ARM Skia NEON patches - 31 - Xfermode: xfer16

src/opts/SkColor_opts_neon.h

Index: src/opts/SkColor_opts_neon.h
diff --git a/src/opts/SkColor_opts_neon.h b/src/opts/SkColor_opts_neon.h
index cd9e8133e20658a7698d3c471f2689691e6a30e0..82ca0072c86596d9ab2a7b9ab8c58f6b1c4a5600 100644
--- a/src/opts/SkColor_opts_neon.h
+++ b/src/opts/SkColor_opts_neon.h
@@ -29,4 +29,32 @@ static inline uint8x8x4_t SkAlphaMulQ_neon8(uint8x8x4_t color, uint16x8_t scale)
     return ret;
 }
 
+static inline uint8x8x4_t SkPixel16ToPixel32_neon8(uint16x8_t vsrc) {

[reed1, 2013/11/06 15:35:38]

ca we add a brief comment above each of these?

e.g.

// input: 8 RGB565 pixels stored (in what packing order?) in vsrc
// output 8 SkPMColor pixels, expanded by bit-replication? (in what packing
order?)

[kevin.petit.not.used.account, 2013/11/06 16:50:05]

If you want. Before I submit another patch I'd like to be sure I precisely
understand what you want.

SkPMColor: thanks to the NEON_* defines it will handle transparently all
possible configurations (little/big endian and ARGB/ABGR for example).

RGB565: I haven't seen anything (in Skia) else than (from high to low) R, G, B.
The code here, even if it uses all the SK_?16_BITS defines, relies on that
assumption.

My proposal is to add something like:

"This function expands 8 pixels from RGB565 (R, G, B from high to low) to
SkPMColor (all possible configurations supported) in the exact same way as
SkPixel16ToPixel32."

And:

"This function packs 8 pixels from SkPMColor (all possible configurations
supported) to RGB565 (R, G, B from high to low) in the exact same way as
SkPixel32ToPixel16."

+
+    uint8x8x4_t ret;
+    uint8x8_t vr, vg, vb;
+
+    vr = vmovn_u16(vshrq_n_u16(vsrc, SK_R16_SHIFT));
+    vg = vmovn_u16(vshrq_n_u16(vshlq_n_u16(vsrc, SK_R16_BITS), SK_R16_BITS + SK_B16_BITS));
+    vb = vmovn_u16(vsrc & vdupq_n_u16(SK_B16_MASK));
+
+    ret.val[NEON_A] = vdup_n_u8(0xFF);
+    ret.val[NEON_R] = vshl_n_u8(vr, 8 - SK_R16_BITS) | vshr_n_u8(vr, 2 * SK_R16_BITS - 8);
+    ret.val[NEON_G] = vshl_n_u8(vg, 8 - SK_G16_BITS) | vshr_n_u8(vg, 2 * SK_G16_BITS - 8);
+    ret.val[NEON_B] = vshl_n_u8(vb, 8 - SK_B16_BITS) | vshr_n_u8(vb, 2 * SK_B16_BITS - 8);
+
+    return ret;
+}
+
+static inline uint16x8_t SkPixel32ToPixel16_neon8(uint8x8x4_t vsrc) {
+
+    uint16x8_t ret;
+
+    ret = vshll_n_u8(vsrc.val[NEON_R], 8);
+    ret = vsriq_n_u16(ret, vshll_n_u8(vsrc.val[NEON_G], 8), SK_R16_BITS);
+    ret = vsriq_n_u16(ret, vshll_n_u8(vsrc.val[NEON_B], 8), SK_R16_BITS + SK_G16_BITS);
+
+    return ret;
+}
+
 #endif /* #ifndef SkColor_opts_neon_DEFINED */