Re: Suspend/Resume on Thinkpad x40
On Sat, 22 Aug 2009, Joe Snikeris wrote: > On Fri, Aug 21, 2009 at 11:53 PM, Ian Smith wrote: > > On Fri, 21 Aug 2009, Joe Snikeris wrote: > > > On Thu, Aug 20, 2009 at 10:58 PM, Ian Smith wrote: [..] > > > At any rate, I gave up on ACPI. I've got suspend-to-ram and > > > suspend-to-disk (hibernation) working perfectly using APM. > > > > > > Does anyone know if there are any disadvantages related to > > > power-saving features when using APM over ACPI? Is powerd able to do > > > its job just as well? > > > > The answer to that (at 5.5-STABLE) used to be 'no', but there is some > > APM code in powerd.c, related to how it determines the AC line state, > > though it's not clear to me if it would require compiling APM in kernel. > > > > Certainly /etc/rc.d/power_profile can't set CPU CX states without ACPI. > > Ok. Just did a search to find out what CPU CX states were. For a > good explanation of the CX states: > > http://lists.freebsd.org/pipermail/freebsd-current/2009-May/006436.html Good hunting; a good overview of current load/power handling on FreeBSD, also posted to -mobile and/or -acpi as I recall. > > Switching speeds relies on the dev.cpu.0.freq and dev.cpu.0.freq_levels > > sysctls - are these available when you're running on APM? If so, try > > running powerd(8) in verbose foreground mode (-v) and see what happens. > > Yes they are. And powerd in verbose foreground mode shows that it is > lowering the frequency all the way down to 75 MHz depending on the > load. Cool. At 75MHz, it should be :) cheers, Ian___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Suspend/Resume on Thinkpad x40
On Fri, Aug 21, 2009 at 11:53 PM, Ian Smith wrote: > On Fri, 21 Aug 2009, Joe Snikeris wrote: > > On Thu, Aug 20, 2009 at 10:58 PM, Ian Smith wrote: > > > On Thu, 20 Aug 2009 14:03:29 -0500 Joe Snikeris > wrote: > > > > > > > First off, I apologize if this is not the right forum for this > > > > question. I was torn between posting this in mobile, ACPI, X11 and > > > > here. If I might get a better response in one of those forums, please > > > > let me know and I'll post there instead. > > > > > > I suspect -mobile might be the best list for this one, perhaps -acpi > > > but that's usually more about development than usage. You might try > > > searching the archives of either for mention of the X40. > > > > > > > I'm having some trouble getting the kinks worked out of the > > > > suspend/resume functionality on my laptop, an IBM Thinkpad X40. It is > > > > mostly working now, but I am still experiencing some strange behavior. > > > > I can suspend and resume from a console just fine (except for the > > > > fact that the console comes up blank and only displays new > > > > characters); however, suspending and resuming in X is problematic. > > > > > > > > The first suspend and resume in X works perfectly, but the next time I > > > > hit suspend, the machine locks up while still displaying whatever I > > > > was doing in X. Note that if I switch to a virtual terminal before > > > > hitting suspend, this problem does not occur. Does anyone have any > > > > suggestions on what I might do to get this resolved? The details of > > > > my machine follows; please let me know if any additional information > > > > would be helpful. > > > > > > My T23 requires hw.syscons.sc_no_suspend_vtswitch=1 to suspend/resume > > > cleanly (7.0), so does my old Compaq Armada (but that's APM, not ACPI) > > > > Thanks Ian. This fixed the problem where suspend would hang the > > second time it was executed in X. > > Good to know another model that this works on, for the archives. > > > At any rate, I gave up on ACPI. I've got suspend-to-ram and > > suspend-to-disk (hibernation) working perfectly using APM. > > > > Does anyone know if there are any disadvantages related to > > power-saving features when using APM over ACPI? Is powerd able to do > > its job just as well? > > The answer to that (at 5.5-STABLE) used to be 'no', but there is some > APM code in powerd.c, related to how it determines the AC line state, > though it's not clear to me if it would require compiling APM in kernel. > > Certainly /etc/rc.d/power_profile can't set CPU CX states without ACPI. Ok. Just did a search to find out what CPU CX states were. For a good explanation of the CX states: http://lists.freebsd.org/pipermail/freebsd-current/2009-May/006436.html > > Switching speeds relies on the dev.cpu.0.freq and dev.cpu.0.freq_levels > sysctls - are these available when you're running on APM? If so, try > running powerd(8) in verbose foreground mode (-v) and see what happens. Yes they are. And powerd in verbose foreground mode shows that it is lowering the frequency all the way down to 75 MHz depending on the load. Cool. Thanks again, Ian. Take care, Joe ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Suspend/Resume on Thinkpad x40
On Fri, 21 Aug 2009, Joe Snikeris wrote: > On Thu, Aug 20, 2009 at 10:58 PM, Ian Smith wrote: > > On Thu, 20 Aug 2009 14:03:29 -0500 Joe Snikeris wrote: > > > > > First off, I apologize if this is not the right forum for this > > > question. I was torn between posting this in mobile, ACPI, X11 and > > > here. If I might get a better response in one of those forums, please > > > let me know and I'll post there instead. > > > > I suspect -mobile might be the best list for this one, perhaps -acpi > > but that's usually more about development than usage. You might try > > searching the archives of either for mention of the X40. > > > > > I'm having some trouble getting the kinks worked out of the > > > suspend/resume functionality on my laptop, an IBM Thinkpad X40. It is > > > mostly working now, but I am still experiencing some strange behavior. > > > I can suspend and resume from a console just fine (except for the > > > fact that the console comes up blank and only displays new > > > characters); however, suspending and resuming in X is problematic. > > > > > > The first suspend and resume in X works perfectly, but the next time I > > > hit suspend, the machine locks up while still displaying whatever I > > > was doing in X. Note that if I switch to a virtual terminal before > > > hitting suspend, this problem does not occur. Does anyone have any > > > suggestions on what I might do to get this resolved? The details of > > > my machine follows; please let me know if any additional information > > > would be helpful. > > > > My T23 requires hw.syscons.sc_no_suspend_vtswitch=1 to suspend/resume > > cleanly (7.0), so does my old Compaq Armada (but that's APM, not ACPI) > > Thanks Ian. This fixed the problem where suspend would hang the > second time it was executed in X. Good to know another model that this works on, for the archives. > At any rate, I gave up on ACPI. I've got suspend-to-ram and > suspend-to-disk (hibernation) working perfectly using APM. > > Does anyone know if there are any disadvantages related to > power-saving features when using APM over ACPI? Is powerd able to do > its job just as well? The answer to that (at 5.5-STABLE) used to be 'no', but there is some APM code in powerd.c, related to how it determines the AC line state, though it's not clear to me if it would require compiling APM in kernel. Certainly /etc/rc.d/power_profile can't set CPU CX states without ACPI. Switching speeds relies on the dev.cpu.0.freq and dev.cpu.0.freq_levels sysctls - are these available when you're running on APM? If so, try running powerd(8) in verbose foreground mode (-v) and see what happens. cheers, Ian___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Suspend/Resume on Thinkpad x40
On Thu, Aug 20, 2009 at 10:58 PM, Ian Smith wrote: > On Thu, 20 Aug 2009 14:03:29 -0500 Joe Snikeris wrote: > > > First off, I apologize if this is not the right forum for this > > question. I was torn between posting this in mobile, ACPI, X11 and > > here. If I might get a better response in one of those forums, please > > let me know and I'll post there instead. > > I suspect -mobile might be the best list for this one, perhaps -acpi > but that's usually more about development than usage. You might try > searching the archives of either for mention of the X40. > > > I'm having some trouble getting the kinks worked out of the > > suspend/resume functionality on my laptop, an IBM Thinkpad X40. It is > > mostly working now, but I am still experiencing some strange behavior. > > I can suspend and resume from a console just fine (except for the > > fact that the console comes up blank and only displays new > > characters); however, suspending and resuming in X is problematic. > > > > The first suspend and resume in X works perfectly, but the next time I > > hit suspend, the machine locks up while still displaying whatever I > > was doing in X. Note that if I switch to a virtual terminal before > > hitting suspend, this problem does not occur. Does anyone have any > > suggestions on what I might do to get this resolved? The details of > > my machine follows; please let me know if any additional information > > would be helpful. > > My T23 requires hw.syscons.sc_no_suspend_vtswitch=1 to suspend/resume > cleanly (7.0), so does my old Compaq Armada (but that's APM, not ACPI) Thanks Ian. This fixed the problem where suspend would hang the second time it was executed in X. At any rate, I gave up on ACPI. I've got suspend-to-ram and suspend-to-disk (hibernation) working perfectly using APM. Does anyone know if there are any disadvantages related to power-saving features when using APM over ACPI? Is powerd able to do its job just as well? > > > > uname -a > > FreeBSD sussman.snikeris.com 7.2-RELEASE-p2 FreeBSD 7.2-RELEASE-p2 #0: > > Wed Jun 24 00:57:44 UTC 2009 > > r...@i386-builder.daemonology.net:/usr/obj/usr/src/sys/GENERIC i386 > > > > > cat /boot/loader.conf > > # Disable boot menu > > beastie_disable="YES" > > autoboot_delay=-1 > > > > # Needed for firefox to display certain HTML5 elements > > sem_load="YES" > > > > # Needed so we can resume from suspend w/ a working display. > > hw.acpi.reset_video=1 > > # Needed so mouse will work on resume > > hint.psm.0.flags="0x3000" > > > > # Possibly needed for successful resume > > hint.apic.0.disabled=1 > > That might be one to try either way, with other combinations. > > > # Needed for sound > > snd_ich_load="YES" > > > > # Needed for wireless (iwi) > > if_iwi_load="YES" > > wlan_load="YES" > > firmware_load="YES" > > iwi_bss_load="YES" > > iwi_ibss_load="YES" > > iwi_monitor_load="YES" > > legal.intel_iwi.license_ack=1 > > > > # Needed for 3d graphics acceleration > > agp_load="YES" > > > > # Needed for cdrecord to work > > atapicam_load="YES" > > > > # See 'man acpi_ibm' > > acpi_ibm_load="YES" > > I've found that vesa_load="YES" helps on my T23 especially if suspending > from a VT rather than in X, referring to your 'console comes up blank.' > Something to try anyway, and I don't think it ever hurts. > > > > cat /etc/rc.conf > > # -- sysinstall generated deltas -- # Sat Jul 4 11:10:49 2009 > > # Created: Sat Jul 4 11:10:49 2009 > > # Enable network daemons for user convenience. > > # Please make all changes to this file, not to /etc/defaults/rc.conf. > > # This file now contains just the overrides from /etc/defaults/rc.conf. > > > > # Set by sysinstall > > hostname="sussman.snikeris.com" > > #ifconfig_em0="DHCP" > > linux_enable="YES" > > moused_enable="YES" > > > > # Allow X to locate mouse and keyboard automatically using HAL > > hald_enable="YES" > > dbus_enable="YES" > > > > keyrate="fast" > > keybell="off" > > > > > cat /etc/sysctl.conf > > # $FreeBSD: src/etc/sysctl.conf,v 1.8.32.1 2009/04/15 03:14:26 kensmith > Exp $ > > # > > # This file is read when going to multi-user and its contents piped thru > > # ``sysctl'' to adjust kernel values. ``man 5 sysctl.conf'' for details. > > # > > > > # Uncomment this to prevent users from seeing information about processes > that > > # are being run under another UID. > > #security.bsd.see_other_uids=0 > > > > hw.acpi.standby_state=S0 > > hw.acpi.suspend_state=S3 > > hw.acpi.sleep_button_state=S3 > > vfs.usermount=1 > > hw.acpi.sleep_delay=3 > > If hw.syscons.sc_no_suspend_vtswitch=1 is no help, I'd try -mobile .. > > cheers, Ian > ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
Re: Suspend/Resume on Thinkpad x40
On Thu, 20 Aug 2009 14:03:29 -0500 Joe Snikeris wrote: > First off, I apologize if this is not the right forum for this > question. I was torn between posting this in mobile, ACPI, X11 and > here. If I might get a better response in one of those forums, please > let me know and I'll post there instead. I suspect -mobile might be the best list for this one, perhaps -acpi but that's usually more about development than usage. You might try searching the archives of either for mention of the X40. > I'm having some trouble getting the kinks worked out of the > suspend/resume functionality on my laptop, an IBM Thinkpad X40. It is > mostly working now, but I am still experiencing some strange behavior. > I can suspend and resume from a console just fine (except for the > fact that the console comes up blank and only displays new > characters); however, suspending and resuming in X is problematic. > > The first suspend and resume in X works perfectly, but the next time I > hit suspend, the machine locks up while still displaying whatever I > was doing in X. Note that if I switch to a virtual terminal before > hitting suspend, this problem does not occur. Does anyone have any > suggestions on what I might do to get this resolved? The details of > my machine follows; please let me know if any additional information > would be helpful. My T23 requires hw.syscons.sc_no_suspend_vtswitch=1 to suspend/resume cleanly (7.0), so does my old Compaq Armada (but that's APM, not ACPI) > > uname -a > FreeBSD sussman.snikeris.com 7.2-RELEASE-p2 FreeBSD 7.2-RELEASE-p2 #0: > Wed Jun 24 00:57:44 UTC 2009 > r...@i386-builder.daemonology.net:/usr/obj/usr/src/sys/GENERIC i386 > > > cat /boot/loader.conf > # Disable boot menu > beastie_disable="YES" > autoboot_delay=-1 > > # Needed for firefox to display certain HTML5 elements > sem_load="YES" > > # Needed so we can resume from suspend w/ a working display. > hw.acpi.reset_video=1 > # Needed so mouse will work on resume > hint.psm.0.flags="0x3000" > > # Possibly needed for successful resume > hint.apic.0.disabled=1 That might be one to try either way, with other combinations. > # Needed for sound > snd_ich_load="YES" > > # Needed for wireless (iwi) > if_iwi_load="YES" > wlan_load="YES" > firmware_load="YES" > iwi_bss_load="YES" > iwi_ibss_load="YES" > iwi_monitor_load="YES" > legal.intel_iwi.license_ack=1 > > # Needed for 3d graphics acceleration > agp_load="YES" > > # Needed for cdrecord to work > atapicam_load="YES" > > # See 'man acpi_ibm' > acpi_ibm_load="YES" I've found that vesa_load="YES" helps on my T23 especially if suspending from a VT rather than in X, referring to your 'console comes up blank.' Something to try anyway, and I don't think it ever hurts. > > cat /etc/rc.conf > # -- sysinstall generated deltas -- # Sat Jul 4 11:10:49 2009 > # Created: Sat Jul 4 11:10:49 2009 > # Enable network daemons for user convenience. > # Please make all changes to this file, not to /etc/defaults/rc.conf. > # This file now contains just the overrides from /etc/defaults/rc.conf. > > # Set by sysinstall > hostname="sussman.snikeris.com" > #ifconfig_em0="DHCP" > linux_enable="YES" > moused_enable="YES" > > # Allow X to locate mouse and keyboard automatically using HAL > hald_enable="YES" > dbus_enable="YES" > > keyrate="fast" > keybell="off" > > > cat /etc/sysctl.conf > # $FreeBSD: src/etc/sysctl.conf,v 1.8.32.1 2009/04/15 03:14:26 kensmith Exp $ > # > # This file is read when going to multi-user and its contents piped thru > # ``sysctl'' to adjust kernel values. ``man 5 sysctl.conf'' for details. > # > > # Uncomment this to prevent users from seeing information about processes > that > # are being run under another UID. > #security.bsd.see_other_uids=0 > > hw.acpi.standby_state=S0 > hw.acpi.suspend_state=S3 > hw.acpi.sleep_button_state=S3 > vfs.usermount=1 > hw.acpi.sleep_delay=3 If hw.syscons.sc_no_suspend_vtswitch=1 is no help, I'd try -mobile .. cheers, Ian ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
