Home
last modified time | relevance | path

Searched refs:ld128 (Results 1 – 13 of 13) sorted by relevance

/external/XNNPACK/scripts/
Dgenerate-qs8-igemm.sh13 …mm/MRx4c8-wasmsimd.c.in -D MR=1 -D VARIANT=LD128 -o src/qs8-igemm/gen/1x4c8-minmax-wasmsimd-ld128.c
14 …mm/MRx4c8-wasmsimd.c.in -D MR=2 -D VARIANT=LD128 -o src/qs8-igemm/gen/2x4c8-minmax-wasmsimd-ld128.c
15 …mm/MRx4c8-wasmsimd.c.in -D MR=3 -D VARIANT=LD128 -o src/qs8-igemm/gen/3x4c8-minmax-wasmsimd-ld128.c
110 …s8-igemm/MRx4c2-sse.c.in -D MR=1 -D SSE=2 -D LD128=1 -o src/qs8-igemm/gen/1x4c2-minmax-sse2-ld128.c
111 …s8-igemm/MRx4c2-sse.c.in -D MR=4 -D SSE=2 -D LD128=1 -o src/qs8-igemm/gen/4x4c2-minmax-sse2-ld128.c
113 …8-igemm/MRx4c2-sse.c.in -D MR=1 -D SSE=3 -D LD128=1 -o src/qs8-igemm/gen/1x4c2-minmax-ssse3-ld128.c
114 …8-igemm/MRx4c2-sse.c.in -D MR=4 -D SSE=3 -D LD128=1 -o src/qs8-igemm/gen/4x4c2-minmax-ssse3-ld128.c
116 …8-igemm/MRx4c2-sse.c.in -D MR=1 -D SSE=4 -D LD128=1 -o src/qs8-igemm/gen/1x4c2-minmax-sse41-ld128.c
117 …8-igemm/MRx4c2-sse.c.in -D MR=4 -D SSE=4 -D LD128=1 -o src/qs8-igemm/gen/4x4c2-minmax-sse41-ld128.c
119 …qs8-igemm/MRx4c2-sse.c.in -D MR=1 -D SSE=5 -D LD128=1 -o src/qs8-igemm/gen/1x4c2-minmax-xop-ld128.c
[all …]
Dgenerate-qs8-gemm.sh13 …/MRx4c8-wasmsimd.c.in -D MR=1 -D VARIANT=LD128 -o src/qs8-gemm/gen/1x4c8-minmax-wasmsimd-ld128.c
14 …/MRx4c8-wasmsimd.c.in -D MR=2 -D VARIANT=LD128 -o src/qs8-gemm/gen/2x4c8-minmax-wasmsimd-ld128.c
15 …/MRx4c8-wasmsimd.c.in -D MR=3 -D VARIANT=LD128 -o src/qs8-gemm/gen/3x4c8-minmax-wasmsimd-ld128.c
114 …/MRx4c2-sse.c.in -D MR=1 -D SSE=2 -D VARIANT=LD128 -o src/qs8-gemm/gen/1x4c2-minmax-sse2-ld128.c
115 …/MRx4c2-sse.c.in -D MR=4 -D SSE=2 -D VARIANT=LD128 -o src/qs8-gemm/gen/4x4c2-minmax-sse2-ld128.c
117 …MRx4c2-sse.c.in -D MR=1 -D SSE=3 -D VARIANT=LD128 -o src/qs8-gemm/gen/1x4c2-minmax-ssse3-ld128.c
118 …MRx4c2-sse.c.in -D MR=4 -D SSE=3 -D VARIANT=LD128 -o src/qs8-gemm/gen/4x4c2-minmax-ssse3-ld128.c
120 …MRx4c2-sse.c.in -D MR=1 -D SSE=4 -D VARIANT=LD128 -o src/qs8-gemm/gen/1x4c2-minmax-sse41-ld128.c
121 …MRx4c2-sse.c.in -D MR=4 -D SSE=4 -D VARIANT=LD128 -o src/qs8-gemm/gen/4x4c2-minmax-sse41-ld128.c
123 …m/MRx4c2-sse.c.in -D MR=1 -D SSE=5 -D VARIANT=LD128 -o src/qs8-gemm/gen/1x4c2-minmax-xop-ld128.c
[all …]
Dgenerate-f32-gemm.sh76 …gen src/f32-gemm/4x8-aarch64-neonfma-ld128.S.in -D INC=0 -o src/f32-gemm/gen/4x8-minmax-aarc…
77 … src/f32-gemm/4x8-aarch64-neonfma-ld128.S.in -D INC=1 -o src/f32-gemm/gen-inc/4x8inc-minmax-…
109 …gen src/f32-gemm/6x8-aarch64-neonfma-ld128.S.in -D INC=0 -o src/f32-gemm/gen/6x8-minmax-aarc…
110 … src/f32-gemm/6x8-aarch64-neonfma-ld128.S.in -D INC=1 -o src/f32-gemm/gen-inc/6x8inc-minmax-…
137 …xngen src/f32-gemm/neon-ld128.c.in -D MR=4 -D NR=8 -D FMA=0 -D INC=0 -D DUP=0 -o src/f32-gemm…
138 …n src/f32-gemm/neon-ld128.c.in -D MR=4 -D NR=8 -D FMA=0 -D INC=1 -D DUP=0 -o src/f32-gemm/gen…
139 …gen src/f32-gemm/neon-ld128.c.in -D MR=4 -D NR=8 -D FMA=1 -D INC=0 -D DUP=0 -o src/f32-gemm/g…
140 … src/f32-gemm/neon-ld128.c.in -D MR=4 -D NR=8 -D FMA=1 -D INC=1 -D DUP=0 -o src/f32-gemm/gen-…
141 …xngen src/f32-gemm/neon-ld128.c.in -D MR=6 -D NR=8 -D FMA=0 -D INC=0 -D DUP=0 -o src/f32-gemm…
142 …n src/f32-gemm/neon-ld128.c.in -D MR=6 -D NR=8 -D FMA=0 -D INC=1 -D DUP=0 -o src/f32-gemm/gen…
[all …]
Dgenerate-f32-igemm.sh122 …ls/xngen src/f32-igemm/neon-ld128.c.in -D MR=4 -D NR=8 -D FMA=0 -D DUP=0 -o src/f32-igemm/gen/…
123 …s/xngen src/f32-igemm/neon-ld128.c.in -D MR=4 -D NR=8 -D FMA=1 -D DUP=0 -o src/f32-igemm/gen/4…
124 …ls/xngen src/f32-igemm/neon-ld128.c.in -D MR=6 -D NR=8 -D FMA=0 -D DUP=0 -o src/f32-igemm/gen/…
125 …s/xngen src/f32-igemm/neon-ld128.c.in -D MR=6 -D NR=8 -D FMA=1 -D DUP=0 -o src/f32-igemm/gen/6…
137 …ols/xngen src/f32-igemm/neon-ld128.c.in -D MR=4 -D NR=8 -D FMA=0 -D DUP=1 -o src/f32-igemm/gen…
138 …s/xngen src/f32-igemm/neon-ld128.c.in -D MR=4 -D NR=8 -D FMA=1 -D DUP=1 -o src/f32-igemm/gen/4…
139 …ols/xngen src/f32-igemm/neon-ld128.c.in -D MR=6 -D NR=8 -D FMA=0 -D DUP=1 -o src/f32-igemm/gen…
140 …s/xngen src/f32-igemm/neon-ld128.c.in -D MR=6 -D NR=8 -D FMA=1 -D DUP=1 -o src/f32-igemm/gen/6…
/external/XNNPACK/
DAndroid.bp601 "src/f32-gemm/gen-inc/4x8inc-minmax-neon-dup-ld128.c",
603 "src/f32-gemm/gen-inc/4x8inc-minmax-neon-lane-ld128.c",
607 "src/f32-gemm/gen-inc/6x8inc-minmax-neon-dup-ld128.c",
609 "src/f32-gemm/gen-inc/6x8inc-minmax-neon-lane-ld128.c",
617 "src/f32-gemm/gen/4x8-minmax-neon-dup-ld128.c",
619 "src/f32-gemm/gen/4x8-minmax-neon-lane-ld128.c",
623 "src/f32-gemm/gen/6x8-minmax-neon-dup-ld128.c",
625 "src/f32-gemm/gen/6x8-minmax-neon-lane-ld128.c",
641 "src/f32-igemm/gen/4x8-minmax-neon-dup-ld128.c",
643 "src/f32-igemm/gen/4x8-minmax-neon-lane-ld128.c",
[all …]
DCMakeLists.txt741 src/f32-gemm/gen-inc/4x8inc-minmax-neon-dup-ld128.c
743 src/f32-gemm/gen-inc/4x8inc-minmax-neon-lane-ld128.c
747 src/f32-gemm/gen-inc/6x8inc-minmax-neon-dup-ld128.c
749 src/f32-gemm/gen-inc/6x8inc-minmax-neon-lane-ld128.c
757 src/f32-gemm/gen/4x8-minmax-neon-dup-ld128.c
759 src/f32-gemm/gen/4x8-minmax-neon-lane-ld128.c
763 src/f32-gemm/gen/6x8-minmax-neon-dup-ld128.c
765 src/f32-gemm/gen/6x8-minmax-neon-lane-ld128.c
781 src/f32-igemm/gen/4x8-minmax-neon-dup-ld128.c
783 src/f32-igemm/gen/4x8-minmax-neon-lane-ld128.c
[all …]
DBUILD.bazel1378 "src/qs8-gemm/gen/1x4c8-minmax-wasmsimd-ld128.c",
1381 "src/qs8-gemm/gen/2x4c8-minmax-wasmsimd-ld128.c",
1384 "src/qs8-gemm/gen/3x4c8-minmax-wasmsimd-ld128.c",
1387 "src/qs8-igemm/gen/1x4c8-minmax-wasmsimd-ld128.c",
1389 "src/qs8-igemm/gen/2x4c8-minmax-wasmsimd-ld128.c",
1391 "src/qs8-igemm/gen/3x4c8-minmax-wasmsimd-ld128.c",
1492 "src/f32-gemm/gen-inc/4x8inc-minmax-neon-dup-ld128.c",
1494 "src/f32-gemm/gen-inc/4x8inc-minmax-neon-lane-ld128.c",
1498 "src/f32-gemm/gen-inc/6x8inc-minmax-neon-dup-ld128.c",
1500 "src/f32-gemm/gen-inc/6x8inc-minmax-neon-lane-ld128.c",
[all …]
/external/XNNPACK/src/qs8-igemm/
DMRx4c8-sse.c.in25 void xnn_qs8_igemm_minmax_ukernel_${MR}x4c8__${ISA}_${"ld128" if LD128 else "ld64"}(
DMRx4c2-sse.c.in25 void xnn_qs8_igemm_minmax_ukernel_${MR}x4c2__${ISA}_${"ld128" if LD128 else "ld64"}(
/external/XNNPACK/src/f32-gemm/gen/
D6x8-minmax-aarch64-neonfma-ld128.S128 # 48 FMA + 6 ld128 A + 4 LDP B
/external/XNNPACK/src/f32-gemm/gen-inc/
D6x8inc-minmax-aarch64-neonfma-ld128.S124 # 48 FMA + 6 ld128 A + 4 LDP B
/external/XNNPACK/src/f32-gemm/
D6x8-aarch64-neonfma-ld128.S.in151 # 48 FMA + 6 ld128 A + 4 LDP B
/external/neon_2_sse/
DNEON_2_SSE.h9848 __m128i ld128; variable
9849 ld128 = vld1q_u8(ptr); //merge two 64-bits in 128 bit
9850 ld128 = _mm_shuffle_epi8(ld128, *(__m128i*)mask8_16_even_odd);
9851 vst1q_u8((v.val), ld128); // v.val[1] = _mm_shuffle_epi32(v.val[0], _SWAP_HI_LOW32);
9859 __m128i ld128; variable
9860 ld128 = vld1q_u16(ptr); //merge two 64-bits in 128 bit
9861 ld128 = _mm_shuffle_epi8(ld128, *(__m128i*) mask8_32_even_odd);
9862 vst1q_u16((v.val), ld128);
9870 __m128i ld128; variable
9871 ld128 = vld1q_u32(ptr); //merge two 64-bits in 128 bit
[all …]