port scale box filter sse2 to gcc

port scale box filter sse2 to gcc

LIBYUV_FLAGS=-1  LIBYUV_WIDTH=1280 LIBYUV_HEIGHT=720 LIBYUV_REPEAT=999 perf
record out/Release/libyuv_unittest --gtest_filter=*.ScaleDownBy3_*

Was
[ RUN      ] libyuvTest.ScaleDownBy3_Box
filter 3 -     6762 us C -     6795 us OPT
[       OK ] libyuvTest.ScaleDownBy3_Box (6803 ms)
    50.18%  libyuv_unittest  libyuv_unittest      [.] ScaleAddRow_C
    24.02%  libyuv_unittest  libyuv_unittest      [.] ScaleAddCols1_C
    23.10%  libyuv_unittest  libyuv_unittest      [.] ScaleCols_C
     1.71%  libyuv_unittest  libc-2.19.so         [.] memset

Now
[ RUN      ] libyuvTest.ScaleDownBy3_Box
filter 3 -     3228 us C -     3223 us OPT
[       OK ] libyuvTest.ScaleDownBy3_Box (3231 ms)
    40.29%  libyuv_unittest  libyuv_unittest      [.] ScaleAddCols1_C
    38.19%  libyuv_unittest  libyuv_unittest      [.] ScaleCols_C
    17.54%  libyuv_unittest  libyuv_unittest      [.] ScaleAddRow_SSE2
     2.56%  libyuv_unittest  libc-2.19.so         [.] memset

TBR=harryjin@google.com
BUG=libyuv:492
Committed: https://chromium.googlesource.com/libyuv/libyuv/+/d70293993fac8161bd48d86d74cd93ad6cad65a0

[fbarchard1, 2015-10-06 23:55:15]

Committed patchset #1 (id:1) manually as
d70293993fac8161bd48d86d74cd93ad6cad65a0 (presubmit successful).

[fbarchard1, 2015-10-06 23:59:38]

LIBYUV_FLAGS=-1  LIBYUV_WIDTH=1280 LIBYUV_HEIGHT=720 LIBYUV_REPEAT=999 perf
record out/Release/libyuv_unittest --gtest_filter=*.ScaleDownBy3_*


Was
[ RUN      ] libyuvTest.ScaleDownBy3_Box
filter 3 -     6762 us C -     6795 us OPT
[       OK ] libyuvTest.ScaleDownBy3_Box (6803 ms)
    50.18%  libyuv_unittest  libyuv_unittest      [.] ScaleAddRow_C             
                                        
    24.02%  libyuv_unittest  libyuv_unittest      [.] ScaleAddCols1_C           
                                        
    23.10%  libyuv_unittest  libyuv_unittest      [.] ScaleCols_C               
                                        
     1.71%  libyuv_unittest  libc-2.19.so         [.] memset                    
                                        
     0.43%  libyuv_unittest  libyuv_unittest      [.] ScalePlane                
                                        
     0.36%  libyuv_unittest  libyuv_unittest      [.] libyuv::TestFilter(int,
int, int, int, libyuv::FilterMode, int, int


Now
[ RUN      ] libyuvTest.ScaleDownBy3_Box
filter 3 -     3228 us C -     3223 us OPT
[       OK ] libyuvTest.ScaleDownBy3_Box (3231 ms)
    40.29%  libyuv_unittest  libyuv_unittest      [.] ScaleAddCols1_C           
                                     
    38.19%  libyuv_unittest  libyuv_unittest      [.] ScaleCols_C               
                                     
    17.54%  libyuv_unittest  libyuv_unittest      [.] ScaleAddRow_SSE2          
                                     
     2.56%  libyuv_unittest  libc-2.19.so         [.] memset                    
                                     
     0.60%  libyuv_unittest  libyuv_unittest      [.] libyuv::TestFilter(int,
int, int, int, libyuv::FilterMode, int, 0.55%  libyuv_unittest  libyuv_unittest 
    [.] ScalePlan

[harryjin, 2015-10-07 05:27:23]

lgtm