Serhiy, I am not able to reproduce your results. I don't get any significant performance improvements with your patch.

22:34:03 [ ~/PythonDev/cpython-baseline ]$ ./python.exe -m timeit "import pickle" "with open('test.pickle4', 'rb', buffering=0) as f: pickle.load(f)"
100 loops, best of 3: 9.26 msec per loop

22:36:13 [ ~/PythonDev/cpython ]$ ./python.exe -m timeit "import pickle" "with open('test.pickle4', 'rb', buffering=0) as f: pickle.load(f)"
100 loops, best of 3: 9.28 msec per loop

I wrote a benchmark to simulate slow reads. But again, there is no performance difference with the patch.

22:24:56 [ ~/PythonDev/benchmarks ]$ ./perf.py -v -b unpickle_file ../cpython-baseline/python.exe ../cpython/python.exe
### unpickle_file ###
Min: 1.103715 -> 1.103666: 1.00x faster
Avg: 1.158279 -> 1.146820: 1.01x faster
Not significant
Stddev: 0.04127 -> 0.03273: 1.2608x smaller