Issues

• Get Segmentation fault when I compile the following FusionDefinition

  csarofeen/pytorch#2582 · Filter by author ftxj opened on Mar 21, 2023
  2

• detect parallel pattern error and provide a more helpful message

  csarofeen/pytorch#2566 · Filter by author liqiangxl opened on Mar 9, 2023
  1

• codegen error: RuntimeError: producer->getMemoryType() == MemoryType::Global || producer->getMemoryType() == MemoryType::Shared INTERNAL ASSERT FAILED

  csarofeen/pytorch#2559 · Filter by author jjsjann123 opened on Mar 8, 2023
  3

  

• Take internal buffer size into account to decide on indexing type size

  csarofeen/pytorch#2558 · Filter by author mmigdal-nv opened on Mar 8, 2023

• Example of Gather/Scatter from CrossEntropyLoss

  csarofeen/pytorch#2556 · Filter by author kevinstephano opened on Mar 8, 2023
  14

  

• ValReplacementMutator fails to mutate an expression even when its output is mutated

  bug
  csarofeen/pytorch#2554 · Filter by author naoyam opened on Mar 7, 2023
  1

• broadcast_in_dim: The size of contiguity must equal to the number of non-broadcasting IterDomains

  csarofeen/pytorch#2549 · Filter by author IvanYashchuk opened on Mar 7, 2023
  6

  

• Improve shared memory reuse

  csarofeen/pytorch#2529 · Filter by author naoyam opened on Feb 28, 2023

• cache var used by each iteration in grid persistent kernel, e.g. weight in layer norm backward

  csarofeen/pytorch#2525 · Filter by author liqiangxl opened on Feb 27, 2023

• nearbyint fails with NVRTC compile error for integer inputs

  csarofeen/pytorch#2524 · Filter by author IvanYashchuk opened on Feb 27, 2023

• Return a copy for integer data types with 'trunc' operation.

  mruberry
  csarofeen/pytorch#2499 · Filter by author rdspring1 opened on Feb 21, 2023

• Add complex specific support for sign operation

  mruberry
  csarofeen/pytorch#2492 · Filter by author kevinstephano opened on Feb 18, 2023

• Persistent buffer with L2 cache

  perf
  csarofeen/pytorch#2469 · Filter by author naoyam opened on Feb 15, 2023
  

• Persistent buffer with Local and Shared

  perf
  csarofeen/pytorch#2468 · Filter by author naoyam opened on Feb 15, 2023
  

• Heuristics tuning of outer grid persistence with iteration domain sizes that are not evenly distributed

  perf
  csarofeen/pytorch#2467 · Filter by author naoyam opened on Feb 15, 2023
  

• Limited vectorization with mixed input types

  perf
  csarofeen/pytorch#2466 · Filter by author naoyam opened on Feb 15, 2023
  

• Outer grid reduction kernels should be optimized as outer grid welford kernels are

  perf
  csarofeen/pytorch#2465 · Filter by author naoyam opened on Feb 15, 2023
  

• Move vectorized domains innermost always

  csarofeen/pytorch#2464 · Filter by author naoyam opened on Feb 15, 2023
  

• views in HF-Bart Self-Attention

  csarofeen/pytorch#2456 · Filter by author kevinstephano opened on Feb 14, 2023
  2

  

• Runtime error when the shape of index tensor is [1] in index_select

  csarofeen/pytorch#2442 · Filter by author ftxj opened on Feb 9, 2023

• Runtime error when using multiple fusion output of native_dropout op

  csarofeen/pytorch#2440 · Filter by author ftxj opened on Feb 9, 2023

• TypeError when defining complex constants

  csarofeen/pytorch#2425 · Filter by author jacobhinkle opened on Feb 7, 2023

• Extend Tensor Dimension support from 8 to 32

  mruberry
  csarofeen/pytorch#2424 · Filter by author kevinstephano opened on Feb 7, 2023
  1

• Compile error in where(x, a, b) with single precision a or b

  csarofeen/pytorch#2403 · Filter by author jacobhinkle opened on Feb 2, 2023
  1

• combined inner outer reduction used in layer norm backward

  csarofeen/pytorch#2399 · Filter by author liqiangxl opened on Feb 1, 2023