/dports/math/openblas/OpenBLAS-0.3.18/kernel/mips/ |
H A D | strsm_kernel_RN_8x8_msa.c | 167 src_c1 *= src_b0; in ssolve_8x8_rn_msa() 169 src_c3 -= src_c1 * src_b1; in ssolve_8x8_rn_msa() 171 src_c5 -= src_c1 * src_b2; in ssolve_8x8_rn_msa() 173 src_c7 -= src_c1 * src_b3; in ssolve_8x8_rn_msa() 175 src_c9 -= src_c1 * src_b4; in ssolve_8x8_rn_msa() 373 src_c1 *= src_b0; in ssolve_8x4_rn_msa() 465 src_c1 *= src_b0; in ssolve_8x2_rn_msa() 568 src_c1 *= src_b0; in ssolve_8x1_rn_msa() 665 src_c1 *= src_b9; in ssolve_4x8_rn_msa() 784 src_c1 *= src_b5; in ssolve_4x4_rn_msa() [all …]
|
H A D | strsm_kernel_RT_8x8_msa.c | 48 LD_SP2(c, 4, src_c0, src_c1); in ssolve_8x8_rt_msa() 277 src_c1 -= src_c3 * src_b8; in ssolve_8x8_rt_msa() 280 src_c1 *= src_b0; in ssolve_8x8_rt_msa() 382 src_c1 -= src_c5 * src_b8; in ssolve_8x4_rt_msa() 387 src_c1 -= src_c3 * src_b4; in ssolve_8x4_rt_msa() 390 src_c1 *= src_b0; in ssolve_8x4_rt_msa() 467 src_c1 *= src_b0; in ssolve_8x2_rt_msa() 567 src_c1 *= src_b0; in ssolve_8x1_rt_msa() 699 src_c1 *= src_b9; in ssolve_4x8_rt_msa() 797 src_c1 *= src_b5; in ssolve_4x4_rt_msa() [all …]
|
H A D | dtrsm_kernel_RT_8x4_msa.c | 214 src_c1 -= src_c9 * src_b8; in dsolve_8x4_rt_msa() 219 src_c1 -= src_c5 * src_b4; in dsolve_8x4_rt_msa() 224 src_c1 *= src_b0; in dsolve_8x4_rt_msa() 310 src_c1 -= src_c5 * src_b2; in dsolve_8x2_rt_msa() 315 src_c1 *= src_b0; in dsolve_8x2_rt_msa() 380 src_c1 *= src_b0; in dsolve_8x1_rt_msa() 473 src_c1 -= src_c5 * src_b8; in dsolve_4x4_rt_msa() 476 src_c1 -= src_c3 * src_b4; in dsolve_4x4_rt_msa() 479 src_c1 *= src_b0; in dsolve_4x4_rt_msa() 533 src_c1 -= src_c3 * src_b2; in dsolve_4x2_rt_msa() [all …]
|
H A D | dtrsm_kernel_RN_8x4_msa.c | 174 src_c1 *= src_b0; in dsolve_8x4_rn_msa() 179 src_c5 -= src_c1 * src_b1; in dsolve_8x4_rn_msa() 189 src_c9 -= src_c1 * src_b2; in dsolve_8x4_rn_msa() 204 src_c13 -= src_c1 * src_b3; in dsolve_8x4_rn_msa() 298 src_c1 *= src_b0; in dsolve_8x2_rn_msa() 303 src_c5 -= src_c1 * src_b1; in dsolve_8x2_rn_msa() 350 src_c1 *= src_b0; in dsolve_8x1_rn_msa() 416 src_c1 *= src_b0; in dsolve_4x4_rn_msa() 419 src_c3 -= src_c1 * src_b1; in dsolve_4x4_rn_msa() 425 src_c5 -= src_c1 * src_b2; in dsolve_4x4_rn_msa() [all …]
|
H A D | dtrsm_kernel_LT_8x4_msa.c | 95 src_c1 -= src_a1 * src_b; in dsolve_8x4_lt_msa() 122 src_c1 -= src_a1 * src_b; in dsolve_8x4_lt_msa() 154 src_c1 -= src_a1 * src_b; in dsolve_8x4_lt_msa() 340 ST_DP(src_c1, c + 2); in dsolve_8x4_lt_msa() 417 src_c1 -= src_a1 * src_b; in dsolve_8x2_lt_msa() 439 src_c1 -= src_a1 * src_b; in dsolve_8x2_lt_msa() 535 ST_DP2(src_c0, src_c1, c, 2); in dsolve_8x2_lt_msa() 723 LD_DP2(c, 2, src_c0, src_c1); in dsolve_4x4_lt_msa() 821 ST_DP2(src_c0, src_c1, c, 2); in dsolve_4x4_lt_msa() 833 LD_DP2(c, 2, src_c0, src_c1); in dsolve_4x2_lt_msa() [all …]
|
H A D | dtrsm_kernel_LN_8x4_msa.c | 94 src_c1 -= src_a1 * src_b; in dsolve_8x4_ln_msa() 121 src_c1 -= src_a1 * src_b; in dsolve_8x4_ln_msa() 153 src_c1 -= src_a1 * src_b; in dsolve_8x4_ln_msa() 327 ST_DP(src_c1, c + 2); in dsolve_8x4_ln_msa() 421 src_c1 -= src_a1 * src_b; in dsolve_8x2_ln_msa() 440 src_c1 -= src_a1 * src_b; in dsolve_8x2_ln_msa() 723 LD_DP2(c, 2, src_c0, src_c1); in dsolve_4x4_ln_msa() 741 src_c1 -= src_a1 * src_b; in dsolve_4x4_ln_msa() 825 ST_DP2(src_c0, src_c1, c, 2); in dsolve_4x4_ln_msa() 837 LD_DP2(c, 2, src_c0, src_c1); in dsolve_4x2_ln_msa() [all …]
|
H A D | strsm_kernel_LN_8x8_msa.c | 50 LD_SP2(c, 4, src_c0, src_c1); in ssolve_8x8_ln_msa() 254 ST_SP(src_c1, c + 4); in ssolve_8x8_ln_msa() 333 LD_SP2(c, 4, src_c0, src_c1); in ssolve_8x4_ln_msa() 345 src_c1 -= src_a1 * src_b0; in ssolve_8x4_ln_msa() 361 src_c1 -= src_a1 * src_b0; in ssolve_8x4_ln_msa() 380 src_c1 -= src_a1 * src_b0; in ssolve_8x4_ln_msa() 483 ST_SP2(src_c0, src_c1, c, 4); in ssolve_8x4_ln_msa() 848 src_c1 = LD_SP(c_nxt1line); in ssolve_4x8_ln_msa() 930 ST_SP(src_c1, c_nxt1line); in ssolve_4x8_ln_msa() 952 src_c1 = LD_SP(c_nxt1line); in ssolve_4x4_ln_msa() [all …]
|
H A D | strsm_kernel_LT_8x8_msa.c | 50 LD_SP2(c, 4, src_c0, src_c1); in ssolve_8x8_lt_msa() 304 ST_SP(src_c1, c + 4); in ssolve_8x8_lt_msa() 329 LD_SP2(c, 4, src_c0, src_c1); in ssolve_8x4_lt_msa() 341 src_c1 -= src_a1 * src_b0; in ssolve_8x4_lt_msa() 444 ST_SP2(src_c0, src_c1, c, 4); in ssolve_8x4_lt_msa() 799 src_c1 = LD_SP(c_nxt1line); in ssolve_4x8_lt_msa() 814 src_c1 -= src_a0 * src_b1; in ssolve_4x8_lt_msa() 833 src_c1 -= src_a0 * src_b1; in ssolve_4x8_lt_msa() 918 ST_SP(src_c1, c_nxt1line); in ssolve_4x8_lt_msa() 939 src_c1 = LD_SP(c_nxt1line); in ssolve_4x4_lt_msa() [all …]
|
/dports/editors/calligra/calligra-3.2.1/libs/pigment/compositeops/ |
H A D | KoOptimizedCompositeOpAlphaDarken32.h | 88 Vc::float_v src_c1; in compositeVector() local 96 KoStreamedMath<_impl>::template fetch_colors_32<src_aligned>(src, src_c1, src_c2, src_c3); in compositeVector() 108 dst_c1 = src_c1; in compositeVector() 117 dst_c1 = src_c1; in compositeVector() 123 dst_c1 = dst_blend * (src_c1 - dst_c1) + dst_c1; in compositeVector() 128 dst_c1(empty_dst_pixels_mask) = src_c1; in compositeVector() 134 dst_c1(not_empty_dst_pixels_mask) = dst_blend * (src_c1 - dst_c1) + dst_c1; in compositeVector()
|
H A D | KoOptimizedCompositeOpAlphaDarken128.h | 67 Vc::float_v src_c1; in compositeVector() local 74 (src_c1, src_c2, src_c3, src_alpha) = data[indexes]; in compositeVector() 103 dst_c1 = (src_c1 - dst_c1) * src_alpha + dst_c1; in compositeVector() 108 dst_c1(empty_dst_pixels_mask) = src_c1; in compositeVector() 112 dst_c1(not_empty_dst_pixels_mask) = (src_c1 - dst_c1) * src_alpha + dst_c1; in compositeVector() 118 dst_c1 = src_c1; in compositeVector()
|
H A D | KoOptimizedCompositeOpOver32.h | 73 Vc::float_v src_c1; in compositeVector() local 82 KoStreamedMath<_impl>::template fetch_colors_32<src_aligned>(src, src_c1, src_c2, src_c3); in compositeVector() 106 dst_c1 = src_blend * (src_c1 - dst_c1) + dst_c1; in compositeVector() 117 dst_c1 = src_c1; in compositeVector()
|
H A D | KoOptimizedCompositeOpOver128.h | 72 Vc::float_v src_c1; in compositeVector() local 78 (src_c1, src_c2, src_c3, src_alpha) = data[indexes]; in compositeVector() 133 dst_c1 = src_blend * (src_c1 - dst_c1) + dst_c1; in compositeVector() 142 dataDest[indexes] = (src_c1, src_c2, src_c3, new_alpha); in compositeVector()
|
/dports/graphics/krita/krita-4.4.8/libs/pigment/compositeops/ |
H A D | KoOptimizedCompositeOpAlphaDarken32.h | 85 Vc::float_v src_c1; in compositeVector() local 93 KoStreamedMath<_impl>::template fetch_colors_32<src_aligned>(src, src_c1, src_c2, src_c3); in compositeVector() 105 dst_c1 = src_c1; in compositeVector() 114 dst_c1 = src_c1; in compositeVector() 120 dst_c1 = dst_blend * (src_c1 - dst_c1) + dst_c1; in compositeVector() 125 dst_c1(empty_dst_pixels_mask) = src_c1; in compositeVector() 131 dst_c1(not_empty_dst_pixels_mask) = dst_blend * (src_c1 - dst_c1) + dst_c1; in compositeVector()
|
H A D | KoOptimizedCompositeOpAlphaDarken128.h | 59 Vc::float_v src_c1; in compositeVector() local 66 tie(src_c1, src_c2, src_c3, src_alpha) = data[indexes]; in compositeVector() 101 dst_c1 = (src_c1 - dst_c1) * src_alpha + dst_c1; in compositeVector() 106 dst_c1(empty_dst_pixels_mask) = src_c1; in compositeVector() 110 dst_c1(not_empty_dst_pixels_mask) = (src_c1 - dst_c1) * src_alpha + dst_c1; in compositeVector() 116 dst_c1 = src_c1; in compositeVector()
|
H A D | KoOptimizedCompositeOpOver128.h | 72 Vc::float_v src_c1; in compositeVector() local 78 tie(src_c1, src_c2, src_c3, src_alpha) = data[indexes]; in compositeVector() 133 dst_c1 = src_blend * (src_c1 - dst_c1) + dst_c1; in compositeVector() 142 dataDest[indexes] = tie(src_c1, src_c2, src_c3, new_alpha); in compositeVector()
|
H A D | KoOptimizedCompositeOpOver32.h | 105 Vc::float_v src_c1; in compositeVector() local 114 KoStreamedMath<_impl>::template fetch_colors_32<src_aligned>(src, src_c1, src_c2, src_c3); in compositeVector() 142 dst_c1 = src_blend * (src_c1 - dst_c1) + dst_c1; in compositeVector() 153 dst_c1 = src_c1; in compositeVector()
|
/dports/lang/clover/mesa-21.3.6/src/amd/llvm/ |
H A D | ac_nir_to_llvm.c | 3265 LLVMValueRef src_c1 = in barycentric_offset() local 3290 temp2 = ac_build_fmad(&ctx->ac, ddy_el, src_c1, temp1); in barycentric_offset() 3317 LLVMValueRef src_c1 = LLVMBuildExtractElement(ctx->ac.builder, sample_pos, ctx->ac.i32_1, ""); in barycentric_at_sample() local 3318 src_c1 = LLVMBuildFSub(ctx->ac.builder, src_c1, halfval, ""); in barycentric_at_sample() 3319 LLVMValueRef coords[] = {src_c0, src_c1}; in barycentric_at_sample()
|
/dports/graphics/libosmesa-gallium/mesa-21.3.6/src/amd/llvm/ |
H A D | ac_nir_to_llvm.c | 3265 LLVMValueRef src_c1 = in barycentric_offset() local 3290 temp2 = ac_build_fmad(&ctx->ac, ddy_el, src_c1, temp1); in barycentric_offset() 3317 LLVMValueRef src_c1 = LLVMBuildExtractElement(ctx->ac.builder, sample_pos, ctx->ac.i32_1, ""); in barycentric_at_sample() local 3318 src_c1 = LLVMBuildFSub(ctx->ac.builder, src_c1, halfval, ""); in barycentric_at_sample() 3319 LLVMValueRef coords[] = {src_c0, src_c1}; in barycentric_at_sample()
|
/dports/graphics/libosmesa/mesa-21.3.6/src/amd/llvm/ |
H A D | ac_nir_to_llvm.c | 3265 LLVMValueRef src_c1 = in barycentric_offset() local 3290 temp2 = ac_build_fmad(&ctx->ac, ddy_el, src_c1, temp1); in barycentric_offset() 3317 LLVMValueRef src_c1 = LLVMBuildExtractElement(ctx->ac.builder, sample_pos, ctx->ac.i32_1, ""); in barycentric_at_sample() local 3318 src_c1 = LLVMBuildFSub(ctx->ac.builder, src_c1, halfval, ""); in barycentric_at_sample() 3319 LLVMValueRef coords[] = {src_c0, src_c1}; in barycentric_at_sample()
|
/dports/graphics/mesa-gallium-xa/mesa-21.3.6/src/amd/llvm/ |
H A D | ac_nir_to_llvm.c | 3265 LLVMValueRef src_c1 = in barycentric_offset() local 3290 temp2 = ac_build_fmad(&ctx->ac, ddy_el, src_c1, temp1); in barycentric_offset() 3317 LLVMValueRef src_c1 = LLVMBuildExtractElement(ctx->ac.builder, sample_pos, ctx->ac.i32_1, ""); in barycentric_at_sample() local 3318 src_c1 = LLVMBuildFSub(ctx->ac.builder, src_c1, halfval, ""); in barycentric_at_sample() 3319 LLVMValueRef coords[] = {src_c0, src_c1}; in barycentric_at_sample()
|
/dports/graphics/mesa-libs/mesa-21.3.6/src/amd/llvm/ |
H A D | ac_nir_to_llvm.c | 3265 LLVMValueRef src_c1 = in barycentric_offset() local 3290 temp2 = ac_build_fmad(&ctx->ac, ddy_el, src_c1, temp1); in barycentric_offset() 3317 LLVMValueRef src_c1 = LLVMBuildExtractElement(ctx->ac.builder, sample_pos, ctx->ac.i32_1, ""); in barycentric_at_sample() local 3318 src_c1 = LLVMBuildFSub(ctx->ac.builder, src_c1, halfval, ""); in barycentric_at_sample() 3319 LLVMValueRef coords[] = {src_c0, src_c1}; in barycentric_at_sample()
|
/dports/graphics/mesa-gallium-va/mesa-21.3.6/src/amd/llvm/ |
H A D | ac_nir_to_llvm.c | 3265 LLVMValueRef src_c1 = in barycentric_offset() local 3290 temp2 = ac_build_fmad(&ctx->ac, ddy_el, src_c1, temp1); in barycentric_offset() 3317 LLVMValueRef src_c1 = LLVMBuildExtractElement(ctx->ac.builder, sample_pos, ctx->ac.i32_1, ""); in barycentric_at_sample() local 3318 src_c1 = LLVMBuildFSub(ctx->ac.builder, src_c1, halfval, ""); in barycentric_at_sample() 3319 LLVMValueRef coords[] = {src_c0, src_c1}; in barycentric_at_sample()
|
/dports/graphics/mesa-dri-gallium/mesa-21.3.6/src/amd/llvm/ |
H A D | ac_nir_to_llvm.c | 3265 LLVMValueRef src_c1 = in barycentric_offset() local 3290 temp2 = ac_build_fmad(&ctx->ac, ddy_el, src_c1, temp1); in barycentric_offset() 3317 LLVMValueRef src_c1 = LLVMBuildExtractElement(ctx->ac.builder, sample_pos, ctx->ac.i32_1, ""); in barycentric_at_sample() local 3318 src_c1 = LLVMBuildFSub(ctx->ac.builder, src_c1, halfval, ""); in barycentric_at_sample() 3319 LLVMValueRef coords[] = {src_c0, src_c1}; in barycentric_at_sample()
|
/dports/graphics/mesa-gallium-vdpau/mesa-21.3.6/src/amd/llvm/ |
H A D | ac_nir_to_llvm.c | 3265 LLVMValueRef src_c1 = in barycentric_offset() local 3290 temp2 = ac_build_fmad(&ctx->ac, ddy_el, src_c1, temp1); in barycentric_offset() 3317 LLVMValueRef src_c1 = LLVMBuildExtractElement(ctx->ac.builder, sample_pos, ctx->ac.i32_1, ""); in barycentric_at_sample() local 3318 src_c1 = LLVMBuildFSub(ctx->ac.builder, src_c1, halfval, ""); in barycentric_at_sample() 3319 LLVMValueRef coords[] = {src_c0, src_c1}; in barycentric_at_sample()
|
/dports/graphics/mesa-dri/mesa-21.3.6/src/amd/llvm/ |
H A D | ac_nir_to_llvm.c | 3265 LLVMValueRef src_c1 = in barycentric_offset() local 3290 temp2 = ac_build_fmad(&ctx->ac, ddy_el, src_c1, temp1); in barycentric_offset() 3317 LLVMValueRef src_c1 = LLVMBuildExtractElement(ctx->ac.builder, sample_pos, ctx->ac.i32_1, ""); in barycentric_at_sample() local 3318 src_c1 = LLVMBuildFSub(ctx->ac.builder, src_c1, halfval, ""); in barycentric_at_sample() 3319 LLVMValueRef coords[] = {src_c0, src_c1}; in barycentric_at_sample()
|