>video=radeonfb.
>>>
>>>I'm getting these errors during Suspend to disk.
>>>
>>>Warning: CPU frequency out of sync: cpufreq and timingcore thinks of
>>>
>>>
>>600000,
>>
>>
>>>is 1400000 kHz.
>>>Debug: sleeping function called from invalid context at
mm/slab.c:1989
>>>in_atomic():1, irqs_disabled():0
>>>[<c01156ac>] __might_sleep+0xa5/0xaf
>>>[<c0135d3f>] __kmalloc+0x40/0x76
>>>[<c023bec9>] acpi_os_allocate+0xa/0xb
>>>[<c024fa1f>] acpi_ut_allocate+0x34/0x57
>>>[<c024f9b0>] acpi_ut_initialize_buffer+0x42/0x7d
>>>[<c024c85e>] acpi_rs_create_byte_stream+0x23/0x39
>>>[<c024dc1e>] acpi_rs_set_srs_method_data+0x1b/0x9d
>>>[<c011ecd4>] __group_send_sig_info+0x36/0x9d
>>>[<c011ee0a>] group_send_sig_info+0x58/0x61
>>>[<c02556fd>] acpi_pci_link_set+0xde/0x155
>>>[<c0255b0f>] irqrouter_resume+0x1c/0x30
>>>[<c027bfff>] sysdev_resume+0x3e/0xc7
>>>[<c027ea69>] device_power_up+0x5/0xa
>>>[<c012b332>] do_magic_resume_2+0x63/0xb9
>>>[<c02db0d9>] copy_done+0x49/0x4c
>>>[<c012b536>] software_suspend+0x98/0xae
>>>[<c0252322>] acpi_system_write_sleep+0x6a/0x85
>>>[<c014650c>] vfs_write+0xb8/0xe4
>>>[<c01465d6>] sys_write+0x3c/0x62
>>>[<c01059fd>] sysenter_past_esp+0x52/0x71
>>>ACPI: PCI interrupt 0000:00:1d.0[A] -> GSI 11 (level, low) -> IRQ 11
>>>PCI: Setting latency timer of device 0000:00:1d.0 to 64
>>>ACPI: PCI interrupt 0000:00:1d.1[B] -> GSI 11 (level, low) -> IRQ 11
>>>PCI: Setting latency timer of device 0000:00:1d.1 to 64
>>>ACPI: PCI interrupt 0000:00:1d.2[C] -> GSI 11 (level, low) -> IRQ 11
>>>PCI: Setting latency timer of device 0000:00:1d.2 to 64
>>>ACPI: PCI interrupt 0000:00:1d.7[D] -> GSI 11 (level, low) -> IRQ 11
>>>PCI: Setting latency timer of device 0000:00:1d.7 to 64
>>>ACPI: PCI interrupt 0000:00:1f.1[A] -> GSI 11 (level, low) -> IRQ 11
>>>ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 7 (level, low) -> IRQ 7
>>>PCI: Setting latency timer of device 0000:00:1f.5 to 64
>>>radeonfb: resumed !
>>>ACPI: PCI interrupt 0000:02:00.0[A] -> GSI 11 (level, low) -> IRQ 11
>>>Fixing swap signatures... ok
>>>
>>>
>>>I also noticed that now, suspend does not work as it should/or was
when
>>>I was using 2.6.7 kernel.
>>>
>>>Right now, it will go to the Freeing memory section.. then the screen
>>>goes blank. (blacklight still on) and then stays that way for quite
>>>sometime before I see the words "radeonfb: resumed" and when that
>>>happens, it's as though time stood still. It won't start writing to
the
>>>disk until I press a key. (I usually just press the spacebar) and
then
>>>it will start writing and works OK from that. (I am writing on my
>>>
>>>
>>laptop
>>
>>
>>>which is been suspended(hibernated) and resumed a few times already.
>>>
>>>Anything changed or broke??
>>>
>>>--
>>>Ow Mun Heng
>>>Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel 2.6.8.1.Aug.29
>>>Neuromancer 01:06:38 up 1 day, 3:25, 1 user, load average: 0.73,
0.88,
>>>0.93
>>>
>>>
>>>-------------------------------------------------------
>>>This SF.Net email is sponsored by BEA Weblogic Workshop
>>>FREE Java Enterprise J2EE developer tools!
>>>Get your free copy of BEA WebLogic Workshop 8.1 today.
>>>http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
>>>_______________________________________________
>>>Acpi-devel mailing list
>>>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>>>https://lists.sourceforge.net/lists/listinfo/acpi-devel
>>>
>>>
>>
>>
>>
>



-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_idP47&alloc_id\x10808&op=click

Re: acpi-20040715+D600+Suspend

[Nathan Bryant]

Li, Shaohua wrote:

>Wednesday, September 01, 2004, Nathan Bryant wrote:
>  
>
>>Oops, that's my bad. Thanks for the quick explanation, Shaohua, looks
>>like we have to move the resume to a special case that runs before
>>sysdevs. If that's ok with everyone, I'll try to make a patch.
>>    
>>
>The limitation of ACPI CA code always confuses people. Run it before
>sysdev means we must enable IRQ. It sounds like risky before we resume
>LAPIC and PIC/IOAPIC. I'm considering resume it after sysdev but before
>any device driver. But this way it still breaks some systems (In
>previous email, we even propose to resume irqrouter before PIC). 
>  
>
I was afraid you were going to say that.

You know there is already some special case code to init the i8259 
earlier for specfic blacklisted laptops. Maybe we can generalize that; 
it is starting to look like we are stuck with special casing all the IRQ 
init. Do you agree or disagree?

Nathan

>Thanks,
>Shaohua
>  
>
>>>>-----Original Message-----
>>>>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:acpi-devel-
>>>>admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Ow Mun Heng
>>>>Sent: Wednesday, September 01, 2004 9:17 AM
>>>>To: Acpi-List
>>>>Subject: [ACPI] acpi-20040715+D600+Suspend
>>>>
>>>>Hi guys,
>>>>
>>>>	Lo and Behold, I managed to get S3 to work on my laptop by
>>>>
>>>>
>>>>        
>>>>
>>>following
>>>
>>>
>>>      
>>>
>>>>instructions on this website.
>>>>
>>>>http://www.loria.fr/~thome/d600/
>>>>
>>>>However, I've not tested it further(as in using the normal kernel
>>>>        
>>>>
>boot
>  
>
>>>>up line in grub.conf) to verify if it is the same whether or not I
>>>>        
>>>>
>use
>  
>
>>>>video=radeonfb.
>>>>
>>>>I'm getting these errors during Suspend to disk.
>>>>
>>>>Warning: CPU frequency out of sync: cpufreq and timingcore thinks of
>>>>
>>>>
>>>>        
>>>>
>>>600000,
>>>
>>>
>>>      
>>>
>>>>is 1400000 kHz.
>>>>Debug: sleeping function called from invalid context at
>>>>        
>>>>
>mm/slab.c:1989
>  
>
>>>>in_atomic():1, irqs_disabled():0
>>>>[<c01156ac>] __might_sleep+0xa5/0xaf
>>>>[<c0135d3f>] __kmalloc+0x40/0x76
>>>>[<c023bec9>] acpi_os_allocate+0xa/0xb
>>>>[<c024fa1f>] acpi_ut_allocate+0x34/0x57
>>>>[<c024f9b0>] acpi_ut_initialize_buffer+0x42/0x7d
>>>>[<c024c85e>] acpi_rs_create_byte_stream+0x23/0x39
>>>>[<c024dc1e>] acpi_rs_set_srs_method_data+0x1b/0x9d
>>>>[<c011ecd4>] __group_send_sig_info+0x36/0x9d
>>>>[<c011ee0a>] group_send_sig_info+0x58/0x61
>>>>[<c02556fd>] acpi_pci_link_set+0xde/0x155
>>>>[<c0255b0f>] irqrouter_resume+0x1c/0x30
>>>>[<c027bfff>] sysdev_resume+0x3e/0xc7
>>>>[<c027ea69>] device_power_up+0x5/0xa
>>>>[<c012b332>] do_magic_resume_2+0x63/0xb9
>>>>[<c02db0d9>] copy_done+0x49/0x4c
>>>>[<c012b536>] software_suspend+0x98/0xae
>>>>[<c0252322>] acpi_system_write_sleep+0x6a/0x85
>>>>[<c014650c>] vfs_write+0xb8/0xe4
>>>>[<c01465d6>] sys_write+0x3c/0x62
>>>>[<c01059fd>] sysenter_past_esp+0x52/0x71
>>>>ACPI: PCI interrupt 0000:00:1d.0[A] -> GSI 11 (level, low) -> IRQ 11
>>>>PCI: Setting latency timer of device 0000:00:1d.0 to 64
>>>>ACPI: PCI interrupt 0000:00:1d.1[B] -> GSI 11 (level, low) -> IRQ 11
>>>>PCI: Setting latency timer of device 0000:00:1d.1 to 64
>>>>ACPI: PCI interrupt 0000:00:1d.2[C] -> GSI 11 (level, low) -> IRQ 11
>>>>PCI: Setting latency timer of device 0000:00:1d.2 to 64
>>>>ACPI: PCI interrupt 0000:00:1d.7[D] -> GSI 11 (level, low) -> IRQ 11
>>>>PCI: Setting latency timer of device 0000:00:1d.7 to 64
>>>>ACPI: PCI interrupt 0000:00:1f.1[A] -> GSI 11 (level, low) -> IRQ 11
>>>>ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 7 (level, low) -> IRQ 7
>>>>PCI: Setting latency timer of device 0000:00:1f.5 to 64
>>>>radeonfb: resumed !
>>>>ACPI: PCI interrupt 0000:02:00.0[A] -> GSI 11 (level, low) -> IRQ 11
>>>>Fixing swap signatures... ok
>>>>
>>>>
>>>>I also noticed that now, suspend does not work as it should/or was
>>>>        
>>>>
>when
>  
>
>>>>I was using 2.6.7 kernel.
>>>>
>>>>Right now, it will go to the Freeing memory section.. then the screen
>>>>goes blank. (blacklight still on) and then stays that way for quite
>>>>sometime before I see the words "radeonfb: resumed" and when that
>>>>happens, it's as though time stood still. It won't start writing to
>>>>        
>>>>
>the
>  
>
>>>>disk until I press a key. (I usually just press the spacebar) and
>>>>        
>>>>
>then
>  
>
>>>>it will start writing and works OK from that. (I am writing on my
>>>>
>>>>
>>>>        
>>>>
>>>laptop
>>>
>>>
>>>      
>>>
>>>>which is been suspended(hibernated) and resumed a few times already.
>>>>
>>>>Anything changed or broke??
>>>>
>>>>--
>>>>Ow Mun Heng
>>>>Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel 2.6.8.1.Aug.29
>>>>Neuromancer 01:06:38 up 1 day, 3:25, 1 user, load average: 0.73,
>>>>        
>>>>
>0.88,
>  
>
>>>>0.93
>>>>
>>>>
>>>>-------------------------------------------------------
>>>>This SF.Net email is sponsored by BEA Weblogic Workshop
>>>>FREE Java Enterprise J2EE developer tools!
>>>>Get your free copy of BEA WebLogic Workshop 8.1 today.
>>>>http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
>>>>_______________________________________________
>>>>Acpi-devel mailing list
>>>>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>>>>https://lists.sourceforge.net/lists/listinfo/acpi-devel
>>>>
>>>>
>>>>        
>>>>
>>>
>>>      
>>>
>
>
>  
>



-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

RE: acpi-20040715+D600+Suspend

[Li, Shaohua]

On Wednesday, September 01, 2004, Nathan Bryant wrote:
>>
>>
>I was afraid you were going to say that.
>
>You know there is already some special case code to init the i8259
>earlier for specfic blacklisted laptops. Maybe we can generalize that;
>it is starting to look like we are stuck with special casing all the
IRQ
>init. Do you agree or disagree?
Don't know if it can apply for all cases. IIRC, there is a similar case
(in SMP, resume MTRR) about resume failure in IRQ disabled. Pavel, can
you show us a light per the MTRR case?

Thanks,
Shaohua
>>>>>-----Original Message-----
>>>>>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:acpi-devel-
>>>>>admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Ow Mun Heng
>>>>>Sent: Wednesday, September 01, 2004 9:17 AM
>>>>>To: Acpi-List
>>>>>Subject: [ACPI] acpi-20040715+D600+Suspend
>>>>>
>>>>>Hi guys,
>>>>>
>>>>>	Lo and Behold, I managed to get S3 to work on my laptop by
>>>>>
>>>>>
>>>>>
>>>>>
>>>>following
>>>>
>>>>
>>>>
>>>>
>>>>>instructions on this website.
>>>>>
>>>>>http://www.loria.fr/~thome/d600/
>>>>>
>>>>>However, I've not tested it further(as in using the normal kernel
>>>>>
>>>>>
>>boot
>>
>>
>>>>>up line in grub.conf) to verify if it is the same whether or not I
>>>>>
>>>>>
>>use
>>
>>
>>>>>video=radeonfb.
>>>>>
>>>>>I'm getting these errors during Suspend to disk.
>>>>>
>>>>>Warning: CPU frequency out of sync: cpufreq and timingcore thinks
of
>>>>>
>>>>>
>>>>>
>>>>>
>>>>600000,
>>>>
>>>>
>>>>
>>>>
>>>>>is 1400000 kHz.
>>>>>Debug: sleeping function called from invalid context at
>>>>>
>>>>>
>>mm/slab.c:1989
>>
>>
>>>>>in_atomic():1, irqs_disabled():0
>>>>>[<c01156ac>] __might_sleep+0xa5/0xaf
>>>>>[<c0135d3f>] __kmalloc+0x40/0x76
>>>>>[<c023bec9>] acpi_os_allocate+0xa/0xb
>>>>>[<c024fa1f>] acpi_ut_allocate+0x34/0x57
>>>>>[<c024f9b0>] acpi_ut_initialize_buffer+0x42/0x7d
>>>>>[<c024c85e>] acpi_rs_create_byte_stream+0x23/0x39
>>>>>[<c024dc1e>] acpi_rs_set_srs_method_data+0x1b/0x9d
>>>>>[<c011ecd4>] __group_send_sig_info+0x36/0x9d
>>>>>[<c011ee0a>] group_send_sig_info+0x58/0x61
>>>>>[<c02556fd>] acpi_pci_link_set+0xde/0x155
>>>>>[<c0255b0f>] irqrouter_resume+0x1c/0x30
>>>>>[<c027bfff>] sysdev_resume+0x3e/0xc7
>>>>>[<c027ea69>] device_power_up+0x5/0xa
>>>>>[<c012b332>] do_magic_resume_2+0x63/0xb9
>>>>>[<c02db0d9>] copy_done+0x49/0x4c
>>>>>[<c012b536>] software_suspend+0x98/0xae
>>>>>[<c0252322>] acpi_system_write_sleep+0x6a/0x85
>>>>>[<c014650c>] vfs_write+0xb8/0xe4
>>>>>[<c01465d6>] sys_write+0x3c/0x62
>>>>>[<c01059fd>] sysenter_past_esp+0x52/0x71
>>>>>ACPI: PCI interrupt 0000:00:1d.0[A] -> GSI 11 (level, low) -> IRQ
11
>>>>>PCI: Setting latency timer of device 0000:00:1d.0 to 64
>>>>>ACPI: PCI interrupt 0000:00:1d.1[B] -> GSI 11 (level, low) -> IRQ
11
>>>>>PCI: Setting latency timer of device 0000:00:1d.1 to 64
>>>>>ACPI: PCI interrupt 0000:00:1d.2[C] -> GSI 11 (level, low) -> IRQ
11
>>>>>PCI: Setting latency timer of device 0000:00:1d.2 to 64
>>>>>ACPI: PCI interrupt 0000:00:1d.7[D] -> GSI 11 (level, low) -> IRQ
11
>>>>>PCI: Setting latency timer of device 0000:00:1d.7 to 64
>>>>>ACPI: PCI interrupt 0000:00:1f.1[A] -> GSI 11 (level, low) -> IRQ
11
>>>>>ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 7 (level, low) -> IRQ 7
>>>>>PCI: Setting latency timer of device 0000:00:1f.5 to 64
>>>>>radeonfb: resumed !
>>>>>ACPI: PCI interrupt 0000:02:00.0[A] -> GSI 11 (level, low) -> IRQ
11
>>>>>Fixing swap signatures... ok
>>>>>
>>>>>
>>>>>I also noticed that now, suspend does not work as it should/or was
>>>>>
>>>>>
>>when
>>
>>
>>>>>I was using 2.6.7 kernel.
>>>>>
>>>>>Right now, it will go to the Freeing memory section.. then the
screen
>>>>>goes blank. (blacklight still on) and then stays that way for quite
>>>>>sometime before I see the words "radeonfb: resumed" and when that
>>>>>happens, it's as though time stood still. It won't start writing to
>>>>>
>>>>>
>>the
>>
>>
>>>>>disk until I press a key. (I usually just press the spacebar) and
>>>>>
>>>>>
>>then
>>
>>
>>>>>it will start writing and works OK from that. (I am writing on my
>>>>>
>>>>>
>>>>>
>>>>>
>>>>laptop
>>>>
>>>>
>>>>
>>>>
>>>>>which is been suspended(hibernated) and resumed a few times
already.
>>>>>
>>>>>Anything changed or broke??
>>>>>
>>>>>--
>>>>>Ow Mun Heng
>>>>>Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel 2.6.8.1.Aug.29
>>>>>Neuromancer 01:06:38 up 1 day, 3:25, 1 user, load average: 0.73,
>>>>>
>>>>>
>>0.88,
>>
>>
>>>>>0.93
>>>>>
>>>>>
>>>>>-------------------------------------------------------
>>>>>This SF.Net email is sponsored by BEA Weblogic Workshop
>>>>>FREE Java Enterprise J2EE developer tools!
>>>>>Get your free copy of BEA WebLogic Workshop 8.1 today.
>>>>>http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
>>>>>_______________________________________________
>>>>>Acpi-devel mailing list
>>>>>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>>>>>https://lists.sourceforge.net/lists/listinfo/acpi-devel
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>
>>
>>
>>



-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_idP47&alloc_id\x10808&op=click

RE: acpi-20040715+D600+Suspend

[Ow Mun Heng]

On Wed, 2004-09-01 at 11:12, Li, Shaohua wrote:
> On Wednesday, September 01, 2004, Nathan Bryant wrote:
> >>
> >>
> >I was afraid you were going to say that.
> >
> >You know there is already some special case code to init the i8259
> >earlier for specfic blacklisted laptops. Maybe we can generalize that;
> >it is starting to look like we are stuck with special casing all the
> IRQ
> >init. Do you agree or disagree?
> Don't know if it can apply for all cases. IIRC, there is a similar case
> (in SMP, resume MTRR) about resume failure in IRQ disabled. Pavel, can
> you show us a light per the MTRR case?

Okay.. this is too deep for me to understand.
The latest resume from hibernate ended up with a segmentation fault and
placed me in a virtual term and it says INIT ??? I didn't copy it down
since I didn't have a pen/paper handy. (it says it's retrying at every
30sec interval)

I just tried booting up with the normal kernel boot parameter "vga=0x342
resume=/dev/hdx", the resume still has the same thing.

Let me know how I can help.

The Keyboard press to get it to start writing to the swap, is it
relevent?


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

RE: acpi-20040715+D600+Suspend

[Pallipadi, Venkatesh]

May be I am completely off-target.
Can we add something in do_irq, not to call any device driver 
irq handlers until the complete resume is done and enable 
irqs before link device resume?

Thanks,
Venki 

>-----Original Message-----
>From: Li, Shaohua 
>Sent: Tuesday, August 31, 2004 8:12 PM
>To: Nathan Bryant; Pavel Machek
>Cc: acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org; Brown, Len; Pallipadi, Venkatesh
>Subject: RE: [ACPI] acpi-20040715+D600+Suspend
>
>On Wednesday, September 01, 2004, Nathan Bryant wrote:
>>>
>>>
>>I was afraid you were going to say that.
>>
>>You know there is already some special case code to init the i8259
>>earlier for specfic blacklisted laptops. Maybe we can generalize that;
>>it is starting to look like we are stuck with special casing 
>all the IRQ
>>init. Do you agree or disagree?
>Don't know if it can apply for all cases. IIRC, there is a 
>similar case (in SMP, resume MTRR) about resume failure in IRQ 
>disabled. Pavel, can you show us a light per the MTRR case?
>
>Thanks,
>Shaohua
>>>>>>-----Original Message-----
>>>>>>From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:acpi-devel-
>>>>>>admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Ow Mun Heng
>>>>>>Sent: Wednesday, September 01, 2004 9:17 AM
>>>>>>To: Acpi-List
>>>>>>Subject: [ACPI] acpi-20040715+D600+Suspend
>>>>>>
>>>>>>Hi guys,
>>>>>>
>>>>>>	Lo and Behold, I managed to get S3 to work on my laptop by
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>following
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>>instructions on this website.
>>>>>>
>>>>>>http://www.loria.fr/~thome/d600/
>>>>>>
>>>>>>However, I've not tested it further(as in using the normal kernel
>>>>>>
>>>>>>
>>>boot
>>>
>>>
>>>>>>up line in grub.conf) to verify if it is the same whether or not I
>>>>>>
>>>>>>
>>>use
>>>
>>>
>>>>>>video=radeonfb.
>>>>>>
>>>>>>I'm getting these errors during Suspend to disk.
>>>>>>
>>>>>>Warning: CPU frequency out of sync: cpufreq and 
>timingcore thinks of
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>600000,
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>>is 1400000 kHz.
>>>>>>Debug: sleeping function called from invalid context at
>>>>>>
>>>>>>
>>>mm/slab.c:1989
>>>
>>>
>>>>>>in_atomic():1, irqs_disabled():0
>>>>>>[<c01156ac>] __might_sleep+0xa5/0xaf
>>>>>>[<c0135d3f>] __kmalloc+0x40/0x76
>>>>>>[<c023bec9>] acpi_os_allocate+0xa/0xb
>>>>>>[<c024fa1f>] acpi_ut_allocate+0x34/0x57
>>>>>>[<c024f9b0>] acpi_ut_initialize_buffer+0x42/0x7d
>>>>>>[<c024c85e>] acpi_rs_create_byte_stream+0x23/0x39
>>>>>>[<c024dc1e>] acpi_rs_set_srs_method_data+0x1b/0x9d
>>>>>>[<c011ecd4>] __group_send_sig_info+0x36/0x9d
>>>>>>[<c011ee0a>] group_send_sig_info+0x58/0x61
>>>>>>[<c02556fd>] acpi_pci_link_set+0xde/0x155
>>>>>>[<c0255b0f>] irqrouter_resume+0x1c/0x30
>>>>>>[<c027bfff>] sysdev_resume+0x3e/0xc7
>>>>>>[<c027ea69>] device_power_up+0x5/0xa
>>>>>>[<c012b332>] do_magic_resume_2+0x63/0xb9
>>>>>>[<c02db0d9>] copy_done+0x49/0x4c
>>>>>>[<c012b536>] software_suspend+0x98/0xae
>>>>>>[<c0252322>] acpi_system_write_sleep+0x6a/0x85
>>>>>>[<c014650c>] vfs_write+0xb8/0xe4
>>>>>>[<c01465d6>] sys_write+0x3c/0x62
>>>>>>[<c01059fd>] sysenter_past_esp+0x52/0x71
>>>>>>ACPI: PCI interrupt 0000:00:1d.0[A] -> GSI 11 (level, 
>low) -> IRQ 11
>>>>>>PCI: Setting latency timer of device 0000:00:1d.0 to 64
>>>>>>ACPI: PCI interrupt 0000:00:1d.1[B] -> GSI 11 (level, 
>low) -> IRQ 11
>>>>>>PCI: Setting latency timer of device 0000:00:1d.1 to 64
>>>>>>ACPI: PCI interrupt 0000:00:1d.2[C] -> GSI 11 (level, 
>low) -> IRQ 11
>>>>>>PCI: Setting latency timer of device 0000:00:1d.2 to 64
>>>>>>ACPI: PCI interrupt 0000:00:1d.7[D] -> GSI 11 (level, 
>low) -> IRQ 11
>>>>>>PCI: Setting latency timer of device 0000:00:1d.7 to 64
>>>>>>ACPI: PCI interrupt 0000:00:1f.1[A] -> GSI 11 (level, 
>low) -> IRQ 11
>>>>>>ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 7 (level, low) -> IRQ 7
>>>>>>PCI: Setting latency timer of device 0000:00:1f.5 to 64
>>>>>>radeonfb: resumed !
>>>>>>ACPI: PCI interrupt 0000:02:00.0[A] -> GSI 11 (level, 
>low) -> IRQ 11
>>>>>>Fixing swap signatures... ok
>>>>>>
>>>>>>
>>>>>>I also noticed that now, suspend does not work as it should/or was
>>>>>>
>>>>>>
>>>when
>>>
>>>
>>>>>>I was using 2.6.7 kernel.
>>>>>>
>>>>>>Right now, it will go to the Freeing memory section.. 
>then the screen
>>>>>>goes blank. (blacklight still on) and then stays that way 
>for quite
>>>>>>sometime before I see the words "radeonfb: resumed" and when that
>>>>>>happens, it's as though time stood still. It won't start 
>writing to
>>>>>>
>>>>>>
>>>the
>>>
>>>
>>>>>>disk until I press a key. (I usually just press the spacebar) and
>>>>>>
>>>>>>
>>>then
>>>
>>>
>>>>>>it will start writing and works OK from that. (I am writing on my
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>laptop
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>>which is been suspended(hibernated) and resumed a few 
>times already.
>>>>>>
>>>>>>Anything changed or broke??
>>>>>>
>>>>>>--
>>>>>>Ow Mun Heng
>>>>>>Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel 2.6.8.1.Aug.29
>>>>>>Neuromancer 01:06:38 up 1 day, 3:25, 1 user, load average: 0.73,
>>>>>>
>>>>>>
>>>0.88,
>>>
>>>
>>>>>>0.93
>>>>>>
>>>>>>
>>>>>>-------------------------------------------------------
>>>>>>This SF.Net email is sponsored by BEA Weblogic Workshop
>>>>>>FREE Java Enterprise J2EE developer tools!
>>>>>>Get your free copy of BEA WebLogic Workshop 8.1 today.
>>>>>>http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
>>>>>>_______________________________________________
>>>>>>Acpi-devel mailing list
>>>>>>Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
>>>>>>https://lists.sourceforge.net/lists/listinfo/acpi-devel
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>>
>>>
>>>
>>>
>>>
>
>


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_idP47&alloc_id\x10808&op=click

Re: acpi-20040715+D600+Suspend

[Pavel Machek]

Hi!

> 	Lo and Behold, I managed to get S3 to work on my laptop by following
> instructions on this website.
> 
> http://www.loria.fr/~thome/d600/
> 
> However, I've not tested it further(as in using the normal kernel boot
> up line in grub.conf) to verify if it is the same whether or not I use
> video=radeonfb.
> 
> I'm getting these errors during Suspend to disk.
> 
> Warning: CPU frequency out of sync: cpufreq and timingcore thinks of 600000, is 1400000 kHz.
> Debug: sleeping function called from invalid context at mm/slab.c:1989
> in_atomic():1, irqs_disabled():0

Turn off CONFIG_PREEMPT. Or better test latest -rcX-mmX.
										Pavel
-- 
When do you have heart between your knees?


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

Re: acpi-20040715+D600+Suspend

[Pavel Machek]

Hi!

> >I was afraid you were going to say that.
> >
> >You know there is already some special case code to init the i8259
> >earlier for specfic blacklisted laptops. Maybe we can generalize that;
> >it is starting to look like we are stuck with special casing all the
> IRQ
> >init. Do you agree or disagree?
> Don't know if it can apply for all cases. IIRC, there is a similar case
> (in SMP, resume MTRR) about resume failure in IRQ disabled. Pavel, can
> you show us a light per the MTRR case?

I do not know about MTRR much... System should work without MTRR,
right?

Anyway that specific blacklisted laptop is toshiba 4030cdt, right?
That's *really* old hack. I guess it could be safely killed these
days, as we call driver model during S1 anyway.
								Pavel
-- 
When do you have heart between your knees?


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

Re: acpi-20040715+D600+Suspend

[Ow Mun Heng]

On Thu, 2004-09-02 at 22:30, Pavel Machek wrote:
> Hi!
> 
> > 	Lo and Behold, I managed to get S3 to work on my laptop by following
> > instructions on this website.
> > 
> > http://www.loria.fr/~thome/d600/
> > 
> > However, I've not tested it further(as in using the normal kernel boot
> > up line in grub.conf) to verify if it is the same whether or not I use
> > video=radeonfb.
> > 
> > I'm getting these errors during Suspend to disk.
> > 
> > Warning: CPU frequency out of sync: cpufreq and timingcore thinks of 600000, is 1400000 kHz.
> > Debug: sleeping function called from invalid context at mm/slab.c:1989
> > in_atomic():1, irqs_disabled():0
> 
> Turn off CONFIG_PREEMPT. Or better test latest -rcX-mmX.

Hey.. pavel.. thanks to your suggestion, my weekend was wrecked (good
natured sarcasm). I found out the hard way that there rc1-mm1 kernel,
(the only one which worked fairly well) has quite a biting bug that
caused data loss. 

I had my root partition (reiserfs crash), I lost a few trivial files (i
hope) and had to do a rebuild-tree.

After that, it was all downhill, I had rpm errors, my database was
currupted (or so I thought) and tar wouldn't finish on some of those .db
files (I was trying to back up evolution as I anticipated a re-install)

Anyway.. I'm now back on 2.6.8.1 (after finding out that this is a known
bug)

Anyway again, I only found that 2.6.9-rc1-mm1 works in the sense that it
didn't need a "keyboard press for it to initiate writing/suspend to
disk"

All the rest didn't work. That's meaning 2.6.9/2.6.9-rc1-mm2/2.6.8.1

I see that 2.6.9-rc1-mm3 is out, I will test it. (hopefully it doen't
bite)
 
> 										Pavel

-- 
Ow Mun Heng
Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel 2.6.8.1.Aug.29 
Neuromancer 11:18:26 up 1:05, 6 users, load average: 1.54, 0.90, 0.52 


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

Re: acpi-20040715+D600+Suspend

[Pavel Machek]

Hi!

> > Turn off CONFIG_PREEMPT. Or better test latest -rcX-mmX.
> 
> Hey.. pavel.. thanks to your suggestion, my weekend was wrecked (good
> natured sarcasm). I found out the hard way that there rc1-mm1 kernel,
> (the only one which worked fairly well) has quite a biting bug that
> caused data loss. 

Sorry for that.

> I had my root partition (reiserfs crash), I lost a few trivial files (i
> hope) and had to do a rebuild-tree.

Hmm, do not use reiserfs ;-).

									Pavel


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

Re: acpi-20040715+D600+Suspend

[Ow Mun Heng]

On Tue, 2004-09-07 at 01:13, Pavel Machek wrote:
> Hi!
> 
> > > Turn off CONFIG_PREEMPT. Or better test latest -rcX-mmX.
> > 
> > Hey.. pavel.. thanks to your suggestion, my weekend was wrecked (good
> > natured sarcasm). I found out the hard way that there rc1-mm1 kernel,
> > (the only one which worked fairly well) has quite a biting bug that
> > caused data loss. 
> 
> Sorry for that.

As I mentioned. it was good natured sarcasm.
I'm back on 2.6.7 with the interactive patch and an old acpi code.

Just an update, I've tried 2.6.9 yesterday, still had the irq_disabled
problem. (though did not need a keyboard press to activate writing to
swap)

2.6.9-rc1-mm3 compiled OK once i8042 is patched (or reversed patched)
but it won't perform modules_install. Problem with "ln" and multiple
directories.


> 
> > I had my root partition (reiserfs crash), I lost a few trivial files (i
> > hope) and had to do a rebuild-tree.
> 
> Hmm, do not use reiserfs ;-).
Hmm.. isn't reiserfs better?
I dunno.
Suse uses it as their root fs.
I've only switched to reiserfs for my FC2 install. (and frankly, like
since I went to FC2 3+ months ago, I've had like 3 reiserfs
--rebuild-checks) compared to like NONE  when I was on RH9/ext3 for ~8
months.

What's the majority of ppl using? I know this is Off-topic. sorry..


XFS for my /home 20GB dir. It handles fairly well. (as long as there's
no sudden power loss, as I read XFS caches the data aggressively and not
to sync with the disk)

-- 
Ow Mun Heng
Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel
2.6.7-2.jul1-interactive 
Neuromancer 11:15:19 up 3:04, 7 users, load average: 1.38, 0.84, 0.64 


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

Re: acpi-20040715+D600+Suspend

[Cameron Patrick]

Ow Mun Heng wrote:

> > Hmm, do not use reiserfs ;-).
> Hmm.. isn't reiserfs better?
> I dunno.
> Suse uses it as their root fs.
> I've only switched to reiserfs for my FC2 install. (and frankly, like
> since I went to FC2 3+ months ago, I've had like 3 reiserfs
> --rebuild-checks) compared to like NONE  when I was on RH9/ext3 for ~8
> months.

> What's the majority of ppl using? I know this is Off-topic. sorry..

In my experience, ext3 is the best in terms of not losing any data.  I
tend to use reiserfs because it's faster, though...  I've never had to
do a reiserfs rebuild.

> XFS for my /home 20GB dir. It handles fairly well. (as long as there's
> no sudden power loss, as I read XFS caches the data aggressively and not
> to sync with the disk)

I used to use XFS.  On a few occasions it crashed hard and would
refuse to mount again (needed an xfs_recover).  I ended up with many
corrupted files full of NULL bytes, pretty much every time the machine
lost power or crashed (yay Nvidia drivers) :(

Cameron.



-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

Re: acpi-20040715+D600+Suspend

[Pavel Machek]

Hi!

> > > I had my root partition (reiserfs crash), I lost a few trivial files (i
> > > hope) and had to do a rebuild-tree.
> > 
> > Hmm, do not use reiserfs ;-).
> Hmm.. isn't reiserfs better?
> I dunno.
> Suse uses it as their root fs.
> I've only switched to reiserfs for my FC2 install. (and frankly, like
> since I went to FC2 3+ months ago, I've had like 3 reiserfs
> --rebuild-checks) compared to like NONE  when I was on RH9/ext3 for ~8
> months.
> 
> What's the majority of ppl using? I know this is Off-topic. sorry..

For suspend work, ext2 is the safest. I'm now using ext3 on less
important machines -- at least fsck works there.

								Pavel


-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click

Re: acpi-20040715+D600+Suspend

[Nigel Cunningham]

Hi.

On Tue, 2004-09-07 at 19:34, Pavel Machek wrote:
> > > > I had my root partition (reiserfs crash), I lost a few trivial files (i
> > > > hope) and had to do a rebuild-tree.
> > > Hmm, do not use reiserfs ;-).
> > Hmm.. isn't reiserfs better?
> > I dunno.
> > Suse uses it as their root fs.
> > I've only switched to reiserfs for my FC2 install. (and frankly, like
> > since I went to FC2 3+ months ago, I've had like 3 reiserfs
> > --rebuild-checks) compared to like NONE  when I was on RH9/ext3 for ~8
> > months.
> > 
> > What's the majority of ppl using? I know this is Off-topic. sorry..
> 
> For suspend work, ext2 is the safest. I'm now using ext3 on less
> important machines -- at least fsck works there.

Ditto here. On the occasions when I have damaged my filesystem, I've had
much better success recovering while using ext3. I tried xfs, but gave
up on it quickly because of the zeroing-out-files issue.

Regards,

Nigel
-- 
Nigel Cunningham
Pastoral Worker
Christian Reformed Church of Tuggeranong
PO Box 1004, Tuggeranong, ACT 2901

Many today claim to be tolerant. True tolerance, however, can cope with others
being intolerant.



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM. 
Deadline: Sept. 13. Go here: http://sf.net/ppc_contest.php

Re: acpi-20040715+D600+Suspend

[Ow Mun Heng]

On Fri, 2004-09-10 at 18:06, Nigel Cunningham wrote:
> Hi.
> 
> On Tue, 2004-09-07 at 19:34, Pavel Machek wrote:
> > > > > I had my root partition (reiserfs crash), I lost a few trivial files (i
> > > > > hope) and had to do a rebuild-tree.
> > > > Hmm, do not use reiserfs ;-).
> > > Hmm.. isn't reiserfs better?
> > > I dunno.
> > > Suse uses it as their root fs.
> > > I've only switched to reiserfs for my FC2 install. (and frankly, like
> > > since I went to FC2 3+ months ago, I've had like 3 reiserfs
> > > --rebuild-checks) compared to like NONE  when I was on RH9/ext3 for ~8
> > > months.
> > > 
> > > What's the majority of ppl using? I know this is Off-topic. sorry..
> > 
> > For suspend work, ext2 is the safest. I'm now using ext3 on less
> > important machines -- at least fsck works there.
> 
> Ditto here. On the occasions when I have damaged my filesystem, I've had
> much better success recovering while using ext3. I tried xfs, but gave
> up on it quickly because of the zeroing-out-files issue.
Okay.. But here's another question. Is it recommended that _all_
partitions be ext3 or only the root partition??

-- 
Ow Mun Heng
Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel
2.6.7-2.jul1-interactive 
Neuromancer 18:17:08 up 22:56, 8 users, load average: 2.17, 2.72, 2.64 


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM. 
Deadline: Sept. 13. Go here: http://sf.net/ppc_contest.php

Re: acpi-20040715+D600+Suspend

[Nigel Cunningham]

Hi.

On Fri, 2004-09-10 at 20:17, Ow Mun Heng wrote:
> > Ditto here. On the occasions when I have damaged my filesystem, I've had
> > much better success recovering while using ext3. I tried xfs, but gave
> > up on it quickly because of the zeroing-out-files issue.
> Okay.. But here's another question. Is it recommended that _all_
> partitions be ext3 or only the root partition??

I'd use ext3 only.

Nigel
-- 
Nigel Cunningham
Pastoral Worker
Christian Reformed Church of Tuggeranong
PO Box 1004, Tuggeranong, ACT 2901

Many today claim to be tolerant. True tolerance, however, can cope with others
being intolerant.



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM. 
Deadline: Sept. 13. Go here: http://sf.net/ppc_contest.php

Re: acpi-20040715+D600+Suspend

[Pavel Machek]

Hi!

> > > For suspend work, ext2 is the safest. I'm now using ext3 on less
> > > important machines -- at least fsck works there.
> > 
> > Ditto here. On the occasions when I have damaged my filesystem, I've had
> > much better success recovering while using ext3. I tried xfs, but gave
> > up on it quickly because of the zeroing-out-files issue.
> Okay.. But here's another question. Is it recommended that _all_
> partitions be ext3 or only the root partition??

All partitions with valuable data :-). Actually, ext2 is even better
idea. It should work with any filesystem, it is just that ext2/ext3
has very good filesystem check... And I managed to find bugs even if
fsck.ext2. You'll probably be luckier, but...

								Pavel


-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM. 
Deadline: Sept. 13. Go here: http://sf.net/ppc_contest.php

Re: acpi-20040715+D600+Suspend

[Stefan Seyfried]

On Fri, Sep 10, 2004 at 12:46:41PM +0200, Pavel Machek wrote:

> All partitions with valuable data :-). Actually, ext2 is even better
> idea. It should work with any filesystem, it is just that ext2/ext3
> has very good filesystem check... And I managed to find bugs even if
> fsck.ext2. You'll probably be luckier, but...

So to lighten the mood here a bit, i'll confess that i have all data
(/ and /home) on reiserfs, only /boot is on ext2 to speed up grub at
resume time (if the filesystem is very dirty at suspend, i guess grub
has to playback the journal in its little mind which can take quite
some seconds for both kernel and initrd).
I have to admit that i do daily backups, though, but this is a good
idea with _any_ filesystem.
-- 
Stefan Seyfried



-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM. 
Deadline: Sept. 13. Go here: http://sf.net/ppc_contest.php

RE: acpi-20040715+D600+Suspend

[Li Shaohua]

[-- Attachment #1: Type: text/plain, Size: 4656 bytes --]

Could you please try the attached workaround? If it really helps, we
will do cleanup soon. Untested, if there is any typo, please correct it.

Thanks,
Shaohua
> >-----Original Message-----
> >From: acpi-devel-admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org [mailto:acpi-devel-
> >admin-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org] On Behalf Of Ow Mun Heng
> >Sent: Wednesday, September 01, 2004 9:17 AM
> >To: Acpi-List
> >Subject: [ACPI] acpi-20040715+D600+Suspend
> >
> >Hi guys,
> >
> >       Lo and Behold, I managed to get S3 to work on my laptop by
> following
> >instructions on this website.
> >
> >http://www.loria.fr/~thome/d600/
> >
> >However, I've not tested it further(as in using the normal kernel
> boot
> >up line in grub.conf) to verify if it is the same whether or not I
> use
> >video=radeonfb.
> >
> >I'm getting these errors during Suspend to disk.
> >
> >Warning: CPU frequency out of sync: cpufreq and timingcore thinks of
> 600000,
> >is 1400000 kHz.
> >Debug: sleeping function called from invalid context at
> mm/slab.c:1989
> >in_atomic():1, irqs_disabled():0
> > [<c01156ac>] __might_sleep+0xa5/0xaf
> > [<c0135d3f>] __kmalloc+0x40/0x76
> > [<c023bec9>] acpi_os_allocate+0xa/0xb
> > [<c024fa1f>] acpi_ut_allocate+0x34/0x57
> > [<c024f9b0>] acpi_ut_initialize_buffer+0x42/0x7d
> > [<c024c85e>] acpi_rs_create_byte_stream+0x23/0x39
> > [<c024dc1e>] acpi_rs_set_srs_method_data+0x1b/0x9d
> > [<c011ecd4>] __group_send_sig_info+0x36/0x9d
> > [<c011ee0a>] group_send_sig_info+0x58/0x61
> > [<c02556fd>] acpi_pci_link_set+0xde/0x155
> > [<c0255b0f>] irqrouter_resume+0x1c/0x30
> > [<c027bfff>] sysdev_resume+0x3e/0xc7
> > [<c027ea69>] device_power_up+0x5/0xa
> > [<c012b332>] do_magic_resume_2+0x63/0xb9
> > [<c02db0d9>] copy_done+0x49/0x4c
> > [<c012b536>] software_suspend+0x98/0xae
> > [<c0252322>] acpi_system_write_sleep+0x6a/0x85
> > [<c014650c>] vfs_write+0xb8/0xe4
> > [<c01465d6>] sys_write+0x3c/0x62
> > [<c01059fd>] sysenter_past_esp+0x52/0x71
> >ACPI: PCI interrupt 0000:00:1d.0[A] -> GSI 11 (level, low) -> IRQ 11
> >PCI: Setting latency timer of device 0000:00:1d.0 to 64
> >ACPI: PCI interrupt 0000:00:1d.1[B] -> GSI 11 (level, low) -> IRQ 11
> >PCI: Setting latency timer of device 0000:00:1d.1 to 64
> >ACPI: PCI interrupt 0000:00:1d.2[C] -> GSI 11 (level, low) -> IRQ 11
> >PCI: Setting latency timer of device 0000:00:1d.2 to 64
> >ACPI: PCI interrupt 0000:00:1d.7[D] -> GSI 11 (level, low) -> IRQ 11
> >PCI: Setting latency timer of device 0000:00:1d.7 to 64
> >ACPI: PCI interrupt 0000:00:1f.1[A] -> GSI 11 (level, low) -> IRQ 11
> >ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 7 (level, low) -> IRQ 7
> >PCI: Setting latency timer of device 0000:00:1f.5 to 64
> >radeonfb: resumed !
> >ACPI: PCI interrupt 0000:02:00.0[A] -> GSI 11 (level, low) -> IRQ 11
> >Fixing swap signatures... ok
> >
> >
> >I also noticed that now, suspend does not work as it should/or was
> when
> >I was using 2.6.7 kernel.
> >
> >Right now, it will go to the Freeing memory section.. then the screen
> >goes blank. (blacklight still on) and then stays that way for quite
> >sometime before I see the words "radeonfb: resumed" and when that
> >happens, it's as though time stood still. It won't start writing to
> the
> >disk until I press a key. (I usually just press the spacebar) and
> then
> >it will start writing and works OK from that. (I am writing on my
> laptop
> >which is been suspended(hibernated) and resumed a few times already.
> >
> >Anything changed or broke??
> >
> >--
> >Ow Mun Heng
> >Fedora GNU/Linux Core 2 on D600 1.4Ghz CPU kernel 2.6.8.1.Aug.29
> >Neuromancer 01:06:38 up 1 day, 3:25, 1 user, load average: 0.73,
> 0.88,
> >0.93
> >
> >
> >-------------------------------------------------------
> >This SF.Net email is sponsored by BEA Weblogic Workshop
> >FREE Java Enterprise J2EE developer tools!
> >Get your free copy of BEA WebLogic Workshop 8.1 today.
> >http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click
> >_______________________________________________
> >Acpi-devel mailing list
> >Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> >https://lists.sourceforge.net/lists/listinfo/acpi-devel
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by BEA Weblogic Workshop
> FREE Java Enterprise J2EE developer tools!
> Get your free copy of BEA WebLogic Workshop 8.1 today.
> http://ads.osdn.com/?ad_idP47&alloc_id\x10808&op=ick
> _______________________________________________
> Acpi-devel mailing list
> Acpi-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/acpi-devel
> 

[-- Attachment #2: wake1.patch --]
[-- Type: text/x-patch, Size: 2615 bytes --]

===== arch/i386/kernel/i8259.c 1.37 vs edited =====
--- 1.37/arch/i386/kernel/i8259.c	2004-09-03 03:20:49 +08:00
+++ edited/arch/i386/kernel/i8259.c	2004-09-15 11:19:10 +08:00
@@ -288,7 +288,34 @@ static int __init i8259A_init_sysfs(void
 	return error;
 }
 
-device_initcall(i8259A_init_sysfs);
+late_initcall(i8259A_init_sysfs);
+
+static int dumy_8259_resume(struct sys_device *dev)
+{
+	outb(0xff, PIC_MASTER_IMR);	/* mask all of 8259A-1 */
+	outb(0xff, PIC_SLAVE_IMR);	/* mask all of 8259A-2 */
+	return 0;
+}
+
+static struct sysdev_class dumy_8259_class = {
+	set_kset_name("dumy_8259"),
+	.resume = dumy_8259_resume,
+};
+
+static struct sys_device dumy_i8259A = {
+	.id	= 0,
+	.cls	= &dumy_8259_class,
+};
+
+static int __init dumy_i8259A_init_sysfs(void)
+{
+	int error = sysdev_class_register(&dumy_8259_class);
+	if (!error)
+		error = sysdev_register(&dumy_i8259A);
+	return error;
+}
+
+fs_initcall(dumy_i8259A_init_sysfs);
 
 void init_8259A(int auto_eoi)
 {
===== arch/i386/kernel/io_apic.c 1.110 vs edited =====
--- 1.110/arch/i386/kernel/io_apic.c	2004-08-31 15:55:07 +08:00
+++ edited/arch/i386/kernel/io_apic.c	2004-09-15 11:18:57 +08:00
@@ -2374,7 +2374,33 @@ static int __init ioapic_init_sysfs(void
 	return 0;
 }
 
-device_initcall(ioapic_init_sysfs);
+late_initcall(ioapic_init_sysfs);
+
+static int dumy_ioapic_resume(struct sys_device *dev)
+{
+	clear_IO_APIC();
+	return 0;
+}
+
+static struct sysdev_class dumy_ioapic_class = {
+	set_kset_name("dumy_ioapic"),
+	.resume = dumy_ioapic_resume,
+};
+
+static struct sys_device dumy_ioapic = {
+	.id	= 0,
+	.cls	= &dumy_ioapic_class,
+};
+
+static int __init dumy_ioapic_init_sysfs(void)
+{
+	int error = sysdev_class_register(&dumy_ioapic_class);
+	if (!error)
+		error = sysdev_register(&dumy_ioapic);
+	return error;
+}
+
+fs_initcall(dumy_ioapic_init_sysfs);
 
 /* --------------------------------------------------------------------------
                           ACPI-based IOAPIC Configuration
===== drivers/acpi/pci_link.c 1.31 vs edited =====
--- 1.31/drivers/acpi/pci_link.c	2004-08-05 03:55:16 +08:00
+++ edited/drivers/acpi/pci_link.c	2004-09-15 11:09:03 +08:00
@@ -714,9 +714,13 @@ irqrouter_resume(
 {
 	struct list_head        *node = NULL;
 	struct acpi_pci_link    *link = NULL;
+	int			flags;
 
 	ACPI_FUNCTION_TRACE("irqrouter_resume");
 
+	local_save_flags(flags);
+	local_irq_enable();
+
 	list_for_each(node, &acpi_link.entries) {
 
 		link = list_entry(node, struct acpi_pci_link, node);
@@ -727,6 +731,7 @@ irqrouter_resume(
 
 		acpi_pci_link_resume(link);
 	}
+	local_irq_restore(flags);
 	return_VALUE(0);
 }