Hi, My Dell laptop, a Latitude D610, no longer resumes with kernel 2.6.25. During resume, there's some hard disk activity, but the screen never turns on. I don't think that the system started normally (ie, that it's only a screen problem) because pressing the power button shortly doesn't power down the laptop. It worked fine with linux-image-2.6.24-1-686 2.6.24-7. It also fails to work with linux-image-2.6.26-rc9-686 2.6.26~rc9-1~experimental.1~snapshot.11825. lspci: 00:00.0 Host bridge: Intel Corporation Mobile 915GM/PM/GMS/910GML Express Processor to DRAM Controller (rev 03) 00:01.0 PCI bridge: Intel Corporation Mobile 915GM/PM Express PCI Express Root Port (rev 03) 00:1c.0 PCI bridge: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) PCI Express Port 1 (rev 03) 00:1d.0 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #1 (rev 03) 00:1d.1 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #2 (rev 03) 00:1d.2 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #3 (rev 03) 00:1d.3 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #4 (rev 03) 00:1d.7 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB2 EHCI Controller (rev 03) 00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev d3) 00:1e.2 Multimedia audio controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) AC'97 Audio Controller (rev 03) 00:1f.0 ISA bridge: Intel Corporation 82801FBM (ICH6M) LPC Interface Bridge (rev 03) 00:1f.2 IDE interface: Intel Corporation 82801FBM (ICH6M) SATA Controller (rev 03) 01:00.0 VGA compatible controller: ATI Technologies Inc M22 [Mobility Radeon X300] 02:00.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5751 Gigabit Ethernet PCI Express (rev 01) 03:01.0 CardBus bridge: Texas Instruments PCI6515 Cardbus Controller 03:01.5 Communication controller: Texas Instruments PCI6515 SmartCard Controller 03:03.0 Network controller: Intel Corporation PRO/Wireless 2200BG Network Connection (rev 05)
reassign 491057 acpi-support 0.109-5 clone 491057 -1 reassign -1 pm-utils 1.1.2.3-1 thanks After more investigation using 2.6.26-rc9: I could suspend and resume using acpi-support's legacy mode. The trick is to shutdown alsa (which acpi-support legacy mode does), and then add the following modules to the exclude list: snd_intel8x0 snd_ac97_codec It should probably be investigated whether those modules should be added in some blacklist in acpi-support (I haven't tested whether snd_intel8x0 or snd_ac97_codec is at fault). (that's why I'm reassigning to acpi-support) I couldn't get hal+pm-utils to work. Apparently, pm-utils doesn't shutdown alsa, so I can't remove the module. Even after shutting down alsa, killing mixer_applet2, and removing the modules manually, the laptop suspends, but doesn't resume (short disk activity, then nothing), when suspended using pm-suspend.
# originally sent to the wrong bug. correcting and sending to #491202 severity 491057 important severity 491202 serious thanks I did more work on that. It suspends/resumes fine with acpi-support's legacy mode. With pm-utils, it still doesn't work. I tried to debug that by building a custom kernel with PM_TRACE enabled. It pointed me to the following kernel modules, successively: serio_raw, psmouse, atkbd. I added those modules to SUSPEND_MODULES, but it still didn't help. By googling almost randomly, I also tried to "unbind" i8042. Still doesn't help. At this point, I'm out of ideas. During the last failure, it apparently generated a Magic number that doesn't match any device: [ 1.246379] registered taskstats [ 1.246662] Magic number: 0:657:383 [ 1.246814] hash matches drivers/base/power/main.c:174 [ 1.254593] rtc_cmos 00:06: setting system clock t o 2028-02-05 14:21:46 UTC (1833373306) I'm not sure of the severity of this bug. My laptop worked fine with etch, using acpi-support, which was sort of the "default" suspend/resume infrastructure. Now, everybody (including acpi-support, by default) is switching to pm-utils, which is the "default" for lenny. So for me at least, it's a clear/severe regression from etch to lenny, hence the severity bump. Also, I'm a bit puzzled that I couldn't find any comment from the maintainer in any of the severity>normal open bugs on pm-utils.
Lucas Nussbaum schrieb: What is the actual problem? Does the display simply stay black, but the machine is otherwise accessible (e.g. via ssh) or does the machine lock up completely? If acpi-support is working for you, could you please post a list loaded modules right before the suspend (or better, the diff between pm-utils and acpi-support). Does your computer require any quirks (lshal | grep quirk)? Can you experiment with a different set of quirks [1]? Cheers, Michael [1] http://people.freedesktop.org/~hughsient/quirk/quirk-suspend-index.html
It locks up completely. I have to power it down by pressing the power button for 10s. Will report on that later. power_management.quirk.no_fb = true (bool) power_management.quirk.vbe_post = true (bool) power_management.quirk.vbestate_restore = true (bool) I tried to play with those earlier, but couldn't reach anything conclusive.
After a lot more testing: acpi-support's legacy mode still fails sometimes (but rarely -- about once every 20 suspend/resumes cycles). It never failed with 2.6.24, so there's a kernel regression somewhere. But using pm-suspend, it fails a lot more frequently. Using pm-suspend --quirk-vbe-post --quirk-vbestate-restore (quirks suggested by lshal | grep quirks), it fails about one third of the times. I also tried with --quirk-vbemode-restore, with no improvement. I think that all of this might be related to the fact that my laptop (Dell Latitude D610) is shipped using either an intel video card, or an ATI one (my case). Maybe the work on this laptop was done by someone with an Intel video card.
tags 491202 moreinfo help thanks Lucas Nussbaum wrote: What about the latest 2.6.26-3 kernel in unstable? Which graphics card is it exactly, what driver do you use (open/closed source)? Do you use a framebuffer (radeonfb)? With ATI cards and the opensource driver I had good experiences with using the s3_bios+s3_mode quirk. Looks very much like a kernel regression and I'm not sure if there is that much that can be done within pm-utils. Nonetheless it would be interesting what acpi-support (legacy mode) does differently so it doesn't fail that often. CCing Bart, maybe he can share his insight. Cheers, Michael
Same problem. 01:00.0 VGA compatible controller: ATI Technologies Inc M22 [Mobility Radeon X300] I use the opensource driver, no framebuffer. Well it used to work very well for me as well ;) -- | Lucas Nussbaum | lucas@lucas-nussbaum.net http://www.lucas-nussbaum.net/ | | jabber: lucas@nussbaum.fr GPG: 1024D/023B3F4F |
severity 491202 important thanks Hi Lucas, I'm downgrading the severity of this bug report back to important. According to your first email, suspend/resume worked fine with 2.6.24, but started to fail with 2.6.25. So it seems obvious to me that this is a kernel regression. Cheers, Michael
Well, I must admit that after lots of tests, I'm not sure anymore of when the regression started. I tried to bisect the kernel, but, since the failure is random, and it sometimes manages to suspend up to 10 times correctly, it's a rather long process. I'll do more tests when I get my internet connection back (I just moved to a new place) and follow up on the bug by then.
Dear submitter, as the package pm-utils has just been removed from the Debian archive unstable we hereby close the associated bug reports. We are sorry that we couldn't deal with your issue properly. For details on the removal, please see https://bugs.debian.org/1058701 The version of this package that was in Debian prior to this removal can still be found using https://snapshot.debian.org/. Please note that the changes have been done on the master archive and will not propagate to any mirrors until the next dinstall run at the earliest. This message was generated automatically; if you believe that there is a problem with it please contact the archive administrators by mailing ftpmaster@ftp-master.debian.org. Debian distribution maintenance software pp. Thorsten Alteholz (the ftpmaster behind the curtain)