swscale/range_convert: saturate output instead of limiting input

For bit depths <= 14, the result is saturated to 15 bits.
For bit depths > 14, the result is saturated to 19 bits.

x86_64:
chrRangeFromJpeg8_1920_c:    2126.5   2127.4  (1.00x)
chrRangeFromJpeg16_1920_c:   2331.4   2325.2  (1.00x)
chrRangeToJpeg8_1920_c:      3163.0   3166.9  (1.00x)
chrRangeToJpeg16_1920_c:     3163.7   2152.4  (1.47x)
lumRangeFromJpeg8_1920_c:    1262.2   1263.0  (1.00x)
lumRangeFromJpeg16_1920_c:   1079.5   1080.5  (1.00x)
lumRangeToJpeg8_1920_c:      1860.5   1886.8  (0.99x)
lumRangeToJpeg16_1920_c:     1910.2   1077.0  (1.77x)

aarch64 A55:
chrRangeFromJpeg8_1920_c:   28836.2  28835.2  (1.00x)
chrRangeFromJpeg16_1920_c:  28840.1  28839.8  (1.00x)
chrRangeToJpeg8_1920_c:     44196.2  23074.7  (1.92x)
chrRangeToJpeg16_1920_c:    36527.3  17318.9  (2.11x)
lumRangeFromJpeg8_1920_c:   15388.5  15389.7  (1.00x)
lumRangeFromJpeg16_1920_c:  15389.3  15388.2  (1.00x)
lumRangeToJpeg8_1920_c:     23069.7  19227.8  (1.20x)
lumRangeToJpeg16_1920_c:    19227.8  15387.0  (1.25x)

aarch64 A76:
chrRangeFromJpeg8_1920_c:    6334.7   6324.4  (1.00x)
chrRangeFromJpeg16_1920_c:   6336.0   6339.9  (1.00x)
chrRangeToJpeg8_1920_c:     11474.5   9656.0  (1.19x)
chrRangeToJpeg16_1920_c:     9640.5   6340.4  (1.52x)
lumRangeFromJpeg8_1920_c:    4453.2   4422.0  (1.01x)
lumRangeFromJpeg16_1920_c:   4414.2   4420.9  (1.00x)
lumRangeToJpeg8_1920_c:      6645.0   5949.1  (1.12x)
lumRangeToJpeg16_1920_c:     6005.2   4446.8  (1.35x)

NOTE: all simd optimizations for range_convert have been disabled
      except for x86, which already had the same behaviour.
      they will be re-enabled when they are fixed for each architecture.

Author: ramiropolla

libswscale/aarch64/swscale.c

@@ -225,6 +225,10 @@ void ff_chrRangeToJpeg_neon(int16_t *dstU, int16_t *dstV, int width);
 
 av_cold void ff_sws_init_range_convert_aarch64(SwsInternal *c)
 {
+    /* This code is currently disabled because of changes in the base
+     * implementation of these functions. This code should be enabled
+     * again once those changes are ported to this architecture. */
+#if 0
     int cpu_flags = av_get_cpu_flags();
 
     if (have_neon(cpu_flags)) {
@@ -238,6 +242,7 @@ av_cold void ff_sws_init_range_convert_aarch64(SwsInternal *c)
             }
         }
     }
+#endif
 }
 
 av_cold void ff_sws_init_swscale_aarch64(SwsInternal *c)

libswscale/loongarch/swscale_init_loongarch.c

@@ -26,6 +26,10 @@
 
 av_cold void ff_sws_init_range_convert_loongarch(SwsInternal *c)
 {
+    /* This code is currently disabled because of changes in the base
+     * implementation of these functions. This code should be enabled
+     * again once those changes are ported to this architecture. */
+#if 0
     int cpu_flags = av_get_cpu_flags();
 
     if (have_lsx(cpu_flags)) {
@@ -52,6 +56,7 @@ av_cold void ff_sws_init_range_convert_loongarch(SwsInternal *c)
         }
     }
 #endif // #if HAVE_LASX
+#endif
 }
 
 av_cold void ff_sws_init_swscale_loongarch(SwsInternal *c)

libswscale/riscv/swscale.c

@@ -28,6 +28,10 @@ void ff_range_chr_from_jpeg_16_rvv(int16_t *, int16_t *, int);
 
 av_cold void ff_sws_init_range_convert_riscv(SwsInternal *c)
 {
+    /* This code is currently disabled because of changes in the base
+     * implementation of these functions. This code should be enabled
+     * again once those changes are ported to this architecture. */
+#if 0
 #if HAVE_RVV
     int flags = av_get_cpu_flags();
 
@@ -47,6 +51,7 @@ av_cold void ff_sws_init_range_convert_riscv(SwsInternal *c)
         c->chrConvertRange = convs[from].chr;
     }
 #endif
+#endif
 }
 
 #define RVV_INPUT(name) \

libswscale/swscale.c

@@ -160,8 +160,10 @@ static void chrRangeToJpeg_c(int16_t *dstU, int16_t *dstV, int width)
 {
     int i;
     for (i = 0; i < width; i++) {
-        dstU[i] = (FFMIN(dstU[i], 30775) * 4663 - 9289992) >> 12; // -264
-        dstV[i] = (FFMIN(dstV[i], 30775) * 4663 - 9289992) >> 12; // -264
+        int U = (dstU[i] * 4663 - 9289992) >> 12; // -264
+        int V = (dstV[i] * 4663 - 9289992) >> 12; // -264
+        dstU[i] = FFMIN(U, (1 << 15) - 1);
+        dstV[i] = FFMIN(V, (1 << 15) - 1);
     }
 }
 
@@ -177,8 +179,10 @@ static void chrRangeFromJpeg_c(int16_t *dstU, int16_t *dstV, int width)
 static void lumRangeToJpeg_c(int16_t *dst, int width)
 {
     int i;
-    for (i = 0; i < width; i++)
-        dst[i] = (FFMIN(dst[i], 30189) * 19077 - 39057361) >> 14;
+    for (i = 0; i < width; i++) {
+        int Y = (dst[i] * 19077 - 39057361) >> 14;
+        dst[i] = FFMIN(Y, (1 << 15) - 1);
+    }
 }
 
 static void lumRangeFromJpeg_c(int16_t *dst, int width)
@@ -194,8 +198,10 @@ static void chrRangeToJpeg16_c(int16_t *_dstU, int16_t *_dstV, int width)
     int32_t *dstU = (int32_t *) _dstU;
     int32_t *dstV = (int32_t *) _dstV;
     for (i = 0; i < width; i++) {
-        dstU[i] = ((int)(FFMIN(dstU[i], 30775 << 4) * 4663U - (9289992 << 4))) >> 12; // -264
-        dstV[i] = ((int)(FFMIN(dstV[i], 30775 << 4) * 4663U - (9289992 << 4))) >> 12; // -264
+        int U = ((int)(dstU[i] * 4663U - (9289992 << 4))) >> 12; // -264
+        int V = ((int)(dstV[i] * 4663U - (9289992 << 4))) >> 12; // -264
+        dstU[i] = FFMIN(U, (1 << 19) - 1);
+        dstV[i] = FFMIN(V, (1 << 19) - 1);
     }
 }
 
@@ -215,7 +221,8 @@ static void lumRangeToJpeg16_c(int16_t *_dst, int width)
     int i;
     int32_t *dst = (int32_t *) _dst;
     for (i = 0; i < width; i++) {
-        dst[i] = ((int)(FFMIN(dst[i], 30189 << 4) * 4769U - (39057361 << 2))) >> 12;
+        int Y = ((int)(dst[i] * 4769U - (39057361 << 2))) >> 12;
+        dst[i] = FFMIN(Y, (1 << 19) - 1);
     }
 }
 

libswscale/x86/range_convert.asm

@@ -40,9 +40,6 @@ lum_from_offset:    times 4 dd 33561947
 
 SECTION .text
 
-; NOTE: there is no need to clamp the input when converting to jpeg range
-;       (like we do in the C code) because packssdw will saturate the output.
-
 ;-----------------------------------------------------------------------------
 ; lumConvertRange
 ;