Updated EEVDF Linux CPU Scheduler Patches Posted That Plan To Replace CFS

[phoronix]

Phoronix: Updated EEVDF Linux CPU Scheduler Patches Posted That Plan To Replace CFS

Intel engineer Peter Zijlstra on Wednesday posted the latest patches for the EEVDF scheduler, the Earliest Eligible Virtual Deadline First approach that is based on a research paper from the late 90's. Ultimately the hope is for EEVDF to replace the existing CFS scheduler code...

https://www.phoronix.com/news/EEVDF-Scheduler-Linux-EO-May

[Kjell]

Benchmarks?

[V1tol]

Benchmarks?

Were in previous article and more detailed can be found in patchset. Using this for some time and it is working extremely good under heavy load. And definitely much better than PDS/BMQ on which some software (like games) can hang unexpectedly.

[Kjell]

Would be interesting to see how EEVDF performs against other schedulers like BORE, Task Type/TT (formally Cacule) and CFS. Not sure if PDS & BMQ are still relevant today

[timofonic]

Would be interesting to see how EEVDF performs against other schedulers like BORE, Task Type/TT (formally Cacule) and CFS. Not sure if PDS & BMQ are still relevant today

I agree! I would love a comparison benchmark!

[ptr1337]

Would be interesting to see how EEVDF performs against other schedulers like BORE, Task Type/TT (formally Cacule) and CFS. Not sure if PDS & BMQ are still relevant today

In terms of throughput there won't be much difference. A bunch from your mentioned schedulers are nowadays not well maintained anymore.
TT gets rarely updates and I just port it over and over to every new kernel version.
PDS/BMQ has a lot of bugs. May he will rewrite it soon, who knows, but it its current state it is often not really use able and the performance improvement as it was some years ago, when CFS was not really fine for desktops.

Anyways, the developer of the BORE scheduler made a little graph, with the tool from Julia, which is the creator of the "NEST" Scheduling.
Here the link:

https://gitlab.inria.fr/schedgraph/schedgraph/-/tree/main

The graph comparison in scheduler characteristics. BORE, CFS, EEVDF and the BORE-EEVDF variant got there compared. This was on a 5800X.

https://imgur.com/a/ByXpvuW

On CachyOS with the default kernel we use nowadays the "eevdf-bore" variant, since it does improve the heavy load responsives and still provides the latency enhancements from EEVDF. Also we use the latency nice interface really much together with our ananicy-cpp and its rules.

Looking forward that this gets merged, using it since around 2 months and love it.

[Vistaus]

/me waits for people to complain about the fact that this is based on a research paper from the 90's

[Kjell]

How can you use 2 schedulers simultaneously (eevdf-bore) ptr1337?

[ptr1337]

How can you use 2 schedulers simultaneously (eevdf-bore) ptr1337?

If you use the "default" BORE Scheduler, you are also using "two" schedulers with CFS. BORE does not replace the complete base scheduling modell, it does enhance it.
Since the BORE developer is in our team, we started very early to adjust BORE to EEVDF.

You can find it in our kernel-patches directory.

In my eyes it is not really smart to replace a complete scheduler, like PDS/BMQ does since the CFS/EEVDF Scheduler gets way more tested against many hardware and other regressions. So improving the scheduling for desktops is just a better idea.

[binarybanana]

I'm always amazed how smooth my system boots when I screw something up and use my old -ck based rescue kernel. With the stock kernel it looks like the system stutters starting up services while with -ck it's smooth as butter. I don't know if it's actually faster but the smoothess is hypnotizing so it certainly feels that way. It's really weird. Not sure if it's really just the MuQSS scheduler or something else in the patch set.