During a warm reset via kexec, the system bypasses the driver removal sequence, meaning that the remove() callback is not invoked. If a QAT device is not shutdown properly, the device driver will fail to load in a newly rebooted kernel.
This might result in output like the following after the kexec reboot:
420xx 0000:01:00.0: Failed to power up the device 420xx 0000:01:00.0: Failed to initialize device 420xx 0000:01:00.0: Resetting device qat_dev0 420xx 0000:01:00.0: probe with driver 420xx failed with error -14
Implement the shutdown() handler that hooks into the reboot notifier list. This brings down the QAT device and ensures it is shut down properly.
Cc: stable@vger.kernel.org Fixes: fcf60f4bcf54 ("crypto: qat - add support for 420xx devices") Reviewed-by: Ahsan Atta ahsan.atta@intel.com Reviewed-by: Andy Shevchenko andriy.shevchenko@linux.intel.com Signed-off-by: Giovanni Cabiddu giovanni.cabiddu@intel.com --- drivers/crypto/intel/qat/qat_420xx/adf_drv.c | 8 ++++++++ 1 file changed, 8 insertions(+)
diff --git a/drivers/crypto/intel/qat/qat_420xx/adf_drv.c b/drivers/crypto/intel/qat/qat_420xx/adf_drv.c index 8084aa0f7f41..b4731f02deb8 100644 --- a/drivers/crypto/intel/qat/qat_420xx/adf_drv.c +++ b/drivers/crypto/intel/qat/qat_420xx/adf_drv.c @@ -186,11 +186,19 @@ static void adf_remove(struct pci_dev *pdev) adf_cleanup_accel(accel_dev); }
+static void adf_shutdown(struct pci_dev *pdev) +{ + struct adf_accel_dev *accel_dev = adf_devmgr_pci_to_accel_dev(pdev); + + adf_dev_down(accel_dev); +} + static struct pci_driver adf_driver = { .id_table = adf_pci_tbl, .name = ADF_420XX_DEVICE_NAME, .probe = adf_probe, .remove = adf_remove, + .shutdown = adf_shutdown, .sriov_configure = adf_sriov_configure, .err_handler = &adf_err_handler, };