Resume -> higher power drain?
Subsystems (2)
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi!
I just observed a strange phenomenon on my new T42p (latest BIOS). I can successfully suspend and resume to and from S3 and S4! Kernel ist vanilla 2.6.11.8 using ACPI support scripts from the Ubuntu acpi-support package (removes critical modules, does vbe save and restore, etc...)
So far very fine and indeed good work.
Before suspend I usually see a power consumption of about 16W including WLAN, which is nicely low.
But strangely after resume from either S3 or S4 I have see a power consumption of 19W+. This means ~3W more or in other words, roughly 20% less runtime.
This is pretty annoying - not a real show-stoppper but it would be nicer without ;)
I have no idea how to diagnose this. My first guess is that some devices are not resumed correctly but I have no real idea how for example gather device states before and after suspend, or other means to diagnose this.
So, if anyone has an idea how I can diagnose this and help this issue to be solved, let me know. Or is there already a solution floating around?
Thanks!
CU
nils faerber
- --
kernel concepts Tel: +49-271-771091-12
Dreisbachstr. 24 Fax: +49-271-771091-19
D-57250 Netphen Mob: +49-176-21024535
- --
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFCfMhpJXeIURG1qHgRAt/BAKCJezug3PTVtwBwhmXAlowh50V8hQCg3plR 4XsEhFqbj6KggagEH4L0DK4=
=1BGX
-----END PGP SIGNATURE-----
-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Hi,
Quoted 7 lines · Nils Faerber
I just observed a strange phenomenon on my new T42p (latest BIOS). [...]
Before suspend I usually see a power consumption of about 16W including WLAN, which is nicely low.
But strangely after resume from either S3 or S4 I have see a power consumption of 19W+. This means ~3W more or in other words, roughly 20% less runtime.
I didn't measure the power consumption of my Samsung P35, but after resume from S3 it gets hotter under the same load than before suspend. So I can somewhat confirm your report.
Regards,
Carl-Daniel
Signature · 8 lines
http://www.hailfinger.org/ ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Am Samstag, den 07.05.2005, 15:59 +0200 schrieb Carl-Daniel Hailfinger:
Quoted 11 lines
Hi,
Quoted 7 lines · Nils Faerber
I just observed a strange phenomenon on my new T42p (latest BIOS). [...]
Before suspend I usually see a power consumption of about 16W including WLAN, which is nicely low.
But strangely after resume from either S3 or S4 I have see a power consumption of 19W+. This means ~3W more or in other words, roughly 20% less runtime.
Quoted 4 lines
I didn't measure the power consumption of my Samsung P35, but after resume from S3 it gets hotter under the same load than before suspend. So I can somewhat confirm your report.
I can also confirm this behaviour with my acer tm4002lmi. it also has ja power consumption ~16W and after resuming it has round about 20W.
Signature · 5 lines
Hendrik Juergens eMail: hjuerg-ULHALamj7Vx/[email protected] ICQ: #63741909 Jabber: [email protected] PGP: http://speedbur.nnga.info/pubkey.asc
Hi,
Quoted 12 lines · Hendrik Jürgens
Am Samstag, den 07.05.2005, 15:59 +0200 schrieb Carl-Daniel Hailfinger:
Quoted 11 lines
Hi,
Quoted 7 lines · Nils Faerber
I just observed a strange phenomenon on my new T42p (latest BIOS). [...]
Before suspend I usually see a power consumption of about 16W including WLAN, which is nicely low.
But strangely after resume from either S3 or S4 I have see a power consumption of 19W+. This means ~3W more or in other words, roughly 20% less runtime.
Quoted 6 lines
Quoted 4 lines
I didn't measure the power consumption of my Samsung P35, but after resume from S3 it gets hotter under the same load than before suspend. So I can somewhat confirm your report.
I can also confirm this behaviour with my acer tm4002lmi. it also has ja power consumption ~16W and after resuming it has round about 20W.
On a P35, a friend of mine determined that suspend to mem leads to a higher power consumption if X was running. But only when using the ATI driver. When using the vesa driver, the power consumption is increased to a higher rate all the time...
Other than the video driver, you can check whether PCI devices changed their status (they shouldn't) by doing "lspci -vvv > lspci_before.txt" before suspend, "lspci -vvv > lspci_after.txt" after suspend, and checking "diff -u lspci_before.txt lspci_after.txt".
Also, does the bus master activity rates typically seen (check /proc/acpi/processor/*/power every once in a while) differ between pre-suspend and post-suspend?
Dominik
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 4 lines · Dominik Brodowski
On a P35, a friend of mine determined that suspend to mem leads to a higher power consumption if X was running. But only when using the ATI driver. When using the vesa driver, the power consumption is increased to a higher rate all the time...
You can decrease power with the x.org drivers and using the
Option "DynamicClocks" "on"
in the device sectio of xorg.conf. This will enable dynamic clock scaling in the Radeon Fire GL and save quite some power.
Quoted 4 lines
Other than the video driver, you can check whether PCI devices changed their status (they shouldn't) by doing "lspci -vvv > lspci_before.txt" before suspend, "lspci -vvv > lspci_after.txt" after suspend, and checking "diff -u lspci_before.txt lspci_after.txt".
Just did that, no change so it seems PCI is safe.
Quoted 3 lines
Also, does the bus master activity rates typically seen (check
/proc/acpi/processor/*/powerevery once in a while) differ between pre-suspend and post-suspend?
This again is interesting!
What does this bus master activity indicate?
Before suspend I see the CPU sporadically changing from C2 to C3 and back and bus master activity to have values between 00000000 and some higher values on activity, like 00200000 or even f0000000.
After suspend/resume this changed!
The CPU is now fixed at C2 and does not switch between C2 and C3. Also the bus master activity is ffffffff in idle state and changes to strange values on activity, like f7ffffff - quite different than before.
So, good spot!
And where can we go from here?
Is this already the possible cause?
Quoted 1 line
Dominik
CU
nils faerber
- --
kernel concepts Tel: +49-271-771091-12
Dreisbachstr. 24 Fax: +49-271-771091-19
D-57250 Netphen Mob: +49-176-21024535
- --
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFCfN57JXeIURG1qHgRAuKhAJoDzbAK6oN/2l+1/t6h5/mM3adGGQCfUFBy 1uLU7TtifkeGf2DVmKqE8K8=
=hprp
-----END PGP SIGNATURE-----
-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 7 lines · Nils Faerber
Quoted 3 lines · Dominik Brodowski
Quoted 3 lines
Also, does the bus master activity rates typically seen (check
/proc/acpi/processor/*/powerevery once in a while) differ between pre-suspend and post-suspend?This again is interesting!
What does this bus master activity indicate?
Quoted 8 lines
Before suspend I see the CPU sporadically changing from C2 to C3 and back and bus master activity to have values between 00000000 and some higher values on activity, like 00200000 or even
f0000000.After suspend/resume this changed!
The CPU is now fixed at C2 and does not switch between C2 and C3. Also the bus master activity is ffffffff in idle state and changes to strange values on activity, likef7ffffff- quite different than before.
OK, just fixed this ;)
The problem seemed to be related to suspend/resume scripts, so to say. I did enable vbetool to save/restore VESA state and doing vbetool post.
Using s3_bios resume this is not necessary!
And after disabling vbetool usage at all the bus master activity is back to normal again after resume!
So at least for my T42p I can recommend NOT to use vbetool at all for suspend/resume but rather to use s3_bios!
Quoted 1 line
Is this already the possible cause?
Pityly it was not, at least not alone.
The CPU is now cooler again, where I got around 47°C before the fix it is now back to normal 42°C and power consumption is roughly 1W down to 18W, but still 2W more than before.
So, still 2W to fix ;)
Any further ideas?
Quoted 1 line
Quoted 1 line
Quoted 1 line
Dominik
Many thanks so far, really good pointers!
CU
nils faerber
- --
kernel concepts Tel: +49-271-771091-12
Dreisbachstr. 24 Fax: +49-271-771091-19
D-57250 Netphen Mob: +49-176-21024535
- --
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFCfOPvJXeIURG1qHgRArRiAJ9AsYcjijlShoDRXoIG9cr3f+YcKgCfYznK NLz/DjwmAzScMZ0ifac0hC0=
=YIou
-----END PGP SIGNATURE-----
-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Hi!
Quoted 15 lines
I just observed a strange phenomenon on my new T42p (latest BIOS). I can successfully suspend and resume to and from S3 and S4! Kernel ist vanilla 2.6.11.8 using ACPI support scripts from the Ubuntu acpi-support package (removes critical modules, does vbe save and restore, etc...)
So far very fine and indeed good work.Before suspend I usually see a power consumption of about 16W including WLAN, which is nicely low.
But strangely after resume from either S3 or S4 I have see a power consumption of 19W+. This means ~3W more or in other words, roughly 20% less runtime.
This is pretty annoying - not a real show-stoppper but it would be nicer without ;)
Quoted 6 lines
I have no idea how to diagnose this. My first guess is that some devices are not resumed correctly but I have no real idea how for example gather device states before and after suspend, or other means to diagnose this.
So, if anyone has an idea how I can diagnose this and help this issue to be solved, let me know. Or is there already a solution floating around?
Is cpufreq working correctly after resume?
Pavel
Signature · 8 lines
Boycott Kodak -- for their patent abuse against Java. ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Hi!
Quoted 20 lines
Quoted 6 lines
Quoted 3 lines
Quoted 3 lines
Also, does the bus master activity rates typically seen (check
/proc/acpi/processor/*/powerevery once in a while) differ between pre-suspend and post-suspend?This again is interesting!
What does this bus master activity indicate?Quoted 8 lines
Before suspend I see the CPU sporadically changing from C2 to C3 and back and bus master activity to have values between 00000000 and some higher values on activity, like 00200000 or even
f0000000.After suspend/resume this changed!
The CPU is now fixed at C2 and does not switch between C2 and C3. Also the bus master activity is ffffffff in idle state and changes to strange values on activity, likef7ffffff- quite different than before.OK, just fixed this ;)
The problem seemed to be related to suspend/resume scripts, so to say. I did enable vbetool to save/restore VESA state and doing vbetool post.
Quoted 7 lines
Using
s3_biosresume this is not necessary!And after disabling vbetool usage at all the bus master activity is back to normal again after resume!
So at least for my T42p I can recommend NOT to use vbetool at all for suspend/resume but rather to use
s3_bios!
Can you suggest a patch to Documentation/power/video.txt?
Pavel
Signature · 8 lines
Boycott Kodak -- for their patent abuse against Java. ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 1 line · Pavel Machek
Hi!
Hi!
Quoted 5 lines
Quoted 5 lines
The problem seemed to be related to suspend/resume scripts, so to say. I did enable vbetool to save/restore VESA state and doing vbetool post.
Using
s3_biosresume this is not necessary!
Quoted 7 lines
Quoted 5 lines
And after disabling vbetool usage at all the bus master activity is back to normal again after resume!
So at least for my T42p I can recommend NOT to use vbetool at all for suspend/resume but rather to use
s3_bios!Can you suggest a patch to
Documentation/power/video.txt?
I changed the file a little accordingly, a diff is attached.
I think it would be a good idea to start some compatibitlity documentation, i.e. which device is known to work with which options/tricks. I appended a "compatibility matrix" to the end of the file.
Tel me if you find this documentation useful or what can/should be added/changed.
Quoted 1 line
Pavel
CU
nils faerber
- --
kernel concepts Tel: +49-271-771091-12
Dreisbachstr. 24 Fax: +49-271-771091-19
D-57250 Netphen Mob: +49-176-21024535
- --
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFCfUbdJXeIURG1qHgRAvmuAJ4/Q1a7HCdeNmpC3MtsGzQ3Z1Xl0QCdGlmz T5SUaUZCiGtraA/Zc4zIcFY=
=aKR1
-----END PGP SIGNATURE-----
*** linux-2.6.11.8/Documentation/power/video.txt-o Sun May 8 00:48:38 2005
--- linux-2.6.11.8/Documentation/power/video.txt Sun May 8 00:45:58 2005
***************
*** 36,43 ****
Now, if you pass acpi_sleep=something, and it does not work with your bios, you'll get hard crash during resume. Be carefull.
! You may have system where none of above works. At that point you
either invent another ugly hack that works, or write proper driver for your video card (good luck getting docs :-(). Maybe suspending from X (proper X, knowing your hardware, not XF68_FBcon) might have better chance of working.
--- 36,84 ----
Now, if you pass acpi_sleep=something, and it does not work with your bios, you'll get hard crash during resume. Be carefull.
! You may have a system where none of above works. At that point you
either invent another ugly hack that works, or write proper driver for your video card (good luck getting docs :-(). Maybe suspending from X (proper X, knowing your hardware, not XF68_FBcon) might have better chance of working.
+
+
+ Other well known solutions to get your video back on resume using vbetool.
+ Vbetool can execude the real-mode video BIOS extension and use standard
+ video BIOS calls to alter the hardware state of your video hardware. There
+ are two ways to get video to resume:
+
+ * save your current video setting before going to suspend and restore it on
+ resume, like
+ vbetool save > /tmp/.vbe-settings
+ [go to sleep]
+ vbetool restore < /tmp/.vbe-settings
+ On many systems this will bring back your complete video hardware, + including backlight on notebook TFTs. This is known to work with many ATI + radeon systems and Intel Centrino.
+
+ * manually execute the video post method if you BIOS does not do it for you,
+ again vbetool can do this
+ vbetool post
+ This can on some systems already be enough to get your video back.
+
+ Warning: The above should only be tested in a safe environment like a single
+ user shell since it may be necessary to cold-reset your machine, your
+ machine might not resume correctly or simply crash hard.
+
+
+ Compatibility matrix for S3 resume:
+
+ Machine s3_bios s3_mode vbe-save vbe-post recommended
+ -------------------------------------------------------------
+ HP Omnibook xe3s
+ Athlon - - - - no special action required + IBM T42p X ? X X s3_bios, vbe methods result in CPU bus master activity madness + Toshiba 4030cdt X s3_mode
+ Toshiba Satellite
+ P10-554 X X s3_bios,s3_mode
+
+
+ X = working
+ ? = unknown
+ - = not action required
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 1 line · Pavel Machek
Hi!
Hi!
Quoted 8 lines
Quoted 8 lines
Before suspend I usually see a power consumption of about 16W including WLAN, which is nicely low.
But strangely after resume from either S3 or S4 I have see a power consumption of 19W+. This means ~3W more or in other words, roughly 20% less runtime.
This is pretty annoying - not a real show-stoppper but it would be nicer without ;)
Quoted 8 lines
Quoted 6 lines
I have no idea how to diagnose this. My first guess is that some devices are not resumed correctly but I have no real idea how for example gather device states before and after suspend, or other means to diagnose this.
So, if anyone has an idea how I can diagnose this and help this issue to be solved, let me know. Or is there already a solution floating around?
Is cpufreq working correctly after resume?
As to all I can observe, yes.
I have powernowd running and it nicely changes frequency up and down and changes show up in /proc/cpuinfo accordingly (as well as in gkrellm's cpu frequency monitor).
So I guess, yes, it works.
Any other ideas what to check for?
Quoted 1 line
Pavel
CU
nils faerber
- --
kernel concepts Tel: +49-271-771091-12
Dreisbachstr. 24 Fax: +49-271-771091-19
D-57250 Netphen Mob: +49-176-21024535
- --
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFCfUfYJXeIURG1qHgRAoPSAKCBFTkCPSTUyPBCgsV33iwYZSaprQCfWVp/ 6mgly24OKJyMbgcW+twdq54=
=BQFP
-----END PGP SIGNATURE-----
-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Quoted 4 lines · Pavel Machek
Quoted 2 lines
So at least for my T42p I can recommend NOT to use vbetool at all for suspend/resume but rather to use
s3_bios!Can you suggest a patch to
Documentation/power/video.txt?
according to some X guys, vbetool vbestate save/restore is "problematic" at least and should only be used as a measure of last resort, if everything else fails.
That said, a recent X.org radeon driver should post the card, bring the light back on (it does on my Dell D600) and make vbetool unnecessary on ATI radeon chipsets
Signature · 9 lines
Stefan Seyfried ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Hi!
Quoted 13 lines
Quoted 5 lines
Quoted 5 lines
The problem seemed to be related to suspend/resume scripts, so to say. I did enable vbetool to save/restore VESA state and doing vbetool post.
Using
s3_biosresume this is not necessary!Quoted 7 lines
Quoted 5 lines
And after disabling vbetool usage at all the bus master activity is back to normal again after resume!
So at least for my T42p I can recommend NOT to use vbetool at all for suspend/resume but rather to use
s3_bios!Can you suggest a patch to
Documentation/power/video.txt?
Quoted 7 lines
I changed the file a little accordingly, a diff is attached.
I think it would be a good idea to start some compatibitlity documentation, i.e. which device is known to work with which options/tricks. I appended a "compatibility matrix" to the end of the file.Tel me if you find this documentation useful or what can/should be added/changed.
I do not think we should list multiple ways to make it work. Thats just too complicated. Just one "best" method should be listed...
Oh and pleas make patches diff -u...
Pavel
Signature · 8 lines
Boycott Kodak -- for their patent abuse against Java. ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 1 line · Pavel Machek
Hi!
Hi!
[...]
Quoted 8 lines
Quoted 8 lines
Quoted 3 lines
Quoted 2 lines
So at least for my T42p I can recommend NOT to use vbetool at all for suspend/resume but rather to use
s3_bios!Can you suggest a patch to
Documentation/power/video.txt?I changed the file a little accordingly, a diff is attached.
I think it would be a good idea to start some compatibitlity documentation, i.e. which device is known to work with which options/tricks. I appended a "compatibility matrix" to the end of the file.
Quoted 5 lines
Quoted 2 lines
Tel me if you find this documentation useful or what can/should be added/changed.
I do not think we should list multiple ways to make it work. Thats just too complicated. Just one "best" method should be listed...
OK, I removed the extra "X"'es from the table ;)
Quoted 1 line
Oh and pleas make patches diff -u...
Attached ;)
Do you have another idea what might cause the still extra ~2W power drain I see after resume? Or any idea how to trace/find that?
Quoted 1 line
Pavel
Kind regards
nils faerber
- --
kernel concepts Tel: +49-271-771091-12
Dreisbachstr. 24 Fax: +49-271-771091-19
D-57250 Netphen Mob: +49-176-21024535
- --
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFCf0NLJXeIURG1qHgRAjAqAJsExKqTJEz5+gq9sEmEhda82MzOxACeJGFa sRRiYWOBp90nUtUEOLF6/gw=
=FAlU
-----END PGP SIGNATURE-----
Documentation/power/video.txt renamed from Documentation/power/video.txt-o +42 -1
@@ -36,8 +36,49 @@ Now, if you pass acpi_sleep=something, and it does not work with your bios, you'll get hard crash during resume. Be carefull. -You may have system where none of above works. At that point you +You may have a system where none of above works. At that point you either invent another ugly hack that works, or write proper driver for your video card (good luck getting docs :-(). Maybe suspending from X (proper X, knowing your hardware, not XF68_FBcon) might have better chance of working. + + +Other well known solutions to get your video back on resume using vbetool. +Vbetool can execude the real-mode video BIOS extension and use standard +video BIOS calls to alter the hardware state of your video hardware. There +are two ways to get video to resume: + +* save your current video setting before going to suspend and restore it on + resume, like + vbetool save > /tmp/.vbe-settings + [go to sleep] + vbetool restore < /tmp/.vbe-settings + On many systems this will bring back your complete video hardware, + including backlight on notebook TFTs. This is known to work with many ATI + radeon systems and Intel Centrino. + +* manually execute the video post method if you BIOS does not do it for you, + again vbetool can do this + vbetool post + This can on some systems already be enough to get your video back. + +Warning: The above should only be tested in a safe environment like a single +user shell since it may be necessary to cold-reset your machine, your +machine might not resume correctly or simply crash hard. + + +Compatibility matrix for S3 resume: + +Machine s3_bios s3_mode vbe-save vbe-post recommended +------------------------------------------------------------- +HP Omnibook xe3s + Athlon - - - - no special action required +IBM T42p X s3_bios, vbe methods result in CPU bus master activity madness +Toshiba 4030cdt X s3_mode +Toshiba Satellite + P10-554 X X s3_bios,s3_mode + + +X = working +? = unknown +- = not action required
Quoted 16 lines · Nils Faerber
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1Quoted 1 line · Pavel Machek
Hi!
Hi!
[...]
Quoted 8 lines
Quoted 8 lines
Quoted 3 lines
Quoted 2 lines
So at least for my T42p I can recommend NOT to use vbetool at all for suspend/resume but rather to use
s3_bios!Can you suggest a patch to
Documentation/power/video.txt?I changed the file a little accordingly, a diff is attached.
I think it would be a good idea to start some compatibitlity documentation, i.e. which device is known to work with which options/tricks. I appended a "compatibility matrix" to the end of the file.
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines
Tel me if you find this documentation useful or what can/should be added/changed.
I do not think we should list multiple ways to make it work. Thats just too complicated. Just one "best" method should be listed...
OK, I removed the extra "X"'es from the table ;)
Put one on for D600 for S3_Bios.. didn't try S3_mode
but right now, I reverted to using suspend2 instead. 15 secs to hibernate only.
Signature · 13 lines
Ow Mun Heng Gentoo/Linux on DELL D600 1.4Ghz 98% Microsoft(tm) Free!! Neuromancer 19:12:19 up 2:47, 5 users, load average: 0.41, 0.60, 0.71 ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Hi!
Quoted 9 lines
[...]
Quoted 8 lines
Quoted 8 lines
Quoted 3 lines
Quoted 2 lines
So at least for my T42p I can recommend NOT to use vbetool at all for suspend/resume but rather to use
s3_bios!Can you suggest a patch to
Documentation/power/video.txt?I changed the file a little accordingly, a diff is attached.
I think it would be a good idea to start some compatibitlity documentation, i.e. which device is known to work with which options/tricks. I appended a "compatibility matrix" to the end of the file.
Quoted 7 lines
Quoted 5 lines
Quoted 2 lines
Tel me if you find this documentation useful or what can/should be added/changed.
I do not think we should list multiple ways to make it work. Thats just too complicated. Just one "best" method should be listed...
OK, I removed the extra "X"'es from the table ;)
Hmm, sorry for not asking earlier. Do you think you could make it relative to recent version of video.txt? [attached]
Quoted 6 lines
Quoted 1 line
Oh and pleas make patches diff -u...
Attached ;)
Do you have another idea what might cause the still extra ~2W power drain I see after resume? Or any idea how to trace/find that?
You might want to try suspending with minimal drivers. If that does not cure ~2W up, try searching for heat source after resume that was not there before. 2W should be enough to be felt...
Signature · 1 line
Boycott Kodak -- for their patent abuse against Java.
Video issues with S3 resume ~~~~~~~~~~~~~~~~~~~~~~~~~~~ 2003-2005, Pavel Machek
During S3 resume, hardware needs to be reinitialized. For most devices, this is easy, and kernel driver knows how to do it. Unfortunately there's one exception: video card. Those are usually initialized by BIOS, and kernel does not have enough information to boot video card. (Kernel usually does not even contain video card driver -- vesafb and vgacon are widely used).
This is not problem for swsusp, because during swsusp resume, BIOS is run normally so video card is normally initialized. S3 has absolutely no chance of working with SMP/HT. Be sure it to turn it off before testing (swsusp should work ok, OTOH).
There are a few types of systems where video works after S3 resume:
(1) systems where video state is preserved over S3.
(2) systems where it is possible to call the video BIOS during S3 resume. Unfortunately, it is not correct to call the video BIOS at that point, but it happens to work on some machines. Use acpi_sleep=s3_bios.
(3) systems that initialize video card into vga text mode and where the BIOS works well enough to be able to set video mode. Use acpi_sleep=s3_mode on these.
(4) on some systems s3_bios kicks video into text mode, and acpi_sleep=s3_bios,s3_mode is needed.
(5) radeon systems, where X can soft-boot your video card. You'll need a new enough X, and a plain text console (no vesafb or radeonfb). See
http://www.doesi.gmxhome.de/linux/tm800s3/s3.html for more information. Alternatively, you should use vbetool (6) instead.
(6) other radeon systems, where vbetool is enough to bring system back to life. It needs text console to be working. Do vbetool vbestate save > /tmp/delme; echo 3 > /proc/acpi/sleep; vbetool post; vbetool vbestate restore < /tmp/delme; setfont <whatever>, and your video should work.
(7) on some systems, it is possible to boot most of kernel, and then POSTing bios works. Ole Rohne has patch to do just that at
http://dev.gentoo.org/~marineam/patch-radeonfb-2.6.11-rc2-mm2.
Now, if you pass acpi_sleep=something, and it does not work with your bios, you'll get a hard crash during resume. Be careful. Also it is safest to do your experiments with plain old VGA console. The vesafb and radeonfb (etc) drivers have a tendency to crash the machine during resume.
You may have a system where none of above works. At that point you either invent another ugly hack that works, or write proper driver for your video card (good luck getting docs :-(). Maybe suspending from X (proper X, knowing your hardware, not XF68_FBcon) might have better chance of working.
Table of known working systems:
Model hack (or "how to do it")
------------------------------------------------------------------------------
Acer Aspire 1406LC ole's late BIOS init (7), turn off DRI
Acer TM 242FX vbetool (6)
Acer TM C300 vga=normal (only suspend on console, not in X), vbetool (6)
Acer TM 4052LCi s3_bios (2)
Acer TM 636Lci s3_bios vga=normal (2)
Acer TM 650 (Radeon M7) vga=normal plus boot-radeon (5) gets text console back
Acer TM 660 ??? (*)
Acer TM 800 vga=normal, X patches, see webpage (5) or vbetool (6)
Acer TM 803 vga=normal, X patches, see webpage (5) or vbetool (6)
Acer TM 803LCi vga=normal, vbetool (6)
Arima W730a vbetool needed (6)
Asus L2400D s3_mode (3)(***) (S1 also works OK)
Asus L3350M (SiS 740) (6)
Asus L3800C (Radeon M7) s3_bios (2) (S1 also works OK)
Asus M6887Ne vga=normal, s3_bios (2), use radeon driver instead of fglrx in x.org
Athlon64 desktop prototype s3_bios (2)
Compal CL-50 ??? (*)
Compaq Armada E500 - P3-700 none (1) (S1 also works OK)
Compaq Evo N620c vga=normal, s3_bios (2)
Dell 600m, ATI R250 Lf none (1), but needs xorg-x11-6.8.1.902-1
Dell D600, ATI RV250 vga=normal and X, or try vbestate (6)
Dell Inspiron 4000 ??? (*)
Dell Inspiron 500m ??? (*)
Dell Inspiron 600m ??? (*)
Dell Inspiron 8200 ??? (*)
Dell Inspiron 8500 ??? (*)
Dell Inspiron 8600 ??? (*)
eMachines athlon64 machines vbetool needed (6) (someone please get me model #s)
HP NC6000 s3_bios, may not use radeonfb (2); or vbetool (6)
HP NX7000 ??? (*)
HP Pavilion ZD7000 vbetool post needed, need open-source nv driver for X
HP Omnibook XE3 athlon version none (1)
HP Omnibook XE3GC none (1), video is S3 Savage/IX-MV
IBM TP T20, model 2647-44G none (1), video is S3 Inc. 86C270-294 Savage/IX-MV, vesafb gets "interesting" but X work.
IBM TP A31 / Type 2652-M5G s3_mode (3) [works ok with BIOS 1.04 2002-08-23, but not at all with BIOS 1.11 2004-11-05 :-(]
IBM TP R32 / Type 2658-MMG none (1)
IBM TP R40 2722B3G ??? (*)
IBM TP R50p / Type 1832-22U s3_bios (2)
IBM TP R51 none (1)
IBM TP T30 236681A ??? (*)
IBM TP T40 / Type 2373-MU4 none (1)
IBM TP T40p none (1)
IBM TP R40p s3_bios (2)
IBM TP T41p s3_bios (2), switch to X after resume
IBM TP T42 s3_bios (2)
IBM ThinkPad T42p (2373-GTG) s3_bios (2)
IBM TP X20 ??? (*)
IBM TP X30 s3_bios (2)
IBM TP X31 / Type 2672-XXH none (1), use radeontool (http://fdd.com/software/radeon/) to turn off backlight.
IBM Thinkpad X40 Type 2371-7JG s3_bios,s3_mode (4)
Medion MD4220 ??? (*)
Samsung P35 vbetool needed (6)
Sharp PC-AR10 (ATI rage) none (1)
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 Satellite 4030CDT s3_mode (3)
Toshiba Satellite 4080XCDT s3_mode (3) Toshiba Satellite 4090XCDT ??? (*) Toshiba Satellite P10-554 s3_bios,s3_mode (4)(****)
Uniwill 244IIO ??? (*)
(*) from http://www.ubuntulinux.org/wiki/HoaryPMResults, not sure
which options to use. If you know, please tell me.
(***) To be tested with a newer kernel.
(****) Not with SMP kernel, UP only.
VBEtool details
~~~~~~~~~~~~~~~
(with thanks to Carl-Daniel Hailfinger)
First, boot into X and run the following script ONCE:
#!/bin/bash
statedir=/root/s3/state
mkdir -p $statedir
chvt 2
sleep 1
vbetool vbestate save >$statedir/vbe
To suspend and resume properly, call the following script as root:
#!/bin/bash
statedir=/root/s3/state
curcons=fgconsole
fuser /dev/tty$curcons 2>/dev/null|xargs ps -o comm= -p|grep -q X && chvt 2 cat /dev/vcsa >$statedir/vcsa
sync
echo 3 >/proc/acpi/sleep
sync
vbetool post
vbetool vbestate restore <$statedir/vbe
cat $statedir/vcsa >/dev/vcsa
rckbd restart
chvt $[curcons%6+1]
chvt $curcons
Unless you change your graphics card or other hardware configuration, the state once saved will be OK for every resume afterwards.
NOTE: The "rckbd restart" command may be different for your distribution. Simply replace it with the command you would use to set the fonts on screen.
Quoted 3 lines · Stefan Seyfried
That said, a recent X.org radeon driver should post the card, bring the light back on (it does on my Dell D600) and make vbetool unnecessary on ATI radeon chipsets
How recent? Does this driver also turn off the backlight during S3?
Sincerely,
Brix
Signature · 2 lines
Henrik Brix Andersen <[email protected]> Gentoo Linux
Quoted 10 lines · Stefan Seyfried
Quoted 4 lines · Pavel Machek
Quoted 2 lines
So at least for my T42p I can recommend NOT to use vbetool at all for suspend/resume but rather to use
s3_bios!Can you suggest a patch to
Documentation/power/video.txt?according to some X guys, vbetool vbestate save/restore is "problematic" at least and should only be used as a measure of last resort, if everything else fails.
The main advantage of saving/restoring the VBE state is that the code to do so is likely to actually be there, which is something that you can't say about POSTing. As far as I can tell, there shouldn't be any issue in making vbestate calls from text mode, especially in situations where the alternative is to have no graphics functionality at all. It would be interesting to look at video card state to see what the difference is afterwards...
Signature · 9 lines
Matthew Garrett | [email protected] ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Quoted 6 lines · Henrik Brix Andersen
Quoted 3 lines · Stefan Seyfried
That said, a recent X.org radeon driver should post the card, bring the light back on (it does on my Dell D600) and make vbetool unnecessary on ATI radeon chipsets
How recent? Does this driver also turn off the backlight during S3?
Unlikely. At the moment, X doesn't know that the system is suspending.
Signature · 9 lines
Matthew Garrett | [email protected] ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Quoted 9 lines · Stefan Seyfried
Quoted 4 lines · Pavel Machek
Quoted 2 lines
So at least for my T42p I can recommend NOT to use vbetool at all for suspend/resume but rather to use
s3_bios!Can you suggest a patch to
Documentation/power/video.txt?
Quoted 7 lines
according to some X guys, vbetool vbestate save/restore is "problematic" at least and should only be used as a measure of last resort, if everything else fails.
That said, a recent X.org radeon driver should post the card, bring the light back on (it does on my Dell D600) and make vbetool unnecessary on ATI radeon chipsets
A recent X.org radeon driver will switch on the backlight and crash afterwards on my Samsung P35, exactly the thing "vbetool post" does. However, POSTing is not enough for my card, I have to restore the vesa state to actually see something on screen.
Regards,
Carl-Daniel
Signature · 8 lines
http://www.hailfinger.org/ ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 7 lines · Matthew Garrett
Quoted 5 lines · Henrik Brix Andersen
Quoted 3 lines · Stefan Seyfried
That said, a recent X.org radeon driver should post the card, bring the light back on (it does on my Dell D600) and make vbetool unnecessary on ATI radeon chipsets
How recent? Does this driver also turn off the backlight during S3?
Unlikely. At the moment, X doesn't know that the system is suspending.
And additionally switching the light off is what seems to work most reliably on most platforms - just the way back seems to be more critical ;)
CU
nils faerber
- --
kernel concepts Tel: +49-271-771091-12
Dreisbachstr. 24 Fax: +49-271-771091-19
D-57250 Netphen Mob: +49-176-21024535
- --
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFCf2u0JXeIURG1qHgRAh8rAKDP7rABItff09qN+3aelZvfGaKkjACgw127 kOpzujOrjDBwN1Ct5T64PFQ=
=gzy+
-----END PGP SIGNATURE-----
-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.
Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Quoted 3 lines · Ow Mun Heng
Quoted 1 line
OK, I removed the extra "X"'es from the table ;)
Put one on for D600 for
S3_Bios.. didn't tryS3_mode
This is not a generally "known good" solution :-(
All my D600 and D800 reboot instead of resuming with _any_ acpi_sleep= parameter. They only work with vga=normal and the X.org radeon driver (D600) or the binary NVidia driver (D800) brings back the gfx card.
Signature · 9 lines
Stefan Seyfried ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Quoted 6 lines · Henrik Brix Andersen
Quoted 3 lines · Stefan Seyfried
That said, a recent X.org radeon driver should post the card, bring the light back on (it does on my Dell D600) and make vbetool unnecessary on ATI radeon chipsets
How recent?
seife@strolchi:~> X -version
X Window System Version 6.8.2
Release Date: 9 February 2005
Quoted 1 line
Does this driver also turn off the backlight during S3?
Well, i have never had problems with the light not turning off during S3 except for one very crappy SHARP machine, but that one turned it off on lid close, so that was no problem. So usually the problem is getting the light back _on_ after resume :-)
Signature · 9 lines
Stefan Seyfried ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Quoted 5 lines · Matthew Garrett
Quoted 3 lines · Stefan Seyfried
according to some X guys, vbetool vbestate save/restore is "problematic" at least and should only be used as a measure of last resort, if everything else fails.
Again, to be clear: this is not my wisdom but from people who know the radeon chips prettty good (Matthias Hopf, among others)
Quoted 3 lines
The main advantage of saving/restoring the VBE state is that the code to do so is likely to actually be there, which is something that you can't say about POSTing. As far as I can tell, there shouldn't be any issue in
The radeon X driver apparently knows enough about the chips to re-post them. Excerpt from my X log:
(WW) RADEON(0): zero MEMSIZE, probably at D3cold. Re-POSTing via int10. (II) RADEON(0): Primary V_BIOS segment is: 0xc000
Quoted 2 lines
making vbestate calls from text mode, especially in situations where the alternative is to have no graphics functionality at all. It would be
Thats what i meant with "only be used as a measure of last resort" :-)
Quoted 2 lines
interesting to look at video card state to see what the difference is afterwards...
Probably. Note that i have no knowledge about graphics cards at all, i just tell what i have been told ;-)
Signature · 9 lines
Stefan Seyfried ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Quoted 10 lines · Stefan Seyfried
Quoted 3 lines · Matthew Garrett
The main advantage of saving/restoring the VBE state is that the code to do so is likely to actually be there, which is something that you can't say about POSTing. As far as I can tell, there shouldn't be any issue in
The radeon X driver apparently knows enough about the chips to re-post them. Excerpt from my X log:
(WW) RADEON(0): zero MEMSIZE, probably at D3cold. Re-POSTing via int10. (II) RADEON(0): PrimaryV_BIOSsegment is:0xc000
Right, but that's only possible if the video BIOS code is still there. It often isn't, at which point your X server is executing random code. That's generally a bad thing.
Signature · 9 lines
Matthew Garrett | [email protected] ------------------------------------------------------- This SF.Net email is sponsored by: NEC IT Guy Games. Get your fingers limbered up and give it your best shot. 4 great events, 4 opportunities to win big! Highest score wins.NEC IT Guy Games. Play to win an NEC 61 plasma display. Visit http://www.necitguy.com/?r=20
Quoted 19 lines · Stefan Seyfried
Quoted 6 lines · Matthew Garrett
Quoted 3 lines · Stefan Seyfried
according to some X guys, vbetool vbestate save/restore is "problematic" at least and should only be used as a measure of last resort, if everything else fails.
Again, to be clear: this is not my wisdom but from people who know the radeon chips prettty good (Matthias Hopf, among others)
Quoted 3 lines
The main advantage of saving/restoring the VBE state is that the code to do so is likely to actually be there, which is something that you can't say about POSTing. As far as I can tell, there shouldn't be any issue in
Quoted 4 lines
The radeon X driver apparently knows enough about the chips to re-post them. Excerpt from my X log:
(WW) RADEON(0): zero MEMSIZE, probably at D3cold. Re-POSTing via int10. (II) RADEON(0): PrimaryV_BIOSsegment is:0xc000
Can you please ask the gurus: What is the difference between POSTing with the X driver and POSTing with vbetool, if both versions use int10? And can I have the "better int10" of the X driver as an isolated piece of code so that resume will eventually work even if X is not running?
Regards,
Carl-Daniel
Signature · 8 lines
http://www.hailfinger.org/ ------------------------------------------------------- This SF.Net email is sponsored by Oracle Space Sweepstakes Want to be the first software developer in space? Enter now for the Oracle Space Sweepstakes! http://ads.osdn.com/?ad_id=7393&alloc_id=16281&op=click