Announcing a new version of MuQSS and a -ck release to go with it in concert with mainline releasing 4.8.5
4.8-ck5 patchset:
http://ck.kolivas.org/patches/4.0/4.8/4.8-ck5/
MuQSS by itself for 4.8:
4.8-sched-MuQSS_120.patch
MuQSS by itself for 4.7:
4.7-sched-MuQSS_120.patch
Git tree:
https://github.com/ckolivas/linux
This is a fairly substantial update to MuQSS which includes bugfixes for the previous version, performance enhancements, new features, and completed documentation. This will likely be the first publicly announced version on LKML.
EDIT: Announce here: LKML
New features:
- MuQSS is now a tickless scheduler. That means it can maintain its guaranteed low latency even in a build configured with a low Hz tick rate. To that end, it is now defaulting to 100Hz, and it is recommended to use this as the default choice for it leads to more throughput and power savings as well.
- Improved performance for single threaded workloads with CPU frequency scaling.
- Full NoHZ now supported. This disables ticks on busy CPUs instead of just idle ones. Unlike mainline, MuQSS can do this virtually all the time, regardless of how many tasks are currently running. However this option is for very specific use cases (compute servers running specific workloads) and not for regular desktops or servers.
- Numerous other configuration options that were previously disabled from mainline are now allowed again (though not recommended for regular users.)
- Completed documentation can now be found in Documentation/scheduler/sched-MuQSS.txt
Bugfixes:
- Fix for the various stalls some people were still experiencing, along with the softirq pending warnings.
- Fix for some loss of CPU for heavily sched_yielding tasks.
- Fix for the BFQ warning (-ck only)
Enjoy!
お楽しみ下さい
-ck
Saturday 29 October 2016
Monday 24 October 2016
Interbench benchmarks for MuQSS 116
As mentioned in my previous post, I recently upgraded interbench which is a benchmark application I invented/wrote to assess perceptible latency in the setting of various loads. The updates were to make the results meaningful on today's larger ram/multicore machines where the load scales accordingly.
The results for mainline 4.8.4 and 4.8.4-ck4 on a multithreaded hexcore (init 1) can be found here:
http://ck.kolivas.org/patches/muqss/Benchmarks/20161024/
and are copied below. I do not have swap on this machine so the "memload" was not performed. This is a 3.6GHz hexcore with 64GB ram and fast Intel SSDs so to show any difference on this is nice. To make it easier, I've highlighted it in colours similar to the throughput benchmarks I posted previously. Blue means within 1% of each other, red means significantly worse and green significantly better.
As you can see, the only times mainline is better, there is less than 1% difference between them which is within the margins for noise. MuQSS meets more deadlines, gives the benchmarked task more of its desired CPU and has substantially lower max latencies.
I'm reasonably confident that I've been able to maintain the interactivity people have come to expect from BFS in the transition to MuQSS now and have the data to support it above.
Enjoy!
お楽しみ下さい
-ck
The results for mainline 4.8.4 and 4.8.4-ck4 on a multithreaded hexcore (init 1) can be found here:
http://ck.kolivas.org/patches/muqss/Benchmarks/20161024/
and are copied below. I do not have swap on this machine so the "memload" was not performed. This is a 3.6GHz hexcore with 64GB ram and fast Intel SSDs so to show any difference on this is nice. To make it easier, I've highlighted it in colours similar to the throughput benchmarks I posted previously. Blue means within 1% of each other, red means significantly worse and green significantly better.
Load set to 12 processors Using 4008580 loops per ms, running every load for 30 seconds Benchmarking kernel 4.8.4 at datestamp 201610242116 Comment: cfs --- Benchmarking simulated cpu of Audio in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.1 +/- 0.1 0.1 100 100 Video 0.0 +/- 0.0 0.1 100 100 X 0.1 +/- 0.1 0.1 100 100 Burn 0.0 +/- 0.0 0.0 100 100 Write 0.1 +/- 0.1 0.1 100 100 Read 0.1 +/- 0.1 0.1 100 100 Ring 0.0 +/- 0.0 0.1 100 100 Compile 0.0 +/- 0.0 0.0 100 100 --- Benchmarking simulated cpu of Video in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.1 +/- 0.1 0.1 100 100 X 0.1 +/- 0.1 0.1 100 100 Burn 17.4 +/- 19.5 46.3 87 7.62 Write 0.1 +/- 0.1 0.1 100 100 Read 0.1 +/- 0.1 0.1 100 100 Ring 0.0 +/- 0.0 0.0 100 100 Compile 17.4 +/- 19.1 45.9 89.5 6.07 --- Benchmarking simulated cpu of X in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.0 +/- 0.1 1.0 100 99.3 Video 13.4 +/- 25.8 68.0 36.2 27.3 Burn 94.4 +/- 127.0 334.0 12.9 4.37 Write 0.1 +/- 0.4 4.0 97.4 96.4 Read 0.1 +/- 0.7 4.0 96.2 93.8 Ring 0.5 +/- 1.9 9.0 89.3 84.9 Compile 93.3 +/- 127.7 333.0 12.2 4.2 --- Benchmarking simulated cpu of Gaming in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU None 0.0 +/- 0.2 2.2 100 Video 7.9 +/- 21.4 69.3 92.7 X 1.4 +/- 1.6 2.7 98.7 Burn 136.5 +/- 145.3 360.8 42.3 Write 1.8 +/- 2.0 4.4 98.2 Read 11.2 +/- 20.3 47.8 89.9 Ring 8.1 +/- 8.1 8.2 92.5 Compile 152.3 +/- 166.8 346.1 39.6
Load set to 12 processors Using 4008580 loops per ms, running every load for 30 seconds Benchmarking kernel 4.8.4-ck4+ at datestamp 201610242047 Comment: muqss116-int1 --- Benchmarking simulated cpu of Audio in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.0 +/- 0.0 0.0 100 100 Video 0.0 +/- 0.0 0.0 100 100 X 0.0 +/- 0.0 0.0 100 100 Burn 0.0 +/- 0.0 0.0 100 100 Write 0.0 +/- 0.0 0.1 100 100 Read 0.0 +/- 0.0 0.0 100 100 Ring 0.0 +/- 0.0 0.0 100 100 Compile 0.0 +/- 0.1 0.8 100 100 --- Benchmarking simulated cpu of Video in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.0 +/- 0.0 0.0 100 100 X 0.0 +/- 0.0 0.0 100 100 Burn 3.1 +/- 7.2 17.7 100 81.6 Write 0.0 +/- 0.0 0.5 100 100 Read 0.0 +/- 0.0 0.0 100 100 Ring 0.0 +/- 0.0 0.0 100 100 Compile 10.5 +/- 13.3 19.7 100 37.3 --- Benchmarking simulated cpu of X in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.0 +/- 0.1 1.0 100 99.3 Video 3.7 +/- 12.1 56.0 89 82.6 Burn 47.2 +/- 66.5 142.0 16.7 7.58 Write 0.1 +/- 0.5 5.0 97.7 95.7 Read 0.1 +/- 0.7 4.0 95.6 93.5 Ring 0.5 +/- 1.9 12.0 89.8 86 Compile 55.9 +/- 77.6 196.0 18.6 8.12 --- Benchmarking simulated cpu of Gaming in the presence of simulated --- Load Latency +/- SD (ms) Max Latency % Desired CPU None 0.0 +/- 0.1 0.5 100 Video 1.2 +/- 1.2 1.8 98.8 X 1.4 +/- 1.6 2.9 98.7 Burn 130.9 +/- 132.1 160.3 43.3 Write 2.4 +/- 2.5 7.0 97.7 Read 3.2 +/- 3.2 3.6 96.9 Ring 5.9 +/- 6.2 10.3 94.4 Compile 146.5 +/- 149.3 209.2 40.6
As you can see, the only times mainline is better, there is less than 1% difference between them which is within the margins for noise. MuQSS meets more deadlines, gives the benchmarked task more of its desired CPU and has substantially lower max latencies.
I'm reasonably confident that I've been able to maintain the interactivity people have come to expect from BFS in the transition to MuQSS now and have the data to support it above.
Enjoy!
お楽しみ下さい
-ck
linux-4.8-ck4, MuQSS CPU scheduler v0.116
Yet another bugfix release for MuQSS and the -ck patchset with one of the most substantial latency fixes yet. Everyone should upgrade if they're on a previous 4.8 patchset of mine. Sorry about the frequency of these releases but I just can't allow a known buggy release be the latest version.
4.8-ck4 patchset:
http://ck.kolivas.org/patches/4.0/4.8/4.8-ck4/
MuQSS by itself for 4.8:
4.8-sched-MuQSS_116.patch
MuQSS by itself for 4.7:
4.7-sched-MuQSS_116.patch
I'm hoping this is the release that allows me to not push any more -ck versions out till 4.9 is released since it addresses all remaining issues that I know about.
A lingering bug that has been troubling me for some time was leading to occasional massive latencies and thanks to some detective work by Serge Belyshev I was able to narrow it down to a single line fix which dramatically improves worst case latency when measured. Throughput is virtually unchanged. The flow-on effect to other areas was also apparent with sometimes unused CPU cycles and weird stalls on some workloads.
Sched_yield was reverted to the old BFS mechanism again which GPU drivers prefer but it wasn't working previously on MuQSS because of the first bug. The difference is substantial now and drivers (such as nvidia proprietary) and apps that use it a lot (such as the folding @ home client) behave much better now.
The late introduced bugs that got into ck3/muqss115 were reverted.
The results come up quite well now with interbench (my latency under load benchmark) which I have recently updated and should now give sensible values:
https://github.com/ckolivas/interbench
If you're baffled by interbench results, the most important number is %deadlines met which should be as close to 100% as possible followed by max latency which should be as low as possible for each section. In the near future I'll announce an official new release version.
Pedro in the comments section previously was using runqlat from bcc tools to test latencies as well, but after some investigation it became clear to me that the tool was buggy and did not work properly with bfs/muqss either so I've provided a slightly updated version here which should work properly:
runqlat.py
Enjoy!
お楽しみ下さい
-ck
4.8-ck4 patchset:
http://ck.kolivas.org/patches/4.0/4.8/4.8-ck4/
MuQSS by itself for 4.8:
4.8-sched-MuQSS_116.patch
MuQSS by itself for 4.7:
4.7-sched-MuQSS_116.patch
I'm hoping this is the release that allows me to not push any more -ck versions out till 4.9 is released since it addresses all remaining issues that I know about.
A lingering bug that has been troubling me for some time was leading to occasional massive latencies and thanks to some detective work by Serge Belyshev I was able to narrow it down to a single line fix which dramatically improves worst case latency when measured. Throughput is virtually unchanged. The flow-on effect to other areas was also apparent with sometimes unused CPU cycles and weird stalls on some workloads.
Sched_yield was reverted to the old BFS mechanism again which GPU drivers prefer but it wasn't working previously on MuQSS because of the first bug. The difference is substantial now and drivers (such as nvidia proprietary) and apps that use it a lot (such as the folding @ home client) behave much better now.
The late introduced bugs that got into ck3/muqss115 were reverted.
The results come up quite well now with interbench (my latency under load benchmark) which I have recently updated and should now give sensible values:
https://github.com/ckolivas/interbench
If you're baffled by interbench results, the most important number is %deadlines met which should be as close to 100% as possible followed by max latency which should be as low as possible for each section. In the near future I'll announce an official new release version.
Pedro in the comments section previously was using runqlat from bcc tools to test latencies as well, but after some investigation it became clear to me that the tool was buggy and did not work properly with bfs/muqss either so I've provided a slightly updated version here which should work properly:
runqlat.py
Enjoy!
お楽しみ下さい
-ck
Saturday 22 October 2016
linux-4.8-ck3, MuQSS version 0.115
This is mainly a bugfix release for those who had boot failures, TOI patched failures, and warnings. Otherwise it only has minor changes.
http://ck.kolivas.org/patches/4.0/4.8/4.8-ck3/
MuQSS version 0.115 by itself:
4.8-sched-MuQSS_115.patch
Git tree includes branches for MuQSS and -ck:
https://github.com/ckolivas/linux
EDIT: There is a regression in this release as well and you need to either grab the latest 4.8-ck git tree or add the two patches here:
http://ck.kolivas.org/patches/muqss/4.0/4.8/Pending/
Sorry, when enough other problems get fixed I'll release another version pretty soon too.
Enjoy!
お楽しみ下さい
-ck
http://ck.kolivas.org/patches/4.0/4.8/4.8-ck3/
MuQSS version 0.115 by itself:
4.8-sched-MuQSS_115.patch
Git tree includes branches for MuQSS and -ck:
https://github.com/ckolivas/linux
EDIT: There is a regression in this release as well and you need to either grab the latest 4.8-ck git tree or add the two patches here:
http://ck.kolivas.org/patches/muqss/4.0/4.8/Pending/
Sorry, when enough other problems get fixed I'll release another version pretty soon too.
Enjoy!
お楽しみ下さい
-ck
Friday 21 October 2016
lrzip version 0.631
Announcing an updated version of lrzip.
Tarballs:
http://ck.kolivas.org/apps/lrzip/
Git tree:
https://github.com/ckolivas/lrzip
This is a minor bugfix release.
- Encryption complexity has been altered to match CPU speed rate rises that have NOT paralleled Moore's law.
- Some of the command line parameters did not work properly in compatibility mode.
- Compressed files did not retain the same date as the original file.
- The -p parameter did not accept arguments and would not work.
Enjoy!
お楽しみ下さい
-ck
Tarballs:
http://ck.kolivas.org/apps/lrzip/
Git tree:
https://github.com/ckolivas/lrzip
This is a minor bugfix release.
- Encryption complexity has been altered to match CPU speed rate rises that have NOT paralleled Moore's law.
- Some of the command line parameters did not work properly in compatibility mode.
- Compressed files did not retain the same date as the original file.
- The -p parameter did not accept arguments and would not work.
Enjoy!
お楽しみ下さい
-ck
linux-4.8-ck2, MuQSS version 0.114
Announcing an updated version, and the first -ck release with MuQSS as the scheduler, officially retiring BFS from further development, in line with the diminished rate of bug reports with MuQSS. It is clear that the little attention BFS had received over the years apart from rushed synchronisation with mainline had cause a number of bugs to creep in and MuQSS is basically a rewritten evolution of the same code so it makes no sense to maintain both.
http://ck.kolivas.org/patches/4.0/4.8/4.8-ck2/
MuQSS version 0.114 by itself:
4.8-sched-MuQSS_114.patch
Git tree includes branches for MuQSS and -ck:
https://github.com/ckolivas/linux
In addition to the most up to date version of MuQSS replacing BFS, this is the first release with BFQ included. It is configurable and is set by default in -ck though it is entirely optional.
The MuQSS changes since 112 are as follows:
- Added cacheline alignment to atomic variables courtesy of Holger Hoffstätte
- Fixed PPC build courtesy of Serge Belyshev.
- Implemented wake lists for separate CPU packages.
- Send hotplug threads to CPUs even if they're not alive yet since they'll be enabling them.
- Build fixes for uniprocessor.
- A substantial revamp of the sub-tick process accounting, decreasing the number of variables used, simplifying the code, and increasing the resolution to nanosecond accounting. Now even tasks that run for less than 100us will not escape visible accounting.
This release should bring slightly better performance, more so on multi-cpu machines, and fairer accounting/latency.
Enjoy!
お楽しみ下さい
-ck
http://ck.kolivas.org/patches/4.0/4.8/4.8-ck2/
MuQSS version 0.114 by itself:
4.8-sched-MuQSS_114.patch
Git tree includes branches for MuQSS and -ck:
https://github.com/ckolivas/linux
In addition to the most up to date version of MuQSS replacing BFS, this is the first release with BFQ included. It is configurable and is set by default in -ck though it is entirely optional.
The MuQSS changes since 112 are as follows:
- Added cacheline alignment to atomic variables courtesy of Holger Hoffstätte
- Fixed PPC build courtesy of Serge Belyshev.
- Implemented wake lists for separate CPU packages.
- Send hotplug threads to CPUs even if they're not alive yet since they'll be enabling them.
- Build fixes for uniprocessor.
- A substantial revamp of the sub-tick process accounting, decreasing the number of variables used, simplifying the code, and increasing the resolution to nanosecond accounting. Now even tasks that run for less than 100us will not escape visible accounting.
This release should bring slightly better performance, more so on multi-cpu machines, and fairer accounting/latency.
Enjoy!
お楽しみ下さい
-ck
Tuesday 18 October 2016
First MuQSS Throughput Benchmarks
The short version graphical summary:
Red = MuQSS 112 interactive off
Purple = MuQSS 112 interactive on
Blue = CFS
The detail:
http://ck.kolivas.org/patches/muqss/Benchmarks/20161018/
I went on a journey looking for meaningful benchmarks to conduct to assess the scalability aspect as far as I could on my own 12x machine and was really quite depressed to see what the benchmark situation on linux is like. Only the old and completely invalid benchmarks seem to still be hanging around in public sites and promoted, like Reaim, aim7, dbench, volanomark, etc. and none of those are useful scalability benchmarks. Even more depressing was the only ones with any reputation are actually commercial benchmarks costing hundreds of dollars.
This made me wonder out loud just how the heck mainline is even doing scalability improvements if there are precious few valid benchmarks for linux and no one's using them. The most promising ones, like mosbench, need multiple machines and quite a bit of set up to get them going.
I spent a day wading through the phoronix test suite - a site and its suite not normally known for meaningful high performance computing discussion and benchmarks - looking for benchmarks that could be used for meaningful results for multicore scalability assessment and were not too difficult to deploy and came up with the following collection:
John The Ripper - a CPU bound application that is threaded to the number of CPUs and intermittently drops to one thread making for slightly more interesting behaviour than just a fully CPU bound workload.
7-Zip Compression - a valid real world CPU bound application that is threaded but rarely able to spread out to all CPUs making it an interesting light load benchmark.
ebizzy - This emulates a heavy content delivery server load which scales beyond the number of CPUs and emulates what goes on between a http server and database.
Timed Linux Kernel Compilation - A perennial favourite because it is a real world case and very easy to reproduce. Despite numerous complaints about its validity as a benchmark, it is surprisingly consistent in its results and tests many facets of scalability, though does not scale to use all CPUs at all time either.
C-Ray - A ray tracing benchmark that uses massive threading per CPU and is completely CPU bound but overloads all CPUs.
Primesieve - A prime number generator that is threaded to the number of CPUs exactly, is fully CPU bound and is cache intensive.
PostgreSQL pgbench - A meaningful database benchmark that is done at 3 different levels - single threaded, normal loaded and heavily contended, each testing different aspects of scalability.
And here is a set of results comparing 4.8.2 mainline (labelled CFS), MuQSS 112 in interactive mode (MuQSS-int1) and MuQSS 112 in non-interactive mode (MuQSS-int0):
http://ck.kolivas.org/patches/muqss/Benchmarks/20161018/
It's worth noting that there is quite a bit of variance in these benchmarks and some are bordering on the difference being just noise. However there is a clear pattern here - when the load is light, in terms of throughput, CFS outperforms MuQSS. When load is heavy, the heavier it gets, MuQSS outperforms CFS, especially in non-interactive mode. As a friend noted, for the workloads where you wouldn't be running MuQSS in interactive mode, such as a web server, database etc, non-interactive mode is of clear performance benefit. So at least on the hardware I had available to me, on a 12x machine, MuQSS is scaling better than mainline on these workloads as load increases.
The obvious question people will ask is why MuQSS doesn't perform better at light loads, and in fact I have an explanation. The reason is that mainline tends to cling to processes much more so that if it is hovering at low numbers of active processes, they'll all cluster on one CPU or fewer CPUs than being spread out everywhere. This means the CPU benefits more from the turbo modes virtually all newer CPUs have, but it comes at a cost. The latency to tasks is greater because they're competing for CPU time on fewer busy CPUs rather than spreading out to idle cores or threads. It is a design decision in MuQSS, as taken from BFS, to always spread out to any idle CPUs if they're available, to minimise latency, and that's one of the reasons for the interactivity and responsiveness of MuQSS. Of course I am still investigating ways of closing that gap further.
Hopefully I can get some more benchmarks from someone with even bigger hardware, and preferably with more than one physical package since that's when things really start getting interesting. All in all I'm very pleased with the performance of MuQSS in terms of scalability on these results, especially assuming I'm able to maintain the interactivity of BFS which were my dual goals.
There is MUCH more to benchmarking than pure throughput of CPU - which is almost the only thing these benchmarks is checking - but that's what I'm interested in here. I hope that providing my list of easy to use benchmarks and the reasoning behind them can generate interest in some kind of meaningful standard set of benchmarks. I did start out in kernel development originally after writing and being a benchmarker :P
To aid that, I'll give simple instructions here for how to ~imitate the benchmarks and get results like I've produced above.
Download the phoronix test suite from here:
http://www.phoronix-test-suite.com/
The generic tar.gz is perfectly fine. Then extract it and install the relevant benchmarks like so:
Now obviously this is not ideal since you shouldn't run benchmarks on a multiuser login with Xorg and all sorts of other crap running so I actually always run benchmarks at init level 1.
Enjoy!
お楽しみ下さい
-ck
Red = MuQSS 112 interactive off
Purple = MuQSS 112 interactive on
Blue = CFS
The detail:
http://ck.kolivas.org/patches/muqss/Benchmarks/20161018/
I went on a journey looking for meaningful benchmarks to conduct to assess the scalability aspect as far as I could on my own 12x machine and was really quite depressed to see what the benchmark situation on linux is like. Only the old and completely invalid benchmarks seem to still be hanging around in public sites and promoted, like Reaim, aim7, dbench, volanomark, etc. and none of those are useful scalability benchmarks. Even more depressing was the only ones with any reputation are actually commercial benchmarks costing hundreds of dollars.
This made me wonder out loud just how the heck mainline is even doing scalability improvements if there are precious few valid benchmarks for linux and no one's using them. The most promising ones, like mosbench, need multiple machines and quite a bit of set up to get them going.
I spent a day wading through the phoronix test suite - a site and its suite not normally known for meaningful high performance computing discussion and benchmarks - looking for benchmarks that could be used for meaningful results for multicore scalability assessment and were not too difficult to deploy and came up with the following collection:
John The Ripper - a CPU bound application that is threaded to the number of CPUs and intermittently drops to one thread making for slightly more interesting behaviour than just a fully CPU bound workload.
7-Zip Compression - a valid real world CPU bound application that is threaded but rarely able to spread out to all CPUs making it an interesting light load benchmark.
ebizzy - This emulates a heavy content delivery server load which scales beyond the number of CPUs and emulates what goes on between a http server and database.
Timed Linux Kernel Compilation - A perennial favourite because it is a real world case and very easy to reproduce. Despite numerous complaints about its validity as a benchmark, it is surprisingly consistent in its results and tests many facets of scalability, though does not scale to use all CPUs at all time either.
C-Ray - A ray tracing benchmark that uses massive threading per CPU and is completely CPU bound but overloads all CPUs.
Primesieve - A prime number generator that is threaded to the number of CPUs exactly, is fully CPU bound and is cache intensive.
PostgreSQL pgbench - A meaningful database benchmark that is done at 3 different levels - single threaded, normal loaded and heavily contended, each testing different aspects of scalability.
And here is a set of results comparing 4.8.2 mainline (labelled CFS), MuQSS 112 in interactive mode (MuQSS-int1) and MuQSS 112 in non-interactive mode (MuQSS-int0):
http://ck.kolivas.org/patches/muqss/Benchmarks/20161018/
It's worth noting that there is quite a bit of variance in these benchmarks and some are bordering on the difference being just noise. However there is a clear pattern here - when the load is light, in terms of throughput, CFS outperforms MuQSS. When load is heavy, the heavier it gets, MuQSS outperforms CFS, especially in non-interactive mode. As a friend noted, for the workloads where you wouldn't be running MuQSS in interactive mode, such as a web server, database etc, non-interactive mode is of clear performance benefit. So at least on the hardware I had available to me, on a 12x machine, MuQSS is scaling better than mainline on these workloads as load increases.
The obvious question people will ask is why MuQSS doesn't perform better at light loads, and in fact I have an explanation. The reason is that mainline tends to cling to processes much more so that if it is hovering at low numbers of active processes, they'll all cluster on one CPU or fewer CPUs than being spread out everywhere. This means the CPU benefits more from the turbo modes virtually all newer CPUs have, but it comes at a cost. The latency to tasks is greater because they're competing for CPU time on fewer busy CPUs rather than spreading out to idle cores or threads. It is a design decision in MuQSS, as taken from BFS, to always spread out to any idle CPUs if they're available, to minimise latency, and that's one of the reasons for the interactivity and responsiveness of MuQSS. Of course I am still investigating ways of closing that gap further.
Hopefully I can get some more benchmarks from someone with even bigger hardware, and preferably with more than one physical package since that's when things really start getting interesting. All in all I'm very pleased with the performance of MuQSS in terms of scalability on these results, especially assuming I'm able to maintain the interactivity of BFS which were my dual goals.
There is MUCH more to benchmarking than pure throughput of CPU - which is almost the only thing these benchmarks is checking - but that's what I'm interested in here. I hope that providing my list of easy to use benchmarks and the reasoning behind them can generate interest in some kind of meaningful standard set of benchmarks. I did start out in kernel development originally after writing and being a benchmarker :P
To aid that, I'll give simple instructions here for how to ~imitate the benchmarks and get results like I've produced above.
Download the phoronix test suite from here:
http://www.phoronix-test-suite.com/
The generic tar.gz is perfectly fine. Then extract it and install the relevant benchmarks like so:
tar xf phoronix-test-suite-6.6.1.tar.gz
cd phoronix-test-suite
./phoronix-test-suite install build-linux-kernel c-ray compress-7zip ebizzy john-the-ripper pgbench primesieve
./phoronix-test-suite default-run build-linux-kernel c-ray compress-7zip ebizzy john-the-ripper pgbench primesieve
Now obviously this is not ideal since you shouldn't run benchmarks on a multiuser login with Xorg and all sorts of other crap running so I actually always run benchmarks at init level 1.
Enjoy!
お楽しみ下さい
-ck
Monday 17 October 2016
MuQSS - The Multiple Queue Skiplist Scheduler v0.112
Here's an updated version of MuQSS.
For 4.8.*:
4.8-sched-MuQSS_112.patch
For 4.7.*:
4.7-sched-MuQSS_112.patch
Git tree here as 4.7-muqss or 4.8-muqss branches:
https://github.com/ckolivas/linux
It's getting close now to the point where it can replace BFS in -ck releases. Thanks to the many people testing and reporting back, some other misbehaviours were discovered and their associated fixes have been committed.
In particular,
- Balancing across CPUs was not looking at higher and lower scheduling policies correctly (SCHED_ISO, SCHED_IDLEPRIO and realtime policies)
- A serious stall/hang could happen with tasks using sched_yield (such as f@h client and numerous GPU drivers)
- Some minor accounting issues on new tasks with affinity set were fixed
- Overhead was further decreased on task selection
- Spurious preemption on CPUs where the preempted task had already gone are now avoided
- Spurious wakeup on CPUs that were assumed and are no longer idle are avoided
- A potential race in suspending to ram was fixed
- Old unused code from BFS was removed, along with unnecessary intermediate variables.
- Clean ups
- Some work towards actually documenting MuQSS in Documentation/scheduler/sched-MuQSS.txt was done, though incomplete.
Enjoy!
お楽しみ下さい
-ck
For 4.8.*:
4.8-sched-MuQSS_112.patch
For 4.7.*:
4.7-sched-MuQSS_112.patch
Git tree here as 4.7-muqss or 4.8-muqss branches:
https://github.com/ckolivas/linux
It's getting close now to the point where it can replace BFS in -ck releases. Thanks to the many people testing and reporting back, some other misbehaviours were discovered and their associated fixes have been committed.
In particular,
- Balancing across CPUs was not looking at higher and lower scheduling policies correctly (SCHED_ISO, SCHED_IDLEPRIO and realtime policies)
- A serious stall/hang could happen with tasks using sched_yield (such as f@h client and numerous GPU drivers)
- Some minor accounting issues on new tasks with affinity set were fixed
- Overhead was further decreased on task selection
- Spurious preemption on CPUs where the preempted task had already gone are now avoided
- Spurious wakeup on CPUs that were assumed and are no longer idle are avoided
- A potential race in suspending to ram was fixed
- Old unused code from BFS was removed, along with unnecessary intermediate variables.
- Clean ups
- Some work towards actually documenting MuQSS in Documentation/scheduler/sched-MuQSS.txt was done, though incomplete.
Enjoy!
お楽しみ下さい
-ck
Tuesday 11 October 2016
MuQSS - The Multiple Queue Skiplist Scheduler v0.111
Lots of bugfixes, lots of improvements, build fixes, you name it.
For 4.8:
4.8-sched-MuQSS_111.patch
For 4.7:
4.7-sched-MuQSS_111.patch
And in a complete departure from BFS, a git tree (which suits constant development like this, unlike BFS's stable release massive ports):
https://github.com/ckolivas/linux
Look in the pending/ directory to see all the patches that went into this or read the git changelog. In particular numerous warnings were fixed, throughput improved compared to 108, SCHED_ISO was rewritten for multiple queues, potential races/crashes were addressed, and build fixes for different configurations were committed.
I haven't been able to track the bizarre latency issues reported by runqlat and when I try to reproduce it myself I get nonsense values of latency greater than the history of the earth so I suspect an interface bug with BPF reporting values. It doesn't seem to affect actual latency in any way.
EDIT: Updated to version 0.111 which has a fix for suspend/resume.
Enjoy!
お楽しみ下さい
-ck
For 4.8:
4.8-sched-MuQSS_111.patch
For 4.7:
4.7-sched-MuQSS_111.patch
And in a complete departure from BFS, a git tree (which suits constant development like this, unlike BFS's stable release massive ports):
https://github.com/ckolivas/linux
Look in the pending/ directory to see all the patches that went into this or read the git changelog. In particular numerous warnings were fixed, throughput improved compared to 108, SCHED_ISO was rewritten for multiple queues, potential races/crashes were addressed, and build fixes for different configurations were committed.
I haven't been able to track the bizarre latency issues reported by runqlat and when I try to reproduce it myself I get nonsense values of latency greater than the history of the earth so I suspect an interface bug with BPF reporting values. It doesn't seem to affect actual latency in any way.
EDIT: Updated to version 0.111 which has a fix for suspend/resume.
Enjoy!
お楽しみ下さい
-ck
Friday 7 October 2016
MuQSS - The Multiple Queue Skiplist Scheduler v0.108
A new version of the MuQSS CPU scheduler
Incrementals and full patches available for 4.8 and 4.7 respectively here:
http://ck.kolivas.org/patches/muqss/4.0/4.8/
http://ck.kolivas.org/patches/muqss/4.0/4.7/
Yet more minor bugfixes and some important performance enhancements.
This version brings to the table the same locking scheme for trying to wake tasks up as mainline which is advantageous on process busy workloads and many CPUs. This is important because the main reason for moving to multiple runqueues was to minimise lock contention for the global runqueue lock that is in BFS (as mentioned here numerous times before) and this wake up scheme helps make the most of the multiple discrete runqueue locks.
Note this change is much more significant than the last releases so new instability is a possibility. Please report any problems or stacktraces!
There was a workload when I started out that I used lockstat to debug to get an idea of how much lock contention was going on and how long it lasted. Originally with the first incarnations of MuQSS on a 14 second benchmark with thousands of tasks on a 12x CPU it obtained 3 million locks and had almost 300k contentions with the longest contention lasting 80us. Now the same workload grabs the lock just 5k times with only 18 contentions in total and the longest lasted 1us.
This clearly demonstrates that the target endpoint for avoiding lock contention has been achieved. It does not translate into performance improvements on ordinary hardware today because you need ridiculous workloads on many CPUs to even begin deriving advantage from it. However as even our phones now have reached 8 logical CPUs, it will only be a matter of time before 16 threads appears on commodity hardware - a complaint that was directed at BFS when it came out 7 years ago but they still haven't appeared just yet. BFS was shown to be scalable for all workloads up to 16 CPUs, and beyond for certain workloads, but suffered dramatically for others. MuQSS now makes it possible for what was BFS to be useful much further into the future.
Again - MuQSS is aimed primarily at desktop/laptop/mobile device users for the best possible interactivity and responsiveness, and is still very simple in its approach to balancing workloads to CPUs so there are likely to be throughput workloads on mainline that outperform it, though there are almost certainly workloads where the opposite is true.
I've now addressed all planned changes to MuQSS and plan to hopefully only look at bug reports instead of further development from here on for a little while. In my eyes it is now stable enough to replace BFS in the next -ck release barring some unexpected showstopper bug appearing.
EDIT: If you blinked you missed the 107 announcement which was shortly superseded by 108.
EDIT2: Always watch the pending directory for updated pending patches to add.
http://ck.kolivas.org/patches/muqss/4.0/4.8/Pending/
Enjoy!
お楽しみ下さい
-ck
Incrementals and full patches available for 4.8 and 4.7 respectively here:
http://ck.kolivas.org/patches/muqss/4.0/4.8/
http://ck.kolivas.org/patches/muqss/4.0/4.7/
Yet more minor bugfixes and some important performance enhancements.
This version brings to the table the same locking scheme for trying to wake tasks up as mainline which is advantageous on process busy workloads and many CPUs. This is important because the main reason for moving to multiple runqueues was to minimise lock contention for the global runqueue lock that is in BFS (as mentioned here numerous times before) and this wake up scheme helps make the most of the multiple discrete runqueue locks.
Note this change is much more significant than the last releases so new instability is a possibility. Please report any problems or stacktraces!
There was a workload when I started out that I used lockstat to debug to get an idea of how much lock contention was going on and how long it lasted. Originally with the first incarnations of MuQSS on a 14 second benchmark with thousands of tasks on a 12x CPU it obtained 3 million locks and had almost 300k contentions with the longest contention lasting 80us. Now the same workload grabs the lock just 5k times with only 18 contentions in total and the longest lasted 1us.
This clearly demonstrates that the target endpoint for avoiding lock contention has been achieved. It does not translate into performance improvements on ordinary hardware today because you need ridiculous workloads on many CPUs to even begin deriving advantage from it. However as even our phones now have reached 8 logical CPUs, it will only be a matter of time before 16 threads appears on commodity hardware - a complaint that was directed at BFS when it came out 7 years ago but they still haven't appeared just yet. BFS was shown to be scalable for all workloads up to 16 CPUs, and beyond for certain workloads, but suffered dramatically for others. MuQSS now makes it possible for what was BFS to be useful much further into the future.
Again - MuQSS is aimed primarily at desktop/laptop/mobile device users for the best possible interactivity and responsiveness, and is still very simple in its approach to balancing workloads to CPUs so there are likely to be throughput workloads on mainline that outperform it, though there are almost certainly workloads where the opposite is true.
I've now addressed all planned changes to MuQSS and plan to hopefully only look at bug reports instead of further development from here on for a little while. In my eyes it is now stable enough to replace BFS in the next -ck release barring some unexpected showstopper bug appearing.
EDIT: If you blinked you missed the 107 announcement which was shortly superseded by 108.
EDIT2: Always watch the pending directory for updated pending patches to add.
http://ck.kolivas.org/patches/muqss/4.0/4.8/Pending/
Enjoy!
お楽しみ下さい
-ck
Wednesday 5 October 2016
MuQSS - The Multiple Queue Skiplist Scheduler v0.106
Another day and time for yet another release.
There are 0.106 versions and incrementals available for linux-4.7:
http://ck.kolivas.org/patches/muqss/4.0/4.7/
and linux-4.8:
http://ck.kolivas.org/patches/muqss/4.0/4.8
Two large remaining races that could lead to warnings, stalls, or in the worst case, crashes, have been fixed in this version.
Additionally the multiple-runqueue locking has been significantly optimised to take only the runqueues needed for as long as they're needed only and dropped as soon as possible which should bring the lock contention levels down even further. This is a performance enhancement, more so in non-interactive mode, though it will only start being demonstrable if you're lucky enough to have many CPUs.
This version addresses all the known bugs and warnings I've received to date so hopefully I can have a little rest and let people out there actually give it a go. What will you expect if you use this instead of BFS? If I've done this correctly, you will notice absolutely no difference since the idea was to preserve the interactivity and responsiveness of BFS and make it scalable to more CPUs than most people can afford.
Keep the feedback coming, thanks.
Enjoy!
お楽しみ下さい
-ck
There are 0.106 versions and incrementals available for linux-4.7:
http://ck.kolivas.org/patches/muqss/4.0/4.7/
and linux-4.8:
http://ck.kolivas.org/patches/muqss/4.0/4.8
Two large remaining races that could lead to warnings, stalls, or in the worst case, crashes, have been fixed in this version.
Additionally the multiple-runqueue locking has been significantly optimised to take only the runqueues needed for as long as they're needed only and dropped as soon as possible which should bring the lock contention levels down even further. This is a performance enhancement, more so in non-interactive mode, though it will only start being demonstrable if you're lucky enough to have many CPUs.
This version addresses all the known bugs and warnings I've received to date so hopefully I can have a little rest and let people out there actually give it a go. What will you expect if you use this instead of BFS? If I've done this correctly, you will notice absolutely no difference since the idea was to preserve the interactivity and responsiveness of BFS and make it scalable to more CPUs than most people can afford.
Keep the feedback coming, thanks.
Enjoy!
お楽しみ下さい
-ck
Tuesday 4 October 2016
MuQSS - The Multiple Queue Skiplist Scheduler v0.105
I spent the last few days fighting with various lock debugging techniques and the numerous bug reports and am pleased to announce a new version of MuQSS, version 0.105
There are versions and incrementals available for linux-4.7:
http://ck.kolivas.org/patches/muqss/4.0/4.7/
and linux-4.8:
http://ck.kolivas.org/patches/muqss/4.0/4.8
If you've been waiting for me to say it's stable enough to try, then now's your chance for I've addressed all known bugs at this time and it's working well for me.
Most of the issues were to do with races and unstable handling of cross-cpu task movement. No effort went into improving performance from 104 though this version should address many of the crashes and hangs that have been reported with earlier versions.
Additionally there is a pending patch being uploaded for BFS512 which, as per usual, had some last minute issues that only just showed up. If enough users complain loudly enough or more issues show up I might just release another bfs and -ck since it should be stable, especially being one of the last BFS releases.
http://ck.kolivas.org/patches/bfs/4.0/4.8/Pending/
Keep the feedback and bug reports coming. Next I need to put more care into the non-interactive mode of muqss for your enjoyment.
Enjoy!
お楽しみ下さい
-ck
There are versions and incrementals available for linux-4.7:
http://ck.kolivas.org/patches/muqss/4.0/4.7/
and linux-4.8:
http://ck.kolivas.org/patches/muqss/4.0/4.8
If you've been waiting for me to say it's stable enough to try, then now's your chance for I've addressed all known bugs at this time and it's working well for me.
Most of the issues were to do with races and unstable handling of cross-cpu task movement. No effort went into improving performance from 104 though this version should address many of the crashes and hangs that have been reported with earlier versions.
Additionally there is a pending patch being uploaded for BFS512 which, as per usual, had some last minute issues that only just showed up. If enough users complain loudly enough or more issues show up I might just release another bfs and -ck since it should be stable, especially being one of the last BFS releases.
http://ck.kolivas.org/patches/bfs/4.0/4.8/Pending/
Keep the feedback and bug reports coming. Next I need to put more care into the non-interactive mode of muqss for your enjoyment.
Enjoy!
お楽しみ下さい
-ck
Monday 3 October 2016
BFS version 0.512, linux-4.8-ck1, MuQSS for linux-4.8
This is to announce an updated version of BFS for the new stable linux kernel 4.8.
BFS by itself:
4.8-sched-bfs-512.patch
-ck patches with BFS:
4.8-ck1
EDIT: Here's a bugfix post release for the above kernels that I highly recommend you include:
http://ck.kolivas.org/patches/bfs/4.0/4.8/Pending/bfs512-fixes.patch
Following on from the aggressive development towards a new scheduler, this BFS incorporates a number of fixes and performance improvements discovered while working on the Multiple Queue Skiplist Scheduler, MuQSS (pronounced mux) and should be the best performing BFS yet.
Note that this may be the last BFS based -ck release as MuQSS is designed to replace it, being the logical evolution of the same scheduler into a more scalable discrete runqueue design.
For those willing to try it in its current version, an incremental patch can be applied to BFS 512:
bfs512-muqss104.patch
or there is a full patch against 4.8:
4.8-sched-MuQSS_104.patch
Again, MuQSS is still immature code and while I have been running it stably for a few days now, and have spent a lot of time debugging locking issues and stability, it is not intended for production use just yet. Having said that, all testing is most welcome, especially benchmarks and stacktraces if you get any crashes.
I've been asked numerous times why I decided to change the name. There are two major reasons. The first is that it signifies just what a dramatic overhaul to the codebase it is, where it is virtually a new scheduler, even though it uses the same scheduling decision policy as BFS. The second is that I've had many people approach me saying they would like to use BFS for their own production environment but alas the offensive name is a showstopper for them. Additionally I had to choose a name that wasn't being used by anything else which both BFS and brainfuck had been used before.
Enjoy!
お楽しみ下さい
-ck
BFS by itself:
4.8-sched-bfs-512.patch
-ck patches with BFS:
4.8-ck1
EDIT: Here's a bugfix post release for the above kernels that I highly recommend you include:
http://ck.kolivas.org/patches/bfs/4.0/4.8/Pending/bfs512-fixes.patch
Following on from the aggressive development towards a new scheduler, this BFS incorporates a number of fixes and performance improvements discovered while working on the Multiple Queue Skiplist Scheduler, MuQSS (pronounced mux) and should be the best performing BFS yet.
Note that this may be the last BFS based -ck release as MuQSS is designed to replace it, being the logical evolution of the same scheduler into a more scalable discrete runqueue design.
For those willing to try it in its current version, an incremental patch can be applied to BFS 512:
bfs512-muqss104.patch
or there is a full patch against 4.8:
4.8-sched-MuQSS_104.patch
Again, MuQSS is still immature code and while I have been running it stably for a few days now, and have spent a lot of time debugging locking issues and stability, it is not intended for production use just yet. Having said that, all testing is most welcome, especially benchmarks and stacktraces if you get any crashes.
I've been asked numerous times why I decided to change the name. There are two major reasons. The first is that it signifies just what a dramatic overhaul to the codebase it is, where it is virtually a new scheduler, even though it uses the same scheduling decision policy as BFS. The second is that I've had many people approach me saying they would like to use BFS for their own production environment but alas the offensive name is a showstopper for them. Additionally I had to choose a name that wasn't being used by anything else which both BFS and brainfuck had been used before.
Enjoy!
お楽しみ下さい
-ck
Saturday 1 October 2016
MuQSS - The Multiple Queue Skiplist Scheduler v0.105
Announcing a multiple runqueue variant of BFS, with the more mundane name of MuQSS (pronounced mux) for linux 4.7:
Full patch for linux-4.7
4.7-sched-MuQSS_105.patch
Keep watching this blog for newer versions!
Incremental to patch bfs502 to MuQSS 0.1:
bfs502-MuQSS_103.patch
It was inevitable that one day I would find myself tackling the 2 major scalability limitations in BFS and this is the result of it. These two issues were
- The single runqueue which means all CPUs would fight for lock contention over the one runqueue, and
- The O(n) look up which means linear increase in overhead for task lookups as number of processes increases.
Till now I did not have the energy nor time to try and find a solution for number 1. that maintained BFS' scheduling decision algorithm as the single runqueue was actually the reason latency remains bound and deterministic on BFS, capitalising with more CPUs instead of fighting against them for scalability.
This scheduler variant is an evolution of BFS, which hopefully will be mature enough to replace BFS one day when stability is assured. It is able to still use the same scheduling algorithm as BFS meaning latency and responsiveness remains as good as always, but with the per-CPU runqueue and discrete locking, it also means it will scale to any number of CPUs, as the mainline scheduler does.
It does NOT guarantee the best possible throughput as there still is virtually no complex balancing mechanism whatsoever, selecting tasks according to deadline primarily with only CPU cache distances being used to determine which idle CPU to go to, or in non-interactive mode, which overloaded CPU to pull from to fill an idle CPU.
It would be possible, with a lot of effort, to wedge the entire balancing algorithm for scalability from mainline into this, though it will probably offset the deterministic latency that makes it special.
This is a massive rewrite and consequently there are bound to still be race conditions and hidden bugs though I have been running it for a while now with reasonable stability. I'm putting this out there for the braver people to test. There's a lot more to document about it but for now let's just say, give it a try.
Please don't use any lock debugging as it will light up every possible complaint for the time being!
Regarding 4.8, for the time being I will still be releasing BFS for it and incorporate it into -ck
EDIT: Updated to version 0.105 with significant bugfixes.
Enjoy!
お楽しみ下さい
-ck
Subscribe to:
Posts (Atom)