Bug 3177

Summary:	S3 resume: garbled video - Asus M2N/855GM
Product:	ACPI	Reporter:	Pedro C (pedro)
Component:	Power-Sleep-Wake	Assignee:	Len Brown (lenb)
Status:	CLOSED PATCH_ALREADY_AVAILABLE
Severity:	normal	CC:	acpi-bugzilla
Priority:	P2
Hardware:	i386
OS:	Linux
Kernel Version:	2.6.7	Subsystem:
Regression:	---	Bisected commit-id:

Description Pedro C 2004-08-09 10:53:16 UTC

Distribution: Debian Sid
Hardware Environment: Asus M2N Centrino Laptop
Software Environment: Kernel 2.6.7 vanilla with ipw2100 driver as module
Problem Description:

The laptop won't return from an S3 sleep state. It seams to suspend fine, but on
resume the screen stays black. When I use the power button for a long time to
shut it down it sometimes won't boot up again until I take the battery off.

Steps to reproduce:

1. echo -n 3 /proc/acpi/sleep
2. Press the power button

I have USB compiled into the kernel, could this be the problem?

Comment 1 Len Brown 2004-08-11 19:06:43 UTC

are you running the latest ACPI patch (or mm tree?)
perhaps it did resume, but the video did not restore and the screen stayed off?
(try accessing it through the net or serial port)

Comment 2 Pedro C 2004-09-06 10:41:06 UTC

This still happens in 2.6.8.1

Comment 3 Maciej Wyka 2004-10-12 14:01:50 UTC

Try acpi_sleep=s3_bios kernel parameter.

Comment 4 Shaohua 2004-10-15 02:39:41 UTC

Does the patch in http://bugme.osdl.org/show_bug.cgi?id=3390 help?

Comment 5 Pedro C 2004-10-16 09:40:30 UTC

I've tried both the acpi_sleep=s3_bios option and the patch and none of them worked.

Comment 6 Venkatesh Pallipadi 2004-10-16 10:55:18 UTC

Are you using FB in your kernel config?

What graphics card do you have on this system (lspci)? Intel/ATI_Radeon/Nvidia

Can you also use "acpi_sleep=s3_bios acpi_slep=s3_mode" boot parameter.

Comment 7 Pedro C 2004-10-16 11:25:21 UTC

The laptop is a centrino with intel 855GM integrated graphics. I'm not using FB.
I'll try that and the second acpi_sleep option.

Comment 8 Pedro C 2004-10-16 11:37:50 UTC

That option does no good and the kernel FB support didn't seam to be for my board.

Comment 9 Pedro C 2004-10-20 15:51:11 UTC

2.6.9 has improved on this. Now it seams it's coming back, with the video
messed. For some reason it does a shutdown after that. Here's the log from
/var/log/messages:

Oct 20 23:44:38 dijkstra kernel: Stopping tasks:
=======================================================|
Oct 20 23:44:38 dijkstra kernel: Warning: CPU frequency is 1300000, cpufreq
assumed 600000 kHz.
Oct 20 23:44:38 dijkstra kernel: ACPI: PCI interrupt 0000:00:02.0[A] -> GSI 5
(level, low) -> IRQ 5
Oct 20 23:44:38 dijkstra kernel: ACPI: PCI interrupt 0000:00:1f.1[A] -> GSI 5
(level, low) -> IRQ 5
Oct 20 23:44:38 dijkstra kernel: ACPI: PCI interrupt 0000:00:1f.5[B] -> GSI 7
(level, low) -> IRQ 7
Oct 20 23:44:38 dijkstra kernel: ACPI: PCI interrupt 0000:01:03.2[C] -> GSI 5
(level, low) -> IRQ 5
Oct 20 23:44:38 dijkstra kernel: eth0: link down
Oct 20 23:44:38 dijkstra kernel: Restarting tasks...<6>eth0: link up, 100Mbps,
full-duplex, lpa 0x45E1
Oct 20 23:44:38 dijkstra kernel:  done
Oct 20 23:44:38 dijkstra shutdown[6440]: shutting down for system halt

I'm guessing this is just some config thing from debian. 

The only real problem seams to be the video that doesn't come back. Any way
around this?

Comment 10 Pedro C 2004-10-29 15:05:54 UTC

Success!

I got this working now. 

I added this option to my X config:

Option          "VBERestore"            "true"

and my suspend script does:

chvt 7
chvt 1
chvt 7

and with this the image is correctly restored. I'm also using vga16fb and the
two acpi kernel command options. I'm going to try without these.

Comment 11 Len Brown 2004-11-18 21:58:04 UTC

marking as resolved b/c there is nothing to debug here,
wondering if others can benefit from this success story.

Comment 12 Venkatesh Pallipadi 2004-11-19 09:11:31 UTC

This "VBERestore" workaround does help on other systems as well. But, as it is 
a bug in the X (i810) driver, I think I will close this bug.