Bug#501084: Info Requested
Francois, I think this bug might be a side-effect of: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=481766 In the above bug report Bart has prepared a pm-utils hook that might help. Could you try integrating this script on your system and giving it a test drive? Bart may then consider merging this bug or somesuch then. -- Regards, Sheridan Hutchinson sheri...@shezza.org signature.asc Description: OpenPGP digital signature
Bug#501084: Info Requested
Dear Sheridan, On Sunday, December 28, at 16:04, you wrote: I think this bug might be a side-effect of: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=481766 In the above bug report Bart has prepared a pm-utils hook that might help. Could you try integrating this script on your system and giving it a test drive? Bart may then consider merging this bug or somesuch then. It seems to work. However, the situation seems more than fuzzy. Is there a clear reason why this fix is needed ? I could not even figure out what script puts the power management level to 128. The script /etc/init.d/hdparm does not and /etc/acpi/*/90-hdparm.sh seem to to take /etc/laptop-mode/laptop-mode.conf into account as expected ... Regards, -- Francois Fleurethttp://www.idiap.ch/~fleuret -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#501084: Info Requested
Francois Fleuret wrote: It seems to work. However, the situation seems more than fuzzy. Is there a clear reason why this fix is needed ? I could not even figure out what script puts the power management level to 128. The script /etc/init.d/hdparm does not and /etc/acpi/*/90-hdparm.sh seem to to take /etc/laptop-mode/laptop-mode.conf into account as expected ... I think the reason that we have this issue is because the value that the power management value that is set is directly programmed to the hard drive. When the machine suspends or hibernates the drive is powered off and the value is lost. When the machine is powered back up again the drive is loaded with it's default value, which appears in this case is 128. At this point LTM needs to be recalled (which the script you tried does) so it can reapply the power management value to the drive again. At some point LTM will hopefully ship with the pm-util hook (the script) and this will all be transparent to users. -- Regards, Sheridan Hutchinson sheri...@shezza.org signature.asc Description: OpenPGP digital signature
Bug#501084: Info
I'll just clarify that the default value that is loaded is most likely from within the hard drive firmware itself and that this will vary according to each drive. -- Regards, Sheridan Hutchinson sheri...@shezza.org signature.asc Description: OpenPGP digital signature
Bug#501084: Info Requested
Dear Sheridan, On Sunday, December 28, at 18:31, you wrote: I think the reason that we have this issue is because the value that the power management value that is set is directly programmed to the hard drive. When the machine suspends or hibernates the drive is powered off and the value is lost. When the machine is powered back up again the drive is loaded with it's default value, which appears in this case is 128. At this point LTM needs to be recalled (which the script you tried does) so it can reapply the power management value to the drive again. Okay, it seems that it is what happens, at least on my Thinkpad X61s. I removed acpi-support and laptop-mode-tools, put the computer to suspend with 'echo mem /sys/power/state' and when the computer resumed the state was indeed 128 again. BTW, even if /etc/laptop-mode/laptop-mode.conf exists and has been modified, it is removed when removing the package. I don't know if that's a bug or a feature, but it seems inconsistent with the usual behavior of Debian packages. Cheers, -- Francois Fleurethttp://www.idiap.ch/~fleuret -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#501084: Info Requested
Francois Fleuret wrote: Okay, it seems that it is what happens, at least on my Thinkpad X61s. I removed acpi-support and laptop-mode-tools, put the computer to suspend with 'echo mem /sys/power/state' and when the computer resumed the state was indeed 128 again. Good, this seems a nice verification of the process. BTW, even if /etc/laptop-mode/laptop-mode.conf exists and has been modified, it is removed when removing the package. I don't know if that's a bug or a feature, but it seems inconsistent with the usual behavior of Debian packages. I understand that this is desired behaviour, a feature rather than a bug so to speak. If the file was unmodified it would have been removed. -- Regards, Sheridan Hutchinson sheri...@shezza.org signature.asc Description: OpenPGP digital signature
Bug#501084: Info Requested
Dear Sheridan, On Sunday, December 28, at 19:21, you wrote: I understand that this is desired behaviour, a feature rather than a bug so to speak. If the file was unmodified it would have been removed. There must be a misunderstanding. The file /etc/laptop-mode/laptop-mode.conf was removed when the package was removed despite the fact that it *was* modified. Cheers, -- Francois Fleurethttp://www.idiap.ch/~fleuret -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
