Please, hint me a meaningfull performance test

[ulenrich]

I am closely following kernel.org stable-queue. Today I am feeling some performance regression running
linux-3.6.9-rc1

But it is just a feeling. I need some sensible hint howto begin with performance testing!
phoronix-testsuite?
_________________
fun2gen2

[aCOSwt]

Well, it depends a lot on which performances you want to measure on which kind of kernel.

The phoronix test suite looks like a gas factory to me.

This is what I rely on :

1/ Sysbench (thread and mutex tests) for global scheduler / multithreading efficiency
2/ IOZone for Disk IO thoughput
3/ Interbench for scheduler latency

BTW... I do confirm performance regressions with early 3.6 (That started with 3.5)
And... if upstream wants to solve the "kernel mount slow" problem with the ideas they get then... I fear about 3.7...
_________________
In theory there are no differences between theory and practice. In practice, there are.
Don't try to understand my posts. Immanuel Kant never did, he thinks that only music and laughter do not have to mean anything.

[ulenrich]