Hi, I need help in clearing some doubts about performing profiling in CCS ver 2.1. I have a C6201 fixed point DSP chip. It has the following specifications: Cycle time (ns) : 5 MIPS : 1600 I have a function in my program that need to be completed by 2msec. THis is what I have done: 1) In the profiler tab, clock setup: Count = CPU cycles Instruction Cycle Time (ns) = 40 Pipeline Adjustment cycle = 0 2) Run Profile 3) After Profiling I got a value x 4) I calculate the max insrruction cycle that can be allow as follow: max instruction cycle time = 200Mhz X 2msec = 400000 5) If x is less than 400000, my function is able to complete in 2msec Is my way of doing profiling correct? Is my way of intepreting the result correct? regard Chee Wei __________________________________________________

Hi, I need help in clearing some doubts about performing profiling in CCS ver 2.1. I have a C6201 fixed point DSP chip. It has the following specifications: Cycle time (ns) : 5 MIPS : 1600 I have a function in my program that need to be completed by 2msec. THis is what I have done: 1) In the profiler tab, clock setup: Count = CPU cycles Instruction Cycle Time (ns) = 40 Pipeline Adjustment cycle = 0 2) Run Profile 3) After Profiling I got a value x 4) I calculate the max insrruction cycle that can be allow as follow: __________________________________________________

Hi, I got your point. Thanks, but i think very hard to fit in everything to IDRM hehehe. nonetheless your 'theory' must have been correct. regards, -----Original Message----- From: [mailto:] Sent: Tue 2/18/2003 9:51 PM To: #JONATHAN SANITO# Cc: Subject: RE: [c6x] Profiling in CCS Hi again :-) well... by ext. memory, I meant the same..SDRAM etc. only.. And you must be knowing that ext. memory access is preceded by 'Wait cycles', which may vary because of cache... ....... This was what i surmised when i also used SDRAM with cache enabled on EVM 6201. and got different values of profiling all the time. Well... i know i dont sound very rational... but I told myself this thing to my own satisfaction. :-) And in fact , when i moved my data into IDRM ( internal data memory), profiling figures became consistant. Good Luck, And I hope if you get correct answer, it'll be made available on C6x group. :-) Regards, Vishal. "#JONATHAN SANITO#" <> on 02/18/2003 07:18:38 PM To: Vishal Goel/HSS@HSS cc: Subject: RE: [c6x] Profiling in CCS Hi, what do you mean by external memory? you mean the SDRAM of my DSK? what's the cause of this ya, got idea? I believe if there is only 1 program loaded to my DSK, the measurement should have been the same right? regards, Jonathan -----Original Message----- From: [mailto:] Sent: Tue 2/18/2003 9:37 PM To: #JONATHAN SANITO# Cc: Subject: Re: [c6x] Profiling in CCS Hi.. :-) This can happen when u access code or data from external memory... check that out.. :o) regards, Vishal. "#JONATHAN SANITO#" <> on 02/18/2003 12:25:17 PM To: cc: (bcc: Vishal Goel/HSS) Subject: [c6x] Profiling in CCS Hi, I am using CCS 2.0, to program on C6711 DSK. I'm trying to profile my code. But the funny thing I got different results for different execution. Is this normal? I actually set to profile functions that don't have any communication with my host pc. ie no printf/fread and such. only performing computation in the DSK board. but still the results seems to be varies within execution. anyone got an idea? Thanks, Jonathan _____________________________________ Note: If you do a simple "reply" with your email client, only the author of this message will receive your answer. You need to do a "reply all" if you want your answer to be distributed to the entire group. _____________________________________ About this discussion group: To Join: Send an email to To Post: Send an email to To Leave: Send an email to Archives: http://www.yahoogroups.com/group/c6x Other Groups: http://www.dsprelated.com ">http://docs.yahoo.com/info/terms/

Hi, what do you mean by external memory? you mean the SDRAM of my DSK? what's the cause of this ya, got idea? I believe if there is only 1 program loaded to my DSK, the measurement should have been the same right? regards, Jonathan -----Original Message----- From: [mailto:] Sent: Tue 2/18/2003 9:37 PM To: #JONATHAN SANITO# Cc: Subject: Re: [c6x] Profiling in CCS Hi.. :-) This can happen when u access code or data from external memory... check that out.. :o) regards, Vishal. "#JONATHAN SANITO#" <> on 02/18/2003 12:25:17 PM To: cc: (bcc: Vishal Goel/HSS) Subject: [c6x] Profiling in CCS Hi, I am using CCS 2.0, to program on C6711 DSK. I'm trying to profile my code. But the funny thing I got different results for different execution. Is this normal? I actually set to profile functions that don't have any communication with my host pc. ie no printf/fread and such. only performing computation in the DSK board. but still the results seems to be varies within execution. anyone got an idea? Thanks, Jonathan _____________________________________ Note: If you do a simple "reply" with your email client, only the author of this message will receive your answer. You need to do a "reply all" if you want your answer to be distributed to the entire group. _____________________________________ About this discussion group: To Join: Send an email to To Post: Send an email to To Leave: Send an email to Archives: http://www.yahoogroups.com/group/c6x Other Groups: http://www.dsprelated.com ">http://docs.yahoo.com/info/terms/

Hi, I am using CCS 2.0, to program on C6711 DSK. I'm trying to profile my code. But the funny thing I got different results for different execution. Is this normal? I actually set to profile functions that don't have any communication with my host pc. ie no printf/fread and such. only performing computation in the DSK board. but still the results seems to be varies within execution. anyone got an idea? Thanks, Jonathan