qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH 4/9] sparc64: check for pending irq when pil, p


From: Blue Swirl
Subject: Re: [Qemu-devel] [PATCH 4/9] sparc64: check for pending irq when pil, pstate or softint is changed
Date: Wed, 6 Jan 2010 15:54:19 +0000

On Tue, Jan 5, 2010 at 11:19 PM, Igor V. Kovalenko
<address@hidden> wrote:
> From: Igor V. Kovalenko <address@hidden>
>
> Signed-off-by: Igor V. Kovalenko <address@hidden>
> ---
>  target-sparc/op_helper.c |   39 ++++++++++++++++++++++++++++++++++++---
>  1 files changed, 36 insertions(+), 3 deletions(-)
>
> diff --git a/target-sparc/op_helper.c b/target-sparc/op_helper.c
> index a7da0e4..b1978cb 100644
> --- a/target-sparc/op_helper.c
> +++ b/target-sparc/op_helper.c
> @@ -3301,6 +3301,12 @@ static inline void change_pstate(uint64_t new_pstate)
>  void helper_wrpstate(target_ulong new_state)
>  {
>     change_pstate(new_state & 0xf3f);
> +
> +#if !defined(CONFIG_USER_ONLY)
> +    if (env->pstate & PS_IE) {
> +        cpu_check_irqs(env);
> +    }
> +#endif
>  }
>
>  void helper_wrpil(target_ulong new_pil)
> @@ -3328,6 +3334,14 @@ void helper_done(void)
>     change_pstate((tsptr->tstate >> 8) & 0xf3f);
>     PUT_CWP64(env, tsptr->tstate & 0xff);
>     env->tl--;
> +
> +    DPRINTF_PSTATE("... helper_done tl=%d\n", env->tl);
> +
> +#if !defined(CONFIG_USER_ONLY)
> +    if (env->pstate & PS_IE) {
> +        cpu_check_irqs(env);
> +    }
> +#endif
>  }
>
>  void helper_retry(void)
> @@ -3341,21 +3355,40 @@ void helper_retry(void)
>     change_pstate((tsptr->tstate >> 8) & 0xf3f);
>     PUT_CWP64(env, tsptr->tstate & 0xff);
>     env->tl--;
> +
> +    DPRINTF_PSTATE("... helper_retry tl=%d\n", env->tl);
> +
> +#if !defined(CONFIG_USER_ONLY)
> +    if (env->pstate & PS_IE) {
> +        cpu_check_irqs(env);
> +    }
> +#endif
> +}
> +
> +static void do_modify_softint(const char* operation, uint32_t value)
> +{
> +    if (env->softint != value) {
> +        env->softint = value;
> +        DPRINTF_PSTATE(": %s new %08X\n", operation, env->softint);

Uppercase hex again. Otherwise the patch looks OK.

Maybe the #if CONFIG_USER_ONLY/#endif could be avoided if
cpu_check_irqs would be #defined as a dummy macro for CONFIG_USER_ONLY
in cpu.h.




reply via email to

[Prev in Thread] Current Thread [Next in Thread]