Google GitGoogle Git
Sign in
chromium/external/github.com/google/XNNPACK/HEAD
  1. d89c549 Merge pull request #11245 from prasanna8585:fix/convolution-nchw-sparse-packed-weights-overflow by XNNPACK Team · 8 hours ago upstream/master
  2. 358745f Merge pull request #11215 from rahulchaphalkar:f16-wasm-vadd by XNNPACK Team · 17 hours ago
  3. ced54bb Unroll x86 int8_int8_int32_k1 dot kernels by block_k=2 by Esun Kim · 25 hours ago
  4. c6b5253 Merge pull request #11222 from Colm-in-Arm:add_sme2p1_detection by XNNPACK Team · 3 days ago
  5. 068a3f2 fix: guard sparse packed_weights_size against overflow in convolution-nchw.c by prasanna8585 · 4 days ago
  6. 81a0804 Merge pull request #11154 from destro4evr-rgb:fix/litert-conv-pool-arithmetic-h-rank-checks by XNNPACK Team · 4 days ago
  7. 2106f03 Enable lowering 2 bit fully connected for XNNPACK. by Quentin Khan · 4 days ago
  8. 192402e Pin `ko` to full extent in `define_pack_b` schedule given_splits. by Marie White · 4 days ago
  9. f5c01a5 Fix Tensor API - XNNPACK type correspondance during lowering. by Quentin Khan · 4 days ago
  10. 434b6e2 Merge pull request #11240 from destro4evr-rgb:fix-packed-weights-overflow-dwconv-vmulcaddc by XNNPACK Team · 4 days ago
  11. 4010442 Fix packed weights size overflow in create_dwconv_path and create_vmulcaddc_path by destro4evr-rgb · 4 days ago
  12. 9d081ca F16 dwconv config use else after avx512fp16 so avx2 is not used when avx512fp16 is available by Frank Barchard · 4 days ago
  13. 32171f0 litert/tensor: guard AveragePool2D/MaxPool2D/Conv2D/DepthwiseConv2D against under-rank input in arithmetic.h by destro4evr-rgb · 5 days ago
  14. dcbbe8b Merge pull request #10913 from velonica0:fp16_build by XNNPACK Team · 5 days ago
  15. 63e6339 Added rewrite reduce(gather(W, i), R) -> gather(reduce(W, R), i) to ynnpack. by Misha Gutman · 5 days ago
  16. 394025d Fix reference traversal order in ArgmaxPoolingOperatorTester to match kernel tie-breaking. by Dillon Sharlet · 5 days ago
  17. 0c9f46a Support permuted broadcast dimensions in `move_broadcast_to_output`. by Dillon Sharlet · 6 days ago chromium/8065 chromium/8066 chromium/8067 chromium/8068 chromium/8069
  18. 4da00c1 Merge pull request #11209 from aizu-m:create-path-operator-leaks by XNNPACK Team · 6 days ago
  19. c057fa5 Set `is_set_up = true` in `slinky::pipeline::evaluate` by Dillon Sharlet · 7 days ago
  20. 43e5a26 Update XNNPACK dependencies by Dillon Sharlet · 7 days ago
  21. 08d7bd9 KFI-716 Add feature detection support for sme2p1 by Colm Donelan · 7 days ago
  22. 4d7c331 Remove unnecessary bug-prone wrapper by Dillon Sharlet · 7 days ago chromium/8062 chromium/8063 chromium/8064
  23. 94629a0 Remove the unnecessary `-march=...+sve+sve2` from the SME/SME2 microkernel groups. by XNNPACK Team · 8 days ago
  24. 8eb1f54 Merge pull request #11210 from destro4evr-rgb:fix/reduce-nd-workspace-size-overflow by XNNPACK Team · 8 days ago
  25. 416b702 Optimize `get_dot_kernel` more by Dillon Sharlet · 8 days ago
  26. 9375d4f Add scalar and WASM FP16 VADD and VADDC kernels by rahulchaphalkar · 8 days ago
  27. 355f86e Fix convergence check in FakeDynamicQuantize. by Dillon Sharlet · 8 days ago
  28. cc33fc4 [gn] Disable assembly microkernels on Windows and add xnnpack_enable_assembly arg. by Dillon Sharlet · 8 days ago
  29. 1fcdbfa Minor subgraph optimizations by Dillon Sharlet · 8 days ago
  30. dc40b59 Add RVV Zvfh microkernels for f16-qs8, f16-qu8, qs8-f16 vcvt by velonica0 · 9 days ago
  31. 287a812 src/operators/reduce-nd: guard workspace size against integer overflow by destro4evr-rgb · 10 days ago
  32. e9f3094 free partial operator on config and alloc failure in create paths by Aizal Khan · 11 days ago
  33. bf3ee43 Add a getter for a buffer byte size. by Quentin Khan · 11 days ago
  34. 09496c2 Merge pull request #11055 from destro4evr-rgb:fix/resize-nearest-neighbor-sigfpe by XNNPACK Team · 11 days ago
  35. 43c4b39 Refactor `backends/nnpack_common:runner`. by Quentin Khan · 11 days ago
  36. b4086e4 Add incremental update helpers to subgraph_analysis to maintain analysis during rewrites by Dillon Sharlet · 11 days ago
  37. 31435be Update XNNPACK dependencies by Dillon Sharlet · 11 days ago
  38. 9933c50 Remove the `ExternalBuffer` class and use `Buffer` instead. by Quentin Khan · 11 days ago
  39. a345743 litert/xnnpack: guard against zero input dimension in ResizeNearestNeighbor by destro4evr-rgb · 11 days ago
  40. e247b29 Add initialization states to kernels. This is beneficial for kernels with initialization procedures that should execute outside the scope of the kernels. by Marie White · 11 days ago
  41. 5637b04 Add missing invalidations of subgraph analysis by Dillon Sharlet · 12 days ago
  42. e267d43 Fix short-circuit in YNNPack fusion loop to rewrite multiple nodes per pass by Dillon Sharlet · 12 days ago
  43. 60bd2eb Split code that can be shared between XNNPACK and a future YNNPACK runner. by Quentin Khan · 12 days ago
  44. c394a5f Merge pull request #10910 from aizu-m:zero-buffers-valid-batch-size by XNNPACK Team · 12 days ago
  45. 5e28723 Remove redundant AMX tile config loads by keeping track of the current tile config. by Marie White · 12 days ago
  46. 64759b6 Restore QEMU cross-compilation emulator for armhf CMake tests. by Frank Barchard · 12 days ago
  47. d89ef66 Optimize SME int4 and int2 kernels on Apple M4 by Dillon Sharlet · 13 days ago
  48. 65cc459 Change SME int4 and int2 dot kernels to be k=8 and k=16 correspondingly. by Volodymyr Kysenko · 13 days ago
  49. b39959b Align the `m` dimension of `pack_a` outputs by Dillon Sharlet · 13 days ago
  50. 8fc4486 Install qemu-user in standard Docker image for armhf emulation. by Frank Barchard · 13 days ago
  51. 74732d1 Update XNNPACk dependencies by Dillon Sharlet · 13 days ago
  52. ed430be Update help text for --type option in dot_bench.cc by Volodymyr Kysenko · 13 days ago
  53. dd66955 Merge pull request #11177 from destro4evr-rgb:fix/litert-topk-rank0-oob-write by XNNPACK Team · 13 days ago
  54. bc12c14 litert/tensor: guard TopK() against rank-0 input to prevent heap OOB write by destro4evr-rgb · 14 days ago
  55. 5b8b210 pqs8_qc4w_gemm_minmax_test for Kleidi detect bug in x8_packlh_neonsme/neonsme2. by Frank Barchard · 14 days ago
  56. 3c4596a Workaround for KleidiAI SME2 RHS packing on Apple Silicon. by Frank Barchard · 14 days ago chromium/8053 chromium/8054 chromium/8055 chromium/8056 chromium/8057 chromium/8058 chromium/8059 chromium/8060 chromium/8061
  57. eaab2c6 Merge pull request #11123 from rahulchaphalkar:f16-rdmax by XNNPACK Team · 14 days ago
  58. 61050fc Add a getter to retrieve the underlying shared pointer to a tensor buffer. by Quentin Khan · 2 weeks ago
  59. 53aa497 Add more documentation by Dillon Sharlet · 2 weeks ago
  60. 2422532 hold the allocation count in the allocator context by Aizal Khan · 2 weeks ago
  61. b494cf2 restore test allocator with an RAII guard by Aizal Khan · 2 weeks ago
  62. 240f90c use valid_batch_size to free zero buffers in xnn_destroy_operator by Aizal Khan · 2 weeks ago
  63. b3bcd40 Merge pull request #11158 from kejcao:fix/deconvolution-zero-output-reshape by XNNPACK Team · 2 weeks ago
  64. 16b05de Optimize `get_dot_kernel` by lifting some flag logic out of the per-kernel check by Dillon Sharlet · 2 weeks ago
  65. 852a300 Fix `static_size_of_value` for implicit broadcast dimensions by Dillon Sharlet · 2 weeks ago
  66. 6daa0c0 Reassociate pack_b and gather in subgraph fusion. by Dillon Sharlet · 2 weeks ago
  67. fdf2b7f Add a static getter to retrieve buffer implementations' type identifier. by Quentin Khan · 2 weeks ago
  68. b80b85d Reject deconvolution reshape with zero output dimensions by kejcao · 2 weeks ago
  69. 6c19e6b Merge pull request #11141 from digantdesai:hexagon-extra-bytes-fix by XNNPACK Team · 2 weeks ago
  70. d50a82e xnnpack_configurable_defines match xnnpack_kleidiai_defines by Frank Barchard · 3 weeks ago
  71. e93f337 Merge pull request #11142 from destro4evr-rgb:fix/litert-rank0-crash-fc-embedlookup-gathernd by XNNPACK Team · 3 weeks ago
  72. 303dc14 Add ARM SME dot product kernels for int8 activations with int4 and int2 weights. by Volodymyr Kysenko · 3 weeks ago
  73. 47fef97 Simplify away provably true checks by Dillon Sharlet · 3 weeks ago
  74. 426b707 Merge pull request #10791 from aizu-m:deconv-igemm-indirection-output-size by XNNPACK Team · 3 weeks ago
  75. 373f03c Remove internal implementation details from fully-connected subgraph test by Dillon Sharlet · 3 weeks ago
  76. 3a32935 litert/tensor: guard FullyConnected, EmbeddingLookup, GatherNd against rank-0 input tensors by destro4evr-rgb · 3 weeks ago
  77. 54ea1b2 Merge pull request #11131 from destro4evr-rgb:fix/litert-select-rank0-null-deref by XNNPACK Team · 3 weeks ago
  78. bc8d01b Fix -Wundef error and dead Hexagon guard in xnnpack.h by Digant Desai · 3 weeks ago
  79. 735a993 Merge pull request #10986 from damdoo001-arm:damdoo01/int4_qc4w by XNNPACK Team · 3 weeks ago
  80. 175d008 Clean up rebased INT4 integration by Damien Dooley · 3 weeks ago
  81. f742e00 Added kernels to build.gn by Damien Dooley · 3 weeks ago
  82. cd321f8 Guard SME2 QC4W MR before configuration by Damien Dooley · 3 weeks ago
  83. 48e5f8a Addressed feedback re:assert, licence header and yoda condition by Damien Dooley · 3 weeks ago
  84. 8a3f46e Keep generated copyright headers generator-owned by Damien Dooley · 3 weeks ago
  85. 45eeefd Use Arm copyright notice format 2 by Damien Dooley · 3 weeks ago
  86. 510c0d8 Use XNNPACK license for added files by Damien Dooley · 3 weeks ago
  87. bd10b21 Add Arm copyright notices by Damien Dooley · 3 weeks ago
  88. 168e8c9 Gate packed QC4W rewrite on zero point by Damien Dooley · 3 weeks ago
  89. 2b1a0e9 Validate SME2 QC4W packing parameters by Damien Dooley · 3 weeks ago
  90. 76fa737 Integrate SME2 INT4 fully connected kernels by Damien Dooley · 3 weeks ago
  91. d36e64d Remove DotSchedulingTest.OpaqueSplitFactorBoundsAreProvable by Dillon Sharlet · 3 weeks ago
  92. f10c3e9 Align the `m` dimension of `pack_a` outputs by XNNPACK Team · 3 weeks ago
  93. 2919616 Merge pull request #11105 from destro4evr-rgb:fix/litert-xnnpack-scale-array-oob-read by XNNPACK Team · 3 weeks ago
  94. d3bfd20 Merge pull request #10877 from damdoo001-arm:damdoo01/int2_qc2w by XNNPACK Team · 3 weeks ago
  95. 1640a03 Align the `m` dimension of `pack_a` outputs by Dillon Sharlet · 3 weeks ago
  96. 42992df litert/tensor/arithmetic.h: guard Select() against rank-0 a/b with 1-D condition by destro4evr-rgb · 3 weeks ago
  97. def8e5b Included bazel dependency by Damien Dooley · 3 weeks ago
  98. 2f3bec1 Merge pull request #11125 from destro4evr-rgb:fix/litert-split-axis-oob-check by XNNPACK Team · 3 weeks ago
  99. c0b771a Merge pull request #11056 from destro4evr-rgb:fix/litert-mean-slice-shape-empty-null-deref by XNNPACK Team · 3 weeks ago
  100. f356f73 Merge pull request #11114 from kgotlinux:freebsd-ppc-fix by XNNPACK Team · 3 weeks ago