You are not logged in.
Hi,
After this morning upgrade, power consumption of my system has increased from 8-10W to 15-18W at idle and 15-18W to 25-35W while using flash. Their were lot of upgrade today including linux-ck. I get some innocuous boot errors and now am getting another new one.
kernel /vmlinuz-linux-ck root=/dev/sda8 ro elevator=bfq i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 drm.vblankoffdelay=1Linux arch 3.4.0-1-ck #1 SMP PREEMPT Sat Jun 2 09:37:08 EDT 2012 x86_64 GNU/Linux[donnie@arch ~]$ dmesg | grep watchdog
[ 0.310821] NMI watchdog enabled, takes one hw-pmu counter.
[ 7.409081] watchdog: INTCAMT: cannot register miscdev on minor=130 (err=-16).
[ 7.409224] watchdog: error registering /dev/watchdog (err=-16).
[ 7.409331] mei 0000:00:16.0: unable to register watchdog device.
[donnie@arch ~]$ dmesg | grep mux
[ 9.430857] [drm:intel_dsm_platform_mux_info] *ERROR* MUX INFO call failed
[donnie@arch ~]$ dmesg | grep tpm
[ 6.585003] tpm_tis 00:0a: 1.2 TPM (device-id 0x0, rev-id 78)
[ 7.236372] tpm_tis 00:0a: A TPM error (6) occurred attempting to read a pcr value
[ 7.236530] tpm_tis 00:0a: TPM is disabled/deactivated (0x6)New Error
Update 0 Power consumption and temperature is normal on stock kernel but excessive with Linux-ck which I had been using for a while. Also updated the title so that thread gets proper attention.
Update 6/20 Not a linux-ck problem but probably an upstream bug. Please see the bug report https://bugs.archlinux.org/task/29850#comment95791
Last edited by donniezazen (2012-07-04 00:29:59)
Offline
Regarding your "New Error"
https://wiki.archlinux.org/index.php/Al … nel_Update
Offline
Offline
The watchdog error is harmless, i had it on my system.
However you can get rid of it by writing "blacklist mei" to the new file "/etc/modprobe.d/mei.conf".
Offline
Offline
What does powertop say?
Help me to improve ssh-rdp !
Retroarch User? Try my koko-aio shader !
Offline
Is it the ck patchset or the 3.4 tree? Did you try the Linux package in [testing] to verify?
Offline
I reverted back to stock kernel and temp seems to be normal idle power consumption has come down too . It might have to do something with linux-ck.
@graysky I can provide some data to you if it helps sort out problems with ck, if any.
Last edited by donniezazen (2012-06-04 17:03:41)
Offline
Offline
Apples-to-apples comparison? Any modifications to the .config when you compiled the ck kernel? If so, is this new to the 3.4 release? Did it work fine as you reported under the 3.3 series? Finally, are you using v3.4.1-1 of the stock kernel from [testing]?
Last edited by graysky (2012-06-06 22:45:05)
Offline
I have made no modification to ck or stock kernel. I am not using testing repo. I am using stock kernel from stable arch repos. I use your repo-ck to download and install linux-ck.
Linux arch 3.3.7-1-ARCHrepo-ck/linux-ck-corex 3.4.1-2 (ck-corex) [installed]
repo-ck/linux-ck-headers 3.4.1-2 (ck-generic) [installed]It would be devastating if Linux kernel slumps into another power regression in 3.4 series. I recently recovered from aspm coma.
Thanks.
Offline
I am not using testing repo.
OK! Your comparison is _NOT_ a fair one. You are comparing 3.3.7-1-ARCH to 3.4.1-2-ck. You must compare 3.4.1-1-ARCH to 3.4.1-2-ck or you must compare 3.3.7-1-ARCH to 3.3.7-1-ck.
Here is a link (just for you) to the 3.3.7-1-ck kernel and headers:
LINK PLACEHOLD -- do you need x86_64 or i686?
It would be devastating if Linux kernel slumps into another power regression in 3.4 series. I recently recovered from aspm coma.
Fasten your seatbelt as I suspect your problem is not with the the linux-ck kernel but with something in the 3.4 series...
Summary:
1) Compare 3.4.1-1-ARCH to 3.4.1-2-ck
or
2) Compare 3.3.7-1-ARCH to 3.3.7-1-ck
Please remove the 'linux-ck' from the title of your post if indeed it is not to blame.
EDIT: why are you not using the corex version of the headers and of the kernel? They are a matched pair.
repo-ck/linux-ck-corex 3.4.1-2 (ck-corex) [installed]
repo-ck/linux-ck-headers 3.4.1-2 (ck-generic) [installed]
Last edited by graysky (2012-06-07 00:00:18)
Offline
I want to let you know that I do not mean disrespect and do not want to sound flaming. If it did happen, I am sorry for it.
I am fine for now as my system works fine under stable 3.3 series kernel and I do not want to you to do extra work as I will eventually have to upgrade. I will test 3.4 from testing repos and will report bug if any.
Thanks for replying and taking time to compile linux-ck for us.
Offline
You're welcome, no problem. I will gladly post the binaries for 3.3.7-1-corex for you. Just tell me which arch you would like: i686 or x86_64.
EDIT: Ack... I see from your first post it's x86_64.
Last edited by graysky (2012-06-07 01:03:23)
Offline
Actually, 3.3.8 has been released. I can build 3.3.8-1-ck for you in the corex flavor. Gimme a few.
Last edited by graysky (2012-06-07 00:21:30)
Offline
OK... you should be able to install via:
$ sudo pacman -U http://repo-ck.com/donniezazen/linux-ck-corex-3.3.8-1-x86_64.pkg.tar.xz http://repo-ck.com/donniezazen/linux-ck-corex-headers-3.3.8-1-x86_64.pkg.tar.xzThere is also a formal index.html if the pacman method doesn't work... but I see no reason why it wouldn't: http://repo-ck.com/donniezazen
Last edited by graysky (2012-06-07 01:02:09)
Offline
On my system:
3.3-arch = good
3.3-ck = good
3.4-arch = bad (high load + higher temperature)
3.4-ck = good (load not working but temperate = normal)
Pretty strange...
ᶘ ᵒᴥᵒᶅ
Offline
Running linux-lqx 3.4.1-1 and so far so good, now compiling new Liquorix linux-lqx 3.4.2-1 - I will post here if I find any of these problems.
@donniezazen What's your hardware? Have you tried powertop?
I'm using FLOSS module radeonhd -which everybody says increases computer temperature- and except right now that I'm compiling the aforementioned kernel my computer stays cool all the time.
Last edited by msx (2012-06-10 21:33:33)
Enjoying i3wm w/ lifebar + j4-dmenu-desktop + tab_windows / fish shell / Emacs / tmux / Konsole / KDE apps
Arch + Linux-libre kernel: ParabolaGNULinux.org
Offline
Offline
@op - problem still present in 3.4.2-1?
Offline
Offline
Sounds like an upstream bug for sure and seems to be isolated to newer Intel chips. Keep on 3.3.8 I guess.. ![]()
EDIT: my laptop (i7-2620M) gives this error but no change in power consumption or in load temps.
Under 3.3.8:
$ dmesg | grep watchdog
[ 0.170100] NMI watchdog enabled, takes one hw-pmu counter.
[ 0.221216] NMI watchdog enabled, takes one hw-pmu counter.
[ 0.252508] NMI watchdog enabled, takes one hw-pmu counter.
[ 0.283789] NMI watchdog enabled, takes one hw-pmu counter.Under 3.4.2:
$ dmesg | grep watchdog
[ 0.172444] NMI watchdog: enabled, takes one hw-pmu counter.
[ 0.205557] NMI watchdog: enabled, takes one hw-pmu counter.
[ 0.218892] NMI watchdog: enabled, takes one hw-pmu counter.
[ 0.232180] NMI watchdog: enabled, takes one hw-pmu counter.
[ 6.930494] watchdog: INTCAMT: cannot register miscdev on minor=130 (err=-16)
[ 6.930650] watchdog: error registering /dev/watchdog (err=-16)
[ 6.930769] mei 0000:00:16.0: unable to register watchdog device.Last edited by graysky (2012-06-10 10:30:05)
Offline