|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
#include <assert.h> |
|
|
|
#include <arm_neon.h> |
|
|
|
#include <xnnpack/ibilinear.h> |
|
|
|
|
|
void xnn_f16_ibilinear_chw_ukernel__neonfp16arith_p4( |
|
size_t output_pixels, |
|
size_t channels, |
|
const void** restrict input, |
|
size_t input_offset, |
|
const void* restrict weights, |
|
void* restrict output, |
|
size_t input_increment) XNN_OOB_READS |
|
{ |
|
assert(output_pixels != 0); |
|
assert(channels != 0); |
|
assert(input_increment % sizeof(uint16_t) == 0); |
|
|
|
uint16_t* o = (uint16_t*) output; |
|
do { |
|
const uint16_t** i = (const uint16_t**)input; |
|
const uint16_t* w = weights; |
|
size_t p = output_pixels; |
|
|
|
for (; p >= 4; p -= 4) { |
|
const uint16_t* itl0 = (const uint16_t*) ((uintptr_t) i[0] + input_offset); |
|
const uint16_t* ibl0 = (const uint16_t*) ((uintptr_t) i[1] + input_offset); |
|
const uint16_t* itl1 = (const uint16_t*) ((uintptr_t) i[2] + input_offset); |
|
const uint16_t* ibl1 = (const uint16_t*) ((uintptr_t) i[3] + input_offset); |
|
const uint16_t* itl2 = (const uint16_t*) ((uintptr_t) i[4] + input_offset); |
|
const uint16_t* ibl2 = (const uint16_t*) ((uintptr_t) i[5] + input_offset); |
|
const uint16_t* itl3 = (const uint16_t*) ((uintptr_t) i[6] + input_offset); |
|
const uint16_t* ibl3 = (const uint16_t*) ((uintptr_t) i[7] + input_offset); |
|
i += 8; |
|
|
|
const uint16x4x2_t vw = vld2_u16(w); w += 8; |
|
|
|
float16x8_t vtltr = vreinterpretq_f16_u32(vld1q_dup_u32((const void*) itl0)); |
|
float16x8_t vblbr = vreinterpretq_f16_u32(vld1q_dup_u32((const void*) ibl0)); |
|
vtltr = vreinterpretq_f16_u32(vld1q_lane_u32((const void*) itl1, vreinterpretq_u32_f16(vtltr), 1)); |
|
vblbr = vreinterpretq_f16_u32(vld1q_lane_u32((const void*) ibl1, vreinterpretq_u32_f16(vblbr), 1)); |
|
vtltr = vreinterpretq_f16_u32(vld1q_lane_u32((const void*) itl2, vreinterpretq_u32_f16(vtltr), 2)); |
|
vblbr = vreinterpretq_f16_u32(vld1q_lane_u32((const void*) ibl2, vreinterpretq_u32_f16(vblbr), 2)); |
|
vtltr = vreinterpretq_f16_u32(vld1q_lane_u32((const void*) itl3, vreinterpretq_u32_f16(vtltr), 3)); |
|
vblbr = vreinterpretq_f16_u32(vld1q_lane_u32((const void*) ibl3, vreinterpretq_u32_f16(vblbr), 3)); |
|
|
|
const float16x4_t valphah = vreinterpret_f16_u16(vw.val[0]); |
|
const float16x4_t valphav = vreinterpret_f16_u16(vw.val[1]); |
|
|
|
const float16x8_t vldrd = vsubq_f16(vblbr, vtltr); |
|
|
|
const float16x4x2_t vld_t = vuzp_f16(vget_low_f16(vldrd), vget_high_f16(vldrd)); |
|
const float16x4_t vld = vld_t.val[0]; |
|
const float16x4_t vrd = vld_t.val[1]; |
|
|
|
const float16x4x2_t vtl_t = vuzp_f16(vget_low_f16(vtltr), vget_high_f16(vtltr)); |
|
const float16x4_t vtl = vtl_t.val[0]; |
|
const float16x4_t vtr = vtl_t.val[1]; |
|
|
|
const float16x4_t vl = vfma_f16(vtl, vld, valphav); |
|
const float16x4_t vr = vfma_f16(vtr, vrd, valphav); |
|
|
|
const float16x4_t vd = vsub_f16(vr, vl); |
|
const float16x4_t vo = vfma_f16(vl, vd, valphah); |
|
|
|
vst1_u16(o, vreinterpret_u16_f16(vo)); o += 4; |
|
} |
|
if XNN_UNLIKELY(p != 0) { |
|
if (p & 2) { |
|
const uint16_t* itl0 = (const uint16_t*) ((uintptr_t) i[0] + input_offset); |
|
const uint16_t* ibl0 = (const uint16_t*) ((uintptr_t) i[1] + input_offset); |
|
const uint16_t* itl1 = (const uint16_t*) ((uintptr_t) i[2] + input_offset); |
|
const uint16_t* ibl1 = (const uint16_t*) ((uintptr_t) i[3] + input_offset); |
|
i += 4; |
|
|
|
const float16x4_t vw = vreinterpret_f16_u16(vld1_u16(w)); w += 4; |
|
|
|
const float16x4x2_t vwhv = vuzp_f16(vw, vw); |
|
const float16x4_t valphah = vwhv.val[0]; |
|
const float16x4_t valphav = vwhv.val[1]; |
|
|
|
float16x4_t vtltr = vreinterpret_f16_u32(vld1_dup_u32((const void*) itl0)); |
|
float16x4_t vblbr = vreinterpret_f16_u32(vld1_dup_u32((const void*) ibl0)); |
|
|
|
vtltr = vreinterpret_f16_u32(vld1_lane_u32((const void*) itl1, vreinterpret_u32_f16(vtltr), 1)); |
|
vblbr = vreinterpret_f16_u32(vld1_lane_u32((const void*) ibl1, vreinterpret_u32_f16(vblbr), 1)); |
|
|
|
const float16x4_t vldrd = vsub_f16(vblbr, vtltr); |
|
|
|
const float16x4x2_t vld_t = vuzp_f16(vldrd, vldrd); |
|
const float16x4_t vld = vld_t.val[0]; |
|
const float16x4_t vrd = vld_t.val[1]; |
|
|
|
const float16x4x2_t vtl_t = vuzp_f16(vtltr, vtltr); |
|
const float16x4_t vtl = vtl_t.val[0]; |
|
const float16x4_t vtr = vtl_t.val[1]; |
|
|
|
const float16x4_t vl = vfma_f16(vtl, vld, valphav); |
|
const float16x4_t vr = vfma_f16(vtr, vrd, valphav); |
|
|
|
const float16x4_t vd = vsub_f16(vr, vl); |
|
const float16x4_t vo = vfma_f16(vl, vd, valphah); |
|
|
|
vst1_lane_u32((void*) o, vreinterpret_u32_f16(vo), 0); o += 2; |
|
} |
|
if (p & 1) { |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
const uint16_t* itl = (const uint16_t*) ((uintptr_t) i[0] + input_offset); |
|
const uint16_t* ibl = (const uint16_t*) ((uintptr_t) i[1] + input_offset); |
|
i += 2; |
|
|
|
const float16x4_t vw = vreinterpret_f16_u32(vld1_dup_u32((const void*) w)); w += 2; |
|
|
|
const float16x4x2_t vwhv = vuzp_f16(vw, vw); |
|
const float16x4_t valphah = vwhv.val[0]; |
|
const float16x4_t valphav = vwhv.val[1]; |
|
|
|
const float16x4_t vtltr = vreinterpret_f16_u32(vld1_dup_u32((const void*) itl)); |
|
const float16x4_t vblbr = vreinterpret_f16_u32(vld1_dup_u32((const void*) ibl)); |
|
|
|
const float16x4_t vldrd = vsub_f16(vblbr, vtltr); |
|
|
|
const float16x4x2_t vld_t = vuzp_f16(vldrd, vldrd); |
|
const float16x4_t vld = vld_t.val[0]; |
|
const float16x4_t vrd = vld_t.val[1]; |
|
|
|
const float16x4x2_t vtl_t = vuzp_f16(vtltr, vtltr); |
|
const float16x4_t vtl = vtl_t.val[0]; |
|
const float16x4_t vtr = vtl_t.val[1]; |
|
|
|
const float16x4_t vl = vfma_f16(vtl, vld, valphav); |
|
const float16x4_t vr = vfma_f16(vtr, vrd, valphav); |
|
|
|
const float16x4_t vd = vsub_f16(vr, vl); |
|
const float16x4_t vo = vfma_f16(vl, vd, valphah); |
|
|
|
vst1_lane_u16(o, vreinterpret_u16_f16(vo), 0); o += 1; |
|
} |
|
} |
|
|
|
input_offset += input_increment; |
|
} while (--channels != 0); |
|
} |
|
|