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

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
06/04/10 16:34
Read: times


 
#176389 - so exactly when does the hardware check the interrupt regs
Responding to: ???'s previous message
Erik Malund said:
No I/O function (UART, timer, whatsoever) causes an interrupt!
the UART just set TX, RX, the timers just set TF, .....


This help clarify the process in my mind. Thanks!

Erik Malund said:
Then the processor at certain times in the instruction cycle check these flags and the IE and the previous instruction and possibly decides to vector to an ISR.

So, to address your actual question, whatever the EA does, it will never disable the setting of a TF and it will never reset it.

Erik


So to address the question behind the question, when does the uC hardware check the interrupt flag registers? Once per instruction? Once per machine cycle?

The 8051 documentation seems to indicate that there is something magical that happens under two specific circumstances; the instant an interrupt flag is raised and on the RETI instruction. But based on experimentation and the discussion so far, this is clearly a misinterpretation.

Is it that the classic 8051 behaved that way and most or all modern implementations are more sophisticated, checking the interrupt registers regularly?

--David

List of 8 messages in thread
TopicAuthorDate
Interrupt Hardware checked on SETB EA?            01/01/70 00:00      
   Yes...            01/01/70 00:00      
   the way to see it            01/01/70 00:00      
      so exactly when does the hardware check the interrupt regs            01/01/70 00:00      
         VERY derivative dependent            01/01/70 00:00      
         Datasheet should tell you            01/01/70 00:00      
            ports? machine cycle?            01/01/70 00:00      
               Still the datasheet will say.            01/01/70 00:00      

Back to Subject List