From 30bef20e228c3e0cae559150c03b93c735d2bfea Mon Sep 17 00:00:00 2001 From: Alexander Alekhin Date: Sun, 31 Jan 2021 00:12:51 +0000 Subject: [PATCH] js: fix SIMD build --- .../include/opencv2/core/hal/intrin_wasm.hpp | 35 ++++++++++--------- 1 file changed, 18 insertions(+), 17 deletions(-) diff --git a/modules/core/include/opencv2/core/hal/intrin_wasm.hpp b/modules/core/include/opencv2/core/hal/intrin_wasm.hpp index 2f835bb9f8..b4178af8b7 100644 --- a/modules/core/include/opencv2/core/hal/intrin_wasm.hpp +++ b/modules/core/include/opencv2/core/hal/intrin_wasm.hpp @@ -1266,8 +1266,9 @@ OPENCV_HAL_IMPL_WASM_BIN_FUNC(v_uint8x16, v_sub_wrap, wasm_i8x16_sub) OPENCV_HAL_IMPL_WASM_BIN_FUNC(v_int8x16, v_sub_wrap, wasm_i8x16_sub) OPENCV_HAL_IMPL_WASM_BIN_FUNC(v_uint16x8, v_sub_wrap, wasm_i16x8_sub) OPENCV_HAL_IMPL_WASM_BIN_FUNC(v_int16x8, v_sub_wrap, wasm_i16x8_sub) -#if (__EMSCRIPTEN_major__ * 1000000 + __EMSCRIPTEN_minor__ * 1000 + __EMSCRIPTEN_tiny__) >= (2000000) +#if (__EMSCRIPTEN_major__ * 1000000 + __EMSCRIPTEN_minor__ * 1000 + __EMSCRIPTEN_tiny__) >= (1039012) // details: https://github.com/opencv/opencv/issues/18097 ( https://github.com/emscripten-core/emscripten/issues/12018 ) +// 1.39.12: https://github.com/emscripten-core/emscripten/commit/cd801d0f110facfd694212a3c8b2ed2ffcd630e2 inline v_uint8x16 v_mul_wrap(const v_uint8x16& a, const v_uint8x16& b) { uchar a_[16], b_[16]; @@ -1275,7 +1276,7 @@ inline v_uint8x16 v_mul_wrap(const v_uint8x16& a, const v_uint8x16& b) wasm_v128_store(b_, b.val); for (int i = 0; i < 16; i++) a_[i] = (uchar)(a_[i] * b_[i]); - return wasm_v128_load(a_); + return v_uint8x16(wasm_v128_load(a_)); } inline v_int8x16 v_mul_wrap(const v_int8x16& a, const v_int8x16& b) { @@ -1284,7 +1285,7 @@ inline v_int8x16 v_mul_wrap(const v_int8x16& a, const v_int8x16& b) wasm_v128_store(b_, b.val); for (int i = 0; i < 16; i++) a_[i] = (schar)(a_[i] * b_[i]); - return wasm_v128_load(a_); + return v_int8x16(wasm_v128_load(a_)); } #else OPENCV_HAL_IMPL_WASM_BIN_FUNC(v_uint8x16, v_mul_wrap, wasm_i8x16_mul) @@ -1757,8 +1758,8 @@ inline v_uint64x2 v_popcount(const v_uint64x2& a) uint64 a_[2], b_[2] = { 0 }; wasm_v128_store(a_, a.val); for (int i = 0; i < 16; i++) - b_[i / 8] += popCountTable[((uint8*)a_)[i]]; - return wasm_v128_load(b_); + b_[i / 8] += popCountTable[((uint8_t*)a_)[i]]; + return v_uint64x2(wasm_v128_load(b_)); } inline v_uint8x16 v_popcount(const v_int8x16& a) { return v_popcount(v_reinterpret_as_u8(a)); } @@ -1938,11 +1939,11 @@ inline v_int32x4 func(const v_float64x2& a) \ double a_[2]; \ wasm_v128_store(a_, a.val); \ int c_[4]; \ - c_[0] = cfunc(a_[i]); \ - c_[1] = cfunc(a_[i]); \ + c_[0] = cfunc(a_[0]); \ + c_[1] = cfunc(a_[1]); \ c_[2] = 0; \ c_[3] = 0; \ - return wasm_v128_load(c_); \ + return v_int32x4(wasm_v128_load(c_)); \ } OPENCV_HAL_IMPL_WASM_MATH_FUNC(v_round, cvRound) @@ -1960,7 +1961,7 @@ inline v_int32x4 v_round(const v_float64x2& a, const v_float64x2& b) c_[1] = cvRound(a_[1]); c_[2] = cvRound(b_[0]); c_[3] = cvRound(b_[1]); - return wasm_v128_load(c_); + return v_int32x4(wasm_v128_load(c_)); } #define OPENCV_HAL_IMPL_WASM_TRANSPOSE4x4(_Tpvec, suffix) \ @@ -2461,7 +2462,7 @@ inline v_float32x4 v_cvt_f32(const v_float64x2& a) c_[1] = (float)(a_[1]); c_[2] = 0; c_[3] = 0; - return wasm_v128_load(c_); + return v_float32x4(wasm_v128_load(c_)); } inline v_float32x4 v_cvt_f32(const v_float64x2& a, const v_float64x2& b) @@ -2474,7 +2475,7 @@ inline v_float32x4 v_cvt_f32(const v_float64x2& a, const v_float64x2& b) c_[1] = (float)(a_[1]); c_[2] = (float)(b_[0]); c_[3] = (float)(b_[1]); - return wasm_v128_load(c_); + return v_float32x4(wasm_v128_load(c_)); } inline v_float64x2 v_cvt_f64(const v_int32x4& a) @@ -2488,7 +2489,7 @@ inline v_float64x2 v_cvt_f64(const v_int32x4& a) double c_[2]; c_[0] = (double)(a_[0]); c_[1] = (double)(a_[1]); - return wasm_v128_load(c_); + return v_float64x2(wasm_v128_load(c_)); #endif } @@ -2503,7 +2504,7 @@ inline v_float64x2 v_cvt_f64_high(const v_int32x4& a) double c_[2]; c_[0] = (double)(a_[2]); c_[1] = (double)(a_[3]); - return wasm_v128_load(c_); + return v_float64x2(wasm_v128_load(c_)); #endif } @@ -2514,7 +2515,7 @@ inline v_float64x2 v_cvt_f64(const v_float32x4& a) double c_[2]; c_[0] = (double)(a_[0]); c_[1] = (double)(a_[1]); - return wasm_v128_load(c_); + return v_float64x2(wasm_v128_load(c_)); } inline v_float64x2 v_cvt_f64_high(const v_float32x4& a) @@ -2524,7 +2525,7 @@ inline v_float64x2 v_cvt_f64_high(const v_float32x4& a) double c_[2]; c_[0] = (double)(a_[2]); c_[1] = (double)(a_[3]); - return wasm_v128_load(c_); + return v_float64x2(wasm_v128_load(c_)); } inline v_float64x2 v_cvt_f64(const v_int64x2& a) @@ -2537,7 +2538,7 @@ inline v_float64x2 v_cvt_f64(const v_int64x2& a) double c_[2]; c_[0] = (double)(a_[0]); c_[1] = (double)(a_[1]); - return wasm_v128_load(c_); + return v_float64x2(wasm_v128_load(c_)); #endif } @@ -2757,7 +2758,7 @@ inline v_float32x4 v_load_expand(const float16_t* ptr) float a[4]; for (int i = 0; i < 4; i++) a[i] = ptr[i]; - return wasm_v128_load(a); + return v_float32x4(wasm_v128_load(a)); } inline void v_pack_store(float16_t* ptr, const v_float32x4& v)