Re: [PATCH v7 19/20] target/arm: Rebuild hflags for M-profile.

[Peter Maydell]

On Thu, 17 Oct 2019 at 19:51, Richard Henderson
<address@hidden> wrote:
>
> Continue setting, but not relying upon, env->hflags.
>
> Reviewed-by: Alex Bennée <address@hidden>
> Signed-off-by: Richard Henderson <address@hidden>
> ---
> v7: Add rebuilds for v7m_msr and nvic_writel to v7m.ccr.
> ---
>  hw/intc/armv7m_nvic.c  | 1 +
>  target/arm/m_helper.c  | 6 ++++++
>  target/arm/translate.c | 5 ++++-
>  3 files changed, 11 insertions(+), 1 deletion(-)
>
> diff --git a/hw/intc/armv7m_nvic.c b/hw/intc/armv7m_nvic.c
> index 8e93e51e81..a3993e7b72 100644
> --- a/hw/intc/armv7m_nvic.c
> +++ b/hw/intc/armv7m_nvic.c
> @@ -1604,6 +1604,7 @@ static void nvic_writel(NVICState *s, uint32_t offset, 
> uint32_t value,
>          }
>
>          cpu->env.v7m.ccr[attrs.secure] = value;
> +        arm_rebuild_hflags(&cpu->env);
>          break;
>      case 0xd24: /* System Handler Control and State (SHCSR) */
>          if (!arm_feature(&cpu->env, ARM_FEATURE_V7)) {

This seems fragile -- we have to remember to add in
a call to arm_rebuild_hflags() for every individual
case of a memory-mapped system register that we choose
to cache in tb flags. It doesn't seem consistent with
the choice for A-profile to rebuild hflags for pretty
much any sysreg write. Maybe it would be better to just
always rebuild hflags at the end of nvic_sysreg_write() ?

thanks
-- PMM