Re: [Qemu-devel] [RFC PATCH v7 08/21] replay: interrupts and exceptions

["Pavel Dovgaluk" <Pavel.Dovgaluk@ispras.ru>]

> From: Paolo Bonzini [mailto:paolo.bonzini@gmail.com] On Behalf Of Paolo Bonzini
> On 14/01/2015 10:07, Pavel Dovgaluk wrote:
> >> From: Paolo Bonzini [mailto:pbonzini@redhat.com]
> >> On 12/01/2015 13:40, Pavel Dovgaluk wrote:
> >>>>>
> >>>>> Perhaps check the replay_interrupt() outside, in an && with "if
> >>>>> (unlikely(interrupt_request))"?
> >>> You mean that I should wrap whole condition into "unlikely"?
> >>>
> >>
> >> No, I wanted to have a single check of "replay_interrupt()" and/or
> >> "replay_has_interrupt()".
> >>
> >> BTW, I think this is incorrect:
> >>
> >>> +                    if ((replay_mode != REPLAY_MODE_PLAY
> >>> +                            || replay_has_interrupt())
> >>> +                        && cc->cpu_exec_interrupt(cpu, interrupt_request)) {
> >>> +                        replay_interrupt();
> >>
> >> because cc->cpu_exec_interrupt() can exit with cpu_loop_exit(cpu).
> >
> > Haven't found any. Do you have an example?
> 
> Yes:
> 
>     cpu_svm_check_intercept_param ->
>     helper_svm_check_intercept_param ->
>     helper_vmexit -> cpu_loop_exit

Thanks.

> >>     if (replay_mode == REPLAY_MODE_PLAY && !replay_has_interrupt()) {
> >>         /* do nothing */
> >>     } else if (interrupt_request & CPU_INTERRUPT_HALT) {
> >>         replay_interrupt();
> >>         ...
> >>         cpu_loop_exit(cpu);
> >>     } else if (interrupt_request & CPU_INTERRUPT_INIT) {
> >>         replay_interrupt();
> >>         ...
> >>         cpu_loop_exit(cpu);
> >>     } else {
> >>         replay_interrupt();
> >>         if (cc->cpu_exec_interrupt(cpu, interrupt_request)) {
> >>             next_tb = 0;
> >>         }
> >>     }
> >
> > Is it normal that processing of the reset request does not execute cpu_loop_exit(cpu)?
> 
> I think it is okay.  INIT executes cpu_loop_exit() on x86 because
> processors other than the boot processor are halted after they receive INIT.

Then I cannot put everything in one if-else chain because it will change the behavior of the code.
After processing RESET branch we can also process hardware interrupts (in unmodified code).

Pavel Dovgalyuk