2014-06-17 17:00 GMT+02:00 Iñigo Muguruza <[email protected]>: > Dear OpenRISC community, > > I write this mail to inform about an issue I have found when I analyzed the > outcome data coming from or1200 and mor1kx profiling. I am working in making > a OpenRISC tutorial to help people install and execute the diverse > open-soruce SoC tools. In this tutorial a profiling test for or1200/mor1kx > is made. Surprisingly, the results are really different, inducting to think > about a fault/bug. (results in the attached txts) > > In the mor1kx, _reset and or1k_dmu_disable routines, takes really a big > percentage of the execution, while in or1200 is does not happen the same. > > The tools that I have used to make the profiling are the next ones: > > - De0_nano board > - mor1kx and or1200 .sof generated with quartus tools. (changing defines > from orpsoc-defines.v) > - Simple test code, basically two decrementing loops that are called by the > main. It also prints messages at the end of each loop (see prof-test.c from > the link below). > - OpenOCD > - Telnet client for running the test. > - OpenOCD patch. Enables to execute profiling in or1200/mor1kx. gmon file > data writing for big endian targets. http://openocd.zylin.com/#/c/2168/ > > The procedure, is the one described in the "profiling" section of this > tutorial: > > http://wiki.mintsoc.org/doku.php?id=run_programs_on_the_altera_de0_nano_board#profiling > > Best regards, > > Inigo Muguruza > > _______________________________________________ > OpenRISC mailing list > [email protected] > http://lists.openrisc.net/listinfo/openrisc >
Hi Iñigo, I don't have answer to your questions. However, as you work on this, can you try to apply what is suggested here: http://openocd.zylin.com/#/c/2168/1/src/target/target.c,cm Thanks, Franck. _______________________________________________ OpenRISC mailing list [email protected] http://lists.openrisc.net/listinfo/openrisc
