Lines Matching refs:src_tmp
575 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
576 + SIMD_MASK mask = SIMD_CMPGT_F32(SIMD_SET0_F32, src_tmp);
577 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_F32(src_tmp, alpha_data), mask));
675 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
676 + SIMD_F32 exp_tmp = SIMD_SUB_N_F32(SIMD_EXP_F32(src_tmp), 1.0f);
677 + SIMD_MASK mask = SIMD_CMPLE_F32(src_tmp, SIMD_SET0_F32);
678 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_N_F32(exp_tmp, alpha), mask));
685 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
686 + SIMD_F32 exp_tmp = SIMD_SUB_N_F32(SIMD_EXP_F32(SIMD_DIV_N_F32(src_tmp, alpha)), 1.0f);
687 + SIMD_MASK mask = SIMD_CMPLE_F32(src_tmp, SIMD_SET0_F32);
688 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_N_F32(exp_tmp, alpha), mask));
696 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
697 + SIMD_MASK mask0 = SIMD_CMPLE_F32(src_tmp, SIMD_MOV_F32(lambd));
698 + SIMD_MASK mask1 = SIMD_CMPLE_F32(SIMD_MOV_F32(neg_lambd), src_tmp);
700 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MOV_F32(0.0f), mask));
723 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
724 + SIMD_F32 divisor_tmp = SIMD_ADD_F32(SIMD_MOV_F32(1.0f), SIMD_ABS_F32(src_tmp));
725 + SIMD_ST_F32(dst + index, SIMD_DIV_F32(src_tmp, divisor_tmp));
3572 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
3573 + SIMD_MASK mask = SIMD_CMPGT_F32(SIMD_SET0_F32, src_tmp);
3574 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_F32(src_tmp, alpha_data), mask));
3672 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
3673 + SIMD_F32 exp_tmp = SIMD_SUB_N_F32(SIMD_EXP_F32(src_tmp), 1.0f);
3674 + SIMD_MASK mask = SIMD_CMPLE_F32(src_tmp, SIMD_SET0_F32);
3675 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_N_F32(exp_tmp, alpha), mask));
3682 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
3683 + SIMD_F32 exp_tmp = SIMD_SUB_N_F32(SIMD_EXP_F32(SIMD_DIV_N_F32(src_tmp, alpha)), 1.0f);
3684 + SIMD_MASK mask = SIMD_CMPLE_F32(src_tmp, SIMD_SET0_F32);
3685 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_N_F32(exp_tmp, alpha), mask));
3693 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
3694 + SIMD_MASK mask0 = SIMD_CMPLE_F32(src_tmp, SIMD_MOV_F32(lambd));
3695 + SIMD_MASK mask1 = SIMD_CMPLE_F32(SIMD_MOV_F32(neg_lambd), src_tmp);
3697 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MOV_F32(0.0f), mask));
3720 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
3721 + SIMD_F32 divisor_tmp = SIMD_ADD_F32(SIMD_MOV_F32(1.0f), SIMD_ABS_F32(src_tmp));
3722 + SIMD_ST_F32(dst + index, SIMD_DIV_F32(src_tmp, divisor_tmp));
7156 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
7157 + SIMD_MASK mask = SIMD_CMPGT_F32(SIMD_SET0_F32, src_tmp);
7158 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_F32(src_tmp, alpha_data), mask));
7256 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
7257 + SIMD_F32 exp_tmp = SIMD_SUB_N_F32(SIMD_EXP_F32(src_tmp), 1.0f);
7258 + SIMD_MASK mask = SIMD_CMPLE_F32(src_tmp, SIMD_SET0_F32);
7259 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_N_F32(exp_tmp, alpha), mask));
7266 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
7267 + SIMD_F32 exp_tmp = SIMD_SUB_N_F32(SIMD_EXP_F32(SIMD_DIV_N_F32(src_tmp, alpha)), 1.0f);
7268 + SIMD_MASK mask = SIMD_CMPLE_F32(src_tmp, SIMD_SET0_F32);
7269 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_N_F32(exp_tmp, alpha), mask));
7277 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
7278 + SIMD_MASK mask0 = SIMD_CMPLE_F32(src_tmp, SIMD_MOV_F32(lambd));
7279 + SIMD_MASK mask1 = SIMD_CMPLE_F32(SIMD_MOV_F32(neg_lambd), src_tmp);
7281 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MOV_F32(0.0f), mask));
7304 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
7305 + SIMD_F32 divisor_tmp = SIMD_ADD_F32(SIMD_MOV_F32(1.0f), SIMD_ABS_F32(src_tmp));
7306 + SIMD_ST_F32(dst + index, SIMD_DIV_F32(src_tmp, divisor_tmp));
10297 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
10298 + SIMD_MASK mask = SIMD_CMPGT_F32(SIMD_SET0_F32, src_tmp);
10299 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_F32(src_tmp, alpha_data), mask));
10397 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
10398 + SIMD_F32 exp_tmp = SIMD_SUB_N_F32(SIMD_EXP_F32(src_tmp), 1.0f);
10399 + SIMD_MASK mask = SIMD_CMPLE_F32(src_tmp, SIMD_SET0_F32);
10400 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_N_F32(exp_tmp, alpha), mask));
10407 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
10408 + SIMD_F32 exp_tmp = SIMD_SUB_N_F32(SIMD_EXP_F32(SIMD_DIV_N_F32(src_tmp, alpha)), 1.0f);
10409 + SIMD_MASK mask = SIMD_CMPLE_F32(src_tmp, SIMD_SET0_F32);
10410 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MUL_N_F32(exp_tmp, alpha), mask));
10418 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
10419 + SIMD_MASK mask0 = SIMD_CMPLE_F32(src_tmp, SIMD_MOV_F32(lambd));
10420 + SIMD_MASK mask1 = SIMD_CMPLE_F32(SIMD_MOV_F32(neg_lambd), src_tmp);
10422 + SIMD_ST_F32(dst + index, SIMD_BLEND_F32(src_tmp, SIMD_MOV_F32(0.0f), mask));
10445 + SIMD_F32 src_tmp = SIMD_LD_F32(src + index);
10446 + SIMD_F32 divisor_tmp = SIMD_ADD_F32(SIMD_MOV_F32(1.0f), SIMD_ABS_F32(src_tmp));
10447 + SIMD_ST_F32(dst + index, SIMD_DIV_F32(src_tmp, divisor_tmp));