RE: [PATCH 7/9] iommu/vt-d: Remove the 'force_on' variable

[Tian, Kevin]

> From: Baolu Lu <baolu.lu@xxxxxxxxxxxxxxx>
> Sent: Friday, June 12, 2026 10:17 PM
> 
> On 6/4/2026 1:15 PM, Kevin Tian wrote:
> >   /*
> > - * set to 1 to panic kernel if can't successfully enable VT-d
> > + * set to 1 to avoid kernel panic if can't successfully enable VT-d
> >    * (used when kernel is launched w/ TXT)
> >    */
> 
> It might be cleaner to replace it entirely with a new comment. For
> example:
> 
> /*
>   * Skip forcing iommu enablement and avoid tboot-related kernel panics
>   * during initialization when set to 1 (via intel_iommu=tboot_noforce).
>   */
> int intel_iommu_tboot_noforce;

+1

> >
> > +static inline bool dmar_is_force_on(void)
> > +{
> > +	return dmar_state == DMAR_ENABLED_FORCE;
> > +}
> 
> This helper is only meaningful after
> 
> tboot_force_iommu();
> platform_optin_force_iommu();
> 
> have been called. It would be better to add a brief comment about this?
> 

IMHO it's obvious. The callers of this helper are limited and well spotted.