libwebp: update to 0.5.0

This commit imports libwebp 0.5.0, including AUTHORS, COPYING, ChangeLog,
NEWS, PATENTS, README and src directories. In src, only includes header
and source files.

The patches required to build it in Qt will follow in separate
commit(s).

Change-Id: I96b4961ba63c75cc7fbab158c36a0f403f254c14
Reviewed-by: aavit <eirik.aavitsland@theqtcompany.com>

1 files changed, 189 insertions, 293 deletions

diff --git a/src/3rdparty/libwebp/src/dsp/enc_mips32.c b/src/3rdparty/libwebp/src/dsp/enc_mips32.c
index 6cede18..fd10143 100644
--- a/src/3rdparty/libwebp/src/dsp/enc_mips32.c
+++ b/src/3rdparty/libwebp/src/dsp/enc_mips32.c
@@ -17,13 +17,10 @@
 
 #if defined(WEBP_USE_MIPS32)
 
+#include "./mips_macro.h"
 #include "../enc/vp8enci.h"
 #include "../enc/cost.h"
 
-#if defined(__GNUC__) && defined(__ANDROID__) && LOCAL_GCC_VERSION == 0x409
-#define WORK_AROUND_GCC
-#endif
-
 static const int kC1 = 20091 + (1 << 16);
 static const int kC2 = 35468;
 
@@ -59,61 +56,61 @@ static const int kC2 = 35468;
 // MUL and STORE macros inlined
 // a = clip_8b(a) is replaced with: a = max(a, 0); a = min(a, 255)
 // temp0..temp15 holds tmp[0]..tmp[15]
-// A..D - offsets in bytes to load from ref and store to dst buffer
+// A - offset in bytes to load from ref and store to dst buffer
 // TEMP0, TEMP4, TEMP8 and TEMP12 - registers for corresponding tmp elements
-#define HORIZONTAL_PASS(A, B, C, D, TEMP0, TEMP4, TEMP8, TEMP12)              \
-  "addiu   %[" #TEMP0 "],    %[" #TEMP0 "],    4             \n\t"            \
-  "addu    %[temp16],      %[" #TEMP0 "],    %[" #TEMP8 "]   \n\t"            \
-  "subu    %[temp17],      %[" #TEMP0 "],    %[" #TEMP8 "]   \n\t"            \
-  "mul     %[" #TEMP0 "],    %[" #TEMP4 "],    %[kC2]        \n\t"            \
-  "mul     %[" #TEMP8 "],    %[" #TEMP12 "],   %[kC1]        \n\t"            \
-  "mul     %[" #TEMP4 "],    %[" #TEMP4 "],    %[kC1]        \n\t"            \
-  "mul     %[" #TEMP12 "],   %[" #TEMP12 "],   %[kC2]        \n\t"            \
-  "sra     %[" #TEMP0 "],    %[" #TEMP0 "],    16            \n\t"            \
-  "sra     %[" #TEMP8 "],    %[" #TEMP8 "],    16            \n\t"            \
-  "sra     %[" #TEMP4 "],    %[" #TEMP4 "],    16            \n\t"            \
-  "sra     %[" #TEMP12 "],   %[" #TEMP12 "],   16            \n\t"            \
-  "subu    %[temp18],      %[" #TEMP0 "],    %[" #TEMP8 "]   \n\t"            \
-  "addu    %[temp19],      %[" #TEMP4 "],    %[" #TEMP12 "]  \n\t"            \
-  "addu    %[" #TEMP0 "],    %[temp16],      %[temp19]       \n\t"            \
-  "addu    %[" #TEMP4 "],    %[temp17],      %[temp18]       \n\t"            \
-  "subu    %[" #TEMP8 "],    %[temp17],      %[temp18]       \n\t"            \
-  "subu    %[" #TEMP12 "],   %[temp16],      %[temp19]       \n\t"            \
-  "lw      %[temp20],      0(%[args])                        \n\t"            \
-  "sra     %[" #TEMP0 "],    %[" #TEMP0 "],    3             \n\t"            \
-  "sra     %[" #TEMP4 "],    %[" #TEMP4 "],    3             \n\t"            \
-  "sra     %[" #TEMP8 "],    %[" #TEMP8 "],    3             \n\t"            \
-  "sra     %[" #TEMP12 "],   %[" #TEMP12 "],   3             \n\t"            \
-  "lbu     %[temp16],      " #A "(%[temp20])                 \n\t"            \
-  "lbu     %[temp17],      " #B "(%[temp20])                 \n\t"            \
-  "lbu     %[temp18],      " #C "(%[temp20])                 \n\t"            \
-  "lbu     %[temp19],      " #D "(%[temp20])                 \n\t"            \
-  "addu    %[" #TEMP0 "],    %[temp16],      %[" #TEMP0 "]   \n\t"            \
-  "addu    %[" #TEMP4 "],    %[temp17],      %[" #TEMP4 "]   \n\t"            \
-  "addu    %[" #TEMP8 "],    %[temp18],      %[" #TEMP8 "]   \n\t"            \
-  "addu    %[" #TEMP12 "],   %[temp19],      %[" #TEMP12 "]  \n\t"            \
-  "slt     %[temp16],      %[" #TEMP0 "],    $zero           \n\t"            \
-  "slt     %[temp17],      %[" #TEMP4 "],    $zero           \n\t"            \
-  "slt     %[temp18],      %[" #TEMP8 "],    $zero           \n\t"            \
-  "slt     %[temp19],      %[" #TEMP12 "],   $zero           \n\t"            \
-  "movn    %[" #TEMP0 "],    $zero,          %[temp16]       \n\t"            \
-  "movn    %[" #TEMP4 "],    $zero,          %[temp17]       \n\t"            \
-  "movn    %[" #TEMP8 "],    $zero,          %[temp18]       \n\t"            \
-  "movn    %[" #TEMP12 "],   $zero,          %[temp19]       \n\t"            \
-  "addiu   %[temp20],      $zero,          255               \n\t"            \
-  "slt     %[temp16],      %[" #TEMP0 "],    %[temp20]       \n\t"            \
-  "slt     %[temp17],      %[" #TEMP4 "],    %[temp20]       \n\t"            \
-  "slt     %[temp18],      %[" #TEMP8 "],    %[temp20]       \n\t"            \
-  "slt     %[temp19],      %[" #TEMP12 "],   %[temp20]       \n\t"            \
-  "movz    %[" #TEMP0 "],    %[temp20],      %[temp16]       \n\t"            \
-  "movz    %[" #TEMP4 "],    %[temp20],      %[temp17]       \n\t"            \
-  "lw      %[temp16],      8(%[args])                        \n\t"            \
-  "movz    %[" #TEMP8 "],    %[temp20],      %[temp18]       \n\t"            \
-  "movz    %[" #TEMP12 "],   %[temp20],      %[temp19]       \n\t"            \
-  "sb      %[" #TEMP0 "],    " #A "(%[temp16])               \n\t"            \
-  "sb      %[" #TEMP4 "],    " #B "(%[temp16])               \n\t"            \
-  "sb      %[" #TEMP8 "],    " #C "(%[temp16])               \n\t"            \
-  "sb      %[" #TEMP12 "],   " #D "(%[temp16])               \n\t"
+#define HORIZONTAL_PASS(A, TEMP0, TEMP4, TEMP8, TEMP12)                       \
+  "addiu   %[" #TEMP0 "],    %[" #TEMP0 "],    4               \n\t"          \
+  "addu    %[temp16],      %[" #TEMP0 "],    %[" #TEMP8 "]     \n\t"          \
+  "subu    %[temp17],      %[" #TEMP0 "],    %[" #TEMP8 "]     \n\t"          \
+  "mul     %[" #TEMP0 "],    %[" #TEMP4 "],    %[kC2]          \n\t"          \
+  "mul     %[" #TEMP8 "],    %[" #TEMP12 "],   %[kC1]          \n\t"          \
+  "mul     %[" #TEMP4 "],    %[" #TEMP4 "],    %[kC1]          \n\t"          \
+  "mul     %[" #TEMP12 "],   %[" #TEMP12 "],   %[kC2]          \n\t"          \
+  "sra     %[" #TEMP0 "],    %[" #TEMP0 "],    16              \n\t"          \
+  "sra     %[" #TEMP8 "],    %[" #TEMP8 "],    16              \n\t"          \
+  "sra     %[" #TEMP4 "],    %[" #TEMP4 "],    16              \n\t"          \
+  "sra     %[" #TEMP12 "],   %[" #TEMP12 "],   16              \n\t"          \
+  "subu    %[temp18],      %[" #TEMP0 "],    %[" #TEMP8 "]     \n\t"          \
+  "addu    %[temp19],      %[" #TEMP4 "],    %[" #TEMP12 "]    \n\t"          \
+  "addu    %[" #TEMP0 "],    %[temp16],      %[temp19]         \n\t"          \
+  "addu    %[" #TEMP4 "],    %[temp17],      %[temp18]         \n\t"          \
+  "subu    %[" #TEMP8 "],    %[temp17],      %[temp18]         \n\t"          \
+  "subu    %[" #TEMP12 "],   %[temp16],      %[temp19]         \n\t"          \
+  "lw      %[temp20],      0(%[args])                          \n\t"          \
+  "sra     %[" #TEMP0 "],    %[" #TEMP0 "],    3               \n\t"          \
+  "sra     %[" #TEMP4 "],    %[" #TEMP4 "],    3               \n\t"          \
+  "sra     %[" #TEMP8 "],    %[" #TEMP8 "],    3               \n\t"          \
+  "sra     %[" #TEMP12 "],   %[" #TEMP12 "],   3               \n\t"          \
+  "lbu     %[temp16],      0+" XSTR(BPS) "*" #A "(%[temp20])   \n\t"          \
+  "lbu     %[temp17],      1+" XSTR(BPS) "*" #A "(%[temp20])   \n\t"          \
+  "lbu     %[temp18],      2+" XSTR(BPS) "*" #A "(%[temp20])   \n\t"          \
+  "lbu     %[temp19],      3+" XSTR(BPS) "*" #A "(%[temp20])   \n\t"          \
+  "addu    %[" #TEMP0 "],    %[temp16],      %[" #TEMP0 "]     \n\t"          \
+  "addu    %[" #TEMP4 "],    %[temp17],      %[" #TEMP4 "]     \n\t"          \
+  "addu    %[" #TEMP8 "],    %[temp18],      %[" #TEMP8 "]     \n\t"          \
+  "addu    %[" #TEMP12 "],   %[temp19],      %[" #TEMP12 "]    \n\t"          \
+  "slt     %[temp16],      %[" #TEMP0 "],    $zero             \n\t"          \
+  "slt     %[temp17],      %[" #TEMP4 "],    $zero             \n\t"          \
+  "slt     %[temp18],      %[" #TEMP8 "],    $zero             \n\t"          \
+  "slt     %[temp19],      %[" #TEMP12 "],   $zero             \n\t"          \
+  "movn    %[" #TEMP0 "],    $zero,          %[temp16]         \n\t"          \
+  "movn    %[" #TEMP4 "],    $zero,          %[temp17]         \n\t"          \
+  "movn    %[" #TEMP8 "],    $zero,          %[temp18]         \n\t"          \
+  "movn    %[" #TEMP12 "],   $zero,          %[temp19]         \n\t"          \
+  "addiu   %[temp20],      $zero,          255                 \n\t"          \
+  "slt     %[temp16],      %[" #TEMP0 "],    %[temp20]         \n\t"          \
+  "slt     %[temp17],      %[" #TEMP4 "],    %[temp20]         \n\t"          \
+  "slt     %[temp18],      %[" #TEMP8 "],    %[temp20]         \n\t"          \
+  "slt     %[temp19],      %[" #TEMP12 "],   %[temp20]         \n\t"          \
+  "movz    %[" #TEMP0 "],    %[temp20],      %[temp16]         \n\t"          \
+  "movz    %[" #TEMP4 "],    %[temp20],      %[temp17]         \n\t"          \
+  "lw      %[temp16],      8(%[args])                          \n\t"          \
+  "movz    %[" #TEMP8 "],    %[temp20],      %[temp18]         \n\t"          \
+  "movz    %[" #TEMP12 "],   %[temp20],      %[temp19]         \n\t"          \
+  "sb      %[" #TEMP0 "],    0+" XSTR(BPS) "*" #A "(%[temp16]) \n\t"          \
+  "sb      %[" #TEMP4 "],    1+" XSTR(BPS) "*" #A "(%[temp16]) \n\t"          \
+  "sb      %[" #TEMP8 "],    2+" XSTR(BPS) "*" #A "(%[temp16]) \n\t"          \
+  "sb      %[" #TEMP12 "],   3+" XSTR(BPS) "*" #A "(%[temp16]) \n\t"
 
 // Does one or two inverse transforms.
 static WEBP_INLINE void ITransformOne(const uint8_t* ref, const int16_t* in,
@@ -130,10 +127,10 @@ static WEBP_INLINE void ITransformOne(const uint8_t* ref, const int16_t* in,
     VERTICAL_PASS(4, 20, 12, 28, temp12, temp8,  temp9,  temp10, temp11)
     VERTICAL_PASS(6, 22, 14, 30, temp20, temp12, temp13, temp14, temp15)
 
-    HORIZONTAL_PASS( 0,  1,  2,  3, temp0, temp4, temp8,  temp12)
-    HORIZONTAL_PASS(16, 17, 18, 19, temp1, temp5, temp9,  temp13)
-    HORIZONTAL_PASS(32, 33, 34, 35, temp2, temp6, temp10, temp14)
-    HORIZONTAL_PASS(48, 49, 50, 51, temp3, temp7, temp11, temp15)
+    HORIZONTAL_PASS(0, temp0, temp4, temp8,  temp12)
+    HORIZONTAL_PASS(1, temp1, temp5, temp9,  temp13)
+    HORIZONTAL_PASS(2, temp2, temp6, temp10, temp14)
+    HORIZONTAL_PASS(3, temp3, temp7, temp11, temp15)
 
     : [temp0]"=&r"(temp0), [temp1]"=&r"(temp1), [temp2]"=&r"(temp2),
       [temp3]"=&r"(temp3), [temp4]"=&r"(temp4), [temp5]"=&r"(temp5),
@@ -241,46 +238,54 @@ static int QuantizeBlock(int16_t in[16], int16_t out[16],
   return 0;
 }
 
+static int Quantize2Blocks(int16_t in[32], int16_t out[32],
+                           const VP8Matrix* const mtx) {
+  int nz;
+  nz  = QuantizeBlock(in + 0 * 16, out + 0 * 16, mtx) << 0;
+  nz |= QuantizeBlock(in + 1 * 16, out + 1 * 16, mtx) << 1;
+  return nz;
+}
+
 #undef QUANTIZE_ONE
 
 // macro for one horizontal pass in Disto4x4 (TTransform)
 // two calls of function TTransform are merged into single one
-// A..D - offsets in bytes to load from a and b buffers
+// A - offset in bytes to load from a and b buffers
 // E..H - offsets in bytes to store first results to tmp buffer
 // E1..H1 - offsets in bytes to store second results to tmp buffer
-#define HORIZONTAL_PASS(A, B, C, D, E, F, G, H, E1, F1, G1, H1)   \
-  "lbu    %[temp0],  " #A "(%[a])            \n\t"                \
-  "lbu    %[temp1],  " #B "(%[a])            \n\t"                \
-  "lbu    %[temp2],  " #C "(%[a])            \n\t"                \
-  "lbu    %[temp3],  " #D "(%[a])            \n\t"                \
-  "lbu    %[temp4],  " #A "(%[b])            \n\t"                \
-  "lbu    %[temp5],  " #B "(%[b])            \n\t"                \
-  "lbu    %[temp6],  " #C "(%[b])            \n\t"                \
-  "lbu    %[temp7],  " #D "(%[b])            \n\t"                \
-  "addu   %[temp8],  %[temp0],    %[temp2]   \n\t"                \
-  "subu   %[temp0],  %[temp0],    %[temp2]   \n\t"                \
-  "addu   %[temp2],  %[temp1],    %[temp3]   \n\t"                \
-  "subu   %[temp1],  %[temp1],    %[temp3]   \n\t"                \
-  "addu   %[temp3],  %[temp4],    %[temp6]   \n\t"                \
-  "subu   %[temp4],  %[temp4],    %[temp6]   \n\t"                \
-  "addu   %[temp6],  %[temp5],    %[temp7]   \n\t"                \
-  "subu   %[temp5],  %[temp5],    %[temp7]   \n\t"                \
-  "addu   %[temp7],  %[temp8],    %[temp2]   \n\t"                \
-  "subu   %[temp2],  %[temp8],    %[temp2]   \n\t"                \
-  "addu   %[temp8],  %[temp0],    %[temp1]   \n\t"                \
-  "subu   %[temp0],  %[temp0],    %[temp1]   \n\t"                \
-  "addu   %[temp1],  %[temp3],    %[temp6]   \n\t"                \
-  "subu   %[temp3],  %[temp3],    %[temp6]   \n\t"                \
-  "addu   %[temp6],  %[temp4],    %[temp5]   \n\t"                \
-  "subu   %[temp4],  %[temp4],    %[temp5]   \n\t"                \
-  "sw     %[temp7],  " #E "(%[tmp])          \n\t"                \
-  "sw     %[temp2],  " #H "(%[tmp])          \n\t"                \
-  "sw     %[temp8],  " #F "(%[tmp])          \n\t"                \
-  "sw     %[temp0],  " #G "(%[tmp])          \n\t"                \
-  "sw     %[temp1],  " #E1 "(%[tmp])         \n\t"                \
-  "sw     %[temp3],  " #H1 "(%[tmp])         \n\t"                \
-  "sw     %[temp6],  " #F1 "(%[tmp])         \n\t"                \
-  "sw     %[temp4],  " #G1 "(%[tmp])         \n\t"
+#define HORIZONTAL_PASS(A, E, F, G, H, E1, F1, G1, H1)                  \
+  "lbu    %[temp0],  0+" XSTR(BPS) "*" #A "(%[a])  \n\t"                \
+  "lbu    %[temp1],  1+" XSTR(BPS) "*" #A "(%[a])  \n\t"                \
+  "lbu    %[temp2],  2+" XSTR(BPS) "*" #A "(%[a])  \n\t"                \
+  "lbu    %[temp3],  3+" XSTR(BPS) "*" #A "(%[a])  \n\t"                \
+  "lbu    %[temp4],  0+" XSTR(BPS) "*" #A "(%[b])  \n\t"                \
+  "lbu    %[temp5],  1+" XSTR(BPS) "*" #A "(%[b])  \n\t"                \
+  "lbu    %[temp6],  2+" XSTR(BPS) "*" #A "(%[b])  \n\t"                \
+  "lbu    %[temp7],  3+" XSTR(BPS) "*" #A "(%[b])  \n\t"                \
+  "addu   %[temp8],  %[temp0],    %[temp2]         \n\t"                \
+  "subu   %[temp0],  %[temp0],    %[temp2]         \n\t"                \
+  "addu   %[temp2],  %[temp1],    %[temp3]         \n\t"                \
+  "subu   %[temp1],  %[temp1],    %[temp3]         \n\t"                \
+  "addu   %[temp3],  %[temp4],    %[temp6]         \n\t"                \
+  "subu   %[temp4],  %[temp4],    %[temp6]         \n\t"                \
+  "addu   %[temp6],  %[temp5],    %[temp7]         \n\t"                \
+  "subu   %[temp5],  %[temp5],    %[temp7]         \n\t"                \
+  "addu   %[temp7],  %[temp8],    %[temp2]         \n\t"                \
+  "subu   %[temp2],  %[temp8],    %[temp2]         \n\t"                \
+  "addu   %[temp8],  %[temp0],    %[temp1]         \n\t"                \
+  "subu   %[temp0],  %[temp0],    %[temp1]         \n\t"                \
+  "addu   %[temp1],  %[temp3],    %[temp6]         \n\t"                \
+  "subu   %[temp3],  %[temp3],    %[temp6]         \n\t"                \
+  "addu   %[temp6],  %[temp4],    %[temp5]         \n\t"                \
+  "subu   %[temp4],  %[temp4],    %[temp5]         \n\t"                \
+  "sw     %[temp7],  " #E "(%[tmp])                \n\t"                \
+  "sw     %[temp2],  " #H "(%[tmp])                \n\t"                \
+  "sw     %[temp8],  " #F "(%[tmp])                \n\t"                \
+  "sw     %[temp0],  " #G "(%[tmp])                \n\t"                \
+  "sw     %[temp1],  " #E1 "(%[tmp])               \n\t"                \
+  "sw     %[temp3],  " #H1 "(%[tmp])               \n\t"                \
+  "sw     %[temp6],  " #F1 "(%[tmp])               \n\t"                \
+  "sw     %[temp4],  " #G1 "(%[tmp])               \n\t"
 
 // macro for one vertical pass in Disto4x4 (TTransform)
 // two calls of function TTransform are merged into single one
@@ -362,10 +367,10 @@ static int Disto4x4(const uint8_t* const a, const uint8_t* const b,
   int temp0, temp1, temp2, temp3, temp4, temp5, temp6, temp7, temp8;
 
   __asm__ volatile(
-    HORIZONTAL_PASS( 0,  1,  2,  3,    0,  4,  8, 12,    64,  68,  72,  76)
-    HORIZONTAL_PASS(16, 17, 18, 19,   16, 20, 24, 28,    80,  84,  88,  92)
-    HORIZONTAL_PASS(32, 33, 34, 35,   32, 36, 40, 44,    96, 100, 104, 108)
-    HORIZONTAL_PASS(48, 49, 50, 51,   48, 52, 56, 60,   112, 116, 120, 124)
+    HORIZONTAL_PASS(0,   0,  4,  8, 12,    64,  68,  72,  76)
+    HORIZONTAL_PASS(1,  16, 20, 24, 28,    80,  84,  88,  92)
+    HORIZONTAL_PASS(2,  32, 36, 40, 44,    96, 100, 104, 108)
+    HORIZONTAL_PASS(3,  48, 52, 56, 60,   112, 116, 120, 124)
     "mthi   $zero                             \n\t"
     "mtlo   $zero                             \n\t"
     VERTICAL_PASS( 0, 16, 32, 48,     64, 80,  96, 112,   0,  8, 16, 24)
@@ -405,41 +410,41 @@ static int Disto16x16(const uint8_t* const a, const uint8_t* const b,
 
 // macro for one horizontal pass in FTransform
 // temp0..temp15 holds tmp[0]..tmp[15]
-// A..D - offsets in bytes to load from src and ref buffers
+// A - offset in bytes to load from src and ref buffers
 // TEMP0..TEMP3 - registers for corresponding tmp elements
-#define HORIZONTAL_PASS(A, B, C, D, TEMP0, TEMP1, TEMP2, TEMP3)   \
-  "lw     %[" #TEMP1 "],  0(%[args])                     \n\t"    \
-  "lw     %[" #TEMP2 "],  4(%[args])                     \n\t"    \
-  "lbu    %[temp16],    " #A "(%[" #TEMP1 "])            \n\t"    \
-  "lbu    %[temp17],    " #A "(%[" #TEMP2 "])            \n\t"    \
-  "lbu    %[temp18],    " #B "(%[" #TEMP1 "])            \n\t"    \
-  "lbu    %[temp19],    " #B "(%[" #TEMP2 "])            \n\t"    \
-  "subu   %[temp20],    %[temp16],    %[temp17]          \n\t"    \
-  "lbu    %[temp16],    " #C "(%[" #TEMP1 "])            \n\t"    \
-  "lbu    %[temp17],    " #C "(%[" #TEMP2 "])            \n\t"    \
-  "subu   %[" #TEMP0 "],  %[temp18],    %[temp19]        \n\t"    \
-  "lbu    %[temp18],    " #D "(%[" #TEMP1 "])            \n\t"    \
-  "lbu    %[temp19],    " #D "(%[" #TEMP2 "])            \n\t"    \
-  "subu   %[" #TEMP1 "],  %[temp16],    %[temp17]        \n\t"    \
-  "subu   %[" #TEMP2 "],  %[temp18],    %[temp19]        \n\t"    \
-  "addu   %[" #TEMP3 "],  %[temp20],    %[" #TEMP2 "]    \n\t"    \
-  "subu   %[" #TEMP2 "],  %[temp20],    %[" #TEMP2 "]    \n\t"    \
-  "addu   %[temp20],    %[" #TEMP0 "],  %[" #TEMP1 "]    \n\t"    \
-  "subu   %[" #TEMP0 "],  %[" #TEMP0 "],  %[" #TEMP1 "]  \n\t"    \
-  "mul    %[temp16],    %[" #TEMP2 "],  %[c5352]         \n\t"    \
-  "mul    %[temp17],    %[" #TEMP2 "],  %[c2217]         \n\t"    \
-  "mul    %[temp18],    %[" #TEMP0 "],  %[c5352]         \n\t"    \
-  "mul    %[temp19],    %[" #TEMP0 "],  %[c2217]         \n\t"    \
-  "addu   %[" #TEMP1 "],  %[" #TEMP3 "],  %[temp20]      \n\t"    \
-  "subu   %[temp20],    %[" #TEMP3 "],  %[temp20]        \n\t"    \
-  "sll    %[" #TEMP0 "],  %[" #TEMP1 "],  3              \n\t"    \
-  "sll    %[" #TEMP2 "],  %[temp20],    3                \n\t"    \
-  "addiu  %[temp16],    %[temp16],    1812               \n\t"    \
-  "addiu  %[temp17],    %[temp17],    937                \n\t"    \
-  "addu   %[temp16],    %[temp16],    %[temp19]          \n\t"    \
-  "subu   %[temp17],    %[temp17],    %[temp18]          \n\t"    \
-  "sra    %[" #TEMP1 "],  %[temp16],    9                \n\t"    \
-  "sra    %[" #TEMP3 "],  %[temp17],    9                \n\t"
+#define HORIZONTAL_PASS(A, TEMP0, TEMP1, TEMP2, TEMP3)                  \
+  "lw     %[" #TEMP1 "],  0(%[args])                           \n\t"    \
+  "lw     %[" #TEMP2 "],  4(%[args])                           \n\t"    \
+  "lbu    %[temp16],    0+" XSTR(BPS) "*" #A "(%[" #TEMP1 "])  \n\t"    \
+  "lbu    %[temp17],    0+" XSTR(BPS) "*" #A "(%[" #TEMP2 "])  \n\t"    \
+  "lbu    %[temp18],    1+" XSTR(BPS) "*" #A "(%[" #TEMP1 "])  \n\t"    \
+  "lbu    %[temp19],    1+" XSTR(BPS) "*" #A "(%[" #TEMP2 "])  \n\t"    \
+  "subu   %[temp20],    %[temp16],    %[temp17]                \n\t"    \
+  "lbu    %[temp16],    2+" XSTR(BPS) "*" #A "(%[" #TEMP1 "])  \n\t"    \
+  "lbu    %[temp17],    2+" XSTR(BPS) "*" #A "(%[" #TEMP2 "])  \n\t"    \
+  "subu   %[" #TEMP0 "],  %[temp18],    %[temp19]              \n\t"    \
+  "lbu    %[temp18],    3+" XSTR(BPS) "*" #A "(%[" #TEMP1 "])  \n\t"    \
+  "lbu    %[temp19],    3+" XSTR(BPS) "*" #A "(%[" #TEMP2 "])  \n\t"    \
+  "subu   %[" #TEMP1 "],  %[temp16],    %[temp17]              \n\t"    \
+  "subu   %[" #TEMP2 "],  %[temp18],    %[temp19]              \n\t"    \
+  "addu   %[" #TEMP3 "],  %[temp20],    %[" #TEMP2 "]          \n\t"    \
+  "subu   %[" #TEMP2 "],  %[temp20],    %[" #TEMP2 "]          \n\t"    \
+  "addu   %[temp20],    %[" #TEMP0 "],  %[" #TEMP1 "]          \n\t"    \
+  "subu   %[" #TEMP0 "],  %[" #TEMP0 "],  %[" #TEMP1 "]        \n\t"    \
+  "mul    %[temp16],    %[" #TEMP2 "],  %[c5352]               \n\t"    \
+  "mul    %[temp17],    %[" #TEMP2 "],  %[c2217]               \n\t"    \
+  "mul    %[temp18],    %[" #TEMP0 "],  %[c5352]               \n\t"    \
+  "mul    %[temp19],    %[" #TEMP0 "],  %[c2217]               \n\t"    \
+  "addu   %[" #TEMP1 "],  %[" #TEMP3 "],  %[temp20]            \n\t"    \
+  "subu   %[temp20],    %[" #TEMP3 "],  %[temp20]              \n\t"    \
+  "sll    %[" #TEMP0 "],  %[" #TEMP1 "],  3                    \n\t"    \
+  "sll    %[" #TEMP2 "],  %[temp20],    3                      \n\t"    \
+  "addiu  %[temp16],    %[temp16],    1812                     \n\t"    \
+  "addiu  %[temp17],    %[temp17],    937                      \n\t"    \
+  "addu   %[temp16],    %[temp16],    %[temp19]                \n\t"    \
+  "subu   %[temp17],    %[temp17],    %[temp18]                \n\t"    \
+  "sra    %[" #TEMP1 "],  %[temp16],    9                      \n\t"    \
+  "sra    %[" #TEMP3 "],  %[temp17],    9                      \n\t"
 
 // macro for one vertical pass in FTransform
 // temp0..temp15 holds tmp[0]..tmp[15]
@@ -483,10 +488,10 @@ static void FTransform(const uint8_t* src, const uint8_t* ref, int16_t* out) {
       { (const int*)src, (const int*)ref, (const int*)out };
 
   __asm__ volatile(
-    HORIZONTAL_PASS( 0,  1,  2,  3, temp0,  temp1,  temp2,  temp3)
-    HORIZONTAL_PASS(16, 17, 18, 19, temp4,  temp5,  temp6,  temp7)
-    HORIZONTAL_PASS(32, 33, 34, 35, temp8,  temp9,  temp10, temp11)
-    HORIZONTAL_PASS(48, 49, 50, 51, temp12, temp13, temp14, temp15)
+    HORIZONTAL_PASS(0, temp0,  temp1,  temp2,  temp3)
+    HORIZONTAL_PASS(1, temp4,  temp5,  temp6,  temp7)
+    HORIZONTAL_PASS(2, temp8,  temp9,  temp10, temp11)
+    HORIZONTAL_PASS(3, temp12, temp13, temp14, temp15)
     "lw   %[temp20],    8(%[args])                     \n\t"
     VERTICAL_PASS(0,  8, 16, 24, temp0, temp4, temp8,  temp12)
     VERTICAL_PASS(2, 10, 18, 26, temp1, temp5, temp9,  temp13)
@@ -508,118 +513,7 @@ static void FTransform(const uint8_t* src, const uint8_t* ref, int16_t* out) {
 #undef VERTICAL_PASS
 #undef HORIZONTAL_PASS
 
-// Forward declaration.
-extern int VP8GetResidualCostMIPS32(int ctx0, const VP8Residual* const res);
-
-int VP8GetResidualCostMIPS32(int ctx0, const VP8Residual* const res) {
-  int n = res->first;
-  // should be prob[VP8EncBands[n]], but it's equivalent for n=0 or 1
-  int p0 = res->prob[n][ctx0][0];
-  const uint16_t* t = res->cost[n][ctx0];
-  int cost;
-  const int const_2 = 2;
-  const int const_255 = 255;
-  const int const_max_level = MAX_VARIABLE_LEVEL;
-  int res_cost;
-  int res_prob;
-  int res_coeffs;
-  int res_last;
-  int v_reg;
-  int b_reg;
-  int ctx_reg;
-  int cost_add, temp_1, temp_2, temp_3;
-
-  if (res->last < 0) {
-    return VP8BitCost(0, p0);
-  }
-
-  cost = (ctx0 == 0) ? VP8BitCost(1, p0) : 0;
-
-  res_cost = (int)res->cost;
-  res_prob = (int)res->prob;
-  res_coeffs = (int)res->coeffs;
-  res_last = (int)res->last;
-
-  __asm__ volatile(
-    ".set   push                                                           \n\t"
-    ".set   noreorder                                                      \n\t"
-
-    "sll    %[temp_1],     %[n],              1                            \n\t"
-    "addu   %[res_coeffs], %[res_coeffs],     %[temp_1]                    \n\t"
-    "slt    %[temp_2],     %[n],              %[res_last]                  \n\t"
-    "bnez   %[temp_2],     1f                                              \n\t"
-    " li    %[cost_add],   0                                               \n\t"
-    "b      2f                                                             \n\t"
-    " nop                                                                  \n\t"
-  "1:                                                                      \n\t"
-    "lh     %[v_reg],      0(%[res_coeffs])                                \n\t"
-    "addu   %[b_reg],      %[n],              %[VP8EncBands]               \n\t"
-    "move   %[temp_1],     %[const_max_level]                              \n\t"
-    "addu   %[cost],       %[cost],           %[cost_add]                  \n\t"
-    "negu   %[temp_2],     %[v_reg]                                        \n\t"
-    "slti   %[temp_3],     %[v_reg],          0                            \n\t"
-    "movn   %[v_reg],      %[temp_2],         %[temp_3]                    \n\t"
-    "lbu    %[b_reg],      1(%[b_reg])                                     \n\t"
-    "li     %[cost_add],   0                                               \n\t"
-
-    "sltiu  %[temp_3],     %[v_reg],          2                            \n\t"
-    "move   %[ctx_reg],    %[v_reg]                                        \n\t"
-    "movz   %[ctx_reg],    %[const_2],        %[temp_3]                    \n\t"
-    //  cost += VP8LevelCost(t, v);
-    "slt    %[temp_3],     %[v_reg],          %[const_max_level]           \n\t"
-    "movn   %[temp_1],     %[v_reg],          %[temp_3]                    \n\t"
-    "sll    %[temp_2],     %[v_reg],          1                            \n\t"
-    "addu   %[temp_2],     %[temp_2],         %[VP8LevelFixedCosts]        \n\t"
-    "lhu    %[temp_2],     0(%[temp_2])                                    \n\t"
-    "sll    %[temp_1],     %[temp_1],         1                            \n\t"
-    "addu   %[temp_1],     %[temp_1],         %[t]                         \n\t"
-    "lhu    %[temp_3],     0(%[temp_1])                                    \n\t"
-    "addu   %[cost],       %[cost],           %[temp_2]                    \n\t"
-
-    //  t = res->cost[b][ctx];
-    "sll    %[temp_1],     %[ctx_reg],        7                            \n\t"
-    "sll    %[temp_2],     %[ctx_reg],        3                            \n\t"
-    "addu   %[cost],       %[cost],           %[temp_3]                    \n\t"
-    "addu   %[temp_1],     %[temp_1],         %[temp_2]                    \n\t"
-    "sll    %[temp_2],     %[b_reg],          3                            \n\t"
-    "sll    %[temp_3],     %[b_reg],          5                            \n\t"
-    "sub    %[temp_2],     %[temp_3],         %[temp_2]                    \n\t"
-    "sll    %[temp_3],     %[temp_2],         4                            \n\t"
-    "addu   %[temp_1],     %[temp_1],         %[temp_3]                    \n\t"
-    "addu   %[temp_2],     %[temp_2],         %[res_cost]                  \n\t"
-    "addiu  %[n],          %[n],              1                            \n\t"
-    "addu   %[t],          %[temp_1],         %[temp_2]                    \n\t"
-    "slt    %[temp_1],     %[n],              %[res_last]                  \n\t"
-    "bnez   %[temp_1],     1b                                              \n\t"
-    " addiu %[res_coeffs], %[res_coeffs],     2                            \n\t"
-   "2:                                                                     \n\t"
-
-    ".set   pop                                                            \n\t"
-    : [cost]"+r"(cost), [t]"+r"(t), [n]"+r"(n), [v_reg]"=&r"(v_reg),
-      [ctx_reg]"=&r"(ctx_reg), [b_reg]"=&r"(b_reg), [cost_add]"=&r"(cost_add),
-      [temp_1]"=&r"(temp_1), [temp_2]"=&r"(temp_2), [temp_3]"=&r"(temp_3)
-    : [const_2]"r"(const_2), [const_255]"r"(const_255), [res_last]"r"(res_last),
-      [VP8EntropyCost]"r"(VP8EntropyCost), [VP8EncBands]"r"(VP8EncBands),
-      [const_max_level]"r"(const_max_level), [res_prob]"r"(res_prob),
-      [VP8LevelFixedCosts]"r"(VP8LevelFixedCosts), [res_coeffs]"r"(res_coeffs),
-      [res_cost]"r"(res_cost)
-    : "memory"
-  );
-
-  // Last coefficient is always non-zero
-  {
-    const int v = abs(res->coeffs[n]);
-    assert(v != 0);
-    cost += VP8LevelCost(t, v);
-    if (n < 15) {
-      const int b = VP8EncBands[n + 1];
-      const int ctx = (v == 1) ? 1 : 2;
-      const int last_p0 = res->prob[b][ctx][0];
-      cost += VP8BitCost(0, last_p0);
-    }
-  }
-  return cost;
-}
+#if !defined(WORK_AROUND_GCC)
 
 #define GET_SSE_INNER(A, B, C, D)                               \
   "lbu     %[temp0],    " #A "(%[a])                 \n\t"      \
@@ -645,7 +539,6 @@ int VP8GetResidualCostMIPS32(int ctx0, const VP8Residual* const res) {
   GET_SSE_INNER(C, C + 1, C + 2, C + 3)   \
   GET_SSE_INNER(D, D + 1, D + 2, D + 3)
 
-#if !defined(WORK_AROUND_GCC)
 static int SSE16x16(const uint8_t* a, const uint8_t* b) {
   int count;
   int temp0, temp1, temp2, temp3, temp4, temp5, temp6, temp7;
@@ -653,29 +546,29 @@ static int SSE16x16(const uint8_t* a, const uint8_t* b) {
   __asm__ volatile(
      "mult   $zero,    $zero                            \n\t"
 
-     GET_SSE(  0,   4,   8,  12)
-     GET_SSE( 16,  20,  24,  28)
-     GET_SSE( 32,  36,  40,  44)
-     GET_SSE( 48,  52,  56,  60)
-     GET_SSE( 64,  68,  72,  76)
-     GET_SSE( 80,  84,  88,  92)
-     GET_SSE( 96, 100, 104, 108)
-     GET_SSE(112, 116, 120, 124)
-     GET_SSE(128, 132, 136, 140)
-     GET_SSE(144, 148, 152, 156)
-     GET_SSE(160, 164, 168, 172)
-     GET_SSE(176, 180, 184, 188)
-     GET_SSE(192, 196, 200, 204)
-     GET_SSE(208, 212, 216, 220)
-     GET_SSE(224, 228, 232, 236)
-     GET_SSE(240, 244, 248, 252)
+     GET_SSE( 0 * BPS, 4 +  0 * BPS, 8 +  0 * BPS, 12 +  0 * BPS)
+     GET_SSE( 1 * BPS, 4 +  1 * BPS, 8 +  1 * BPS, 12 +  1 * BPS)
+     GET_SSE( 2 * BPS, 4 +  2 * BPS, 8 +  2 * BPS, 12 +  2 * BPS)
+     GET_SSE( 3 * BPS, 4 +  3 * BPS, 8 +  3 * BPS, 12 +  3 * BPS)
+     GET_SSE( 4 * BPS, 4 +  4 * BPS, 8 +  4 * BPS, 12 +  4 * BPS)
+     GET_SSE( 5 * BPS, 4 +  5 * BPS, 8 +  5 * BPS, 12 +  5 * BPS)
+     GET_SSE( 6 * BPS, 4 +  6 * BPS, 8 +  6 * BPS, 12 +  6 * BPS)
+     GET_SSE( 7 * BPS, 4 +  7 * BPS, 8 +  7 * BPS, 12 +  7 * BPS)
+     GET_SSE( 8 * BPS, 4 +  8 * BPS, 8 +  8 * BPS, 12 +  8 * BPS)
+     GET_SSE( 9 * BPS, 4 +  9 * BPS, 8 +  9 * BPS, 12 +  9 * BPS)
+     GET_SSE(10 * BPS, 4 + 10 * BPS, 8 + 10 * BPS, 12 + 10 * BPS)
+     GET_SSE(11 * BPS, 4 + 11 * BPS, 8 + 11 * BPS, 12 + 11 * BPS)
+     GET_SSE(12 * BPS, 4 + 12 * BPS, 8 + 12 * BPS, 12 + 12 * BPS)
+     GET_SSE(13 * BPS, 4 + 13 * BPS, 8 + 13 * BPS, 12 + 13 * BPS)
+     GET_SSE(14 * BPS, 4 + 14 * BPS, 8 + 14 * BPS, 12 + 14 * BPS)
+     GET_SSE(15 * BPS, 4 + 15 * BPS, 8 + 15 * BPS, 12 + 15 * BPS)
 
     "mflo    %[count]                                   \n\t"
     : [temp0]"=&r"(temp0), [temp1]"=&r"(temp1), [temp2]"=&r"(temp2),
       [temp3]"=&r"(temp3), [temp4]"=&r"(temp4), [temp5]"=&r"(temp5),
       [temp6]"=&r"(temp6), [temp7]"=&r"(temp7), [count]"=&r"(count)
     : [a]"r"(a), [b]"r"(b)
-    : "memory", "hi" , "lo"
+    : "memory", "hi", "lo"
   );
   return count;
 }
@@ -687,21 +580,21 @@ static int SSE16x8(const uint8_t* a, const uint8_t* b) {
   __asm__ volatile(
      "mult   $zero,    $zero                            \n\t"
 
-     GET_SSE(  0,   4,   8,  12)
-     GET_SSE( 16,  20,  24,  28)
-     GET_SSE( 32,  36,  40,  44)
-     GET_SSE( 48,  52,  56,  60)
-     GET_SSE( 64,  68,  72,  76)
-     GET_SSE( 80,  84,  88,  92)
-     GET_SSE( 96, 100, 104, 108)
-     GET_SSE(112, 116, 120, 124)
+     GET_SSE( 0 * BPS, 4 +  0 * BPS, 8 +  0 * BPS, 12 +  0 * BPS)
+     GET_SSE( 1 * BPS, 4 +  1 * BPS, 8 +  1 * BPS, 12 +  1 * BPS)
+     GET_SSE( 2 * BPS, 4 +  2 * BPS, 8 +  2 * BPS, 12 +  2 * BPS)
+     GET_SSE( 3 * BPS, 4 +  3 * BPS, 8 +  3 * BPS, 12 +  3 * BPS)
+     GET_SSE( 4 * BPS, 4 +  4 * BPS, 8 +  4 * BPS, 12 +  4 * BPS)
+     GET_SSE( 5 * BPS, 4 +  5 * BPS, 8 +  5 * BPS, 12 +  5 * BPS)
+     GET_SSE( 6 * BPS, 4 +  6 * BPS, 8 +  6 * BPS, 12 +  6 * BPS)
+     GET_SSE( 7 * BPS, 4 +  7 * BPS, 8 +  7 * BPS, 12 +  7 * BPS)
 
     "mflo    %[count]                                   \n\t"
     : [temp0]"=&r"(temp0), [temp1]"=&r"(temp1), [temp2]"=&r"(temp2),
       [temp3]"=&r"(temp3), [temp4]"=&r"(temp4), [temp5]"=&r"(temp5),
       [temp6]"=&r"(temp6), [temp7]"=&r"(temp7), [count]"=&r"(count)
     : [a]"r"(a), [b]"r"(b)
-    : "memory", "hi" , "lo"
+    : "memory", "hi", "lo"
   );
   return count;
 }
@@ -713,17 +606,17 @@ static int SSE8x8(const uint8_t* a, const uint8_t* b) {
   __asm__ volatile(
      "mult   $zero,    $zero                            \n\t"
 
-     GET_SSE( 0,   4,  16,  20)
-     GET_SSE(32,  36,  48,  52)
-     GET_SSE(64,  68,  80,  84)
-     GET_SSE(96, 100, 112, 116)
+     GET_SSE(0 * BPS, 4 + 0 * BPS, 1 * BPS, 4 + 1 * BPS)
+     GET_SSE(2 * BPS, 4 + 2 * BPS, 3 * BPS, 4 + 3 * BPS)
+     GET_SSE(4 * BPS, 4 + 4 * BPS, 5 * BPS, 4 + 5 * BPS)
+     GET_SSE(6 * BPS, 4 + 6 * BPS, 7 * BPS, 4 + 7 * BPS)
 
     "mflo    %[count]                                   \n\t"
     : [temp0]"=&r"(temp0), [temp1]"=&r"(temp1), [temp2]"=&r"(temp2),
       [temp3]"=&r"(temp3), [temp4]"=&r"(temp4), [temp5]"=&r"(temp5),
       [temp6]"=&r"(temp6), [temp7]"=&r"(temp7), [count]"=&r"(count)
     : [a]"r"(a), [b]"r"(b)
-    : "memory", "hi" , "lo"
+    : "memory", "hi", "lo"
   );
   return count;
 }
@@ -735,42 +628,45 @@ static int SSE4x4(const uint8_t* a, const uint8_t* b) {
   __asm__ volatile(
      "mult   $zero,    $zero                            \n\t"
 
-     GET_SSE(0, 16, 32, 48)
+     GET_SSE(0 * BPS, 1 * BPS, 2 * BPS, 3 * BPS)
 
     "mflo    %[count]                                   \n\t"
     : [temp0]"=&r"(temp0), [temp1]"=&r"(temp1), [temp2]"=&r"(temp2),
       [temp3]"=&r"(temp3), [temp4]"=&r"(temp4), [temp5]"=&r"(temp5),
       [temp6]"=&r"(temp6), [temp7]"=&r"(temp7), [count]"=&r"(count)
     : [a]"r"(a), [b]"r"(b)
-    : "memory", "hi" , "lo"
+    : "memory", "hi", "lo"
   );
   return count;
 }
 
-#endif  // WORK_AROUND_GCC
+#undef GET_SSE
+#undef GET_SSE_INNER
 
-#undef GET_SSE_MIPS32
-#undef GET_SSE_MIPS32_INNER
-
-#endif  // WEBP_USE_MIPS32
+#endif  // !WORK_AROUND_GCC
 
 //------------------------------------------------------------------------------
 // Entry point
 
 extern void VP8EncDspInitMIPS32(void);
 
-void VP8EncDspInitMIPS32(void) {
-#if defined(WEBP_USE_MIPS32)
+WEBP_TSAN_IGNORE_FUNCTION void VP8EncDspInitMIPS32(void) {
   VP8ITransform = ITransform;
+  VP8FTransform = FTransform;
   VP8EncQuantizeBlock = QuantizeBlock;
+  VP8EncQuantize2Blocks = Quantize2Blocks;
   VP8TDisto4x4 = Disto4x4;
   VP8TDisto16x16 = Disto16x16;
-  VP8FTransform = FTransform;
 #if !defined(WORK_AROUND_GCC)
   VP8SSE16x16 = SSE16x16;
   VP8SSE8x8 = SSE8x8;
   VP8SSE16x8 = SSE16x8;
   VP8SSE4x4 = SSE4x4;
 #endif
-#endif  // WEBP_USE_MIPS32
 }
+
+#else  // !WEBP_USE_MIPS32
+
+WEBP_DSP_INIT_STUB(VP8EncDspInitMIPS32)
+
+#endif  // WEBP_USE_MIPS32