Problems With 'apm -s' And Xwindows

[danr]

Has anyone else had any problems executing 'apm -s' from X?  I have found that when I do this, upon wakeup X runs really slowly, and I get a load of errors about timeouts with read and write pointers.  If I use the suspend button, I have no problems.  Similarly, running 'apm -s' from the console wakes up with no problems.

So it appears that pressing the suspend button does something more than just call apm.x, and this extra behaviour prevents the X problem from occurring.  Does anyone know anything about this?  Does it do something with the framebuffer?

I'd like to get this working so that I can control automatic suspends if I wakeup the Z in the night to do some night-time processing!

I'm running RC12 on an SL-860.

Dan

[pgas]

'xset dpms force off' ?

[danr]

'xset dpms force off' ?
[div align=\"right\"][a href=\"index.php?act=findpost&pid=135391\"][{POST_SNAPBACK}][/a][/div]

That does the trick  - I never knew that X has its own power management.  I'm still curious to know why the power off button works both for console and X, however.

Dan

[karlto]

I think I recall reading about this problem in an earlier thread, and the solution above from pgas was the answer there too.

All the apm stuff is actually in the kernel, so I guess the userland programs are just interfacing with that, rather than apm and xset actually doing any power management. I would presume that pressing the power button would also shortcut straight to the kernel apm...