Benchmark of a set of functions most important for web browser
---
All results are presented in millions of pixels per second
'=' - normal pixman calls
'-' - direct call to the blitter, bypassing pixman dispatch
      logic
L1  - small 1024x1 rectangle, always blitted at the same
      memory location with small drift in horizontal direction
M   - large 1856x1080 rectangle, always blitted at the same
      memory location with small drift in horizontal direction
HT  - random rectangles with 32x32 average size are copied from
      one 1920x1080 buffer to another, traversing from left to right
      and from top to bottom
VT  - random rectangles with 32x32 average size are copied from
      one 1920x1080 buffer to another, traversing from top to bottom
      and from left to right
R   - random rectangles with 32x32 average size are copied from
      random locations of one 1920x1080 buffer to another
---
reference memcpy speed = 519.0MB/s (129.8MP/s for 32bpp pixels)
---
over_8888_0565 = L1: 119.68 M: 89.49 HT: 70.28 VT: 63.83 R: 54.78
over_8888_0888 = L1: 119.90 M: 89.56 HT: 70.51 VT: 63.76 R: 54.59
 src_0888_0888 = L1: 216.06 M:128.28 HT:100.17 VT: 58.05 R: 54.39
 src_8888_0565 = L1: 142.23 M: 98.36 HT: 72.59 VT: 48.95 R: 45.94
 src_0565_0565 = L1:  55.99 M: 54.60 HT: 31.11 VT: 26.38 R: 25.08