Bugzilla – Bug 357598
CPU frequency scaling not working after upgrade to latest kernel (2.6.22.16-0.1-default)
Last modified: 2008-02-16 11:31:05 UTC
Hi, today after installing lastest kernel update (2.6.22.16-0.1-default) CPU frequency scaling stop working. Here is the output from dmesg: Jan 31 13:46:28 diablo syslog-ng[2556]: syslog-ng version 1.6.12 starting Jan 31 13:46:30 diablo rchal: CPU frequency scaling is not supported by your processor. Jan 31 13:46:30 diablo rchal: boot with 'CPUFREQ=no' in to avoid this warning. Jan 31 13:46:30 diablo rchal: Cannot load cpufreq governors - No cpufreq driver available Jan 31 13:46:33 diablo kernel: klogd 1.4.1, log source = /proc/kmsg started. Jan 31 13:46:33 diablo kernel: powernow: This module only works with AMD K7 CPUs I had to install previous kernel version 2.6.22.13-0.3 to solve this problem. This problem has occurred on my notebook HP Compaq nc6320. Regards, Lubos Klokner
The same problem on my Lenovo 3000 N 100 (BYG 0768). After updating to kernel 2.6.22.16-0.1-default frequency scaling dont work throughout 100 %.
Same problem on my DELL XPS M1330 (Centrino Duo). Just to understand, did someone tested the kernel before releasing it? It's one of the now too many times a patch causes more serious issues than what it fixes. I hope in an urgent fix, and I increase the severity. With kind regards, Alberto
Same here, ThinkPad R61 8919-6VG, Intel Core 2 Duo CPU T7300 @ 2.00GHz stepping 0a. I found some strange lines in the output of dmesg: ACPI Error (dsopcode-0250): No pointer back to NS node in buffer obj dfb55254 [20070126] ACPI Error (psparse-0537): Method parse/execution failed [\_PR_.CPU0._PDC] (Node dffc7658), AE_AML_ ACPI: CPU0 (power states: C1[C1] C2[C2] C3[C3]) ACPI: Processor [CPU0] (supports 8 throttling states) ACPI Error (dsopcode-0250): No pointer back to NS node in buffer obj dff81cf4 [20070126] ACPI Error (psparse-0537): Method parse/execution failed [\_PR_.CPU1._PDC] (Node dffc7608), AE_AML_ ACPI: CPU1 (power states: C1[C1] C2[C2] C3[C3]) ACPI: Processor [CPU1] (supports 8 throttling states)
Same here: Samsung R65 (Centrino Duo)
Same on my Benq Joybook R55 g22 <4>Using IPI No-Shortcut mode <6>Freeing unused kernel memory: 228k freed <4>ACPI Error (dsopcode-0250): No pointer back to NS node in buffer obj dfb6d5ec [20070126] <4>ACPI Error (psparse-0537): Method parse/execution failed [\_PR_.CPU0._PDC] (Node dffe0694), AE_AML_INTERNAL <6>ACPI: CPU0 (power states: C1[C1] C2[C2] C3[C3]) <6>ACPI: Processor [CPU0] (supports 8 throttling states) <4>ACPI Error (dsopcode-0250): No pointer back to NS node in buffer obj dfb6d63c [20070126] <4>ACPI Error (psparse-0537): Method parse/execution failed [\_PR_.CPU1._PDC] (Node dffe0644), AE_AML_INTERNAL <6>ACPI: CPU1 (power states: C1[C1] C2[C2] C3[C3]) <6>ACPI: Processor [CPU1] (supports 8 throttling states) <4>Marking TSC unstable due to: possible TSC halt in C2.
Same on Fujitsu Siemens Celsius H240
same problem on my hp pavilion dv5194ea.. centrino duo
*** Bug 357622 has been marked as a duplicate of this bug. ***
*** Bug 357516 has been marked as a duplicate of this bug. ***
I've got the same problem on a HP Compaq nx7400.. Core2Duo I hope no one minds that I marked the bugs above as duplicate of this one .. (If a regular dude like me shouldnt do such stuff, please tell me..) Output from hwinfo --cpu 01: None 00.0: 10103 CPU [Created at cpu.301] Unique ID: rdCR.j8NaKXDZtZ6 Hardware Class: cpu Arch: X86-64 Vendor: "GenuineIntel" Model: 6.15.6 "Intel(R) Core(TM)2 CPU T7200 @ 2.00GHz" Features: fpu,vme,de,pse,tsc,msr,pae,mce,cx8,apic,sep,mtrr,pge,mca,cmov,pat,pse36,clflush,dts,acpi,mmx,fxsr,sse,sse2,ss,ht,tm,syscall,nx,lm,constant_tsc,pni,monitor,ds_cpl,vmx,est,tm2,ssse3,cx16,xtpr,lahf_lm Clock: 1994 MHz BogoMips: 3994.41 Cache: 4096 kb Units/Processor: 2 Config Status: cfg=new, avail=yes, need=no, active=unknown 02: None 01.0: 10103 CPU [Created at cpu.301] Unique ID: wkFv.j8NaKXDZtZ6 Hardware Class: cpu Arch: X86-64 Vendor: "GenuineIntel" Model: 6.15.6 "Intel(R) Core(TM)2 CPU T7200 @ 2.00GHz" Features: fpu,vme,de,pse,tsc,msr,pae,mce,cx8,apic,sep,mtrr,pge,mca,cmov,pat,pse36,clflush,dts,acpi,mmx,fxsr,sse,sse2,ss,ht,tm,syscall,nx,lm,constant_tsc,pni,monitor,ds_cpl,vmx,est,tm2,ssse3,cx16,xtpr,lahf_lm Clock: 1994 MHz BogoMips: 3990.16 Cache: 4096 kb Units/Processor: 2 Config Status: cfg=new, avail=yes, need=no, active=unknown
I wrote it in my duplicate bug report: The following modules are missing in the update kernel: speedstep-centrino speedstep-ich speedstep-lib speedstep-smi
Ohh well, that is my fault. I accidently removed the centrino driver for x86_64. Instead of only removing an old flag: # CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI is not set this one also was unset accidently: CONFIG_X86_SPEEDSTEP_CENTRINO=m Can you try this one: ftp.suse.com/pub/people/trenn/10.3_cpufreq_centrino_x86_64 I hope there are no dependencies and the config got reset again, please double check with: zcat /proc/config.gz |less and search for CONFIG_X86_SPEEDSTEP_CENTRINO, it should be set to =m. Matthias: speedstep-centrino speedstep-ich speedstep-lib speedstep-smi speedstep-ich and speedstep-smi are old drivers and should not be available in x86_64 kernels anyway. But speedstep-centrino and speedstep-lib should... Comment #3 and #5 look like there is another bug, maybe triggered by the kernel config change? Maybe this happens when the acpi cpufreq driver is tried to be loaded? Do these ACPI error messages go away with the new kernel?
Hi, Same here with HP nx9420 Core Duo T2400 and 32bit default kernel... rchal: CPU frequency scaling is not supported by your processor. Regards. Philippe.
Hi, i've got the same problem on a Fujitsu-Siemens Amilo Pi 1556 and a 32bit Genuine Intel(R) CPU T2400 @ 1.83GHz. <Output of dmesg|grep ACPI> ACPI Error (dsopcode-0250): No pointer back to NS node in buffer obj dfe09d1c [20070126] ACPI Error (psparse-0537): Method parse/execution failed [\_PR_.CPU0._PDC] (Node dffe2824), AE_AML_INTERNAL ACPI: CPU0 (power states: C1[C1] C2[C2] C3[C3]) ACPI: Processor [CPU0] (supports 8 throttling states) ACPI Error (dsopcode-0250): No pointer back to NS node in buffer obj dfe09d94 [20070126] ACPI Error (psparse-0537): Method parse/execution failed [\_PR_.CPU1._PDC] (Node dffe27d4), AE_AML_INTERNAL ACPI: CPU1 (power states: C1[C1] C2[C2] C3[C3]) ACPI: Processor [CPU1] (supports 8 throttling states) </Output of dmesg|grep ACPI> Regard, Thomas
(In reply to comment #12 from Thomas Renninger) > Can you try this one: > ftp.suse.com/pub/people/trenn/10.3_cpufreq_centrino_x86_64 I've tested it on my NEC VERSA M360 laptop and I get at boot time : frecb006692:~ # grep -i freq /var/log/messages Feb 1 15:15:29 frecb006692 rchal: CPU frequency scaling is not supported by your processor. Feb 1 15:15:29 frecb006692 rchal: boot with 'CPUFREQ=no' in to avoid this warning. Feb 1 15:15:29 frecb006692 rchal: Cannot load cpufreq governors - No cpufreq driver available Feb 1 15:15:40 frecb006692 kernel: ipw3945: Radio Frequency Kill Switch is On: Feb 1 15:15:43 frecb006692 powersaved[3291]: WARNING (CpufreqManagement:51) No capability cpufreq_control Feb 1 15:15:43 frecb006692 powersaved[3291]: WARNING (CpufreqManagement:51) No capability cpufreq_control Knowing that on my laptop, the module that should drive the dynamic cpu frequency management is acpi_cpufreq, here is the result of trying to insert it: frecb006692:~ # modprobe acpi_cpufreq FATAL: Error inserting acpi_cpufreq (/lib/modules/2.6.22.16-0.1-default/kernel/arch/x86_64/kernel/cpufreq/acpi-cpufreq.ko): No such device
(In reply to comment #12 from Thomas Renninger) > Ohh well, that is my fault. > I accidently removed the centrino driver for x86_64. > Instead of only removing an old flag: > # CONFIG_X86_SPEEDSTEP_CENTRINO_ACPI is not set > > this one also was unset accidently: > CONFIG_X86_SPEEDSTEP_CENTRINO=m > > Can you try this one: > ftp.suse.com/pub/people/trenn/10.3_cpufreq_centrino_x86_64 > I hope there are no dependencies and the config got reset again, please double > check with: > zcat /proc/config.gz |less > and search for CONFIG_X86_SPEEDSTEP_CENTRINO, it should be set to =m. As a matter of fact is it! But anyway I still there is no cpufreq support, at least from what I see in kpowersave and what I saw during bootup :( > > Do these ACPI error messages go away with the new kernel? > ACPI Error (dsopcode-0250): No pointer back to NS node in buffer obj ffff81007cc35160 [20070126] ACPI Error (psparse-0537): Method parse/execution failed [\_PR_.CPU0._PDC] (Node ffff810037fc5690), AE_AML_INTERNAL ACPI: Processor [CPU0] (supports 8 throttling states) ACPI Error (dsopcode-0250): No pointer back to NS node in buffer obj ffff810037f8fce0 [20070126] ACPI Error (psparse-0537): Method parse/execution failed [\_PR_.CPU1._PDC] (Node ffff810037fc5610), AE_AML_INTERNAL ACPI: Processor [CPU1] (supports 8 throttling states)
Its happening in my Dell Latitude D620 with the Intel T2300 1.66G Core2Duo. 32-bit.
Just for kicks I tried a modprobe on all the speedstep modules. They all failed (no such device) except speedstep-lib. That was on an Intel T2300 (32bit Core2Duo) which has 8 stepping levels according to Intel's brag sheet.
Same in Dell 640m (with 64 bit Intel T7200 Core2Duo) and OpenSUSE 32 bit version. How can I downgrade the kernel??? Thanks!
Wow... I never had a bug where that much people answering that fast... I was wrong and kernel config is right. speedstep-centrino must not exist on x86_64 as it only supports some Pentium M i386 machines for some time. I expect an ACPI bug got introduced, could you try these two kernels, one should work, the other one not. Please try first one of these: ftp.suse.com/pub/people/trenn/10.3_cpufreq_without_sizeOf_fix then one of these: ftp.suse.com/pub/people/trenn/10.3_cpufreq_without_psd_fix You can stop if the first one works (what I expect..). > How can I downgrade the kernel??? It's only cpufreq not working, or is it worse? Just wait some days and you can fetch a fixed 2.6.22 kernel from here: http://ftp.suse.com/pub/projects/kernel/kotd/SL103_BRANCH/
(In reply to comment #20 from Thomas Renninger) > > How can I downgrade the kernel??? > It's only cpufreq not working, or is it worse? > Just wait some days and you can fetch a fixed 2.6.22 kernel from here: > http://ftp.suse.com/pub/projects/kernel/kotd/SL103_BRANCH/ How long till we get a fix through the updater?
(In reply to comment #20 from Thomas Renninger) > Wow... > I never had a bug where that much people answering that fast... Hehe > I expect an ACPI bug got introduced, could you try these two kernels, one > should work, the other one not. > Please try first one of these: > ftp.suse.com/pub/people/trenn/10.3_cpufreq_without_sizeOf_fix > then one of these: > ftp.suse.com/pub/people/trenn/10.3_cpufreq_without_psd_fix > > You can stop if the first one works (what I expect..). FYI: The first one did indeed work for me :) If there are specific information you could need, I'll happily provide them. Thanks for your quick work on this :)
Hi Thomas, thanks for your help. Yes, the first kernel fix the problem (ftp.suse.com/pub/people/trenn/10.3_cpufreq_without_sizeOf_fix). It's a long time that I don't use RPM distros, so when I typed (the first time) sudo rpm -Uvh /home/pascal/Desktop/kernel-default-2.6.22.16-0.1.i586.rpm i got this: Preparing... ########################################### [100%] package kernel-default-2.6.22.16-0.1 is already installed file /boot/System.map-2.6.22.16-0.1-default from install of kernel-default-2.6.22.16-0.1 conflicts with file from package kernel-default-2.6.22.16-0.1 file /boot/symsets-2.6.22.16-0.1-default.tar.gz from install of kernel-default-2.6.22.16-0.1 conflicts with file from package kernel-default-2.6.22.16-0.1 file /boot/symtypes-2.6.22.16-0.1-default.gz from install of kernel-default-2.6.22.16-0.1 conflicts with file from package kernel-default-2.6.22.16-0.1 file /boot/symvers-2.6.22.16-0.1-default.gz from install of kernel-default-2.6.22.16-0.1 conflicts with file from package kernel-default-2.6.22.16-0.1 file /boot/vmlinux-2.6.22.16-0.1-default.gz from install of kernel-default-2.6.22.16-0.1 conflicts with file from package kernel-default-2.6.22.16-0.1 file /boot/vmlinuz-2.6.22.16-0.1-default from install of kernel-default-2.6.22.16-0.1 conflicts with file from package kernel-default-2.6.22.16-0.1 after I tried with this: sudo rpm -Uvh /home/pascal/Desktop/kernel-default-2.6.22.16-0.1.i586.rpm --force and all was fine. Just a little question, will I have problems in the next kernel upgrade???
(In reply to comment #20 from Thomas Renninger) > Please try first one of these: > ftp.suse.com/pub/people/trenn/10.3_cpufreq_without_sizeOf_fix > yep, the first one fixed it. vielen dank Thomas
Argh, a new problem arose with that first kernel: My Intel PRO/Wireless 3945ABG driven by ipw3945 does not survive a suspend to ram anymore. I know, that in the release notes of 10.3 it is said, that the ipw driver does not support suspend, but in my case suspend to ram did work, as it resumed and reconnected to network with no user interaction required, with the last kernels, even with the cpufreq-problem kernel and even with the one from ftp://ftp.suse.com/pub/people/trenn/10.3_cpufreq_centrino_x86_64/ (see comment 12). (btw: wlan didnt survive a suspend to disk - had to rcnetwork restart to reanimate the network to function).
Comment #23, Pascal: Using --force here is fine. There are files which had to be overridden, you do not come around it... Comment #25, Alexander: I doubt this has anything to do with this kernel, it pretty much is the same than the update kernel (may have got some very few add-on fixes meanwhile which will go into next update kernel). Patch has been ripped out. An automatic rebuild should get triggered and you should find a kernel (may take a day or more) here soon: http://ftp.suse.com/pub/projects/kernel/kotd/SL103_BRANCH/ if you do a: rpm --qp --changelog kernel-xy.rpm |less you should find this entry: - patches.fixes/acpica_sizeof.patch: Delete.
Automatic rebuild - does this mean that it should make it into the Online Updates system? Just call me lazy - I'd like to do it through zypper rather than manually.
We will be making another kernel update for 10.3 in the next week or two, including this regression fix.
@Thomas: Using 10.3_cpufreq_without_sizeOf_fix, the frequency scaling is back. Thanks!
Just to add some information. I've also noticed this error : Marking TSC unstable due to TSC unsynchronised time.c detected 2393.999 Mhz. With previous release of the kernel there's no line like this and freq was correctly detected as a 24001.0 Mhz. I suppose this as more to see with acpi , did it ? also cpufreq didn't work. Tanks for the rapid resolution.
Using both fixes, but there are no changes after installing. System Benq Joybook R55 G22 Intel T2400 1.8 Ghz
Complementary information : I've just try a the 2.6.22-16-47.1 from the download.opensuse.com/repositories/Kernel:/SL103_BRANCH/openSUSE_10.3/ And it work for cpuscaling But unsable as there no ipw3945 module nor nvidia for this one. Waiting impatiently the new official kernel build ..
ipw3945 and nvidia modules work from the current official kernel update
I'm not surprising with this ... Normal feature, the trouble the official kernel have is anormal... There's some post on the kernel list about timer.c, acpi (it seems some patch have be removed by error, and reinclude on 2.2.2008) Would be nice if we can have a new kernel this way. And/or in wiki documentation a simple way to go back. in Yast it's possible for kernel to choose a previous version aka 2.6.22-13 but not for the nvidia or ipw3945 drivers : this render the undo way unpossible. Or did I forget something ?
Christoph: If cpufreq still does not work, it is another problem (and I doubt it is a regression then). Pls open a new bug report for that. Also it would be great to give the next OpenSUSE 11.0 Alpha2 or one of the early Betas a test. If your problem isn't something really simple it won't get fixed for 10.3. I won't take (and normally do not, adding the patch that was the cause for this bug was wrong) the risk to fix a single machine while others might break on a released product. If you want to be sure your machine works flawless in 11.0, now is the time to try, as soon as 11.0 is released, it's too late.
*** Bug 358235 has been marked as a duplicate of this bug. ***
*** Bug 358354 has been marked as a duplicate of this bug. ***
*** Bug 358385 has been marked as a duplicate of this bug. ***
this bug still present on 3 notebook that i have. last update doesn't fix it. it's impossible to use a laptop at a temperature of 90 degrees... And I'm not alone with this problem. Why then bug closed? i post it in 358235, that closed as duplicate.
Dear Artyom, you should read https://bugzilla.novell.com/page.cgi?id=fields.html#resolution Resolved say bug resolution is in progress. So for the same trouble it just bother the suse's guy to open new bug. (I think we have suffisant number of them) Like us just wait and praid that QA team give us a new kernel this week. ;-)
A complementary note. If you have a dual core cpu like my centrino you could perharps disable the second core. After that the cpufreq is working. Not a real solution, but is thermal & battery saver workaround.
Bruno: apologize. didn't know that "resolved" is interpreted in such a way here. look forward to, thanks to the excellent distribution. if needed, I'm ready to test possible solutions in a short time.
(In reply to comment #38 from Greg Kroah-Hartman) > *** Bug 358354 has been marked as a duplicate of this bug. *** > Please note: Bug 358354 is a bug in kernel 2.6.16.54-0.2.5-default of SuSE Linux 10.1 Bug 357598 is a bug in kernel 2.6.22.16-0.1-default of openSuSE 10.3
Additionnal infos : After disabling the second core I've a faster cpu than the hardware permit ... Funny cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 15 model name : Intel(R) Core(TM)2 Duo CPU T7700 @ 2.40GHz stepping : 10 cpu MHz : 2600.000 cache size : 4096 KB fpu : yes fpu_exception : yes cpuid level : 10 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss tm syscall nx lm constant_tsc up pni monitor ds_cpl vmx est tm2 ssse3 cx16 xtpr lahf_lm bogomips : 4793.02 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: 2.6Ghz versus 2.40Ghz
Re #46: that's a feature in some recent Intel CPUs to make one core faster if the other is idle because the temperature is lower then. But it should work too transparently if both cores are enabled but one are idle with recent kernels (certainly in 10.3)
I have just released the 2.6.22.16-0.2 kernel to the update channels to fix this issue. We cross checked that the bug is fixed in-house again before. ;)
Linux c-3po 2.6.22.16-0.2-default #1 SMP 2008/02/01 19:36:55 UTC x86_64 x86_64 x86_64 GNU/Linux Cpufreq is now working with the new kernel even with a Marking TSC unstable due to TSCs unsynchronized time.c: Detected 2393.998 MHz processor. As I can see the rest is functionnal. Thank you opensuse guys ! For me you can close this bug !
Can someone (one is enough :) ) please attach acpidump output. Just call: acpidump >/tmp/acpidump and upload the /tmp/acpidump file. Intel's ACPICA developers want to have a closer look at this as the patch is already in their mainline code (and will pop up in the next mainline kernel then). Please be sure that cpufreq is now fixed with the new kernel (e.g. Christoph still has no cpufreq (and never had?)). Alexander, maybe we could use your acpidump as reference to fix this upstream (simply deleting the patch is affecting other (probably very specific) systems and is also not a solution). The error messages related to the sizeOf patch you saw in dmesg were these, correct?: ACPI Error (dsopcode-0250): No pointer back to NS node in buffer obj ffff81007cc35160 [20070126] ACPI Error (psparse-0537): Method parse/execution failed [\_PR_.CPU0._PDC] (Node ffff810037fc5690), AE_AML_INTERNAL ACPI: Processor [CPU0] (supports 8 throttling states) ACPI Error (dsopcode-0250): No pointer back to NS node in buffer obj ffff810037f8fce0 [20070126] ACPI Error (psparse-0537): Method parse/execution failed [\_PR_.CPU1._PDC] (Node ffff810037fc5610), AE_AML_INTERNAL ACPI: Processor [CPU1] (supports 8 throttling states) and are now gone after the sizeOf patch has been ripped out of our 2.6.22 10.3 kernel, right?
Created attachment 193689 [details] Output of acpidump Hello Thomas, here is the output of acpidump. I'm running the latest kernel from the openSUSE Update Repo: kernel-default-2.6.22.16-0.2 which does support cpufreq (again). If there's anything else I could do, please let me know. :)
Created attachment 193714 [details] Output acpidump CentrinoDualCore T7700 sur kernel Linux c-3po 2.6.22.16-0.2-default #1 SMP 2008/02/01 19:36:55 UTC x86_64 x86_64 x86_64 GNU/Linux
Created attachment 193715 [details] AcpiDump content on AMD X2 6400 with kernel Linux c-3po 2.6.22.16-0.2-default #1 SMP 2008/02/01 19:36:55 UTC x86_64 x86_64 x86_64 GNU/Linux
Created attachment 193720 [details] acpidump Here there's my acpidump too, if it can be useful. The system is a DELL XPS m1330. uname -a gives: Linux Chamaleon 2.6.22.16-0.2-default #1 SMP 2008/02/01 19:36:55 UTC x86_64 x86_64 x86_64 GNU/Linux Regards, Alberto
Is this really resolved now? reassigning just to make sure...
CPUFREQ scaling works fine again with 2.6.22.16-0.2 kernel on my HP nc6320.
All functionnality work with my Dell Precision M4300 with 2.6.22.16-02 x86_64 (Centrino DualCore T7700), on Sony laptop PCG-4A1L (P3 Mobile), Dell 510m (Centrino 1.5Ghz) and AMD x2 6400 workstation.
*** Bug 359828 has been marked as a duplicate of this bug. ***
(In reply to comment #55 from Greg Kroah-Hartman) > Is this really resolved now? reassigning just to make sure... > Frequency scaling is working again on my Lenovo 3000 N 100 (BYG 0768) with kernel 2.6.22.16-0.2-default. Thank you SuSE-Team. But bug 358354 which is marked as a duplicate of this bug (Comment#38) is still existing. Is there a new kernel panned too ?
Norman: The 10.1 problem is another one and it will get addressed (see the other bug, which one is not a duplicate anymore and got reopened). Please do not open this bug anymore, the problem should be solved. You can still ask for info (which kernel has it, or whatever...)
*** Bug 358923 has been marked as a duplicate of this bug. ***
*** Bug 360201 has been marked as a duplicate of this bug. ***
Can someone please give this kernel a test: ftp.suse.com/pub/people/trenn/10_3_with_sizeof_fixed This is a 10.3 kernel with a modified sizeof patch, that should have this issue fixed. Please install, boot it once and check in dmesg that there are no acpi errors left (compared to the broken update kernel): dmesg |grep -i acpi If this won't get tested/fixed it will appear mainline soon (part of current acpica code that will get merged into the kernel soon), so please help.
CPUFREQ works fine :) # uname -r 2.6.22.18-0.1-default # dmesg | grep -i acpi BIOS-e820: 000000007f7e5600 - 000000007f7f8000 (ACPI NVS) ACPI: RSDP 000F7BD0, 0024 (r2 HP ) ACPI: XSDT 7F7E57C4, 0074 (r1 HPQOEM SLIC-MPC 1 HP 1) ACPI: FACP 7F7E5684, 00F4 (r4 HP 30A2 3 HP 1) ACPI: DSDT 7F7E5AC0, F9BB (r1 HP nx7400 10000 MSFT 100000E) ACPI: FACS 7F7F7E80, 0040 ACPI: SLIC 7F7E5838, 0176 (r1 HPQOEM SLIC-MPC 1 HP 1) ACPI: HPET 7F7E59B0, 0038 (r1 HP 30A2 1 HP 1) ACPI: APIC 7F7E59E8, 0068 (r1 HP 30A2 1 HP 1) ACPI: MCFG 7F7E5A50, 003C (r1 HP 30A2 1 HP 1) ACPI: TCPA 7F7E5A8C, 0032 (r2 HP 30A2 1 HP 1) ACPI: SSDT 7F7F547B, 0326 (r1 HP HPQSAT 1 MSFT 100000E) ACPI: SSDT 7F7F5FF6, 025F (r1 HP Cpu0Tst 3000 INTL 20060317) ACPI: SSDT 7F7F6255, 00A6 (r1 HP Cpu1Tst 3000 INTL 20060317) ACPI: SSDT 7F7F62FB, 04D7 (r1 HP CpuPm 3000 INTL 20060317) ACPI: PM-Timer IO Port: 0x1008 ACPI: Local APIC address 0xfee00000 ACPI: LAPIC (acpi_id[0x01] lapic_id[0x00] enabled) ACPI: LAPIC (acpi_id[0x02] lapic_id[0x01] enabled) ACPI: LAPIC_NMI (acpi_id[0x01] high edge lint[0x1]) ACPI: LAPIC_NMI (acpi_id[0x02] high edge lint[0x1]) ACPI: IOAPIC (id[0x01] address[0xfec00000] gsi_base[0]) ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl) ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level) ACPI: IRQ0 used by override. ACPI: IRQ2 used by override. ACPI: IRQ9 used by override. ACPI: HPET id: 0x8086a201 base: 0xfed00000 Using ACPI (MADT) for SMP configuration information ACPI: Core revision 20070126 tbxface-0587 [00] tb_load_namespace : ACPI Tables successfully acquired evxfevnt-0091 [00] enable : Transition to ACPI mode successful ACPI: bus type pci registered ACPI: EC: Look up EC in DSDT ACPI: Interpreter enabled ACPI: (supports S0 S3 S4 S5) ACPI: Using IOAPIC for interrupt routing ACPI: PCI Root Bridge [C002] (0000:00) PCI quirk: region 1000-107f claimed by ICH6 ACPI/GPIO/TCO ACPI: PCI Interrupt Routing Table [\_SB_.C002._PRT] ACPI: PCI Interrupt Routing Table [\_SB_.C002.C098._PRT] ACPI: PCI Interrupt Routing Table [\_SB_.C002.C104._PRT] ACPI: PCI Interrupt Routing Table [\_SB_.C002.C114._PRT] ACPI: PCI Interrupt Routing Table [\_SB_.C002.C11A._PRT] ACPI: PCI Interrupt Link [C110] (IRQs *10 11) ACPI: PCI Interrupt Link [C111] (IRQs *10 11) ACPI: PCI Interrupt Link [C112] (IRQs 10 *11) ACPI: PCI Interrupt Link [C113] (IRQs *10 11) ACPI: PCI Interrupt Link [C12C] (IRQs *10 11) ACPI: PCI Interrupt Link [C12D] (IRQs 10 11) *0, disabled. ACPI: PCI Interrupt Link [C12E] (IRQs 10 11) *0, disabled. ACPI Exception (pci_link-0184): AE_NOT_FOUND, Evaluating _PRS [20070126] ACPI: Power Resource [C21D] (on) ACPI: Power Resource [C226] (off) ACPI: Power Resource [C320] (off) ACPI: Power Resource [C321] (off) ACPI: Power Resource [C322] (off) ACPI: Power Resource [C323] (off) pnp: PnP ACPI init ACPI: bus type pnp registered pnp: PnP ACPI: found 12 devices ACPI: ACPI bus type pnp unregistered PCI: Using ACPI for IRQ routing ACPI: RTC can wake from S4 ACPI: PCI Interrupt 0000:00:1c.0[A] -> GSI 16 (level, low) -> IRQ 16 ACPI: PCI Interrupt 0000:00:1c.1[B] -> GSI 17 (level, low) -> IRQ 17 ACPI: PCI Interrupt 0000:00:1c.3[D] -> GSI 19 (level, low) -> IRQ 19 ACPI: PCI Interrupt 0000:02:06.0[A] -> GSI 18 (level, low) -> IRQ 18 ACPI: SSDT 7F7F5869, 023D (r1 HP Cpu0Ist 3000 INTL 20060317) ACPI: SSDT 7F7F5B2B, 04CB (r1 HP Cpu0Cst 3001 INTL 20060317) ACPI: CPU0 (power states: C1[C1] C2[C2]) ACPI: Processor [CPU0] (supports 8 throttling states) ACPI: SSDT 7F7F57A1, 00C8 (r1 HP Cpu1Ist 3000 INTL 20060317) ACPI: SSDT 7F7F5AA6, 0085 (r1 HP Cpu1Cst 3000 INTL 20060317) ACPI: CPU1 (power states: C1[C1] C2[C2]) ACPI: Processor [CPU1] (supports 8 throttling states) ACPI: Thermal Zone [TZ0] (29 C) ACPI: Thermal Zone [TZ1] (33 C) ACPI: Thermal Zone [TZ2] (21 C) ACPI: Thermal Zone [TZ3] (24 C) ACPI: Thermal Zone [TZ4] (75 C) ACPI: PCI Interrupt 0000:00:1f.2[B] -> GSI 17 (level, low) -> IRQ 17 ACPI: PCI Interrupt 0000:00:1f.1[A] -> GSI 16 (level, low) -> IRQ 16 ACPI: Transitioning device [C324] to D3 ACPI: Transitioning device [C324] to D3 ACPI: Fan [C324] (off) ACPI: Transitioning device [C325] to D3 ACPI: Transitioning device [C325] to D3 ACPI: Fan [C325] (off) ACPI: Transitioning device [C326] to D3 ACPI: Transitioning device [C326] to D3 ACPI: Fan [C326] (off) ACPI: Transitioning device [C327] to D3 ACPI: Transitioning device [C327] to D3 ACPI: Fan [C327] (off) ACPI: PCI Interrupt 0000:00:1d.7[A] -> GSI 20 (level, low) -> IRQ 20 sda:<6>ACPI: PCI Interrupt 0000:00:1d.0[A] -> GSI 20 (level, low) -> IRQ 20 ACPI: PCI Interrupt 0000:00:1d.1[B] -> GSI 17 (level, low) -> IRQ 17 sda1 sda2 sda3 sda4 <<6>ACPI: PCI Interrupt 0000:00:1d.2[C] -> GSI 18 (level, low) -> IRQ 18 ACPI: PCI Interrupt 0000:00:1d.3[D] -> GSI 19 (level, low) -> IRQ 19 ACPI: PCI Interrupt 0000:10:00.0[A] -> GSI 17 (level, low) -> IRQ 17 ACPI: AC Adapter [C1BC] (on-line) ACPI: Power Button (FF) [PWRF] ACPI: Sleep Button (CM) [C243] ACPI: Battery Slot [C1BE] (battery present) ACPI: Battery Slot [C1BD] (battery absent) ACPI: Lid Switch [C23C] ACPI: PCI Interrupt 0000:00:1b.0[A] -> GSI 16 (level, low) -> IRQ 16 ACPI: PCI Interrupt 0000:02:06.1[B] -> GSI 19 (level, low) -> IRQ 19 ACPI: PCI Interrupt 0000:02:0e.0[A] -> GSI 16 (level, low) -> IRQ 16
(In reply to comment #63 from Thomas Renninger) > Can someone please give this kernel a test: > ftp.suse.com/pub/people/trenn/10_3_with_sizeof_fixed > > This is a 10.3 kernel with a modified sizeof patch, that should have this issue > fixed. > Please install, boot it once and check in dmesg that there are no acpi errors > left (compared to the broken update kernel): > dmesg |grep -i acpi Done on my NEC VERSA M360, and everything seems to work fine # uname -a Linux frecb006692 2.6.22.18-0.1-default #1 SMP 2008/02/12 23:44:25 UTC x86_64 x86_64 x86_64 GNU/Linux # dmesg |grep -i acpi BIOS-e820: 000000003f670000 - 000000003f700000 (ACPI NVS) ACPI: RSDP 000F6AB0, 0014 (r0 PTLTD ) ACPI: RSDT 3F67E710, 0048 (r1 PTLTD Capell00 6040000 LTP 0) ACPI: FACP 3F685DEE, 0074 (r1 INTEL CALISTGA 6040000 LOHR 5A) ACPI: DSDT 3F67FBDA, 6214 (r1 INTEL CALISTGA 6040000 INTL 20050624) ACPI: FACS 3F686FC0, 0040 ACPI: APIC 3F685E62, 0068 (r1 INTEL CALISTGA 6040000 LOHR 5A) ACPI: HPET 3F685ECA, 0038 (r1 INTEL CALISTGA 6040000 LOHR 5A) ACPI: MCFG 3F685F02, 003C (r1 INTEL CALISTGA 6040000 LOHR 5A) ACPI: BOOT 3F685FD8, 0028 (r1 PTLTD $SBFTBL$ 6040000 LTP 1) ACPI: APIC 3F685F70, 0068 (r1 PTLTD APIC 6040000 LTP 0) ACPI: SSDT 3F67F587, 064F (r1 SataRe SataPri 1000 INTL 20050624) ACPI: SSDT 3F67EEEB, 069C (r1 SataRe SataSec 1000 INTL 20050624) ACPI: SSDT 3F67E758, 051A (r1 PmRef CpuPm 3000 INTL 20050624) ACPI: BIOS bug: multiple APIC/MADT found, using 0 ACPI: If "acpi_apic_instance=2" works better, notify linux-acpi@vger.kernel.org ACPI: PM-Timer IO Port: 0x1008 ACPI: Local APIC address 0xfee00000 ACPI: LAPIC (acpi_id[0x00] lapic_id[0x00] enabled) ACPI: LAPIC (acpi_id[0x01] lapic_id[0x01] enabled) ACPI: LAPIC_NMI (acpi_id[0x00] high edge lint[0x1]) ACPI: LAPIC_NMI (acpi_id[0x01] high edge lint[0x1]) ACPI: IOAPIC (id[0x01] address[0xfec00000] gsi_base[0]) ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 dfl dfl) ACPI: INT_SRC_OVR (bus 0 bus_irq 9 global_irq 9 high level) ACPI: IRQ0 used by override. ACPI: IRQ2 used by override. ACPI: IRQ9 used by override. ACPI: HPET id: 0x8086a201 base: 0xfed00000 Using ACPI (MADT) for SMP configuration information ACPI: Core revision 20070126 tbxface-0587 [00] tb_load_namespace : ACPI Tables successfully acquired evxfevnt-0091 [00] enable : Transition to ACPI mode successful ACPI: bus type pci registered ACPI: EC: Look up EC in DSDT Initializing Device/Processor/Thermal objects by executing _INI methods:<4>ACPI: System BIOS is requesting _OSI(Linux) ACPI: Please test with "acpi_osi=!Linux" Please send dmidecode to linux-acpi@vger.kernel.org ACPI: Interpreter enabled ACPI: (supports S0 S3 S4 S5) ACPI: Using IOAPIC for interrupt routing ACPI: PCI Root Bridge [PCI0] (0000:00) PCI quirk: region 1000-107f claimed by ICH6 ACPI/GPIO/TCO ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.RP01._PRT] ACPI: PCI Interrupt Routing Table [\_SB_.PCI0.PCIB._PRT] ACPI: PCI Interrupt Link [LNKA] (IRQs 1 3 4 5 6 7 *10 12 14 15) ACPI: PCI Interrupt Link [LNKB] (IRQs 1 3 4 5 6 7 11 12 14 15) *10 ACPI: PCI Interrupt Link [LNKC] (IRQs 1 3 4 5 6 7 *10 12 14 15) ACPI: PCI Interrupt Link [LNKD] (IRQs 1 3 4 5 6 7 11 12 14 15) *10 ACPI: PCI Interrupt Link [LNKE] (IRQs 1 3 4 5 6 7 *10 12 14 15) ACPI: PCI Interrupt Link [LNKF] (IRQs 1 3 4 5 6 7 11 12 14 15) *0, disabled. ACPI: PCI Interrupt Link [LNKG] (IRQs 1 3 4 5 6 7 *10 12 14 15) ACPI: PCI Interrupt Link [LNKH] (IRQs 1 3 4 5 6 7 11 12 14 15) *10 ACPI: Power Resource [QFAN] (off) pnp: PnP ACPI init ACPI: bus type pnp registered pnp: PnP ACPI: found 13 devices ACPI: ACPI bus type pnp unregistered PCI: Using ACPI for IRQ routing ACPI: RTC can wake from S4 ACPI: PCI Interrupt 0000:00:1c.0[A] -> GSI 17 (level, low) -> IRQ 17 ACPI: PCI Interrupt 0000:03:07.0[A] -> GSI 16 (level, low) -> IRQ 16 ACPI: SSDT 3F67EC72, 01E5 (r1 PmRef Cpu0Ist 3000 INTL 20050624) ACPI: Processor [CPU0] (supports 8 throttling states) ACPI: SSDT 3F67EE57, 0094 (r1 PmRef Cpu1Ist 3000 INTL 20050624) ACPI: Processor [CPU1] (supports 8 throttling states) ACPI: Thermal Zone [THRM] (50 C) ACPI: PCI Interrupt 0000:00:1f.1[B] -> GSI 19 (level, low) -> IRQ 19 ACPI: PCI Interrupt 0000:00:1f.2[B] -> GSI 19 (level, low) -> IRQ 19 ACPI: Transitioning device [FAN] to D3 ACPI: Transitioning device [FAN] to D3 ACPI: Fan [FAN] (off) ACPI: PCI Interrupt 0000:00:1d.0[A] -> GSI 23 (level, low) -> IRQ 23 ACPI: PCI Interrupt 0000:00:1d.1[B] -> GSI 19 (level, low) -> IRQ 19 ACPI: PCI Interrupt 0000:00:1d.7[A] -> GSI 23 (level, low) -> IRQ 23 ACPI: PCI Interrupt 0000:00:1d.2[C] -> GSI 18 (level, low) -> IRQ 18 ACPI: PCI Interrupt 0000:00:1d.3[D] -> GSI 16 (level, low) -> IRQ 16 ACPI: AC Adapter [ACAD] (off-line) ACPI: Power Button (FF) [PWRF] ACPI: Lid Switch [LID0] ACPI: Power Button (CM) [PWRB] ACPI: PCI Interrupt 0000:00:1b.0[A] -> GSI 22 (level, low) -> IRQ 22 ACPI: PCI Interrupt 0000:03:00.0[A] -> GSI 19 (level, low) -> IRQ 19 ACPI: PCI Interrupt 0000:03:07.1[C] -> GSI 18 (level, low) -> IRQ 18 ACPI: PCI Interrupt 0000:02:00.0[A] -> GSI 16 (level, low) -> IRQ 16 parport_pc 00:0a: reported by Plug and Play ACPI ACPI: Battery Slot [BAT1] (battery present) ACPI: PCI Interrupt 0000:00:1f.3[B] -> GSI 19 (level, low) -> IRQ 19
Created attachment 195192 [details] dmesg output Intel Core2Duo 4400: Shuttle SG31G2
Created attachment 195196 [details] dmesg output AthlonX2 Dual Core BE-2350 Shuttle SS21T
Created attachment 195206 [details] dmesg outpu Toshiba Satellite A100 Core2Duo
(In reply to comment #63 from Thomas Renninger) > Can someone please give this kernel a test: > ftp.suse.com/pub/people/trenn/10_3_with_sizeof_fixed > > This is a 10.3 kernel with a modified sizeof patch, that should have this issue > fixed. > Please install, boot it once and check in dmesg that there are no acpi errors > left (compared to the broken update kernel): > dmesg |grep -i acpi > > If this won't get tested/fixed it will appear mainline soon (part of current > acpica code that will get merged into the kernel soon), so please help. > Hi Thomas, I've tried on three different servers - no problems so far. Acpi output is attached (sorry for the mailstorm) Regards, Tom wrt@nomad:~> uname -a Linux nomad 2.6.22.18-0.1-default #1 SMP 2008/02/12 23:44:25 UTC i686 i686 i386 GNU/Linux wrt@nomad:~> wrt@nomad:~> wrt@nomad:~> ssh 192.168.1.11 -l root Password: Last login: Fri Feb 15 20:38:15 2008 from 192.168.1.21 Have a lot of fun... workstation:~ # uname -a Linux workstation 2.6.22.18-0.1-default #1 SMP 2008/02/12 23:44:25 UTC i686 i686 i386 GNU/Linux workstation:~ # exit logout Connection to 192.168.1.11 closed. wrt@nomad:~> wrt@nomad:~> wrt@nomad:~> ssh 192.168.1.15 -l root Password: Last login: Fri Feb 15 20:37:15 2008 from 192.168.1.21 Have a lot of fun... outside:~ # outside:~ # uname -a Linux outside 2.6.22.18-0.1-default #1 SMP 2008/02/12 23:44:25 UTC x86_64 x86_64 x86_64 GNU/Linux outside:~ # outside:~ # outside:~ # exit logout Connection to 192.168.1.15 closed. wrt@nomad:~>
Created attachment 195267 [details] Dmesg acpi on Dell Precision M4300 (CentrinoDualCore T7700) With 2.6.22-18 : acpi ok cpu throttling ok ... Well everything is fine .