Hi All, Apologies - this is a similar post to one I previously made but the question was tagged onto the end of a previous question so it was never answered.
I am developing an algorithm that will run on the DSP (with codec engine), and an application that hands the algo video and displays the results on the ARM (linux). My problem is this: + I need to develop the algorithm with CCS, in Windows. In this environment I have no support from the codec engine (it can only be built under linux?), so I must provide my own video (via RTDX), take care of peripheral initialisation, etc. + It seems that the correct development route is as follows... 1. Develop algo in CCS, test with data from RTDX, stdio, whatever. Manually initialise peripherals I need (like EMIF) the 'hard' way (uboot would do this on the ARM). 2. Once I am happy the algorithm is good, reboot into linux, build the algo into a codec engine app. Test. 3. If it doesn't work, reboot into windows, attach the debugger via JTAG. Fix any problems. 4. Reboot to linux, recompile engine app, retest. Goto 3. My problem is that it seems silly... ridiculously silly to have to keep switching the OS to rebuild and re-test. I get the feeling I am missing something because I can't believe anybody would ever suggest this is the right way to do it. Is it really necessary to use 2 different OSs to build and test 1 exe? Thanks very much for responses in advance. Ben
_______________________________________________ Davinci-linux-open-source mailing list Davinci-linux-open-source@linux.davincidsp.com http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
