I want to see the performance between with and without C extension. I downloaded the latest 2.1.3 version. And make three builds using Python 2.6.6:

  1. rm -rf build ; python setup.py build && rm build/lib*/simplejson/_speedups.so
  2. rm -rf build ; python setup.py build
  3. rm -rf build ; CFLAGS="-march=core2 -O2 -pipe -fomit-frame-pointer" python setup.py build

The first one removes the compiled C extension, the second one is normal, then third one uses the current CFLAGS I use in emerge. And the following lines is how C extensions got compiled,

Without customized CFLAGS
x86_64-pc-linux-gnu-gcc -pthread -fPIC -I/usr/include/python2.6 -c simplejson/_speedups.c -o build/temp.linux-x86_64-2.6/simplejson/_speedups.o
x86_64-pc-linux-gnu-gcc -pthread -shared build/temp.linux-x86_64-2.6/simplejson/_speedups.o -L/usr/lib64 -lpython2.6 -o build/lib.linux-x86_64-2.6/simplejson/_speedups.so

With customized CFLAGS
x86_64-pc-linux-gnu-gcc -pthread -march=core2 -O2 -pipe -fomit-frame-pointer -fPIC -I/usr/include/python2.6 -c simplejson/_speedups.c -o build/temp.linux-x86_64-2.6/simplejson/_speedups.o
x86_64-pc-linux-gnu-gcc -pthread -shared -march=core2 -O2 -pipe -fomit-frame-pointer build/temp.linux-x86_64-2.6/simplejson/_speedups.o -L/usr/lib64 -lpython2.6 -o build/lib.linux-x86_64-2.6/

I use the following code to do the test,

import timeit
t = timeit.Timer('json.loads(json_str)', 'import simplejson as json;json_str=open("test.json","r").read()')
print t.timeit(100)

The test.json is http://googleblog.blogspot.com/feeds/posts/default?alt=json&max-results=500, over 3 MB, 500 entires.

The results is

                            Elapsed time for 100 loads()
Without C extension       : 126.230s
json 1.9 in Python 2.6.6  :  60.616s
With C extension          :   9.945s 
With C extension (CFLAGS) :   7.555s

With C extension is at least 10 times faster. I also put the simplejson 1.9 which in Python 2.6.6 in the result. Without extension, 1.9 -> 2.1.3, twice more slower. I didn't download simplejson 1.9 to double check, but I don't think it's modified for being shipped with Python.