Re: [PATCH for-8.0 4/7] target/ppc: Use QEMU_IOTHREAD_LOCK_GUARD in cpu_interrupt_exittb

[Richard Henderson]

On 11/18/22 02:13, Daniel Henrique Barboza wrote:
> On 11/18/22 06:18, Richard Henderson wrote:
> > In addition, use tcg_enabled instead of !kvm_enabled.
> >
> > Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> > ---
>
> Should we strive for this change (tcg_enabled instead of !kvm_enabled)
> everywhere when applicable? There's a lot of places in the ppc code where
> this can be done.

I think it's better style, since that's generally what is meant.

It's important when the target supports multiple accelerators.  A test for !kvm begs the 
question of why we aren't also testing for e.g. !hvf.  I've noticed a couple of these in 
the code base.


r~

> Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
>
>
> > Cc: qemu-ppc@nongnu.org
> > ---
> >   target/ppc/helper_regs.c | 14 ++++----------
> >   1 file changed, 4 insertions(+), 10 deletions(-)
> >
> > diff --git a/target/ppc/helper_regs.c b/target/ppc/helper_regs.c
> > index c0aee5855b..779e7db513 100644
> > --- a/target/ppc/helper_regs.c
> > +++ b/target/ppc/helper_regs.c
> > @@ -22,6 +22,7 @@
> >   #include "qemu/main-loop.h"
> >   #include "exec/exec-all.h"
> >   #include "sysemu/kvm.h"
> > +#include "sysemu/tcg.h"
> >   #include "helper_regs.h"
> >   #include "power8-pmu.h"
> >   #include "cpu-models.h"
> > @@ -203,17 +204,10 @@ void cpu_interrupt_exittb(CPUState *cs)
> >   {
> >       /*
> >        * We don't need to worry about translation blocks
> > -     * when running with KVM.
> > +     * unless running with TCG.
> >        */
> > -    if (kvm_enabled()) {
> > -        return;
> > -    }
> > -
> > -    if (!qemu_mutex_iothread_locked()) {
> > -        qemu_mutex_lock_iothread();
> > -        cpu_interrupt(cs, CPU_INTERRUPT_EXITTB);
> > -        qemu_mutex_unlock_iothread();
> > -    } else {
> > +    if (tcg_enabled()) {
> > +        QEMU_IOTHREAD_LOCK_GUARD();
> >           cpu_interrupt(cs, CPU_INTERRUPT_EXITTB);
> >       }
> >   }