git update 12 sep 2011

2 files changed, 152 insertions, 45 deletions

diff --git a/pixman/pixman/pixman-arm-neon-asm.S b/pixman/pixman/pixman-arm-neon-asm.S
index 3dc14d79d..3fcd07dc3 100644
--- a/pixman/pixman/pixman-arm-neon-asm.S
+++ b/pixman/pixman/pixman-arm-neon-asm.S
@@ -706,13 +706,55 @@ generate_composite_function_single_scanline \
 
 /******************************************************************************/
 
-/* TODO: expand macros and do better instructions scheduling */
+.macro pixman_composite_over_n_8888_process_pixblock_head
+    /* deinterleaved source pixels in {d0, d1, d2, d3} */
+    /* inverted alpha in {d24} */
+    /* destination pixels in {d4, d5, d6, d7} */
+    vmull.u8    q8, d24, d4
+    vmull.u8    q9, d24, d5
+    vmull.u8    q10, d24, d6
+    vmull.u8    q11, d24, d7
+.endm
+
+.macro pixman_composite_over_n_8888_process_pixblock_tail
+    vrshr.u16   q14, q8, #8
+    vrshr.u16   q15, q9, #8
+    vrshr.u16   q2, q10, #8
+    vrshr.u16   q3, q11, #8
+    vraddhn.u16 d28, q14, q8
+    vraddhn.u16 d29, q15, q9
+    vraddhn.u16 d30, q2, q10
+    vraddhn.u16 d31, q3, q11
+    vqadd.u8    q14, q0, q14
+    vqadd.u8    q15, q1, q15
+.endm
+
 .macro pixman_composite_over_n_8888_process_pixblock_tail_head
-    pixman_composite_over_8888_8888_process_pixblock_tail
+        vrshr.u16   q14, q8, #8
+        vrshr.u16   q15, q9, #8
+        vrshr.u16   q2, q10, #8
+        vrshr.u16   q3, q11, #8
+        vraddhn.u16 d28, q14, q8
+        vraddhn.u16 d29, q15, q9
+        vraddhn.u16 d30, q2, q10
+        vraddhn.u16 d31, q3, q11
     vld4.8      {d4, d5, d6, d7}, [DST_R, :128]!
-    vst4.8      {d28, d29, d30, d31}, [DST_W, :128]!
-    pixman_composite_over_8888_8888_process_pixblock_head
-    cache_preload 8, 8
+        vqadd.u8    q14, q0, q14
+                                    PF add PF_X, PF_X, #8
+                                    PF tst PF_CTL, #0x0F
+                                    PF addne PF_X, PF_X, #8
+                                    PF subne PF_CTL, PF_CTL, #1
+        vqadd.u8    q15, q1, q15
+                                    PF cmp PF_X, ORIG_W
+    vmull.u8    q8, d24, d4
+                                    PF pld, [PF_DST, PF_X, lsl #dst_bpp_shift]
+    vmull.u8    q9, d24, d5
+                                    PF subge PF_X, PF_X, ORIG_W
+    vmull.u8    q10, d24, d6
+                                    PF subges PF_CTL, PF_CTL, #0x10
+    vmull.u8    q11, d24, d7
+                                    PF ldrgeb DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+        vst4.8      {d28, d29, d30, d31}, [DST_W, :128]!
 .endm
 
 .macro pixman_composite_over_n_8888_init
@@ -722,6 +764,7 @@ generate_composite_function_single_scanline \
     vdup.8      d1, d3[1]
     vdup.8      d2, d3[2]
     vdup.8      d3, d3[3]
+    vmvn.8      d24, d3  /* get inverted alpha */
 .endm
 
 generate_composite_function \
@@ -1183,49 +1226,83 @@ generate_composite_function \
     /* mask is in d24 (d25, d26, d27 are unused) */
 
     /* in */
-    vmull.u8    q0, d24, d8
-    vmull.u8    q1, d24, d9
-    vmull.u8    q6, d24, d10
-    vmull.u8    q7, d24, d11
-    vrshr.u16   q10, q0, #8
-    vrshr.u16   q11, q1, #8
-    vrshr.u16   q12, q6, #8
-    vrshr.u16   q13, q7, #8
-    vraddhn.u16 d0, q0, q10
-    vraddhn.u16 d1, q1, q11
-    vraddhn.u16 d2, q6, q12
-    vraddhn.u16 d3, q7, q13
-    vmvn.8      d24, d3  /* get inverted alpha */
+    vmull.u8    q6, d24, d8
+    vmull.u8    q7, d24, d9
+    vmull.u8    q8, d24, d10
+    vmull.u8    q9, d24, d11
+    vrshr.u16   q10, q6, #8
+    vrshr.u16   q11, q7, #8
+    vrshr.u16   q12, q8, #8
+    vrshr.u16   q13, q9, #8
+    vraddhn.u16 d0, q6, q10
+    vraddhn.u16 d1, q7, q11
+    vraddhn.u16 d2, q8, q12
+    vraddhn.u16 d3, q9, q13
+    vmvn.8      d25, d3  /* get inverted alpha */
     /* source:      d0 - blue, d1 - green, d2 - red, d3 - alpha */
     /* destination: d4 - blue, d5 - green, d6 - red, d7 - alpha */
     /* now do alpha blending */
-    vmull.u8    q8, d24, d4
-    vmull.u8    q9, d24, d5
-    vmull.u8    q10, d24, d6
-    vmull.u8    q11, d24, d7
+    vmull.u8    q8, d25, d4
+    vmull.u8    q9, d25, d5
+    vmull.u8    q10, d25, d6
+    vmull.u8    q11, d25, d7
 .endm
 
 .macro pixman_composite_over_n_8_8888_process_pixblock_tail
     vrshr.u16   q14, q8, #8
     vrshr.u16   q15, q9, #8
-    vrshr.u16   q12, q10, #8
-    vrshr.u16   q13, q11, #8
+    vrshr.u16   q6, q10, #8
+    vrshr.u16   q7, q11, #8
     vraddhn.u16 d28, q14, q8
     vraddhn.u16 d29, q15, q9
-    vraddhn.u16 d30, q12, q10
-    vraddhn.u16 d31, q13, q11
+    vraddhn.u16 d30, q6, q10
+    vraddhn.u16 d31, q7, q11
     vqadd.u8    q14, q0, q14
     vqadd.u8    q15, q1, q15
 .endm
 
-/* TODO: expand macros and do better instructions scheduling */
 .macro pixman_composite_over_n_8_8888_process_pixblock_tail_head
-    pixman_composite_over_n_8_8888_process_pixblock_tail
-    vst4.8      {d28, d29, d30, d31}, [DST_W, :128]!
+        vrshr.u16   q14, q8, #8
     vld4.8      {d4, d5, d6, d7}, [DST_R, :128]!
+        vrshr.u16   q15, q9, #8
     fetch_mask_pixblock
-    cache_preload 8, 8
-    pixman_composite_over_n_8_8888_process_pixblock_head
+        vrshr.u16   q6, q10, #8
+                                    PF add PF_X, PF_X, #8
+        vrshr.u16   q7, q11, #8
+                                    PF tst PF_CTL, #0x0F
+        vraddhn.u16 d28, q14, q8
+                                    PF addne PF_X, PF_X, #8
+        vraddhn.u16 d29, q15, q9
+                                    PF subne PF_CTL, PF_CTL, #1
+        vraddhn.u16 d30, q6, q10
+                                    PF cmp PF_X, ORIG_W
+        vraddhn.u16 d31, q7, q11
+                                    PF pld, [PF_DST, PF_X, lsl #dst_bpp_shift]
+    vmull.u8    q6, d24, d8
+                                    PF pld, [PF_MASK, PF_X, lsl #mask_bpp_shift]
+    vmull.u8    q7, d24, d9
+                                    PF subge PF_X, PF_X, ORIG_W
+    vmull.u8    q8, d24, d10
+                                    PF subges PF_CTL, PF_CTL, #0x10
+    vmull.u8    q9, d24, d11
+                                    PF ldrgeb DUMMY, [PF_DST, DST_STRIDE, lsl #dst_bpp_shift]!
+        vqadd.u8    q14, q0, q14
+                                    PF ldrgeb DUMMY, [PF_MASK, MASK_STRIDE, lsl #mask_bpp_shift]!
+        vqadd.u8    q15, q1, q15
+    vrshr.u16   q10, q6, #8
+    vrshr.u16   q11, q7, #8
+    vrshr.u16   q12, q8, #8
+    vrshr.u16   q13, q9, #8
+    vraddhn.u16 d0, q6, q10
+    vraddhn.u16 d1, q7, q11
+    vraddhn.u16 d2, q8, q12
+    vraddhn.u16 d3, q9, q13
+        vst4.8      {d28, d29, d30, d31}, [DST_W, :128]!
+    vmvn.8      d25, d3
+    vmull.u8    q8, d25, d4
+    vmull.u8    q9, d25, d5
+    vmull.u8    q10, d25, d6
+    vmull.u8    q11, d25, d7
 .endm
 
 .macro pixman_composite_over_n_8_8888_init
diff --git a/pixman/pixman/pixman-bits-image.c b/pixman/pixman/pixman-bits-image.c
index f540c76e1..f382c65ad 100644
--- a/pixman/pixman/pixman-bits-image.c
+++ b/pixman/pixman/pixman-bits-image.c
@@ -935,17 +935,16 @@ MAKE_FETCHERS (reflect_r5g6b5,   r5g6b5,   PIXMAN_REPEAT_REFLECT)
 MAKE_FETCHERS (normal_r5g6b5,    r5g6b5,   PIXMAN_REPEAT_NORMAL)
 
 static void
-bits_image_fetch_solid_32 (pixman_image_t * image,
-                           int              x,
-                           int              y,
-                           int              width,
-                           uint32_t *       buffer,
-                           const uint32_t * mask)
+replicate_pixel_32 (bits_image_t *   bits,
+		    int              x,
+		    int              y,
+		    int              width,
+		    uint32_t *       buffer)
 {
     uint32_t color;
     uint32_t *end;
 
-    color = image->bits.fetch_pixel_32 (&image->bits, 0, 0);
+    color = bits->fetch_pixel_32 (bits, x, y);
 
     end = buffer + width;
     while (buffer < end)
@@ -953,18 +952,17 @@ bits_image_fetch_solid_32 (pixman_image_t * image,
 }
 
 static void
-bits_image_fetch_solid_64 (pixman_image_t * image,
-                           int              x,
-                           int              y,
-                           int              width,
-                           uint32_t *       b,
-                           const uint32_t * unused)
+replicate_pixel_64 (bits_image_t *   bits,
+		    int              x,
+		    int              y,
+		    int              width,
+		    uint32_t *       b)
 {
     uint64_t color;
     uint64_t *buffer = (uint64_t *)b;
     uint64_t *end;
 
-    color = image->bits.fetch_pixel_64 (&image->bits, 0, 0);
+    color = bits->fetch_pixel_64 (bits, x, y);
 
     end = buffer + width;
     while (buffer < end)
@@ -972,6 +970,28 @@ bits_image_fetch_solid_64 (pixman_image_t * image,
 }
 
 static void
+bits_image_fetch_solid_32 (pixman_image_t * image,
+                           int              x,
+                           int              y,
+                           int              width,
+                           uint32_t *       buffer,
+                           const uint32_t * mask)
+{
+    replicate_pixel_32 (&image->bits, 0, 0, width, buffer);
+}
+
+static void
+bits_image_fetch_solid_64 (pixman_image_t * image,
+                           int              x,
+                           int              y,
+                           int              width,
+                           uint32_t *       b,
+                           const uint32_t * unused)
+{
+    replicate_pixel_64 (&image->bits, 0, 0, width, b);
+}
+
+static void
 bits_image_fetch_untransformed_repeat_none (bits_image_t *image,
                                             pixman_bool_t wide,
                                             int           x,
@@ -1031,6 +1051,16 @@ bits_image_fetch_untransformed_repeat_normal (bits_image_t *image,
     while (y >= image->height)
 	y -= image->height;
 
+    if (image->width == 1)
+    {
+	if (wide)
+	    replicate_pixel_64 (image, 0, y, width, buffer);
+	else
+	    replicate_pixel_32 (image, 0, y, width, buffer);
+
+	return;
+    }
+
     while (width)
     {
 	while (x < 0)