Announcement

Collapse
No announcement yet.

The Increasing Importance Of ACPI Platform Profiles With Today's Throttle-Happy Hardware

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

  • cbxbiker61
    replied
    Originally posted by ciupenhauer View Post

    Code:
    $ lsmod | grep ideapad_laptop
    ideapad_laptop 28672 0
    platform_profile 16384 1 ideapad_laptop
    It's there alright. Weirdly enough I just noticed that changing the power settings in platform_profile does change the TDP values that ryzenadj -i is reporting, powersave showing 12.5W and performance showing 45-65W, but I can't tell much of a difference between the 2 in regular usage. Now I'm wondering if ryzenadj is reporting false/theoretical values that lenovo is simply ignoring
    I'm using a Python based system monitor called glances. Glances will show current/rated GHz values during usage. I'm sure there are other tools that can do it as well. Run a cpu intensive test program and toggle the modes. If your current GHz changes, then it would appear to be working properly.

    Leave a comment:


  • ciupenhauer
    replied
    Originally posted by cbxbiker61 View Post

    Check to see if either the ideapad_laptop or thinkpad_acpi module get loaded (lsmod). If the ideapad_laptop module gets loaded then it will theoretically work. If the thinkpad_acpi module is loaded the same approach could apply if the thinkpad_acpi.c file was changed in a similar fashion to my patch.

    Note that there are quite a few module dependencies to have the ideapad_laptop module auto load. In my testing Ubuntu seemed to be the best bet for a Distro that pretty much supports my Ideapad out of the box (with the exception of the wifi driver).
    Code:
    $ lsmod | grep ideapad_laptop
    ideapad_laptop 28672 0
    platform_profile 16384 1 ideapad_laptop
    It's there alright. Weirdly enough I just noticed that changing the power settings in platform_profile does change the TDP values that ryzenadj -i is reporting, powersave showing 12.5W and performance showing 45-65W, but I can't tell much of a difference between the 2 in regular usage. Now I'm wondering if ryzenadj is reporting false/theoretical values that lenovo is simply ignoring

    Leave a comment:


  • cbxbiker61
    replied
    Originally posted by ciupenhauer View Post

    Dang, still no ouput. is it possible lenovo slim7 pro doesn't even trigger your code?
    Check to see if either the ideapad_laptop or thinkpad_acpi module get loaded (lsmod). If the ideapad_laptop module gets loaded then it will theoretically work. If the thinkpad_acpi module is loaded the same approach could apply if the thinkpad_acpi.c file was changed in a similar fashion to my patch.

    Note that there are quite a few module dependencies to have the ideapad_laptop module auto load. In my testing Ubuntu seemed to be the best bet for a Distro that pretty much supports my Ideapad out of the box (with the exception of the wifi driver).

    Leave a comment:


  • ciupenhauer
    replied
    Originally posted by cbxbiker61 View Post

    It won't show in dmesg. If you have systemd:

    journalctl -a | grep DYTC
    Sep 19 03:32:02 comer.internal kernel: ideapad_acpi VPC2004:00: DYTC Vendor: LENOVO
    Sep 19 03:32:02 comer.internal kernel: ideapad_acpi VPC2004:00: DYTC Product: 82L5
    Dang, still no ouput. is it possible lenovo slim7 pro doesn't even trigger your code?

    Leave a comment:


  • rlkrlk
    replied
    Originally posted by tildearrow View Post

    Yeah...



    Wait, a laptop with 4 RAM slots?! I have only seen up to 2 on every laptop...



    Good news: they still exist.
    All of my personal laptops (dating back to the Dell Inspiron 8000) have had four DIMM slots. I've always bought used workstation-type ones.

    The Galaxy S8 isn't a current phone. I only bought my current phone (Note 9) a year ago, so I'm planning to keep it going for quite a while longer, but I'd really like a bigger screen (not just long, wide too; even phablets today have very narrow screens). But this is a bit off topic.

    Leave a comment:


  • cbxbiker61
    replied
    Originally posted by ciupenhauer View Post

    Compiled successfully but I can't find DYTC message anywhere in dmesg. Is it supposed to show somewhere else, or any other reason why the message wouldn't appear? Thx
    It won't show in dmesg. If you have systemd:

    journalctl -a | grep DYTC
    Sep 19 03:32:02 comer.internal kernel: ideapad_acpi VPC2004:00: DYTC Vendor: LENOVO
    Sep 19 03:32:02 comer.internal kernel: ideapad_acpi VPC2004:00: DYTC Product: 82L5

    Leave a comment:


  • ciupenhauer
    replied
    Originally posted by cbxbiker61 View Post

    I just posted (an hour ago) the patch to the linux and platforms mailing lists. You can find it there. Yes, please test it out. I've created a whitelist function which will allow you to specify a system that is supported at dytc_version 4. If you apply the patch you will be able to look in your system logs for DYTC Product, that will be the key that you can add to the whitelist function. Please let me know how that works for you. If it works please let me know the DYTC Product key and I should be able to add that to the patch before it gets authorized for inclusion.
    Compiled successfully but I can't find DYTC message anywhere in dmesg. Is it supposed to show somewhere else, or any other reason why the message wouldn't appear? Thx

    Leave a comment:


  • sinepgib
    replied
    Originally posted by Sonadow View Post
    never will.
    Why not?

    Leave a comment:


  • Sonadow
    replied
    Originally posted by Azrael5 View Post
    This means that Linux Oses are unable to manage throttling correctly?
    Never had, never will.

    And installing additional software to accomplish it does not qualify.

    Leave a comment:


  • timrichardson
    replied
    Hans de Goede ... "Junior Member" lol.

    Leave a comment:

Working...
X