Announcement

Collapse
No announcement yet.

AMD Prepares PMF Linux Driver For "Smart PC Solutions Builder"

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

  • #21
    Well, as long as we have to deal with UEFI and PSP as blackboxes anyway, this blob on top doesn't make things worse imho.

    Comment


    • #22
      Originally posted by bridgman View Post
      Would you be OK with a mechanism that lets you say "I am taking responsibility for fine-grained power management and do not expect warranty coverage if something goes wrong" then records the answer permanently in the hardware ? My impression from customer support is that the general answer seems to be "no". The problem is not individual developers experimenting with PM software and letting the smoke out, but thousands or millions of users running power management software written by someone they never met.
      I might also fail to follow proper appliance safety practices, start a kitchen fire, and then demand the company that built the appliance I destroyed replaces it at no extra cost. Wouldn't it be better to build every appliance with a camera that turns it off if it detects you trying to prepare ingredients that the company hasn't manually approved? The problem is not individual chefs experimenting with food preparation, but thousands or millions of users following cooking guides written by someone they never met.

      Comment


      • #23
        Originally posted by timofonic View Post

        I fully agree!

        But Microsoft is a platinum Linux Foundation sponsor these days. Money corrupts everything

        They merged the Trojan's Horse with 6.1 kernel.

        https://www.phoronix.com/news/x86-Platform-Drivers-6.1
        Could you please explain why you think the PMF is a Trojan horse?

        Comment


        • #24
          Originally posted by EphemeralEft View Post
          I might also fail to follow proper appliance safety practices, start a kitchen fire, and then demand the company that built the appliance I destroyed replaces it at no extra cost. Wouldn't it be better to build every appliance with a camera that turns it off if it detects you trying to prepare ingredients that the company hasn't manually approved? The problem is not individual chefs experimenting with food preparation, but thousands or millions of users following cooking guides written by someone they never met.
          I don't think the analogy works - there's a big difference between "burning what you are cooking" and "killing the stove".
          Test signature

          Comment


          • #25
            Originally posted by Jonjolt View Post

            I wasn't even thinking of this, for anyone unaware if you have enough $$$ Intel will customize the chip for you.
            Don't be surprised if 90% of the people here are not even aware of this.

            Comment


            • #26
              Who cares? Either Linux will get proper support for it where you never even notice it, or it'll get cracked and the "feature" becomes moot.

              Comment


              • #27
                Originally posted by bridgman View Post
                These days it's a different story, particularly in mobile products. GPUs (and presumably CPUs) can use much (MUCH) more power that the cooling system can dissipate - even getting heat out of the package is a problem - so getting best performance while not letting the smoke out is much more complicated and multi-dimensional. Even thermal limiting with dozens of sensors is too slow to avoid local meltdowns, so power management also uses distributed current sensors to anticipate thermal rise and start limiting clocks & voltages.

                The key question is who takes responsibility for not letting the smoke out and killing the system. With desktop systems the worst case on the CPU side is usually that you blow up a socketed CPU, but with mobile systems the cost of damage is usually much higher, typically over 50% of the system cost. Right now HW vendors keep fine-grained power management under their control and if something goes wrong they can be pretty sure it was "their fault" rather than something the user did. There are grey areas like dust buildup in cooling solutions but that tends to happen sufficiently slowly that the firmware can adapt to it and still keep the smoke in.
                I don't see how this is a valid reason for a closed source solution. You can basically run any modern CPU without any cooler and it will run (slow like shit of course) because it will send cores to sleep that get too hot.
                That part could be implemented in hardware (for example send cores that reach 100°C to sleep till they reach 80°C) and the rest could be done with open source firmware and who ever thinkers with it has in the worst case a slow CPU. Anyone else could just use the AMD provided firmware and get AMDs intended performance.

                If you overclock your CPU it can still be roasted and AMD explicitly supports overclocking (and voids warranty if it's legal in your country), so where is the problem here?

                Comment


                • #28
                  Originally posted by bridgman View Post

                  I don't think the analogy works - there's a big difference between "burning what you are cooking" and "killing the stove".
                  I implied that the stove would be destroyed. But it's just a hypothetical for the purpose of discussion. Imagine I explicitly said "killing the stove".

                  Comment


                  • #29
                    Originally posted by Anux View Post
                    I don't see how this is a valid reason for a closed source solution. You can basically run any modern CPU without any cooler and it will run (slow like shit of course) because it will send cores to sleep that get too hot.
                    That part could be implemented in hardware (for example send cores that reach 100°C to sleep till they reach 80°C) and the rest could be done with open source firmware and who ever thinkers with it has in the worst case a slow CPU. Anyone else could just use the AMD provided firmware and get AMDs intended performance.

                    If you overclock your CPU it can still be roasted and AMD explicitly supports overclocking (and voids warranty if it's legal in your country), so where is the problem here?
                    The problem is the OEM designs the platform and the cooling solutions so how the "hardware" should react varies greatly on the platform. 100°C may be too late on one platform, but on another, 105°C is fine. The OEM basically provides the limits of their platform and then firmware or hardware can take those limits into account. Setting arbitrary limits that are safe everywhere would artificially limit performance in a lot of cases.

                    You don't have to run PMF and if you don't, as far as I know, you get the default universally safe limits.
                    Last edited by agd5f; 25 September 2023, 05:13 PM.

                    Comment


                    • #30
                      Originally posted by Jonjolt View Post

                      I wasn't even thinking of this, for anyone unaware if you have enough $$$ Intel will customize the chip for you.
                      If you have enough $$$ you can buy Intel and make them bake pancakes for you instead of CPUs. Why would be anyone unaware that with enough money you can buy anything? Like a couple of senators or such. What do they teach nowadays in capitalist schools?

                      Comment

                      Working...
                      X