On 02-06-2025 08:14, Parav Pandit wrote:
When the PCI device is surprise removed, requests may not complete the device as the VQ is marked as broken. Due to this, the disk deletion hangs.
Fix it by aborting the requests when the VQ is broken.
With this fix now fio completes swiftly. An alternative of IO timeout has been considered, however when the driver knows about unresponsive block device, swiftly clearing them enables users and upper layers to react quickly.
Verified with multiple device unplug iterations with pending requests in virtio used ring and some pending with the device.
Fixes: 43bb40c5b926 ("virtio_pci: Support surprise removal of virtio pci device") Cc:stable@vger.kernel.org Reported-by: Li RongQinglirongqing@baidu.com Closes:https://urldefense.com/v3/__https://lore.kernel.org/virtualization/ c45dd68698cd47238c55fb73ca9b4741@baidu.com/__;!!ACWV5N9M2RV99hQ! OKusQ3sGH7nkVTy9MUDtaIS17UDtjabP3Lby9jTBdT3Ur38XLl2_DOFo0eVzx_dNeh16lD3Ss6ItE9eG$ Reviewed-by: Max Gurtovoymgurtovoy@nvidia.com Reviewed-by: Israel Rukshinisraelr@nvidia.com Signed-off-by: Parav Panditparav@nvidia.com
look good to me.
Reviewed-by: Alok Tiwari alok.a.tiwari@oracle.com
Thanks, Alok