Messages in this thread | | | Subject | Re: [rfc] SLQB: YASA | From | Andi Kleen <> | Date | Thu, 03 Apr 2008 10:41:13 +0200 |
| |
Nick Piggin <npiggin@suse.de> writes: > > Nothing really interesting, unfortunately. I have run some tests on > various microbenchmarks like tbench and things like that. But I > don't have many good ideas for more meaningful tests where slab > allocation performance is critial. Any suggestions? :)
Some networking workloads hit slab pretty aggressive (two allocations per packet)
Just be careful with standard loopback, it has a contended lock elsewhere
-Andi
| |