I've added the murmur hash to eina (only the one producing a 32bit hash) in [¹]. I wanted to have a better bench so 1st I needed a better dictionary than just hashing integers. I've wrote a little script [²] launched on the repo that takes edc files and extracts strings and generates a C header file containing an array of strings [³]. It currently uses git ls-files but can be changed to use find or … I've changed eina_bench_hash.c to use that dictionary. It also use the same random seed to run each bench. Full patch is [⁴]. On [⁵], Murmur seems a bit better. On [⁶], eina_hash could be improved.
If you want those commits in svn, please tell me. [¹] http://www.billiob.net/eina_bench/0001-eina-add-murmur3-hash.patch [²] http://www.billiob.net/eina_bench/make_dict.py [³] http://www.billiob.net/eina_bench/dict.h [⁴] http://www.billiob.net/eina_bench/0002-eina-improve-eina_bench_hash-to-test-with-a-dictiona.patch [⁵] http://www.billiob.net/eina_bench/output_Hash_b.png [⁶] http://www.billiob.net/eina_bench/output_Hash_full.png -- Boris Faure ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity and more. Splunk takes this data and makes sense of it. Business sense. IT sense. Common sense. http://p.sf.net/sfu/splunk-d2dcopy1 _______________________________________________ enlightenment-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
