Announcement

Collapse
No announcement yet.

One Of The Reasons Why Linux 5.5 Can Be Running Slower

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

  • #11
    Originally posted by Michael View Post
    Greater than 50% ad block rates.
    I wonder if this is based on page views, or source ip, or something else? If a Phoronix Subscriber is browsing while not logged in and with blockers enabled, does this skew the statistics?

    Comment


    • #12
      Originally posted by juno View Post
      You violate EU law btw.
      EU law doesn't apply on this side of the pond! Internet FTW.

      Comment


      • #13
        Originally posted by Spam View Post
        Try Patreon. Would allow you to send perks too
        Originally posted by Jigglywiggly View Post
        Yeah try patreon, it will likely help a lot.
        Why not librepay instead? https://liberapay.com/
        Last edited by Laughing1; 30 December 2019, 12:22 AM.

        Comment


        • #14

          ​IMHO, this is some of your most useful work.

          Unfortunately most other articles are just graph porn with virtually no real world value, or announcements of a new kernel RC release.

          I'm constantly amazed that you don't use your resources to write actually useful content that your readers can carryout themselves, such as how to optimize their own systems.

          For example in-depth information on optimizing file systems for specific use cases and the performance impact that could potentially have.

          Instead it's constant benchmarks showing default values between one kernel version or distro is 0.5% slower than another. Entirely boring and unactionable information.

          Comment


          • #15
            Originally posted by caligula View Post
            Just curious to know if Michael extracts / checks out the kernel and compiles in tmpfs? It's much faster than on SSD/HDD.
            I haven't measured, but I have not noticed much of a slowdown in kernel compiles on my systems. Do you maybe use a small setting for sysctl vm.dirty_bytes or dirty_ratio? When I do a kernel compile almost all of the output files fit into cache, without needing to do a tmpfs build.

            I'm running a new build now just to see. It's an 8 core 1700X with 32 GB RAM and a btrfs disk array. It's a NAS mostly. It does report that PSI for IO has generally 2 processes stuck on IO (full IO stall). So it is getting a slowdown. But I'm not convinced it'd be worth putting the 20 GB of a complete build directory into tmpfs.

            Comment


            • #16
              Great reporting.

              Comment


              • #17
                Originally posted by Zan Lynx View Post
                But I'm not convinced it'd be worth putting the 20 GB of a complete build directory into tmpfs.
                Is most of that the AMD GPU kernel drivers or something?

                When I do a kernel compile for my system (ARM) the build directory only grows to 6 GB, so I can compile in tmpfs with 4 GB RAM and zram compression.

                Comment


                • #18
                  Always test the stock vanilla kernel from www.kernel.org compiled by yourself!
                  Or call these benchmarks "distribution benchmarks" instead of "linux/kernel benchmarks".

                  Comment


                  • #19
                    Originally posted by ipsirc View Post
                    Always test the stock vanilla kernel from www.kernel.org compiled by yourself!
                    Or call these benchmarks "distribution benchmarks" instead of "linux/kernel benchmarks".
                    My understanding is he's already using here the stock vanilla kernel compiled by himself, as you are asking. The vanilla kernel configuration doesn't compile to something that can be run on a computer; all configuration variations are equally supported by upstream vanilla linux. Since you need to make a lot of choices to enable the stuff needed for the computer you run on, dist kernel configs as a base are the relevant starting point for users, generally speaking. This is not the same thing as the downstream patched kernel used by the dist.
                    Last edited by aphysically; 30 December 2019, 03:48 AM.

                    Comment


                    • #20
                      Originally posted by Michael View Post

                      Sadly, most do not, at least for showing support to make future tests possible.

                      Greater than 50% ad block rates.
                      Less than 1% being premium subscribers.
                      Less than 1% tipping.
                      I navigate in incognito mode, logging in the forums only for commenting or liking posts, am O accounted as ad block group? 😥

                      Comment

                      Working...
                      X