Re: [PATCH v9 08/12] iommu/tegra241-cmdqv: Add a helper to quiesce VCMDQs

[Pranjal Shrivastava <praan@google.com>]

On Tue, Aug 25, 2026 at 01:36:11PM -0300, Jason Gunthorpe wrote:
> > [ ... 73 lines skipped ... ]
> > @@ -467,6 +487,14 @@ static int tegra241_cmdqv_drain_vintf0_lvcmdqs(struct arm_smmu_device *smmu)
> >  	if (!READ_ONCE(vintf->enabled))
> >  		return 0;
> >  
> > +	/*
> > +	 * Gate all vCMDQs by setting the STOP_FLAG in a separate,
> > +	 * initial loop to ensure no new commands can be submitted
> > +	 * to any secondary queue while we are waiting to drain them.
> > +	 */
> > +	tegra241_cmdqv_quiesce_vintf0_lvcmdqs(smmu);
> > +
> > +	/* Now that all queues are safely gated, drain them sequentially. */
> >  	for (lidx = 0; lidx < cmdqv->num_lvcmdqs_per_vintf; lidx++) {
> >  		struct tegra241_vcmdq *vcmdq = vintf->lvcmdqs[lidx];
> >  
> 
>  [Severity: High]
>  If arm_smmu_queue_poll_until_empty() fails during this sequential drain, it
>  returns an error without clearing the CMDQ_PROD_STOP_FLAG that was just set on
>  all queues:
> 
> Make sense to me, the earlier patch added:
> 
> 		ret = arm_smmu_queue_poll_until_empty(smmu, &vcmdq->cmdq.q);
> 		if (ret)
> 			break;
> 

Yes, I plan to treat this the same as the primary CMDQ and move on 
instead of break; I'll address this in v10.

Thanks,
Praan