{"id":116280,"date":"2024-06-02T15:40:23","date_gmt":"2024-06-02T13:40:23","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/novaya-optimizacziya-funkczii-memset-v-glibc"},"modified":"2024-06-02T15:40:23","modified_gmt":"2024-06-02T13:40:23","slug":"novaya-optimizacziya-funkczii-memset-v-glibc","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/novosti-interneta\/novaya-optimizacziya-funkczii-memset-v-glibc","title":{"rendered":"Nj\u00eb optimizim i ri i funksionit memset() n\u00eb glibc","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"
Inxhinieri i Intel, Noah Goldstein, optimizoi funksionin memset() n\u00eb bibliotek\u00ebn glibc. Kjo optimizim ofron nj\u00eb rritje t\u00eb performanc\u00ebs prej rreth 7.5% n\u00eb versionet desktop t\u00eb procesor\u00ebve me arkitektur\u00eb Skylake-X dhe Ice Lake. N\u00eb versionet server, rritja e performanc\u00ebs \u00ebsht\u00eb pak m\u00eb e ul\u00ebt, kryesisht p\u00ebr shkak t\u00eb performanc\u00ebs m\u00eb t\u00eb ul\u00ebt t\u00eb b\u00ebrtham\u00ebs s\u00eb vetme.<\/p>\n
N\u00eb realizimin e m\u00ebparsh\u00ebm t\u00eb funksionit memset() p\u00ebrdorej instruksioni assembler rep stosb. Deri koh\u00ebt e fundit, ky instrukcion punonte mjaft shpejt, fal\u00eb optimizimit t\u00eb brendsh\u00ebm zero-over-zero writeback. Megjithat\u00eb, n\u00eb k\u00ebt\u00eb optimizim u zgjodh nj\u00eb vulnerabilitet potencial, i cili mund t\u00eb \u00e7oj\u00eb n\u00eb nj\u00eb sulm p\u00ebrmes kanaleve an\u00ebsore. Si rezultat, optimizimi zero-over-zero writeback u anulua, \u00e7ka \u00e7oi n\u00eb nj\u00eb r\u00ebnie t\u00eb performanc\u00ebs s\u00eb rep stosb. N\u00eb versionin e ri t\u00eb memset(), instruksioni rep stosb akoma p\u00ebrdoret, por n\u00ebn kushte m\u00eb strikte.<\/p>\n
\u00c7far\u00eb sakt\u00ebsisht ka ndryshuar mund t\u00eb kuptohet nga ndryshimi i komentit t\u00eb m\u00ebposht\u00ebm n\u00eb kod, i cili p\u00ebrshkruan detajet e zbatimit t\u00eb memset()<\/p>\n
Versioni i m\u00ebparsh\u00ebm i p\u00ebrshkrimit:<\/p>\n
\/* memset is implemented as: 1. Use overlapping store to avoid branch. 2. If size is less than VEC, use integer register stores. 3. If size is from VEC_SIZE to 2 * VEC_SIZE, use 2 VEC stores. 4. If size is from 2 * VEC_SIZE to 4 * VEC_SIZE, use 4 VEC stores. 5. On machines ERMS feature, if size is greater or equal than __x86_rep_stosb_threshold then REP STOSB will be used. 6. If size is more to 4 * VEC_SIZE, align to 4 * VEC_SIZE with 4 VEC stores and store 4 * VEC at a time until done. *\/ <\/p>\n
Versioni i ri i p\u00ebrshkrimit:<\/p>\n
\/* memset is implemented as: 1. Use overlapping store to avoid branch. 2. If size is less than VEC, use integer register stores. 3. If size is from VEC_SIZE to 2 * VEC_SIZE, use 2 VEC stores. 4. If size is from 2 * VEC_SIZE to 4 * VEC_SIZE, use 4 VEC stores. 5. If size is more to 4 * VEC_SIZE, align to 1 * VEC_SIZE with 4 VEC stores and store 4 * VEC at a time until done. 6. On machines ERMS feature, if size is range \t [__x86_rep_stosb_threshold, __x86_memset_non_temporal_threshold) \t then REP STOSB will be used. 7. If size >= __x86_memset_non_temporal_threshold, use a \t non-temporal stores. *\/<\/p>\n