Announcement

Collapse
No announcement yet.

Linux Kernel Power Consumption Is Lowered, But Regressions Remain

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Linux Kernel Power Consumption Is Lowered, But Regressions Remain

    Phoronix: Linux Kernel Power Consumption Is Lowered, But Regressions Remain

    As discovered by a Phoronix reader, there is a patch in the Linux 2.6.39.1 kernel that can partially improve the system's power performance. The patch by a Nokia engineer is entitled "cpuidle: menu: fixed wrapping timers at 4.294 seconds" and initial reports have said that it will lower the power consumption compared to the stock 2.6.39 kernel.

    http://www.phoronix.com/vr.php?view=16104

  • #2
    How comes bisecting is taking so long? Having a very broad window (at commit-per-commit basis) and lowering it constantly should narrow it down very fast, right?

    Comment


    • #3
      Originally posted by V!NCENT View Post
      How comes bisecting is taking so long? Having a very broad window (at commit-per-commit basis) and lowering it constantly should narrow it down very fast, right?
      micheael wrote it on zwitter some kernel versions are broken and this stops you in bisecting.

      bisecting is only fast if every version you generate works but what is if you only generate not working versions?

      Comment


      • #4
        Yeah but it's taken a bit too long now?!
        Last concrete finding was that the offending commits that messed up 2.6.38 were from 13-15 January. And unfortunatly that one is not fixed in 2.6.39.1. Perhaps Micheal should have focused on one regression at a time...

        Comment


        • #5
          How comes bisecting is taking so long? Having a very broad window (at commit-per-commit basis) and lowering it constantly should narrow it down very fast, right?

          i blame the Bavarians
          Last edited by 89c51; 06-08-2011, 07:21 PM.

          Comment


          • #6
            Reasons why it's taking so long to bisect:

            1. Needing to manually plug/unplug power adapter...
            2. Still haven't received (or otherwise figured out) any sub-$100 USB UPS/power-meter units for automatic power monitoring so fast desktop/workstations can be used for bisecting in a fully automated manner...
            3. Haven't been motivated by any ISVs or IHVs -- i.e. even any communication from them about correlating our bug detecting efforts. The only ISV communication has been from an interested (and totally independent) company offering up a 48-core system if it means bisecting can be done faster, but it's behind a corporate firewall and there is no power metering...
            4. Due to all of the involved work, it will be a number of articles in order to offset my expenses....
            5. Phoronix is pretty much just me doing all the work... I also need to be producing other content as well on a daily basis.
            6. Yes, the Bavarian visitors/friends did insert an additional few day delay.
            etc...
            Michael Larabel
            http://www.michaellarabel.com/

            Comment


            • #7
              Originally posted by Michael View Post
              Reasons why it's taking so long to bisect:

              1. Needing to manually plug/unplug power adapter...
              2. Still haven't received (or otherwise figured out) any sub-$100 USB UPS/power-meter units for automatic power monitoring so fast desktop/workstations can be used for bisecting in a fully automated manner...
              3. Haven't been motivated by any ISVs or IHVs -- i.e. even any communication from them about correlating our bug detecting efforts. The only ISV communication has been from an interested (and totally independent) company offering up a 48-core system if it means bisecting can be done faster, but it's behind a corporate firewall and there is no power metering...
              4. Due to all of the involved work, it will be a number of articles in order to offset my expenses....
              5. Phoronix is pretty much just me doing all the work... I also need to be producing other content as well on a daily basis.
              6. Yes, the Bavarian visitors/friends did insert an additional few day delay.
              etc...
              Any chance you can update us with your current progress so phoronix readers can assist with running their own bisections?

              Comment


              • #8
                It looks to me like the patch doesn't change things? or only a tiny bit? Still way over <= .37 versions.

                Comment


                • #9
                  This is very disappointing. I would expect such issues to be a top priority for companies like Google (Android) and Canonical and Red Hat.

                  Comment


                  • #10
                    Originally posted by zoomblab View Post
                    This is very disappointing. I would expect such issues to be a top priority for companies like Google (Android) and Canonical and Red Hat.
                    Only if they feel the pain.

                    Android may very well not be affected, and google and Canonical are more into corporate deployments, and I don't think there are too many laptops there...

                    Comment


                    • #11
                      kernel is still up by an approximate 18~20% over older (Linux 2.6.34 era)
                      ufffff!!! Michael, bisecting this shouldn't be your task. You already pointed the problem.

                      Linux is not so small, there is certain amount of manpower, linux foundations, consortium... I cannot accept that nobody is taking care of this awful regression. I even don't understand why they released a kernel which has 10-20% more power consumption?? this makes me seriously doubt about Linux kernel overall quality (until now I have been very confident about it)

                      Comment


                      • #12
                        Howto bisecting

                        Hi michael , I would love you to write a small howto on how i and others can help this, i have a small amount of sparetime and a spare x60 i can do some testing on. But i need help getting startet.

                        Which releases and commits do you take and build and test?

                        Comment


                        • #13
                          Originally posted by Michael View Post
                          Reasons why it's taking so long to bisect:
                          1. Needing to manually plug/unplug power adapter...
                          2. Still haven't received (or otherwise figured out) any sub-$100 USB UPS/power-meter units for automatic power monitoring so fast desktop/workstations can be used for bisecting in a fully automated manner...
                          Why don't you do the bisecting and compiling on one sytem (64 core) and the testing on a normal notebook with battery? That should speed it up a lot.

                          But yes, it would be great to have power performance testing systems!

                          Why doesn't Linuxfoundation pay for some of the costs? And you could try to raise the money with donations, too.

                          Comment


                          • #14
                            Originally posted by Qaridarium View Post
                            micheael wrote it on zwitter some kernel versions are broken and this stops you in bisecting.

                            bisecting is only fast if every version you generate works but what is if you only generate not working versions?
                            Code:
                            $ git bisect skip
                            ?

                            Comment


                            • #15
                              Originally posted by djtm View Post
                              Why don't you do the bisecting and compiling on one sytem (64 core) and the testing on a normal notebook with battery? That should speed it up a lot.

                              But yes, it would be great to have power performance testing systems!

                              Why doesn't Linuxfoundation pay for some of the costs? And you could try to raise the money with donations, too.
                              I don't know the details but, distcc has been used to compile the Linux kernel. If you must run the build process from the laptop then at least you could leverage the processing power of other machines while building.

                              http://code.google.com/p/distcc/

                              Again I don't know the details, but it would seem that you could build the kernel quickly even on a laptop machine if you have a few machines running distcc for it.

                              Comment

                              Working...
                              X