Linux 7.3 Expected To "Flatten The Pick" For Better Scheduling While Gaming & More
([Linux Kernel] 3 Hours Ago
"Flatten The Pick" Patches Queued)
- Reference: 0001644937
- News link: https://www.phoronix.com/news/Linux-7.3-Flattens-The-Pick
- Source link:
Going back to early May there were patches for [1]improving the Linux scheduler to help with gaming performance on old "potato" hardware by [2]providing better cgroup scheduling . Those patches, referred to as the "flatten the pick" patch series, are now slated for introduction in the Linux 7.3 kernel.
Intel engineer Peter Zijlstra explains with the patch moving EEVDF to a single runqueue in the [3]commit message :
"Infamously fair/cgroup isn't working for a number of people; typically the complaint is latencies and/or overhead. The latency issue is due to the intermediate entries that represent a combination of tasks and thereby obfuscate the runnability of tasks.
The approach here is to leave the cgroup hierarchy as is; including the intermediate enqueue/dequeue but move the actual EEVDF runqueue outside. This means things like the shares_weight approximation are fully preserved."
With the patch series is [4]changing the defaukt cgroup_mode to "concur". Peter Zijlstra explained there:
"For all the reasons described in the preceding patches, the way cgroup weight is computed is problematic. However, changing it is bound to also lead to trouble. Esp. since people might have taken to inflating the weight value where they can.
Since things are configurable, change the default and hope this serves more people than it hurts, esp. in the longer run.
Specifically, this prepares for a flattened runqueue, where the hierarchical weight becomes far more important (F_g^d terms), so getting rid of small F_g is imperative."
Tasks, max, up, and switch are the other cgroup_modes introduced.
Benchmarks shown by Peter on an old Intel Sandy Bridge system and with AMD Radeon Polaris graphics show a nice FPS improvement and much lower frametime amid the cgroup noise.
The news this past week is those "flatten the pick" patches have been queued into [5]tip/tip.git's "sched/core" Git branch . With the patches now in a TIP branch, they are expected to be submitted for the upcoming Linux 7.3 merge window in late August.
[1] https://www.phoronix.com/news/Linux-Flatten-The-Pick
[2] https://www.phoronix.com/news/Flatten-The-Pick-v3
[3] https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?h=sched/core&id=85570f10a4c61372c1d437365b9d8dbad512ec6f
[4] https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?h=sched/core&id=fb1050ac8e40aaa014291b5b53612d27890c2b09
[5] https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/log/?h=sched/core
Intel engineer Peter Zijlstra explains with the patch moving EEVDF to a single runqueue in the [3]commit message :
"Infamously fair/cgroup isn't working for a number of people; typically the complaint is latencies and/or overhead. The latency issue is due to the intermediate entries that represent a combination of tasks and thereby obfuscate the runnability of tasks.
The approach here is to leave the cgroup hierarchy as is; including the intermediate enqueue/dequeue but move the actual EEVDF runqueue outside. This means things like the shares_weight approximation are fully preserved."
With the patch series is [4]changing the defaukt cgroup_mode to "concur". Peter Zijlstra explained there:
"For all the reasons described in the preceding patches, the way cgroup weight is computed is problematic. However, changing it is bound to also lead to trouble. Esp. since people might have taken to inflating the weight value where they can.
Since things are configurable, change the default and hope this serves more people than it hurts, esp. in the longer run.
Specifically, this prepares for a flattened runqueue, where the hierarchical weight becomes far more important (F_g^d terms), so getting rid of small F_g is imperative."
Tasks, max, up, and switch are the other cgroup_modes introduced.
Benchmarks shown by Peter on an old Intel Sandy Bridge system and with AMD Radeon Polaris graphics show a nice FPS improvement and much lower frametime amid the cgroup noise.
The news this past week is those "flatten the pick" patches have been queued into [5]tip/tip.git's "sched/core" Git branch . With the patches now in a TIP branch, they are expected to be submitted for the upcoming Linux 7.3 merge window in late August.
[1] https://www.phoronix.com/news/Linux-Flatten-The-Pick
[2] https://www.phoronix.com/news/Flatten-The-Pick-v3
[3] https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?h=sched/core&id=85570f10a4c61372c1d437365b9d8dbad512ec6f
[4] https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?h=sched/core&id=fb1050ac8e40aaa014291b5b53612d27890c2b09
[5] https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/log/?h=sched/core