Re: [PATCH v5 3/7] eal: add lcore variable performance test

2024-09-17 Thread Mattias Rönnblom
On 2024-09-17 17:40, Morten Brørup wrote: + start = rte_rdtsc(); + + for (i = 0; i < ITERATIONS; i++) + update_fun(mods[i & num_mods_mask]); This indexing adds more instructions to be executed than just the update function. The added overhead is the same for all teste

RE: [PATCH v5 3/7] eal: add lcore variable performance test

2024-09-17 Thread Morten Brørup
> + start = rte_rdtsc(); > + > + for (i = 0; i < ITERATIONS; i++) > + update_fun(mods[i & num_mods_mask]); This indexing adds more instructions to be executed than just the update function. The added overhead is the same for all tested access methods, so the absolute differen

[PATCH v5 3/7] eal: add lcore variable performance test

2024-09-17 Thread Mattias Rönnblom
Add basic micro benchmark for lcore variables, in an attempt to assure that the overhead isn't significantly greater than alternative approaches, in scenarios where the benefits aren't expected to show up (i.e., when plenty of cache is available compared to the working set size of the per-lcore dat