branch: master
Commits on master
- cededd8 minor multi cleanup (#5311) 1 year ago
- 8a99514 generalize the uops toposort spec to ptx (#5309) 1 year ago
- ca0ef17 use precise::sin in metal (#5307) 1 year ago
- 5c2ca7b remove UOps.SPECIAL rendering from llvm (#5306) 1 year ago
- 356e5d2 touchup multi dtype in elementwise (#5305) 1 year ago
- 7ddda9f hotfix: cache seen graphs in fusion (#5302) 1 year ago
- 11dfb19 track seen graphs in recursive group (#5301) 1 year ago
- d813617 prescheduling refactor (#5300) 1 year ago
- c1e166c fix dtype mismatch for bool ops in multi (#5299) 1 year ago
- fc03fc0 enable sin on METAL in test_dtype_alu (#5298) 1 year ago
- b369e75 refactor schedule creation (#5297) 1 year ago
- 5292d37 LoadOps.VIEW in the scheduler spec (#5296) 1 year ago
- 1ab7a4c Handling Multiple UnaryOps.BITCAST in Function for Proper Kernel Fusion [run_process_replay] (#5172) 1 year ago
- 43c3f73 handcode_bert_opt.py (#5295) 1 year ago
- d7835a7 hotfix: fix metal with vars (#5294) 1 year ago
- 8a548b0 metal support offset (#5293) 1 year ago
- 1cefbb3 uop graph tests + type_verify cleanup (#5292) 1 year ago
- 341c4a2 hotfix: use dtype.scalar() for rendering cast [run_process_replay] [no_assert] (#5290) 1 year ago
- 87d27c4 minor _broadcast cleanup (#5286) 1 year ago
- 8c03816 fix README example (#5284) 1 year ago
- 2778b60 new memory scheduler (#5278) 1 year ago
- 84b3e3b hcq exec no embedded signal (#5142) 1 year ago
- 0c3a35e Stable Diffusion v2 Inference (#5283) 1 year ago
- e5ba385 remove first contiguous in multi from_sharded (#5121) 1 year ago
- f1ff65e remove "no-nans-fp-math"="true" for LLVM (#5282) 1 year ago
- 3929a9d fix UOp.cmp_tuple for ALU (#5280) 1 year ago
- a9d6a6c verify_lazyop with multi reduce (#5276) 1 year ago
- 16e3b8b uops work from lowerer [run_process_replay] (#5279) 1 year ago
- 622b7bd simpler TinyJit inside TinyJit detection (#5219) 1 year ago
- 04ef0fd fix: message when applegpu tools missiong (#5236) 1 year ago