Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
06/18/07 12:55
Read: times


 
#140946 - please elaborate
Responding to: ???'s previous message
This is where common sequences of instruction are abstracted out to form a new function thus reducing code size. If so, then this type of optimization comes at a cost (speed) and is typically provided as a speed vs space option. As for debugging when this type of optimization is in use; if the debug file format is good enough, then it should not matter what type of optimization has occurred
if 3 functions (a, b and c) 'share' some common code (made common by the optimizer, not the source code) and I want to set a breakpoint when function b passes this 'optimized' code; how do you propose to stop there when only function b passes without any difference in operation/timing.

If you plan to come up with something that changes the operation of code under debug other than when the breakpoint is reached scrap them immediately!!!

Eri9k

List of 80 messages in thread
TopicAuthorDate
What\'s more important optimisations or debugging?            01/01/70 00:00      
   as Ive said before,you have two versions            01/01/70 00:00      
      Fly what you debug; Debug what you fly            01/01/70 00:00      
         Yes and No            01/01/70 00:00      
            and no...            01/01/70 00:00      
               You can not see what is not there            01/01/70 00:00      
                  Yes, that's the challenge            01/01/70 00:00      
               incorrct            01/01/70 00:00      
                  read the whole post            01/01/70 00:00      
                     comments            01/01/70 00:00      
                        I said: Ideally            01/01/70 00:00      
                           That's exactly the problem!            01/01/70 00:00      
                  Correct - but maybe mis-stated            01/01/70 00:00      
                     still not the same            01/01/70 00:00      
                        it would require some modifications elsewhere            01/01/70 00:00      
                           the catch            01/01/70 00:00      
                              sure            01/01/70 00:00      
      Buggy compilers.            01/01/70 00:00      
         not the most common reason            01/01/70 00:00      
            Probability.            01/01/70 00:00      
   Avoid problems            01/01/70 00:00      
   debugger not useful            01/01/70 00:00      
      'debuggers' ....            01/01/70 00:00      
         Re: Bullshit            01/01/70 00:00      
            why are you so determined on using ONE method            01/01/70 00:00      
               my silabs experince            01/01/70 00:00      
                  you must have your 'experience' from a VERY early            01/01/70 00:00      
   Make it right, then make it fast            01/01/70 00:00      
   Optimization misapplied - editorial            01/01/70 00:00      
      what is the difference...            01/01/70 00:00      
         this is the difference            01/01/70 00:00      
            I see, thanks.            01/01/70 00:00      
               with compiler optimizing ...            01/01/70 00:00      
   profile            01/01/70 00:00      
      So he should be able to sort it to give us BOTH!            01/01/70 00:00      
         on further analysis            01/01/70 00:00      
            PIG?            01/01/70 00:00      
               I got used to that misspell ....            01/01/70 00:00      
                  OT: PIC assembly language            01/01/70 00:00      
                     errata            01/01/70 00:00      
                        which you often can not get by with            01/01/70 00:00      
                           Takes the sting out            01/01/70 00:00      
                        PIC assembler is fun!            01/01/70 00:00      
                           well it does make more clones of me            01/01/70 00:00      
                           matter of personal preference, I think            01/01/70 00:00      
                              HT 8051 C too            01/01/70 00:00      
                                 Did you miss...            01/01/70 00:00      
         No hidden agenda            01/01/70 00:00      
            make it "Keil compatible"            01/01/70 00:00      
            "omniscient code generation"            01/01/70 00:00      
               OCG looks interesting            01/01/70 00:00      
               omniscient code generation            01/01/70 00:00      
                  Recursive Calls            01/01/70 00:00      
                     need recursive calls? use a chip that can take it            01/01/70 00:00      
                        RE: Recursive Calls            01/01/70 00:00      
                     RE: Recursive Calls            01/01/70 00:00      
                  Ryan, as I said above...            01/01/70 00:00      
                     optimization 1) and 2)            01/01/70 00:00      
                        procedural abstraction?            01/01/70 00:00      
                           debugging tools            01/01/70 00:00      
                              support for other tools and our own tool offering            01/01/70 00:00      
                                 tools etc.            01/01/70 00:00      
                                    beta testers?            01/01/70 00:00      
                                       advertisement?            01/01/70 00:00      
                                       Link is fine            01/01/70 00:00      
                                          silabs?            01/01/70 00:00      
                                             chips            01/01/70 00:00      
                                                Future direction            01/01/70 00:00      
                                                   do not forget the 'economy ones'            01/01/70 00:00      
                                                      "economy" ICE...            01/01/70 00:00      
                                                         SILabs does not belong            01/01/70 00:00      
                                                            ICE cathegorisation (C) Erik            01/01/70 00:00      
                              my guess            01/01/70 00:00      
                                 the big shots...            01/01/70 00:00      
                           please elaborate            01/01/70 00:00      
                              Debugging challenges            01/01/70 00:00      
   It depends on what you want            01/01/70 00:00      
   Realy What is Important, is there any paradox            01/01/70 00:00      
   AVR or C51 ,PLease Choose ?!!            01/01/70 00:00      
   Embedded System or your noodle Code            01/01/70 00:00      

Back to Subject List