Re: [PATCH] replay: fix replay of the interrupts

[Paolo Bonzini]

On 19/01/21 13:39, Pavel Dovgalyuk wrote:
> Sometimes interrupt event comes at the same time with
> the virtual timers. In this case replay tries to proceed
> the timers, because deadline for them is zero.
> This patch allows processing interrupts and exceptions
> by entering the vCPU execution loop, when deadline is zero,
> but checkpoint associated with virtual timers is not ready
> to be replayed.
>
> Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgalyuk@ispras.ru>
> ---
>   accel/tcg/tcg-cpus-icount.c |    8 +++++++-
>   1 file changed, 7 insertions(+), 1 deletion(-)
>
> diff --git a/accel/tcg/tcg-cpus-icount.c b/accel/tcg/tcg-cpus-icount.c
> index 9f45432275..a6d2bb8a88 100644
> --- a/accel/tcg/tcg-cpus-icount.c
> +++ b/accel/tcg/tcg-cpus-icount.c
> @@ -81,7 +81,13 @@ void icount_handle_deadline(void)
>       int64_t deadline = qemu_clock_deadline_ns_all(QEMU_CLOCK_VIRTUAL,
>                                                     QEMU_TIMER_ATTR_ALL);
>   
> -    if (deadline == 0) {
> +    /*
> +     * Instructions, interrupts, and exceptions are processed in cpu-exec.
> +     * Don't interrupt cpu thread, when these events are waiting
> +     * (i.e., there is no checkpoint)
> +     */
> +    if (deadline == 0
> +        && (replay_mode == REPLAY_MODE_RECORD || replay_has_checkpoint())) {

Should this be replay_mode != REPLAY_MODE_PLAY || replay_has_checkpoint()?

Paolo

>           icount_notify_aio_contexts();
>       }
>   }