ssvb

$ cat test_adds_epu16.c
#include <emmintrin.h>

__m128i x(__m128i a, __m128i b)
{
    return _mm_adds_epu16(a, b);
}

$ gcc -O3 -c test_adds_epu16.c
$ objdump -d test_adds_epu16.o

ssvb

#include <stdint.h>
#include <assert.h>
#include <emmintrin.h>

uint32_t foo(uint32_t a)
{
    __m128i xa = _mm_cvtsi32_si128(a);
    return _mm_cvtsi128_si32(_mm_packus_epi16(xa, xa));
}

ssvb

Some oprofile log snippets for https://plus.google.com/u/0/100242854243155306943/posts/KWcGxqFemfV
Run on ARM Cortex-A8 1GHz, videotestsrc used instead of v4l2src (no camera here)

----------------------------------------------------------------------------

# opcontrol --deinit
# opcontrol --separate=kernel --no-vmlinux
# opcontrol --init
# opcontrol --reset

ssvb

/dts-v1/;

/ {
description = "Chrome OS kernel image with one or more FDT blobs";
#address-cells = <1>;

images {
kernel@1 {
data = /incbin/("arch/arm/boot/zImage");
type = "kernel_noload";

ssvb

Graphics acceleration benchmark on Allwinner A10 (1GHz, 2x16-bit DDR3 @480MHz)
DirectFB 1.4.11 patched with https://github.com/allwinner-ics/lichee_buildroot/blob/master/package/directfb/directfb-1.4.11-vmware.patch
32bpp color depth, 640x480 screen resolution

----------------------------------------------------------------------------------------------------------------------------------------
                                            ./df_dok --noaccel --system              ./df_dok (G2D acceleration)         pixman-0.28.0
----------------------------------------------------------------------------------------------------------------------------------------
Anti-aliased Text                         (  271.765 KChars/sec) [100.0%]          (  139.840 KChars/sec) [100.0%]
Anti-aliased Text (blend)                 (   49.202 KChars/sec) [100.0%]          (   33.455 KChars/sec) [100.0%]
Fill Rectangle                            (  326.699 MPixel/sec) [100.3%]          (* 200.442 MPixel/sec) [  2.0%]     359.

ssvb

$ gcc -v
Using built-in specs.
COLLECT_GCC=gcc-4.7.2
COLLECT_LTO_WRAPPER=/usr/libexec/gcc/x86_64-pc-linux-gnu/4.7.2/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: /var/tmp/portage/sys-devel/gcc-4.7.2/work/gcc-4.7.2/configure --prefix=/usr --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/4.7.2 --includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/4.7.2/include --datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/4.7.2 --mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/4.7.2/man --infodir=/usr/share/gcc-data/x86_64-pc-linux-gnu/4.7.2/info --with-gxx-include-dir=/usr/lib/gcc/x86_64-pc-linux-gnu/4.7.2/include/g++-v4 --host=x86_64-pc-linux-gnu --build=x86_64-pc-linux-gnu --disable-altivec --disable-fixed-point --without-ppl --without-cloog --enable-lto --enable-nls --without-included-gettext --with-system-zlib --enable-obsolete --disable-werror --enable-secureplt --enable-multilib --with-multilib-list=m32,m64 --enable-libmudflap --disable-libssp --enable-libgomp --with-python-dir=/share/gcc-data/x86_64-pc-linux-gnu/4.7.2/py

ssvb

Tasks:  61 total,   2 running,  59 sleeping,   0 stopped,   0 zombie
Cpu(s): 98.7%us,  1.3%sy,  0.0%ni,  0.0%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
Mem:    315004k total,   115304k used,   199700k free,       24k buffers
Swap:        0k total,        0k used,        0k free,    67692k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                                                   
 1417 root      20   0 59264  25m 6956 S 86.6  8.2   2:42.63 retroarch                                                 
  870 root      20   0 41208  15m 7444 R 12.5  5.0   0:29.89 X                                                         
 1037 root      20   0  8968 3128 2484 S  0.3  1.0   0:00.24 sshd                                                      
 1481 root      20   0  2504 1152  920 R  0.3  0.4   0:00.05 top                                                       

ssvb

# Overhead    Command       Shared Object                           Symbol
# ........  .........  ..................  ...............................
#
    38.75%  retroarch  libGL.so.1.5.08004  [.] linear_texel_locations     
    22.67%  retroarch  libGL.so.1.5.08004  [.] _swrast_texture_span       
    13.30%  retroarch  libGL.so.1.5.08004  [.] lerp_rgba_2d               
     7.53%  retroarch  libGL.so.1.5.08004  [.] shade_texture_span         
     6.75%  retroarch  libGL.so.1.5.08004  [.] fetch_texel_2d_f_rgba8888  
     5.96%  retroarch  libGL.so.1.5.08004  [.] sample_2d_linear.isra.27   
     0.85%  retroarch  libGL.so.1.5.08004  [.] pack_row_ubyte_ARGB8888    

ssvb

sun4i ~ # cat /etc/X11/xorg.conf
Section "Device"
    Identifier "FBDEV"
    Driver "fbdev"
    Option "fbdev" "/dev/fb0"
#    Option "ShadowFB" "false"
EndSection

Section "ServerFlags"
    Option "BlankTime" "0"

ssvb

[product]
version = "1.0"
machine = "A10-EVB-V1.1"

[target]
boot_clock = 1008
dcdc2_vol = 1400
dcdc3_vol = 1250
ldo2_vol = 3000
ldo3_vol = 2800