Screen slides down after STR / neomagic
Hello,
Toshiba Libretto; Every time I suspend to RAM an come back to Console or later exit Xorg (it's ok within X), the screen is somewhat displaced downward:
physical screen: video output:
----------------------- - - - - - - - - - -
| | -----------------------
| | |$ |
| | | |
| | | |
----------------------- - - | - - - - - -|
----------------------- snippedFirst noticed this with IIRC kernel 2.6.12, quite a while ago, and now want to do something about it. I don't use (it doesn't occur with) the (neomagic) framebuffer videodriver. Bootoption is vga=5, but choosing another doesn't fix it.
I also recognize previous versions of XFree and IIRC pre 6.9 Xorg thinking about screen geometry being 800x600, while in fact it's 800x480. These 60/120 pixels(?) are about the amount of space (if you think of centering output to screen vertically) the display is misplaced. However, it also happens if I don't startup X at all.
I also saw this: If I suspend, there are some lines printed about the suspend process and if they hit bottom line and start scrolling (after _a lot_ of suspends) screen layout seems to get fixed. sometimes? hmm.
What can I do about it? Add a few lines printk's into the suspend code? ;-)
http://www.mittendorfer.com/rm/temp/config.txt
http://www.mittendorfer.com/rm/temp/dmesg.txt
0000:00:04.0 VGA compatible controller: Neomagic Corporation NM2160 [MagicGraph 128XD] (rev 01)
Thanks for some input,
sl ritch
Quoted 15 lines · Richard Mittendorfer
Hello,
Toshiba Libretto; Every time I suspend to RAM an come back to Console or later exit Xorg (it's ok within X), the screen is somewhat displaced downward:
physical screen: video output:
----------------------- - - - - - - - - - -| | ----------------------- | | |$ | | | | | | | | | ----------------------- - - | - - - - - -| ----------------------- snipped
Quoted 4 lines
First noticed this with IIRC kernel 2.6.12, quite a while ago, and now want to do something about it. I don't use (it doesn't occur with) the (neomagic) framebuffer videodriver. Bootoption is vga=5, but choosing another doesn't fix it.
Did you read Doc*/power/video.txt?
Pavel
Signature · 1 line
Web maintainer for suspend.sf.net (www.sf.net/projects/suspend) wanted...
Also sprach Pavel Machek <[email protected]> (Tue, 7 Mar 2006 22:43:37 +0100):
Quoted 8 lines · Richard Mittendorfer
Quoted 5 lines
Hello,
Toshiba Libretto; Every time I suspend to RAM an come back to Console or later exit Xorg (it's ok within X), the screen is somewhat displaced downward:
Did you read
Doc*/power/video.txt?
Oh, wasn't aware of this file. (Havn't looked there for a while now.) Now I know what went wrong. :-)
Finally the vbetool trick did it.
thx, ritch
Quoted 9 lines · Richard Mittendorfer
Also sprach Pavel Machek <[email protected]> (Tue, 7 Mar 2006 22:43:37 +0100):
Quoted 7 lines · Richard Mittendorfer
Quoted 5 lines
Hello,
Toshiba Libretto; Every time I suspend to RAM an come back to Console or later exit Xorg (it's ok within X), the screen is somewhat displaced downward:
Quoted 6 lines
Quoted 1 line
Did you read
Doc*/power/video.txt?Oh, wasn't aware of this file. (Havn't looked there for a while now.) Now I know what went wrong. :-)
Finally the vbetool trick did it.
Could you
1) try to find out if acpi_sleep=* options can fix it too (they are better for debugging)
2) submit patch for video.txt
3) if possible, download s2ram from www.sf.net/projects/suspend (from CVS) and add your machine to whitelist?
Pavel
Signature · 1 line
Web maintainer for suspend.sf.net (www.sf.net/projects/suspend) wanted...
Also sprach Pavel Machek <[email protected]> (Wed, 8 Mar 2006 12:01:02 +0100):
Quoted 15 lines · Richard Mittendorfer
Quoted 14 lines
Also sprach Pavel Machek <[email protected]> (Tue, 7 Mar 2006 22:43:37 +0100):
Quoted 8 lines · Richard Mittendorfer
Quoted 5 lines
Hello,
Toshiba Libretto; Every time I suspend to RAM an come back to Console or later exit Xorg (it's ok within X), the screen is somewhat displaced downward:
Did you read
Doc*/power/video.txt?Oh, wasn't aware of this file. (Havn't looked there for a while now.) Now I know what went wrong. :-)
Quoted 6 lines
Quoted 1 line
Finally the vbetool trick did it.
Could you
1) try to find out if acpi_sleep=* options can fix it too (they are better for debugging)
acpi_sleep=s3_bios|s3_mode|s3_bios,s3_mode didn't do it.
Quoted 1 line
2) submit patch for video.txt
#include <dontbeatme.h> =)
This patch adds an entry for handling video bios resets during s3 suspends on toshiba's 110ct/100ct neomagic chip. Tested on 110ct; 100ct should have the same chipset.
--- linux-2.6.15/Documentation/power/video.txt.orig 2006-03-08
12:35:33.000000000 +0100
+++ linux-2.6.15/Documentation/power/video.txt 2006-03-08
12:36:40.000000000 +0100
@@ -130,6 +130,7 @@ Sony Vaio PCG-F403 ??? (*)
Sony Vaio PCG-N505SN ??? (*) Sony Vaio vgn-s260 X or boot-radeon can init it (5) Toshiba Libretto L5 none (1) +Toshiba Libretto 100CT/110CT none, see "VBEtool details" below Toshiba Satellite 4030CDT s3_mode (3)
Toshiba Satellite 4080XCDT s3_mode (3)
Toshiba Satellite 4090XCDT ??? (*)
Quoted 2 lines
3) if possible, download s2ram from www.sf.net/projects/suspend (from CVS) and add your machine to whitelist?
I'll look into this.
Quoted 1 line
Pavel
sl ritch