2021-03-23 17:10:07 -04:00
|
|
|
Search.setIndex({docnames:["getting-started/installation","getting-started/tutorials/01-vector-add","getting-started/tutorials/02-fused-softmax","getting-started/tutorials/03-matrix-multiplication","getting-started/tutorials/index","getting-started/tutorials/sg_execution_times","index","programming-guide/chapter-1/introduction","programming-guide/chapter-2/related-work","programming-guide/chapter-3/triton-c","programming-guide/chapter-4/triton-ir"],envversion:{"sphinx.domains.c":2,"sphinx.domains.changeset":1,"sphinx.domains.citation":1,"sphinx.domains.cpp":3,"sphinx.domains.index":1,"sphinx.domains.javascript":2,"sphinx.domains.math":2,"sphinx.domains.python":2,"sphinx.domains.rst":2,"sphinx.domains.std":2,sphinx:56},filenames:["getting-started/installation.rst","getting-started/tutorials/01-vector-add.rst","getting-started/tutorials/02-fused-softmax.rst","getting-started/tutorials/03-matrix-multiplication.rst","getting-started/tutorials/index.rst","getting-started/tutorials/sg_execution_times.rst","index.rst","programming-guide/chapter-1/introduction.rst","programming-guide/chapter-2/related-work.rst","programming-guide/chapter-3/triton-c.rst","programming-guide/chapter-4/triton-ir.rst"],objects:{},objnames:{},objtypes:{},terms:{"0000":3,"1024":1,"10mn":2,"1250":3,"128":3,"182":3,"1823":2,"184":3,"185":3,"186":3,"188":3,"190":[3,5],"191":3,"192":3,"193":3,"194":3,"195":3,"196":3,"198":3,"1983":7,"1984":8,"1989":8,"199":3,"1991":[7,8],"1999":[8,10],"200":3,"2001":10,"2004":8,"2006":8,"2012":8,"2013":[7,10],"2014":7,"2016":[7,8],"2017":7,"2018":[7,8],"2019":[1,8],"202":3,"2021":[7,8],"2048":2,"2141":1,"220":3,"245":3,"2500":3,"256":[2,3],"2mn":2,"3076":1,"3713":1,"3750":3,"3mn":2,"4096":2,"4940":1,"5000":3,"502":[3,5],"512":3,"6250":3,"6724":1,"7500":3,"756":[1,5],"768":3,"781":2,"7mn":2,"8750":3,"896":3,"933":[2,5],"9733":1,"98432":1,"abstract":[7,8,10],"break":8,"byte":2,"case":[2,3,7,8,9,10],"class":[1,2,3,7,8,9],"export":3,"final":10,"float":[1,2,3,7,8,9],"function":[1,2,8,9],"import":[1,2,3,7,8,9],"int":[1,2,3,7,8,9],"return":[1,2,3,10],"short":3,"static":[7,8,9,10],"super":3,"switch":3,"true":[1,2,3],"typeof":3,"var":8,"void":[1,2,3],"while":[0,7,9],For:[1,3,7,8,9,10],One:[3,9],SMs:8,T_S:8,The:[1,2,6,7,8],These:[3,8,9,10],Used:[1,2,3],__expf:2,__global__:[1,2,3,9],_add:1,_align8:3,_dot:3,_softmax:2,_src:[1,2],a100:[3,8],abl:[3,8],about:[1,2,3,6],abov:[1,2,3,8,9],abs:1,absolut:3,academ:7,acc:[3,7,8],acceler:7,access:[1,7,8,10],accomod:[3,9],accordingli:8,account:[3,8],accumul:8,accuraci:7,achiev:[1,7,8,9],across:[7,8],activ:3,actual:[3,7,8],add:[1,5,9,10],added:[9,10],addit:[2,3,4,5,7,9,10],addition:[8,9],address:[2,7],adequ:10,adopt:8,advanc:7,advantag:1,advoc:8,affect:3,affin:8,after:[3,10],again:9,against:[1,2,3,6,10],aggreg:10,aggress:[2,7,8,10],agnost:[7,8],ahead:8,aim:6,algebra:8,algorithm:[3,7,8,9,10],alia:8,alias:10,align8:3,align:10,all:[2,3,4,7,8,9,10],allclos:[2,3],allen1984:8,allen:8,alloc:[1,7],allow:[1,2,7,8,10],almost:2,along:[1,9,10],also:[1,2,3,7,8,9,10],alwai:8,amd:7,amen:8,amount:[2,7],ampl:8,analysi:[7,8,9],analyz:8,ancourt1991:8,ancourt:8,ani:[2,3,8,9],anoth:[2,8,10],ansi:9,anywai:10,apart:8,api:[0,1],appli:[1,2,3,7,8],applic:[8,10],approach:[7,8,9],appropri:9,approxim:2,architectur:[3,7],area:8,arg:[1,2,3],argument:[1,2,3,9,10],arithmet:10,around:2,arrai:[1,8,9,10],arrang:3,art:[7,8],arxiv:[7,8],ask:2,aspect:[8,9],asplo:7,assembl:3,assert:[1,2,3],assign:10,associ:9,assum:2,ast:10,asynchron:7,atol:3,atom:[9,10],atomic_add:9,attribut:[1,9,10],auguin1983:7,auguin:7,auto:[2,8,10],autograd:[1,2],autom:7,automat:[2,3,7,8,9],autotun:[3,8],autotune_config:3,autotune_kei:3,autotune_v:3,avail:[3,7,8],awar:[7,9],axi:[1,2,3,9],back:[1,2,3],backend:10,backward:1,baghdadi2021:[7,8],baghdadi:[7,8],balanc:8,bandwidth:2,base:[7,8,9,10],basic:[1,3,4,8,9],becaus:[2,9],becom:7,been:[7,8],befor:[3,10],begin:8,behavior:8,being:10,believ:8,below:[1,4,8,9,10],bench:0,benchmark:0,benefit:[2,7,8,9,10],best:[1,3,7],better:[3,10],between:[1,7],bin:3,binari:9,bit:[2,10],block:[1,2,3,7,8,9],block_siz:8,blockidx:1,bodi:10,b
|
