Hi Kapil, I've checked some things out now regarding your suspend issues. Here's my comments:
On Mon, December 31, 2007 12:09, Kapil Hari Paranjape wrote: > 1. I no longer need to use "SAVE_VIDEO_PCI_STATE=true" > which was required with "etch". That's probably an issue with upgraded X drivers or some such -- I don't think our handling has changed, but the drivers might have changed. > 2. I ensured that laptop-mode is switched off as the first > step of suspend.d by creating 01-laptop-mode-stop.sh > which runs "invoke-rc.d laptop-mode stop". OK, I've added that. No harm in that as far as I'm concerned. > 3. I ensured that laptop-mode is re-invoked as the *last* > step of the resume.d by creating 95-laptop-mode-start.sh > which runs "invoke-rc.d laptop-mode restart". That was actually already there! :-) > After these three steps suspend-to-ram was working fine but > suspend-to-disk was failing. Then I found that laptop-mode was being > restarted at the *first* step of resumption from suspend-to-disk in > /etc/acpi/hibernate.sh. > > 4. I commented out the line "invoke-rc.d laptop-mode restart" > from the script /etc/acpi/hibernate.sh OK, I've taken that out. It also runs the resume.d scripts, so we shouldn't need that anyway. These changes will be in the next upload, I'll mark the bug fixed but please reopen the bug if I missed anything! > However, I strongly suspect that it is the invocation of laptop-mode > that caused the earlier regression. This makes the following > statements at the end of /etc/default/acpi-support somewhat > mis-leading: > > # Note: to enable "laptop mode" (to spin down your hard drive for longer > # periods of time), install the laptop-mode-tools package and configure > # it in /etc/laptop-mode/laptop-mode.conf. > # > # (Note to upgraders: earlier versions of the acpi-support package > contained > # an option to enable/disable laptop mode. This option has never > actually > # worked, and for that reason it has been removed.) Well, they're true words. They don't refer to the things you've been tinkering with, they refer to the option that was present in earlier acpi-support versions to start laptop mode on battery and stop it on AC power. This has nothing to do with the suspend/resume support. I suspect you find these words misleading because you now see only this laptop mode interaction, and you'd expect that laptop mode stuff would have been removed completely. Is that correct? > Possible reasons: > > "laptop-mode-tools" make a number of changes to the system and some of > them > may be interfering with resume. One would need to break down the > process further to figure out the problem: > > 1. hdparm > 2. cpufreq settings > 3. kernel "laptop-mode" settings > > Perhaps, if someone has time this could be investigated further. It would be possible to pin this down, simply by disabling options in laptop-mode.conf one by one. It would be you who would have to do it though, so, do you have time? ;-) Cheers, Bart
