Re: [PATCH rc v7 5/6] iommu: Fix ATS invalidation timeouts during __iommu_remove_group_pasid()

[Nicolin Chen <nicolinc@nvidia.com>]

On Tue, Apr 21, 2026 at 07:15:34AM +0000, Tian, Kevin wrote:
> > From: Nicolin Chen <nicolinc@nvidia.com>
> > Sent: Sunday, April 19, 2026 7:42 AM
> > 
> > If a device is blocked, its PASID domains are already detached. Repeating
> > iommu_remove_dev_pasid() is unnecessary and might trigger ATS
> > invalidation
> > timeouts.
> > 
> > Skip the iommu_remove_dev_pasid() call upon the blocked flag.
> > 
> > Fixes: c279e83953d9 ("iommu: Introduce
> > pci_dev_reset_iommu_prepare/done()")
> > Cc: stable@vger.kernel.org
> > Reported-by: Kevin Tian <kevin.tian@intel.com>
> 
> It's not reported by me. Instead, I just tried to copy the words from Sashiko (same
> in next patch).

I think you should take the credit, though maybe we could add:

Assisted-by: Gemini:gemini-3.1-pro-preview
 
> btw not sure whether Joerg plans to enable Sashiko auto-reply to the iommu
> list. I feel most suggestions there are positive. 

That sounds helpful to me.

> If it won't happen near term, suggest you find a way to check Sashilo manually
> for future patches.

I use AI reviewer too, but it didn't report these corner cases.
I will add Sashiko (gemini-3.1-pro-preview) to my routine.

Thanks
Nicolin