Hello,

first post on LQ. I have a test machine with a NEW Asus P8H61-i and I3 2120 cpu with 8Gb of ram and a Nvidia GT-430.

I have been testing kernels from Mainline to fix the supend issues with Sandy Bridge Platforms.

Here is where I am at, I have Mythbuntu 10.04 loaded and fully updated I have installed kernel 3.3rc2 but tested 3.3rc1 and the daily kernels prior to that. The kernel installs fine other than some realtek firmware issues. I installed the x-swat ppa and update the nvidia driver it loads fine, I updated alsa 1.0.25 and it compiles and installs fine.

So here is the issue.

I installed ir-keytable as the new kernel has the lirc bridge and imon modules in it. reconfigure lirc 0.8.6 to use devinput and devinput/by-id for the imon vfd and to not load the old modules. ir-keytable sees the imon as rc0. restart lirc and I have no response from the remote with irw or ir-keytable -t.

So I dwnld .0.9.0 lirc compile it and install; stop lirc 0.8.6. I ran lirc .0.9.0 from a cli with a --no daemon and created a lircd.pid it loads fine and waits for a client. open up a new terminal and run irw, still no response.

OK the funky part. Now I suspend the system (WORKS NOW on 3.3rc2) and then resume and lirc now works in mythbuntu and in xbmc and I get a full report via irw for what keys I am pressing. but if I restart the system (reboot) then no reponse from lirc until I suspend and resume.

I have looked at the logs for Xorg, pm-suspend, syslog, checked dmesg etc.. I can post some of those snippets here as well if needed.

I also added sysbus wakeup call for device usb 1-1 in rc.local but this did not do anything to help the issues, It seems none of that is required anymore ACPI enables and disable the power button, usb hid ehci etc on suspend and resume.

I am going to start to change some bios setting on the board one at a time and test and then try adding no apci to kernel boot line and see what happens.

not sure where to go to diagnose this issue so any help would be greatly welcomed.

Regards,

Dave

having a quick look in the syslog (tail) via ssh when I suspend the system I see these three lines

Feb 2 07:12:59 fusion kernel: [ 560.092760] pcieport 0000:00:1c.4: wake-up capability enabled by ACPI
Feb 2 07:12:59 fusion kernel: [ 560.108715] ehci_hcd 0000:00:1d.0: wake-up capability enabled by ACPI
Feb 2 07:12:59 fusion kernel: [ 560.124593] ehci_hcd 0000:00:1a.0: wake-up capability enabled by ACPI

This is where the trouble is I think. ehci_hcd is waking up the 1a device which is the imon. BUT it is already awake at least when I issue a cat /proc/acpi/wakeup the usb device for the imon is enabled and as well I added a sysbus wakeup call in my rc.local for the imon which is on usb 1-1 port. and if I reboot and check that it says enabled not disabled.

So I am not sure if something changed in the new kernel for handling wakeup devices or if something else needs to be changed in Addition now.

i have loaded and unloaded modules trying to pinpoint the one that is the issue with no luck so far.

Thanks,

Dave

posted a bug to launchpad, hopefully it will be accepted. The above issues I can reporduce in Lucid, Maverick, Natty and Oneiric immediately after upgrading to any 3.0 > kernel.

https://bugs.launchpad.net/ubuntu/+s...ux/+bug/925524

rgds,

Dave

got it all patched actually it was pretty easy to build the module outside of the kernel with the installed mainline debs. I thought it would have been a all day affair - LOL. here is a link to how I did. Thanks to Jarod Wilson for the patch and a couple of the threads on-line for the Makefile creation.

https://bugs.launchpad.net/ubuntu/+s...524/comments/7

rgds,

Dave