Re: [PATCH v6 2/3] PCI: Add device-specific reset for Qualcomm devices

Jeff Johnson <[email protected]> Tue, 2 Jun 2026 10:10:13 -0700
Newsgroups dev.linux.lists.mhi,org.infradead.lists.ath11k,org.infradead.lists.ath12k,org.kernel.vger.linux-kernel,org.kernel.vger.linux-pci,org.kernel.vger.linux-wireless
Message-ID <[email protected]>
On 6/2/2026 9:00 AM, Jose Ignacio Tornos Martinez wrote:
> Some Qualcomm PCIe devices (ath11k WiFi, ath12k WiFi, SDX62/SDX65 modems)

Throughout your series you refer to ath11k and ath12k as if they are devices.
They are not. They are device drivers, each of which supports a large number
of devices.

> lack working reset methods for VFIO passthrough scenarios. These devices
> have no FLR capability, advertise NoSoftRst+ (blocking PM reset), and have
> broken bus reset.
> 
> The problem manifests in VFIO passthrough scenarios:
> 
> - ath11k WiFi (17cb:1103): Normal VM operation works fine, including

17cb:1103 is WCN6855

>   clean shutdown/reboot. However, when the VM terminates uncleanly
>   (crash, force-off), VFIO attempts to reset the device before it can
>   be assigned to another VM. Without a working reset method, the device
>   remains in an undefined state, preventing reuse.
> 
> - ath12k WiFi (17cb:1107): Same behavior as ath11k.

17cb:1107 is WCN7850

> 
> - SDX62/SDX65 5G modems (17cb:0308): Never successfully initialize even
>   on first VM assignment without proper reset capability.
...

> @@ -4252,6 +4268,9 @@ static const struct pci_dev_reset_methods pci_dev_reset_methods[] = {
>  		reset_chelsio_generic_dev },
>  	{ PCI_VENDOR_ID_HUAWEI, PCI_DEVICE_ID_HINIC_VF,
>  		reset_hinic_vf_dev },
> +	{ PCI_VENDOR_ID_QCOM, 0x1103, reset_d3cold_d3hot },  /* ath11k */

s/ath11k/WCN6855/

> +	{ PCI_VENDOR_ID_QCOM, 0x1107, reset_d3cold_d3hot },  /* ath12k */

s/ath12k/WCN7850/

> +	{ PCI_VENDOR_ID_QCOM, 0x0308, reset_d3cold_d3hot },  /* SDX62/SDX65 */
>  	{ 0 }
>  };
>  

same guidance applies to the 3/3 patch

/jeff