This is a pretty minor update to this seires. It fixes the things Ilia requested fixed from v1, and makes the series behavioral change neutral. -v/--verbose is back, and the status line in verbose mode works as it did before.
I have also done some benchmarking to get status numbers. I created a set of 10 tests that are based a very simple Test derived class (it sets interpret_result = self.result['result'] = 'pass') Then I ran 10 tests where the command was set to 'bash -c "sleep 1"'. Each test was run once as a warm up, to ensure that all of the pyhton bytecode had been generated before running tests. This gave me as close as possible to a controlled environment, I ran each test 5 times, averaging them, measuirng them with the unix `time` command, and using 'total time'. all tests were run with -c on an IvyBridge with 4 threads. (in seconds, rounded to the nearest thousandth) Master default: 3.493 branch quiet : 3.466 Master verbose: 3.485 branch verbose: 3.483 This is available at my github: https://github.com/dcbaker/piglit.git submit/log-refactor _______________________________________________ Piglit mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/piglit
