On 6/16/2026 8:13 AM, Shameer Kolothum Thodi wrote:
-----Original Message-----
From: Nathan Chen<[email protected]>
Sent: 15 June 2026 21:49
To:[email protected]
Cc: Yi Liu<[email protected]>; Eric Auger<[email protected]>;
Zhenzhong Duan<[email protected]>; Philippe Mathieu-Daudé
<[email protected]>; Zhao Liu<[email protected]>; Alex Williamson
<[email protected]>; Cédric Le Goater<[email protected]>; Matt Ochs
<[email protected]>; Nicolin Chen<[email protected]>; Shameer
Kolothum Thodi<[email protected]>; Nathan Chen
<[email protected]>
Subject: [PATCH v2 3/3] vfio/pci: Warn when ats=on but host device lacks ATS
cap
From: Nathan Chen<[email protected]>
Emit a warning when ats=on is requested but the physical device does not
advertise ATS, since ATS cannot be exposed to the guest in this case.
Signed-off-by: Nathan Chen<[email protected]>
---
hw/vfio/pci.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/hw/vfio/pci.c b/hw/vfio/pci.c index 0f592ada89..49ba6813ee
100644
--- a/hw/vfio/pci.c
+++ b/hw/vfio/pci.c
@@ -2592,6 +2592,7 @@ static void vfio_add_ext_cap(VFIOPCIDevice
*vdev, bool ats_needed) {
PCIDevice *pdev = PCI_DEVICE(vdev);
bool pasid_cap_added = false;
+ bool ats_cap_present = false;
Error *err = NULL;
uint32_t header;
uint16_t cap_id, next, size;
@@ -2680,6 +2681,7 @@ static void vfio_add_ext_cap(VFIOPCIDevice
*vdev, bool ats_needed)
pcie_add_capability(pdev, cap_id, cap_ver, next, size);
break;
case PCI_EXT_CAP_ID_ATS:
+ ats_cap_present = true;
/*
* If ATS is requested and supported according to the kernel, add
* the ATS capability. If not supported according to the kernel
or @@ -
2699,6 +2701,11 @@ static void vfio_add_ext_cap(VFIOPCIDevice *vdev,
bool ats_needed)
error_report_err(err);
}
+ if (vdev->ats == ON_OFF_AUTO_ON && !ats_cap_present) {
Nit: Can we do:
If (ats_needed && !ats_cap_present) here?
Yes, I'll revise this in the next refresh.>
+ warn_report("vfio-pci: ats=on requested, but host device has no "
+ "ATS extended capability");
+ }
+
Also, why this has to be a separate patch? Can't we fold into patch #2?
I'll fold this into patch #2. I had made this separate since it was
suggested it could be a follow up change to the original series, but I
agree it can fit with the preceding patch's changes to vfio_add_ext_cap()
Thanks,
Nathan
