AD535 | BIOS | Booting | Bootloader | C621 | C6211 | C6415 | C671 | C6711 | C6711DSK | C6713 | CCS | Chassaing | COFF | DAT | DM64 | DM642 | DMA | DSK671 | DSK6711 | EDM | EDMA | EMIF | Emulator | EVM | EVM620 | FFT | FIR | GPIO | Halting | HPI | HWI | IDK | JTAG | LDB | LDH | LDW | Linker | LMS | LOG_printf | Matlab | McBSP | MEM_alloc | MIPS | PCI | PCM3003 | Pipeline | Profiling | QDM | Reset | ROM | RTDX | Sampling | SDRAM | Stack | TEB | THS1206 | TMS320C621 | TMS320C6416 | TMS320C6711 | TMS320C6713 | UART | Vector Table | XBUS | XDS560

Dear Friends, I am using EVM6201 and Pentek 4291 DSP boards for measuring the performance of c6X processors. Actually I try DSP operations such as "FFT, FIR, etc" by using ready C codes. But there is a problem about Profile points. When I use profile point to count the number of cycles of a function, I cannot obtain a reasonable number of cycles and time. For example, an unoptimized 256 point FFT code in C language seems to take about 1500 cycles. So it takes a time duration less than the time duration, 42 microseconds for C6701 , which is known as the optimum time which can be achieved by using the optimum assembley code. Our C code is not optimum any more and the optimization keys at compilation are not set. Also I tried the profile point for a statement which has a for loop with an index of 10000 . But the the statistics obtained by Code Composer Studio profile point don't seem to be reasonable. What's is the matter? Is an initialization at the program or a setting at Code Composer Studio needeed? I will be waiting for your comments and suggestions on this subject.... Ozgur

______________________________
New Code Sharing Section now Live on DSPRelated.com. Learn about the Reward Program for Contributors here.



(You need to be a member of c6x -- send a blank email to c6x-subscribe@yahoogroups.com )