Re: [PATCH 1/2] iommu/vt-d: Skip dev-iotlb flush for inaccessible PCIe device without scalable mode

[Baolu Lu <baolu.lu@linux.intel.com>]

On 12/11/25 12:17, Jinhui Guo wrote:
> On Thu, Dec 11, 2025 10:10:52AM +0800, Baolu Lu wrote:
>>> Since __context_flush_dev_iotlb() is only called on
>>> {attach,release}_dev paths (not hot), add pci_device_is_present()
>>> there to skip inaccessible devices and avoid the hard-lock.
>>>
>>> Fixes: 37764b952e1b ("iommu/vt-d: Global devTLB flush when present context entry changed")
>>> Fixes: 81e921fd3216 ("iommu/vt-d: Fix NULL domain on device release")
>> Cc:stable@vger.kernel.org
>>
>>> Signed-off-by: Jinhui Guo<guojinhui.liam@bytedance.com>
>>> ---
>>>    drivers/iommu/intel/pasid.c | 11 +++++++++++
>>>    1 file changed, 11 insertions(+)
>>>
>>> diff --git a/drivers/iommu/intel/pasid.c b/drivers/iommu/intel/pasid.c
>>> index 3e2255057079..b1e8eb6a6504 100644
>>> --- a/drivers/iommu/intel/pasid.c
>>> +++ b/drivers/iommu/intel/pasid.c
>>> @@ -1099,9 +1099,20 @@ int intel_pasid_setup_sm_context(struct device *dev)
>>>     */
>>>    static void __context_flush_dev_iotlb(struct device_domain_info *info)
>>>    {
>>> +	struct pci_dev *pdev;
>>> +
>>>    	if (!info->ats_enabled)
>>>    		return;
>>>    
>>> +	/*
>>> +	 * Skip dev-IOTLB flush for inaccessible PCIe devices to prevent the
>>> +	 * Intel IOMMU from waiting indefinitely for an ATS invalidation that
>>> +	 * cannot complete.
>>> +	 */
>>> +	pdev = dev_is_pci(info->dev) ? to_pci_dev(info->dev) : NULL;
>>> +	if (pdev && !pci_device_is_present(pdev))
>>> +		return;
>> Could simply be
>>
>> 	if (dev_is_pci(info->dev) &&
>> 	    !pci_device_is_present(to_pci_dev(info->dev)))
>> 		return;
>>
>> ?
>>
>>> +
>>>    	qi_flush_dev_iotlb(info->iommu, PCI_DEVID(info->bus, info->devfn),
>>>    			   info->pfsid, info->ats_qdep, 0, MAX_AGAW_PFN_WIDTH);
>>>    
>> Thanks,
>> baolu
> Hi, baolu
> 
> Thanks for your time and suggestions.
> 
> I’ve sent v2 (https://lore.kernel.org/all/20251211035946.2071-1- 
> guojinhui.liam@bytedance.com/)
> with the following changes:
> 
> 1. Simplified the pci_device_is_present() check in __context_flush_dev_iotlb().
> 2. Added Cc:stable@vger.kernel.org to both patches.

Thanks! I would suggest not updating the patch versions so frequently
next time. It is important to provide the reviewers with more time to
comment on your submissions. :-)

Thanks,
baolu