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

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
10/12/11 21:03
Read: times


 
#184196 - yes
Responding to: ???'s previous message
Andy Neil said:
Andy Neil said:
You can never rely on timing with High-Level Languages!

Surely, you've been a member here for long enough to know that?!

eg, see: http://www.8052.com/forum/read/162556 - but, as you say, even assembler is not a solution when pipelines, caches, and suchlike come into play...

I knew that even before I became a member here in 2001, but many thanks to this forum I learned many things.
What I meant by delays in C is writing them with inline assembley which most compilers do not optimize. But looks like we have to change our coding philosophy with ARM. So is interrupt latency fixed on ARM or does it follow pipelines caches etc... I have to experiment with timer interrupts with ARM to see if I get precise time intervals. looks like the ghost of protected modes of Pentiums, winxp and above is going to be transferred to ARMs and ruin our realtime work.
Mahmood

List of 7 messages in thread
TopicAuthorDate
ARM timming            01/01/70 00:00      
   Yes - that *is* a problem!            01/01/70 00:00      
      You can never rely on timing with High-Level Languages!            01/01/70 00:00      
         yes            01/01/70 00:00      
            coding philosophy            01/01/70 00:00      
               "multicore"            01/01/70 00:00      
                  RE: "multicore"            01/01/70 00:00      

Back to Subject List