[PATCH v5 02/17] iommu/arm-smmu-v3: Consolidate the STE generation for abort/bypass

[Will Deacon]

On Fri, Feb 23, 2024 at 02:53:58PM -0400, Jason Gunthorpe wrote:
> On Thu, Feb 15, 2024 at 05:27:09PM +0000, Robin Murphy wrote:
> > > @@ -1583,22 +1595,20 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
> > >   	arm_smmu_write_ste(master, sid, dst, &target);
> > >   }
> > > +/*
> > > + * This can safely directly manipulate the STE memory without a sync sequence
> > > + * because the STE table has not been installed in the SMMU yet.
> > > + */
> > >   static void arm_smmu_init_bypass_stes(struct arm_smmu_ste *strtab,
> > 
> > This name is long out-of-date - if we're refreshing this area, please rename
> > to something relevant to what it actually does, e.g. s/bypass/initial/.
> 
> Done
> 
> > Although frankly I also think that at this point we should just get rid of
> > the disable_bypass parameter altogether - it's been almost entirely
> > meaningless since default domain support was added, and any tenuous cases
> > for wanting inital STEs to be bypass should probably be using RMRs now
> > anyway.
> 
> I can write the patch for this if you and Will agree

Yes, please! I'll be glad to see the back of that option. Since you just
posted v6 of your "part 1", feel free to send a patch which applies on top
of that (rather than having to rebase the whole shebang).

Will