Quoting Ali Saidi <[email protected]>:
On Fri, 8 Jul 2011 10:38:51 -0700, nathan binkert <[email protected]> wrote:
In this case shouldn't the sti instruction cause some form of serialization
in the pipeline and make the cli wait in fetch until it completes?
Otherwise both could go down the pipe and commit together, never allowing
interrupts to happen. But again, I'm not entirely clear how the ISA should
work in respect to this type of instruction.
Is it really necessary for the pipeline to have such a large bubble?
In theory an interrupt should just be like a branch that changes the
processor mode. I believe that some machines even insert an
instruction into the pipeline to do the interrupt.
Nate
_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev
It doesn't have a large bubble, it's committing instructions until
the end. We inject a fault that is the interrupt, so that is exactly
what wer'e doing.
Ali
_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev
This all doesn't change the fact that we would never recognize the interrupt.
Gabe
_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev
