{"id":113552,"date":"2024-02-12T22:02:43","date_gmt":"2024-02-12T20:02:45","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/opublikovan-instrumentarij-zluda-pozvolyayushhij-zapuskat-cuda-prilozheniya-na-gpu-amd"},"modified":"2024-02-12T22:02:43","modified_gmt":"2024-02-12T20:02:45","slug":"opublikovan-instrumentarij-zluda-pozvolyayushhij-zapuskat-cuda-prilozheniya-na-gpu-amd","status":"publish","type":"post","link":"https:\/\/prohoster.info\/cs\/blog\/news\/opublikovan-instrumentarij-zluda-pozvolyayushhij-zapuskat-cuda-prilozheniya-na-gpu-amd","title":{"rendered":"Byl zve\u0159ejn\u011bn n\u00e1stroj ZLUDA, kter\u00fd umo\u017e\u0148uje spou\u0161t\u011bt CUDA aplikace na GPU AMD.","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

Projekt ZLUDA p\u0159ipravil otev\u0159enou implementaci technologie CUDA pro GPU AMD, kter\u00e1 umo\u017e\u0148uje spou\u0161t\u011bn\u00ed nemodifikovan\u00fdch CUDA aplikac\u00ed s v\u00fdkonem bl\u00edzk\u00fdm v\u00fdkonu aplikac\u00ed spou\u0161t\u011bn\u00fdch bez mezivrstvy. Publikovan\u00fd n\u00e1stroj zaji\u0161\u0165uje bin\u00e1rn\u00ed kompatibilitu se st\u00e1vaj\u00edc\u00edmi CUDA aplikacemi sestaven\u00fdmi pomoc\u00ed kompil\u00e1toru CUDA pro GPU NVIDIA. Implementace funguje na rozv\u00edjen\u00e9 stacku ROCm a runtime HIP (Heterogeneous-computing Interface for Portability), vyvinut\u00e9m spole\u010dnost\u00ed AMD. K\u00f3d projektu je naps\u00e1n v jazyce Rust a je \u0161\u00ed\u0159en na z\u00e1klad\u011b licenc\u00ed MIT a Apache 2.0. Podporov\u00e1na je pr\u00e1ce v Linuxu a Windows. <\/p>\n

Vrstva pro organizaci pr\u00e1ce CUDA na syst\u00e9mech s GPU AMD se vyv\u00edjela posledn\u00ed dva roky, ale projekt m\u00e1 del\u0161\u00ed historii a p\u016fvodn\u011b byl vyvinut pro zaji\u0161t\u011bn\u00ed fungov\u00e1n\u00ed CUDA na GPU Intel. Zm\u011bna v politice podpory GPU je vysv\u011btlena t\u00edm, \u017ee na za\u010d\u00e1tku byl v\u00fdvoj\u00e1\u0159 ZLUDA zam\u011bstnancem spole\u010dnosti Intel, av\u0161ak v roce 2021 tato firma pova\u017eovala mo\u017enost spu\u0161t\u011bn\u00ed CUDA aplikac\u00ed na GPU Intel za nezaj\u00edmavou pro byznys a necht\u011bla urychlovat rozvoj iniciativy. <\/p>\n

Na za\u010d\u00e1tku roku 2022 v\u00fdvoj\u00e1\u0159 opustil Intel a uzav\u0159el smlouvu se spole\u010dnost\u00ed AMD na v\u00fdvoj vrstvy pro kompatibilitu s CUDA. Spole\u010dnost AMD b\u011bhem v\u00fdvoje po\u017e\u00e1dala, aby nebyl ve\u0159ejn\u011b propagov\u00e1n z\u00e1jem AMD o projekt ZLUDA a aby nebyly prov\u00e1d\u011bny commity v ve\u0159ejn\u00e9m repozit\u00e1\u0159i ZLUDA. Po dvou letech se spole\u010dnost AMD rozhodla, \u017ee spu\u0161t\u011bn\u00ed CUDA aplikac\u00ed na GPU AMD nen\u00ed pro podnik\u00e1n\u00ed zaj\u00edmav\u00e9, co\u017e podle podm\u00ednek smlouvy umo\u017enilo v\u00fdvoj\u00e1\u0159i zve\u0159ejnit sv\u00e9 pr\u00e1ce. Vzhledem k tomu, \u017ee v\u00fdrobci GPU p\u0159estali projekt financovat, jeho budoucnost nyn\u00ed z\u00e1vis\u00ed na z\u00e1jmu komunity a p\u0159ijet\u00ed nab\u00eddek na spolupr\u00e1ci od dal\u0161\u00edch firem. Bez extern\u00ed podpory se projekt bude pravd\u011bpodobn\u011b vyv\u00edjet pouze v oblastech, kter\u00e9 jsou zaj\u00edmav\u00e9 osobn\u011b pro autora, jako je DLSS (Deep Learning Super Sampling). <\/p>\n

V sou\u010dasn\u00e9 dob\u011b je \u00farove\u0148 kvality implementace hodnocena jako alfa verze. Nicm\u00e9n\u011b, ZLUDA lze ji\u017e pou\u017e\u00edvat pro vykon\u00e1v\u00e1n\u00ed mnoha CUDA aplikac\u00ed, v\u010detn\u011b Geekbench, 3DF Zephyr, Blender, Reality Capture, LAMMPS, NAMD, waifu2x, OpenFOAM a Arnold. Zaji\u0161t\u011bna je minim\u00e1ln\u00ed podpora primitiv a knihoven cuDNN, cuBLAS, cuSPARSE, cuFFT, NCCL a NVML. <\/p>\n

Prvn\u00ed spu\u0161t\u011bn\u00ed aplikac\u00ed CUDA pod ZLUDA trp\u00ed znateln\u00fdmi zpo\u017ed\u011bn\u00edmi, proto\u017ee ZLUDA prov\u00e1d\u00ed kompilaci GPU k\u00f3du. P\u0159i dal\u0161\u00edch spu\u0161t\u011bn\u00edch takov\u00e9 zpo\u017ed\u011bn\u00ed nen\u00ed, proto\u017ee zkompilovan\u00fd k\u00f3d se ukl\u00e1d\u00e1 do cache. B\u011bhem prov\u00e1d\u011bn\u00ed zkompilovan\u00e9ho k\u00f3du je v\u00fdkon t\u00e9m\u011b\u0159 nativn\u00ed. P\u0159i spu\u0161t\u011bn\u00ed bal\u00ed\u010dku Geekbench na GPU AMD Radeon 6800 XT uk\u00e1zal testovac\u00ed set pro CUDA prov\u00e1d\u011bn\u00fd pomoc\u00ed ZLUDA v\u00fdrazn\u011b vy\u0161\u0161\u00ed v\u00fdkon ne\u017e varianta zalo\u017een\u00e1 na OpenCL.

\"Byl<\/center> <\/p>\n

Podpora ofici\u00e1ln\u00edho CUDA Driver API a \u010d\u00e1sti nedokumentovan\u00e9ho API CUDA, kter\u00e1 byla prozkoum\u00e1na pomoc\u00ed reverzn\u00edho in\u017een\u00fdrstv\u00ed, je v ZLUDA implementov\u00e1na prost\u0159ednictv\u00edm nahrazen\u00ed funkc\u00ed obdobn\u00fdmi funkcemi poskytovan\u00fdmi v HIP runtime, co\u017e je ve mnoha ohledech podobn\u00e9 CUDA. Nap\u0159\u00edklad funkce cuDeviceGetAttribute() je nahrazena funkc\u00ed hipDeviceGetAttribute(). Podobn\u00fdm zp\u016fsobem je zaji\u0161t\u011bna i kompatibilita s knihovnami NVIDIA, jako jsou NVML, cuBLAS a cuSPARSE \u2014 pro tyto knihovny ZLUDA poskytuje p\u0159ekladov\u00e9 knihovny se stejn\u00fdm n\u00e1zvem a sadou funkc\u00ed, postaven\u00e9 jako nadstavby nad podobn\u00fdmi knihovnami AMD. <\/p>\n

GPU k\u00f3d aplikac\u00ed, zkompilovan\u00fd do form\u00e1tu PTX (Parallel Thread Execution), je nejprve speci\u00e1ln\u00edm kompil\u00e1torem p\u0159eklad\u00e1n do mezilehl\u00e9ho form\u00e1tu LLVM IR, na jeho\u017e z\u00e1klad\u011b je generov\u00e1n bin\u00e1rn\u00ed k\u00f3d pro GPU AMD.

\"Byl<\/center>
\n
Zdroj: opennet.ru<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"

\u041f\u0440\u043e\u0435\u043a\u0442 ZLUDA \u043f\u043e\u0434\u0433\u043e\u0442\u043e\u0432\u0438\u043b \u043e\u0442\u043a\u0440\u044b\u0442\u0443\u044e \u0440\u0435\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u044e \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0438 CUDA \u0434\u043b\u044f GPU AMD, \u043f\u043e\u0437\u0432\u043e\u043b\u044f\u044e\u0449\u0443\u044e \u0432\u044b\u043f\u043e\u043b\u043d\u044f\u0442\u044c \u043d\u0435\u043c\u043e\u0434\u0438\u0444\u0438\u0446\u0438\u0440\u043e\u0432\u0430\u043d\u043d\u044b\u0435 CUDA-\u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u044f \u0441 \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u044c\u044e, \u0431\u043b\u0438\u0437\u043a\u043e\u0439 \u043a \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u0438 \u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u0439, \u0432\u044b\u043f\u043e\u043b\u043d\u044f\u0435\u043c\u044b\u0445 \u0431\u0435\u0437 \u043f\u0440\u043e\u0441\u043b\u043e\u0435\u043a. \u041e\u043f\u0443\u0431\u043b\u0438\u043a\u043e\u0432\u0430\u043d\u043d\u044b\u0439 \u0438\u043d\u0441\u0442\u0440\u0443\u043c\u0435\u043d\u0442\u0430\u0440\u0438\u0439 \u043e\u0431\u0435\u0441\u043f\u0435\u0447\u0438\u0432\u0430\u0435\u0442 \u0431\u0438\u043d\u0430\u0440\u043d\u0443\u044e \u0441\u043e\u0432\u043c\u0435\u0441\u0442\u0438\u043c\u043e\u0441\u0442\u044c \u0441 \u0441\u0443\u0449\u0435\u0441\u0442\u0432\u0443\u044e\u0449\u0438\u043c\u0438 CUDA-\u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u044f\u043c\u0438, \u0441\u043e\u0431\u0440\u0430\u043d\u043d\u044b\u043c\u0438 \u043f\u0440\u0438 \u043f\u043e\u043c\u043e\u0449\u0438 \u043a\u043e\u043c\u043f\u0438\u043b\u044f\u0442\u043e\u0440\u0430 CUDA \u0434\u043b\u044f GPU NVIDIA. \u0420\u0435\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u044f \u0440\u0430\u0431\u043e\u0442\u0430\u0435\u0442 \u043f\u043e\u0432\u0435\u0440\u0445 \u0440\u0430\u0437\u0432\u0438\u0432\u0430\u0435\u043c\u043e\u0433\u043e \u043a\u043e\u043c\u043f\u0430\u043d\u0438\u0435\u0439 AMD \u0441\u0442\u0435\u043a\u0430 ROCm \u0438 runtime HIP (Heterogeneous-computing Interface for Portability). \u041a\u043e\u0434 […]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":113553,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[702],"tags":[],"class_list":["post-113552","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\n","aioseo_head_json":{"title":"\ud83e\udd47Byl zve\u0159ejn\u011bn n\u00e1stroj ZLUDA, kter\u00fd umo\u017e\u0148uje spou\u0161t\u011bn\u00ed CUDA aplikac\u00ed na GPU AMD | ProHoster","description":"Projekt ZLUDA p\u0159ipravil otev\u0159enou implementaci technologie CUDA pro GPU AMD, kter\u00e1 umo\u017e\u0148uje spou\u0161t\u011bt nemodifikovan\u00e9 aplikace CUDA s v\u00fdkonem bl\u00edzk\u00fdm aplikac\u00edm, kter\u00e9 b\u011b\u017e\u00ed bez.","canonical_url":"https:\/\/prohoster.info\/cs\/blog\/news\/opublikovan-instrumentarij-zluda-pozvolyayushhij-zapuskat-cuda-prilozheniya-na-gpu-amd","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"cs_CZ","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u041e\u043f\u0443\u0431\u043b\u0438\u043a\u043e\u0432\u0430\u043d \u0438\u043d\u0441\u0442\u0440\u0443\u043c\u0435\u043d\u0442\u0430\u0440\u0438\u0439 ZLUDA, \u043f\u043e\u0437\u0432\u043e\u043b\u044f\u044e\u0449\u0438\u0439 \u0437\u0430\u043f\u0443\u0441\u043a\u0430\u0442\u044c CUDA-\u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u044f \u043d\u0430 GPU AMD | ProHoster","og:description":"\u041f\u0440\u043e\u0435\u043a\u0442 ZLUDA \u043f\u043e\u0434\u0433\u043e\u0442\u043e\u0432\u0438\u043b \u043e\u0442\u043a\u0440\u044b\u0442\u0443\u044e \u0440\u0435\u0430\u043b\u0438\u0437\u0430\u0446\u0438\u044e \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0438 CUDA \u0434\u043b\u044f GPU AMD, \u043f\u043e\u0437\u0432\u043e\u043b\u044f\u044e\u0449\u0443\u044e \u0432\u044b\u043f\u043e\u043b\u043d\u044f\u0442\u044c \u043d\u0435\u043c\u043e\u0434\u0438\u0444\u0438\u0446\u0438\u0440\u043e\u0432\u0430\u043d\u043d\u044b\u0435 CUDA-\u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u044f \u0441 \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u044c\u044e, \u0431\u043b\u0438\u0437\u043a\u043e\u0439 \u043a \u043f\u0440\u043e\u0438\u0437\u0432\u043e\u0434\u0438\u0442\u0435\u043b\u044c\u043d\u043e\u0441\u0442\u0438 \u043f\u0440\u0438\u043b\u043e\u0436\u0435\u043d\u0438\u0439, \u0432\u044b\u043f\u043e\u043b\u043d\u044f\u0435\u043c\u044b\u0445 \u0431\u0435\u0437.","og:url":"https:\/\/prohoster.info\/cs\/blog\/news\/opublikovan-instrumentarij-zluda-pozvolyayushhij-zapuskat-cuda-prilozheniya-na-gpu-amd","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2024-02-12T20:02:45+00:00","article:modified_time":"2024-02-12T20:02:45+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":[],"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/113552","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/comments?post=113552"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/posts\/113552\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media\/113553"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/media?parent=113552"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/categories?post=113552"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/cs\/wp-json\/wp\/v2\/tags?post=113552"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}