Solus Linux Experimenting With Automated Profiling/Optimizations

[indepe]

I haven't come across such issues as yet. Given my goal is optimizing Solus, I'm making sure that I'm testing what someone would get from the repo (and then push the optimizations straight to the repo). So the tests must use system provided libraries/binaries (directly or indirectly). So a game doesn't have to be built against local libs to test mesa/xorg-server, but it does if I want to test SDL optimizations for example.

Then I can go back and run all the tests with a different kernel configuration/scheduler and see the impact of that.

I didn't clarify that I was referring to a way of reducing the "margin of error" which you mentioned. However it may be too specific to the kind of (function-level) measurements I am making, and may not apply to your situation.

[Yndoendo]

Just checking...where you referring to function multiversioing (FMV) as the posts before yours did?

I'm looking at or testing multi-platform performance across multiple Instruction Set Architecture (ISA)s. Moving a uCOS 2 project to Linux but would like to profile the software against AMD64, ARM, and RSIC-V.

[Yndoendo]

Nothing falls apart. Added architectures will simply use the generic version of a function. The same situation as if there wasn't any multiversioning. Until, optionally, a version taking advantage of special instruction sets on that architecture, is supplied.

So you (potentially) get optimal performance on each platform, for that function, instead of some compromise.

I'm talking about multiple Instruction Set Architecture (ISA)s. Plan on using Gentoo to profile software across different CPUS and specially would like to see the RISC-V benchmarks.