Skip to content
Branch: master
Find file History
Permalink
Type Name Latest commit message Commit time
..
Failed to load latest commit information.
gemm cpu: gemm: Improve IGEMM C-offset handling on AVX512 Apr 17, 2019
jit_utils src: cpu: remove weak symbols from jitprofiling Apr 16, 2019
rnn cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
xbyak cpu: xbyak: workaround smt identification Mar 1, 2019
CMakeLists.txt build: add nested cmake lists for cpu and common Mar 18, 2019
cpu_barrier.cpp all: updated copyright notice Feb 15, 2018
cpu_barrier.hpp all: updated copyright notice Feb 15, 2018
cpu_batch_normalization_pd.hpp
cpu_batch_normalization_utils.cpp api: move padding info to memory descriptor structure Mar 7, 2019
cpu_batch_normalization_utils.hpp api: make dimensions of type int64_t (mock-up) Mar 7, 2019
cpu_concat.cpp api: getting rid of memory_pd. src/cpu Mar 7, 2019
cpu_concat_pd.hpp api: getting rid of memory_pd. src/cpu Mar 7, 2019
cpu_convolution_pd.hpp
cpu_deconvolution_pd.hpp api: getting rid of memory_pd. src/cpu Mar 7, 2019
cpu_eltwise_pd.hpp
cpu_engine.cpp
cpu_engine.hpp
cpu_inner_product_pd.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
cpu_isa_traits.hpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
cpu_lrn_pd.hpp api: getting rid of memory_pd. src/cpu Mar 7, 2019
cpu_memory_storage.hpp src: cpu: fix critical kw issue Apr 19, 2019
cpu_pooling_pd.hpp
cpu_primitive.hpp ocl: add primitives Apr 18, 2019
cpu_reducer.cpp src: cpu: reducer: fix potental integer overflow Apr 16, 2019
cpu_reducer.hpp cpu: reducer: use memory tracking for scratchpad Dec 21, 2018
cpu_reorder.cpp memory: introduce memory storage Apr 11, 2019
cpu_reorder_pd.hpp common: introduce explicit scratchpad Mar 7, 2019
cpu_shuffle_pd.hpp
cpu_softmax_pd.hpp api: getting rid of memory_pd. src/cpu Mar 7, 2019
cpu_stream.hpp api: introduce stream wait and add cpu_stream_t Apr 12, 2019
cpu_sum.cpp api: getting rid of memory_pd. src/cpu Mar 7, 2019
cpu_sum_pd.hpp api: getting rid of memory_pd. src/cpu Mar 7, 2019
gemm_convolution.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
gemm_convolution.hpp api: make scratchpad provide by a user optional Mar 8, 2019
gemm_convolution_utils.cpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
gemm_convolution_utils.hpp cpu: conv: introduce blocking by spatial in forward int8 gemm convolu… Mar 6, 2019
gemm_inner_product.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
gemm_inner_product.hpp style: src: common: remove redundant this-> Mar 7, 2019
gemm_x8s8s32x_convolution.cpp src: cpu: conv: gemm: fix handling of scale adjustment Apr 16, 2019
gemm_x8s8s32x_convolution.hpp api: remove rounding mode Mar 8, 2019
gemm_x8s8s32x_inner_product.cpp cpu: ip: int8: add check if post op is necessary Mar 13, 2019
gemm_x8s8s32x_inner_product.hpp api: remove rounding mode Mar 8, 2019
jit_avx2_1x1_conv_kernel_f32.cpp memory: introduce memory storage Apr 11, 2019
jit_avx2_1x1_conv_kernel_f32.hpp memory: introduce memory storage Apr 11, 2019
jit_avx2_1x1_convolution.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
jit_avx2_1x1_convolution.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
jit_avx2_conv_kernel_f32.cpp memory: introduce memory storage Apr 11, 2019
jit_avx2_conv_kernel_f32.hpp
jit_avx2_convolution.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
jit_avx2_convolution.hpp
jit_avx512_common_1x1_conv_kernel.cpp memory: introduce memory storage Apr 11, 2019
jit_avx512_common_1x1_conv_kernel.hpp cpu: conv: 1x1: use memory tracking for scratchpad Dec 21, 2018
jit_avx512_common_1x1_convolution.cpp api: remove int16 support Mar 8, 2019
jit_avx512_common_1x1_convolution.hpp
jit_avx512_common_conv_kernel.cpp src: cpu: improved avx512 conv jit kernels input params processing Apr 18, 2019
jit_avx512_common_conv_kernel.hpp cpu: avx512: enable non-unit stride for 3d-convs weights update Mar 14, 2019
jit_avx512_common_conv_winograd_kernel_f32.cpp memory: introduce memory storage Apr 11, 2019
jit_avx512_common_conv_winograd_kernel_f32.hpp memory: introduce memory storage Apr 11, 2019
jit_avx512_common_convolution.cpp src: cpu: improved avx512 conv jit kernels input params processing Apr 18, 2019
jit_avx512_common_convolution.hpp api: remove int16 support Mar 8, 2019
jit_avx512_common_convolution_winograd.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
jit_avx512_common_convolution_winograd.hpp api: make scratchpad provide by a user optional Mar 8, 2019
jit_avx512_common_lrn.cpp
jit_avx512_common_lrn.hpp
jit_avx512_core_fp32_wino_conv_2x3.cpp api: remove int16 support Mar 8, 2019
jit_avx512_core_fp32_wino_conv_2x3.hpp
jit_avx512_core_fp32_wino_conv_4x3.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
jit_avx512_core_fp32_wino_conv_4x3.hpp
jit_avx512_core_fp32_wino_conv_4x3_kernel.cpp api: style: rename wino and rnn_packed related structures and enums Mar 7, 2019
jit_avx512_core_fp32_wino_conv_4x3_kernel.hpp
jit_avx512_core_u8s8s32x_wino_convolution.cpp api: remove rounding mode Mar 8, 2019
jit_avx512_core_u8s8s32x_wino_convolution.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
jit_avx512_core_x8s8s32x_1x1_conv_kernel.cpp src: cpu: conv: gemm: fix handling of scale adjustment Apr 16, 2019
jit_avx512_core_x8s8s32x_1x1_conv_kernel.hpp
jit_avx512_core_x8s8s32x_1x1_convolution.cpp
jit_avx512_core_x8s8s32x_1x1_convolution.hpp src: cpu: x8s8s32x 1x1: do not access bias memory if it is not defined Apr 16, 2019
jit_avx512_core_x8s8s32x_1x1_deconvolution.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
jit_avx512_core_x8s8s32x_conv_kernel.cpp
jit_avx512_core_x8s8s32x_conv_kernel.hpp cpu: conv: int8: depthwise: enable fewer loads in kernel Mar 12, 2019
jit_avx512_core_x8s8s32x_convolution.cpp cpu: conv: int8: depthwise: enable dedicated harness Mar 12, 2019
jit_avx512_core_x8s8s32x_convolution.hpp cpu: conv: int8: depthwise: enable dedicated harness Mar 12, 2019
jit_avx512_core_x8s8s32x_deconvolution.cpp src: cpu: conv: gemm: fix handling of scale adjustment Apr 16, 2019
jit_avx512_core_x8s8s32x_deconvolution.hpp memory: introduce memory storage Apr 11, 2019
jit_generator.hpp src: cpu: eltwise: fix rounding implementation for avx512 Apr 22, 2019
jit_primitive_conf.hpp cpu: conv: int8: depthwise: enable fewer loads in kernel Mar 12, 2019
jit_sse41_1x1_conv_kernel_f32.cpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_sse41_1x1_conv_kernel_f32.hpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_sse41_1x1_convolution.cpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_sse41_1x1_convolution.hpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_sse41_conv_kernel_f32.cpp
jit_sse41_conv_kernel_f32.hpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_sse41_convolution.cpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_sse41_convolution.hpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_transpose_src_utils.cpp api: remove int16 support Mar 8, 2019
jit_transpose_src_utils.hpp cpu: conv: 1x1: remove obsolete structure Jul 30, 2018
jit_uni_1x1_conv_utils.hpp api: remove int16 support Mar 8, 2019
jit_uni_batch_normalization.cpp
jit_uni_batch_normalization.hpp bnorm: fix incorrect usage of anonymous namespaces Apr 8, 2019
jit_uni_batch_normalization_s8.cpp cpu: jit_bn_s8: add avx2 support Apr 12, 2019
jit_uni_batch_normalization_s8.hpp bnorm: fix incorrect usage of anonymous namespaces Apr 8, 2019
jit_uni_dw_conv_kernel_f32.cpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_uni_dw_conv_kernel_f32.hpp
jit_uni_dw_convolution.cpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_uni_dw_convolution.hpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_uni_eltwise.cpp src: cpu: eltwise: fix rounding implementation for avx512 Apr 22, 2019
jit_uni_eltwise.hpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_uni_i8i8_pooling.cpp api: remove rounding mode Mar 8, 2019
jit_uni_i8i8_pooling.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
jit_uni_lrn.cpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_uni_lrn.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
jit_uni_lrn_kernel_f32.cpp
jit_uni_lrn_kernel_f32.hpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_uni_pool_kernel_f32.cpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_uni_pool_kernel_f32.hpp
jit_uni_pooling.cpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_uni_pooling.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
jit_uni_reorder.cpp cpu: second part of sse42 -> sse41: actually change the logic Apr 15, 2019
jit_uni_reorder.hpp api: style: rename TENSOR_MAX_DIMS to MKLDNN_MAX_NDIMS Mar 7, 2019
jit_uni_reorder_utils.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
nchw_pooling.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
nchw_pooling.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
ncsp_batch_normalization.cpp
ncsp_batch_normalization.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
nhwc_pooling.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
nhwc_pooling.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
nspc_batch_normalization.cpp src: cpu: fwd bnorm: use more accurate division w/ scaleshift Mar 15, 2019
nspc_batch_normalization.hpp
ref_batch_normalization.cpp src: cpu: fwd bnorm: use more accurate division w/ scaleshift Mar 15, 2019
ref_batch_normalization.hpp src: add s8 bnorm support for scaleshift Mar 11, 2019
ref_concat.hpp api: getting rid of memory_pd. src/cpu Mar 7, 2019
ref_convolution.cpp api: remove int16 support Mar 8, 2019
ref_convolution.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
ref_deconvolution.cpp
ref_deconvolution.hpp
ref_eltwise.cpp api: remove int16 support Mar 8, 2019
ref_eltwise.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
ref_inner_product.cpp api: remove int16 support Mar 8, 2019
ref_inner_product.hpp style: src: common: remove redundant this-> Mar 7, 2019
ref_lrn.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
ref_lrn.hpp
ref_pooling.cpp api: remove int16 support Mar 8, 2019
ref_pooling.hpp api: getting rid of memory_pd. src/cpu Mar 7, 2019
ref_shuffle.cpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
ref_shuffle.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
ref_softmax.cpp
ref_softmax.hpp cpu: softmax: use dense algorithm for blocked formats Apr 5, 2019
ref_sum.hpp api: getting rid of memory_pd. src/cpu Mar 7, 2019
simple_concat.cpp api: style: rename TENSOR_MAX_DIMS to MKLDNN_MAX_NDIMS Mar 7, 2019
simple_concat.hpp
simple_q10n.hpp api: remove rounding mode Mar 8, 2019
simple_reorder.hpp api: remove int16 support Mar 8, 2019
simple_sum.cpp
simple_sum.hpp api: rename md.layout_desc to md.format_desc (src) Mar 7, 2019
wino_reorder.hpp api: remove rounding mode Mar 8, 2019
You can’t perform that action at this time.