/external/XNNPACK/src/x32-pad/ |
D | x2-neon.c | 50 vst1q_lane_u32(y0, vc, 0); y0 += 1; in xnn_x32_pad_x2__neon() 51 vst1q_lane_u32(y1, vc, 0); y1 += 1; in xnn_x32_pad_x2__neon() 88 vst1q_lane_u32(y0, vc, 0); in xnn_x32_pad_x2__neon() 89 vst1q_lane_u32(y1, vc, 0); in xnn_x32_pad_x2__neon()
|
/external/libvpx/libvpx/vpx_dsp/arm/ |
D | vpx_convolve8_neon.c | 390 vst1q_lane_u32((uint32_t *)(dst + 0 * dst_stride), d0123, 0); in vpx_convolve8_avg_horiz_neon() 391 vst1q_lane_u32((uint32_t *)(dst + 1 * dst_stride), d0123, 2); in vpx_convolve8_avg_horiz_neon() 392 vst1q_lane_u32((uint32_t *)(dst + 2 * dst_stride), d0123, 1); in vpx_convolve8_avg_horiz_neon() 393 vst1q_lane_u32((uint32_t *)(dst + 3 * dst_stride), d0123, 3); in vpx_convolve8_avg_horiz_neon() 477 vst1q_lane_u32((uint32_t *)dst, d0415, 0); in vpx_convolve8_avg_horiz_neon() 479 vst1q_lane_u32((uint32_t *)dst, d0415, 2); in vpx_convolve8_avg_horiz_neon() 481 vst1q_lane_u32((uint32_t *)dst, d2637, 0); in vpx_convolve8_avg_horiz_neon() 483 vst1q_lane_u32((uint32_t *)dst, d2637, 2); in vpx_convolve8_avg_horiz_neon() 485 vst1q_lane_u32((uint32_t *)dst, d0415, 1); in vpx_convolve8_avg_horiz_neon() 487 vst1q_lane_u32((uint32_t *)dst, d0415, 3); in vpx_convolve8_avg_horiz_neon() [all …]
|
/external/XNNPACK/src/q8-igemm/ |
D | 8x8-neon.c | 615 …vst1q_lane_u32(__builtin_assume_aligned(c7, 1), vreinterpretq_u32_u8(vout6x01234567_7x01234567), 2… in xnn_q8_igemm_ukernel_8x8__neon() 616 …vst1q_lane_u32(__builtin_assume_aligned(c6, 1), vreinterpretq_u32_u8(vout6x01234567_7x01234567), 0… in xnn_q8_igemm_ukernel_8x8__neon() 617 …vst1q_lane_u32(__builtin_assume_aligned(c5, 1), vreinterpretq_u32_u8(vout4x01234567_5x01234567), 2… in xnn_q8_igemm_ukernel_8x8__neon() 618 …vst1q_lane_u32(__builtin_assume_aligned(c4, 1), vreinterpretq_u32_u8(vout4x01234567_5x01234567), 0… in xnn_q8_igemm_ukernel_8x8__neon() 619 …vst1q_lane_u32(__builtin_assume_aligned(c3, 1), vreinterpretq_u32_u8(vout2x01234567_3x01234567), 2… in xnn_q8_igemm_ukernel_8x8__neon() 620 …vst1q_lane_u32(__builtin_assume_aligned(c2, 1), vreinterpretq_u32_u8(vout2x01234567_3x01234567), 0… in xnn_q8_igemm_ukernel_8x8__neon() 621 …vst1q_lane_u32(__builtin_assume_aligned(c1, 1), vreinterpretq_u32_u8(vout0x01234567_1x01234567), 2… in xnn_q8_igemm_ukernel_8x8__neon() 622 …vst1q_lane_u32(__builtin_assume_aligned(c0, 1), vreinterpretq_u32_u8(vout0x01234567_1x01234567), 0… in xnn_q8_igemm_ukernel_8x8__neon()
|
D | 4x8-neon.c | 386 …vst1q_lane_u32(__builtin_assume_aligned(c3, 1), vreinterpretq_u32_u8(vout2x01234567_3x01234567), 2… in xnn_q8_igemm_ukernel_4x8__neon() 387 …vst1q_lane_u32(__builtin_assume_aligned(c2, 1), vreinterpretq_u32_u8(vout2x01234567_3x01234567), 0… in xnn_q8_igemm_ukernel_4x8__neon() 388 …vst1q_lane_u32(__builtin_assume_aligned(c1, 1), vreinterpretq_u32_u8(vout0x01234567_1x01234567), 2… in xnn_q8_igemm_ukernel_4x8__neon() 389 …vst1q_lane_u32(__builtin_assume_aligned(c0, 1), vreinterpretq_u32_u8(vout0x01234567_1x01234567), 0… in xnn_q8_igemm_ukernel_4x8__neon()
|
/external/XNNPACK/src/q8-gemm/ |
D | 8x8-neon.c | 577 …vst1q_lane_u32(__builtin_assume_aligned(c0, 1), vreinterpretq_u32_u8(vout0x01234567_1x01234567), 0… in xnn_q8_gemm_ukernel_8x8__neon() 578 …vst1q_lane_u32(__builtin_assume_aligned(c1, 1), vreinterpretq_u32_u8(vout0x01234567_1x01234567), 2… in xnn_q8_gemm_ukernel_8x8__neon() 579 …vst1q_lane_u32(__builtin_assume_aligned(c2, 1), vreinterpretq_u32_u8(vout2x01234567_3x01234567), 0… in xnn_q8_gemm_ukernel_8x8__neon() 580 …vst1q_lane_u32(__builtin_assume_aligned(c3, 1), vreinterpretq_u32_u8(vout2x01234567_3x01234567), 2… in xnn_q8_gemm_ukernel_8x8__neon() 581 …vst1q_lane_u32(__builtin_assume_aligned(c4, 1), vreinterpretq_u32_u8(vout4x01234567_5x01234567), 0… in xnn_q8_gemm_ukernel_8x8__neon() 582 …vst1q_lane_u32(__builtin_assume_aligned(c5, 1), vreinterpretq_u32_u8(vout4x01234567_5x01234567), 2… in xnn_q8_gemm_ukernel_8x8__neon() 583 …vst1q_lane_u32(__builtin_assume_aligned(c6, 1), vreinterpretq_u32_u8(vout6x01234567_7x01234567), 0… in xnn_q8_gemm_ukernel_8x8__neon() 584 …vst1q_lane_u32(__builtin_assume_aligned(c7, 1), vreinterpretq_u32_u8(vout6x01234567_7x01234567), 2… in xnn_q8_gemm_ukernel_8x8__neon()
|
D | 4x8-neon.c | 353 …vst1q_lane_u32(__builtin_assume_aligned(c0, 1), vreinterpretq_u32_u8(vout0x01234567_1x01234567), 0… in xnn_q8_gemm_ukernel_4x8__neon() 354 …vst1q_lane_u32(__builtin_assume_aligned(c1, 1), vreinterpretq_u32_u8(vout0x01234567_1x01234567), 2… in xnn_q8_gemm_ukernel_4x8__neon() 355 …vst1q_lane_u32(__builtin_assume_aligned(c2, 1), vreinterpretq_u32_u8(vout2x01234567_3x01234567), 0… in xnn_q8_gemm_ukernel_4x8__neon() 356 …vst1q_lane_u32(__builtin_assume_aligned(c3, 1), vreinterpretq_u32_u8(vout2x01234567_3x01234567), 2… in xnn_q8_gemm_ukernel_4x8__neon()
|
/external/webp/src/dsp/ |
D | lossless_neon.c | 270 vst1q_lane_u32(&out[i + (LANE)], vreinterpretq_u32_u8(res), (LANE)); \ 294 vst1q_lane_u32(&out[i + (LANE)], vreinterpretq_u32_u8(res), (LANE)); \ 357 vst1q_lane_u32(&out[i + (LANE)], vreinterpretq_u32_u8(res), (LANE)); \ 388 vst1q_lane_u32(&out[i + (LANE)], vreinterpretq_u32_u8(res), (LANE)); \
|
/external/clang/test/CodeGen/ |
D | aarch64-neon-ldst-one.c | 5302 vst1q_lane_u32(a, b, 3); in test_vst1q_lane_u32()
|
D | arm_neon_intrinsics.c | 18201 vst1q_lane_u32(a, b, 3); in test_vst1q_lane_u32()
|
/external/tensorflow/tensorflow/lite/kernels/internal/optimized/ |
D | depthwiseconv_uint8_3x3_filter.h | 48 vst1q_lane_u32(reinterpret_cast<uint32_t*>(dst), reg, lane_num)
|
D | depthwiseconv_uint8_transitional.h | 68 vst1q_lane_u32(reinterpret_cast<uint32_t*>(dst), reg, lane_num)
|
/external/neon_2_sse/ |
D | NEON_2_SSE.h | 9691 _NEON2SSESTORAGE void vst1q_lane_u32(__transfersize(1) uint32_t * ptr, uint32x4_t val, __constrange… 9692 #define vst1q_lane_u32(ptr, val, lane) *(ptr) = (uint32_t) _MM_EXTRACT_EPI32 (val, lane) macro 11627 vst1q_lane_u32(ptr, val->val[0], lane); 11628 vst1q_lane_u32((ptr + 1), val->val[1], lane); 11716 vst1q_lane_u32((ptr + 2), val->val[2], lane);
|