NEON shortcut for flat colour blending into 16-bit

This is a shortcut for the needs descriptor
00000077:03515104_00000000_00000000.  It requires blending a single 32-bit
colour value into a 16-bit framebuffer.
It's used when fading out the screen, eg. when a modal requester pops-up.

The PF JIT produces code for this using 24 instructions/pixel. The NEON
implementation requires 2.1 instructions/pixel. Performance hasn't been
benchmarked, but the improvement is quite visible.

This code has only been tested by inspection of the fading effect described
above, when press+holding a finger on the home screen to pop up the
Shortcuts/Widgets/Folders/Wallpaper requester.

Along with the NEON version, a fallback v5TE implementation is also provided.

This ARM version of col32cb16blend is not fully optimised, but is a reasonable
implementation, and better than the version produced by the JIT. It is here as
a fallback, if NEON is not available.

1 files changed, 78 insertions, 0 deletions

diff --git a/libpixelflinger/col32cb16blend.S b/libpixelflinger/col32cb16blend.S
new file mode 100644
index 00000000..1450bde8
--- /dev/null
+++ b/libpixelflinger/col32cb16blend.S
@@ -0,0 +1,78 @@
+/* libs/pixelflinger/col32cb16blend.S
+**
+** (C) COPYRIGHT 2009 ARM Limited.
+**
+** Licensed under the Apache License, Version 2.0 (the "License"); 
+** you may not use this file except in compliance with the License. 
+** You may obtain a copy of the License at 
+**
+**     http://www.apache.org/licenses/LICENSE-2.0 
+**
+** Unless required by applicable law or agreed to in writing, software 
+** distributed under the License is distributed on an "AS IS" BASIS, 
+** WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 
+** See the License for the specific language governing permissions and 
+** limitations under the License.
+**
+*/
+
+    .text
+    .align
+
+    .global scanline_col32cb16blend_arm
+
+//
+// This function alpha blends a fixed color into a destination scanline, using
+// the formula:
+//
+//     d = s + (((a + (a >> 7)) * d) >> 8)
+//
+// where d is the destination pixel,
+//       s is the source color,
+//       a is the alpha channel of the source color.
+//
+
+// r0 = destination buffer pointer
+// r1 = color value
+// r2 = count
+
+
+scanline_col32cb16blend_arm:
+    push        {r4-r10, lr}                    // stack ARM regs
+
+    mov         r5, r1, lsr #24                 // shift down alpha
+    mov         r9, #0xff                       // create mask
+    add         r5, r5, r5, lsr #7              // add in top bit
+    rsb         r5, r5, #256                    // invert alpha
+    and         r10, r1, #0xff                  // extract red
+    and         r12, r9, r1, lsr #8             // extract green
+    and         r4, r9, r1, lsr #16             // extract blue
+    mov         r10, r10, lsl #5                // prescale red
+    mov         r12, r12, lsl #6                // prescale green
+    mov         r4, r4, lsl #5                  // prescale blue
+    mov         r9, r9, lsr #2                  // create dest green mask
+
+1:
+    ldrh        r8, [r0]                        // load dest pixel
+    subs        r2, r2, #1                      // decrement loop counter
+    mov         r6, r8, lsr #11                 // extract dest red
+    and         r7, r9, r8, lsr #5              // extract dest green
+    and         r8, r8, #0x1f                   // extract dest blue
+
+    smlabb      r6, r6, r5, r10                 // dest red * alpha + src red
+    smlabb      r7, r7, r5, r12                 // dest green * alpha + src green
+    smlabb      r8, r8, r5, r4                  // dest blue * alpha + src blue
+
+    mov         r6, r6, lsr #8                  // shift down red
+    mov         r7, r7, lsr #8                  // shift down green
+    mov         r6, r6, lsl #11                 // shift red into 565
+    orr         r6, r7, lsl #5                  // shift green into 565
+    orr         r6, r8, lsr #8                  // shift blue into 565
+
+    strh        r6, [r0], #2                    // store pixel to dest, update ptr
+    bne         1b                              // if count != 0, loop
+
+    pop         {r4-r10, pc}                    // return
+
+
+