Swap width and height args to add/sub blend functions for consistency

5 files changed, 58 insertions, 58 deletions

diff --git a/libass/ass_bitmap.c b/libass/ass_bitmap.c
index 78894c9..73620dd 100644
--- a/libass/ass_bitmap.c
+++ b/libass/ass_bitmap.c
@@ -579,7 +579,7 @@ int outline_to_bitmap2(ASS_Renderer *render_priv,
  */
 void ass_add_bitmaps_c(uint8_t *dst, intptr_t dst_stride,
                        uint8_t *src, intptr_t src_stride,
-                       intptr_t height, intptr_t width)
+                       intptr_t width, intptr_t height)
 {
     unsigned out;
     uint8_t* end = dst + dst_stride * height;
@@ -595,7 +595,7 @@ void ass_add_bitmaps_c(uint8_t *dst, intptr_t dst_stride,
 
 void ass_sub_bitmaps_c(uint8_t *dst, intptr_t dst_stride,
                        uint8_t *src, intptr_t src_stride,
-                       intptr_t height, intptr_t width)
+                       intptr_t width, intptr_t height)
 {
     short out;
     uint8_t* end = dst + dst_stride * height;
diff --git a/libass/ass_bitmap.h b/libass/ass_bitmap.h
index 71a3f5e..e4c1451 100644
--- a/libass/ass_bitmap.h
+++ b/libass/ass_bitmap.h
@@ -36,7 +36,7 @@ typedef void (*FillGenericTileFunc)(uint8_t *buf, ptrdiff_t stride,
 
 typedef void (*BitmapBlendFunc)(uint8_t *dst, intptr_t dst_stride,
                                 uint8_t *src, intptr_t src_stride,
-                                intptr_t height, intptr_t width);
+                                intptr_t width, intptr_t height);
 typedef void (*BitmapMulFunc)(uint8_t *dst, intptr_t dst_stride,
                               uint8_t *src1, intptr_t src1_stride,
                               uint8_t *src2, intptr_t src2_stride,
diff --git a/libass/ass_func_template.h b/libass/ass_func_template.h
index 6c72021..0eccdfb 100644
--- a/libass/ass_func_template.h
+++ b/libass/ass_func_template.h
@@ -33,10 +33,10 @@ void DECORATE(fill_generic_tile32)(uint8_t *buf, ptrdiff_t stride,
 
 void DECORATE(add_bitmaps)(uint8_t *dst, intptr_t dst_stride,
                            uint8_t *src, intptr_t src_stride,
-                           intptr_t height, intptr_t width);
+                           intptr_t width, intptr_t height);
 void DECORATE(sub_bitmaps)(uint8_t *dst, intptr_t dst_stride,
                            uint8_t *src, intptr_t src_stride,
-                           intptr_t height, intptr_t width);
+                           intptr_t width, intptr_t height);
 void DECORATE(mul_bitmaps)(uint8_t *dst, intptr_t dst_stride,
                            uint8_t *src1, intptr_t src1_stride,
                            uint8_t *src2, intptr_t src2_stride,
diff --git a/libass/ass_render.c b/libass/ass_render.c
index f124114..77c7e30 100644
--- a/libass/ass_render.c
+++ b/libass/ass_render.c
@@ -567,7 +567,7 @@ static void blend_vector_clip(ASS_Renderer *render_priv,
             memcpy(nbuffer, abuffer, ((ah - 1) * as) + aw);
             render_priv->engine->sub_bitmaps(nbuffer + atop * as + aleft, as,
                                              bbuffer + btop * bs + bleft, bs,
-                                             h, w);
+                                             w, h);
         } else {
             // Regular clip
             if (ax + aw < bx || ay + ah < by || ax > bx + bw ||
@@ -2319,7 +2319,7 @@ static void render_and_combine_glyphs(ASS_Renderer *render_priv,
                     unsigned char *buf = dst->buffer + y * dst->stride + x;
                     render_priv->engine->add_bitmaps(buf, dst->stride,
                                                      src->buffer, src->stride,
-                                                     src->h, src->w);
+                                                     src->w, src->h);
                 }
             }
         }
@@ -2353,7 +2353,7 @@ static void render_and_combine_glyphs(ASS_Renderer *render_priv,
                     unsigned char *buf = dst->buffer + y * dst->stride + x;
                     render_priv->engine->add_bitmaps(buf, dst->stride,
                                                      src->buffer, src->stride,
-                                                     src->h, src->w);
+                                                     src->w, src->h);
                 }
             }
         }
diff --git a/libass/x86/blend_bitmaps.asm b/libass/x86/blend_bitmaps.asm
index 3a9b2dd..118feea 100644
--- a/libass/x86/blend_bitmaps.asm
+++ b/libass/x86/blend_bitmaps.asm
@@ -29,16 +29,16 @@ SECTION .text
 ;------------------------------------------------------------------------------
 ; void add_bitmaps( uint8_t *dst, intptr_t dst_stride,
 ;                   uint8_t *src, intptr_t src_stride,
-;                   intptr_t height, intptr_t width );
+;                   intptr_t width, intptr_t height );
 ;------------------------------------------------------------------------------
 
 INIT_XMM
 cglobal add_bitmaps_x86, 6,7
 .skip_prologue:
-    imul r4, r3
-    add r4, r2
-    PUSH r4
-    mov r4, r3
+    imul r5, r3
+    add r5, r2
+    PUSH r5
+    mov r5, r3
 .height_loop:
     xor r6, r6 ; x offset
 .stride_loop:
@@ -49,10 +49,10 @@ cglobal add_bitmaps_x86, 6,7
 .continue:
     mov byte [r0 + r6], r3b
     inc r6
-    cmp r6, r5
+    cmp r6, r4
     jl .stride_loop ; still in scan line
     add r0, r1
-    add r2, r4
+    add r2, r5
     cmp r2, [rsp]
     jl .height_loop
     ADD rsp, gprsize
@@ -61,7 +61,7 @@ cglobal add_bitmaps_x86, 6,7
 %macro ADD_BITMAPS 0
     cglobal add_bitmaps, 6,7
     .skip_prologue:
-        cmp r5, mmsize
+        cmp r4, mmsize
         %if mmsize == 16
             jl add_bitmaps_x86.skip_prologue
         %else
@@ -70,20 +70,20 @@ cglobal add_bitmaps_x86, 6,7
         %if mmsize == 32
             vzeroupper
         %endif
-        imul r4, r3
-        add r4, r2 ; last address
+        imul r5, r3
+        add r5, r2 ; last address
     .height_loop:
         xor r6, r6 ; x offset
     .stride_loop:
-        movu m0, [r0 + r6]
-        paddusb m0, [r2 + r6]
-        movu [r0 + r6], m0
+        movu m1, [r0 + r6]
+        paddusb m1, [r2 + r6]
+        movu [r0 + r6], m1
         add r6, mmsize
-        cmp r6, r5
+        cmp r6, r4
         jl .stride_loop ; still in scan line
         add r0, r1
         add r2, r3
-        cmp r2, r4
+        cmp r2, r5
         jl .height_loop
         RET
 %endmacro
@@ -96,16 +96,16 @@ ADD_BITMAPS
 ;------------------------------------------------------------------------------
 ; void sub_bitmaps( uint8_t *dst, intptr_t dst_stride,
 ;                   uint8_t *src, intptr_t src_stride,
-;                   intptr_t height, intptr_t width );
+;                   intptr_t width, intptr_t height );
 ;------------------------------------------------------------------------------
 
 INIT_XMM
 cglobal sub_bitmaps_x86, 6,10
 .skip_prologue:
-    imul r4, r3
-    add r4, r2 ; last address
-    PUSH r4
-    mov r4, r3
+    imul r5, r3
+    add r5, r2 ; last address
+    PUSH r5
+    mov r5, r3
 .height_loop:
     xor r6, r6 ; x offset
 .stride_loop:
@@ -116,10 +116,10 @@ cglobal sub_bitmaps_x86, 6,10
 .continue:
     mov byte [r0 + r6], r3b
     inc r6
-    cmp r6, r5
+    cmp r6, r4
     jl .stride_loop ; still in scan line
     add r0, r1
-    add r2, r4
+    add r2, r5
     cmp r2, [rsp]
     jl .height_loop
     ADD rsp, gprsize
@@ -130,7 +130,7 @@ cglobal sub_bitmaps_x86, 6,10
 %macro SUB_BITMAPS 0
     cglobal sub_bitmaps, 6,10
     .skip_prologue:
-        cmp r5, mmsize
+        cmp r4, mmsize
         %if mmsize == 16
             jl sub_bitmaps_x86.skip_prologue
         %else
@@ -139,23 +139,23 @@ cglobal sub_bitmaps_x86, 6,10
         %if mmsize == 32
             vzeroupper
         %endif
-        imul r4, r3
-        add r4, r2 ; last address
-        mov r7, r5
+        imul r5, r3
+        add r5, r2 ; last address
+        mov r7, r4
         and r7, -mmsize ; &= (16);
         xor r9, r9
     .height_loop:
         xor r6, r6 ; x offset
     .stride_loop:
-        movu m0, [r0 + r6]
-        movu m1, [r2 + r6]
-        psubusb m0, m1
-        movu [r0 + r6], m0
+        movu m1, [r0 + r6]
+        movu m2, [r2 + r6]
+        psubusb m1, m2
+        movu [r0 + r6], m1
         add r6, mmsize
         cmp r6, r7
         jl .stride_loop ; still in scan line
     .stride_loop2
-        cmp r6, r5
+        cmp r6, r4
         jge .finish
         movzx r8, byte [r0 + r6]
         sub r8b, byte [r2 + r6]
@@ -166,7 +166,7 @@ cglobal sub_bitmaps_x86, 6,10
     .finish
         add r0, r1
         add r2, r3
-        cmp r2, r4
+        cmp r2, r5
         jl .height_loop
         RET
 %endmacro
@@ -221,15 +221,15 @@ cglobal mul_bitmaps, 8,12
 .height_loop:
     xor r8, r8 ; x offset
 .stride_loop:
-    movq xmm0, [r2 + r8]
-    movq xmm1, [r4 + r8]
-    punpcklbw xmm0, xmm2
+    movq xmm1, [r2 + r8]
+    movq xmm2, [r4 + r8]
     punpcklbw xmm1, xmm2
-    pmullw xmm0, xmm1
-    paddw xmm0, xmm3
-    psrlw xmm0, 0x08
-    packuswb xmm0, xmm0
-    movq [r0 + r8], xmm0
+    punpcklbw xmm2, xmm2
+    pmullw xmm1, xmm2
+    paddw xmm1, xmm3
+    psrlw xmm1, 0x08
+    packuswb xmm1, xmm1
+    movq [r0 + r8], xmm1
     add r8, 8
     cmp r8, r9
     jl .stride_loop ; still in scan line
@@ -268,18 +268,18 @@ cglobal mul_bitmaps, 8,12
 .height_loop:
     xor r8, r8 ; x offset
 .stride_loop:
-    vmovdqu xmm0, [r2 + r8]
-    vpermq ymm0, ymm0, 0x10
-    vmovdqu xmm1, [r4 + r8]
+    vmovdqu xmm1, [r2 + r8]
     vpermq ymm1, ymm1, 0x10
-    vpunpcklbw ymm0, ymm0, ymm2
+    vmovdqu xmm2, [r4 + r8]
+    vpermq ymm2, ymm2, 0x10
     vpunpcklbw ymm1, ymm1, ymm2
-    vpmullw ymm0, ymm0, ymm1
-    vpaddw ymm0, ymm0, ymm3
-    vpsrlw ymm0, ymm0, 0x08
-    vextracti128 xmm4, ymm0, 0x1
-    vpackuswb ymm0, ymm0, ymm4
-    vmovdqa [r0 + r8], xmm0
+    vpunpcklbw ymm2, ymm2, ymm2
+    vpmullw ymm1, ymm1, ymm2
+    vpaddw ymm1, ymm1, ymm3
+    vpsrlw ymm1, ymm1, 0x08
+    vextracti128 xmm4, ymm1, 0x1
+    vpackuswb ymm1, ymm1, ymm4
+    vmovdqa [r0 + r8], xmm1
     add r8, 16
     cmp r8, r9
     jl .stride_loop ; still in scan line