On 4/2/24 10:07 PM, Jerry Snitselaar wrote:
> Currenty if the user driver is probed on a workqueue configured for
> another driver with SVA not enabled on the system, it will print
> out a number of probe failing messages like the following:
>
> [ 264.831140] user: probe of wq13.0 failed with error -95
>
> On some systems, such as GNR, the number of messages can
> reach over 100.
>
> Move the SVA feature check to be after the driver name match
> check.
>
> Cc: Vinod Koul <vkoul@xxxxxxxxxx>
> Cc: dmaengine@xxxxxxxxxxxxxxx
> Cc: linux-kernel@xxxxxxxxxxxxxxx
> Signed-off-by: Jerry Snitselaar <jsnitsel@xxxxxxxxxx>
Seems reasonable. Thanks!
Reviewed-by: Dave Jiang <dave.jiang@xxxxxxxxx>
> ---
> drivers/dma/idxd/cdev.c | 17 +++++++++--------
> 1 file changed, 9 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/dma/idxd/cdev.c b/drivers/dma/idxd/cdev.c
> index 8078ab9acfbc..a4b771781afc 100644
> --- a/drivers/dma/idxd/cdev.c
> +++ b/drivers/dma/idxd/cdev.c
> @@ -517,6 +517,14 @@ static int idxd_user_drv_probe(struct idxd_dev *idxd_dev)
> if (idxd->state != IDXD_DEV_ENABLED)
> return -ENXIO;
>
> + mutex_lock(&wq->wq_lock);
> +
> + if (!idxd_wq_driver_name_match(wq, dev)) {
> + idxd->cmd_status = IDXD_SCMD_WQ_NO_DRV_NAME;
> + rc = -ENODEV;
> + goto wq_err;
> + }
> +
> /*
> * User type WQ is enabled only when SVA is enabled for two reasons:
> * - If no IOMMU or IOMMU Passthrough without SVA, userspace
> @@ -532,14 +540,7 @@ static int idxd_user_drv_probe(struct idxd_dev *idxd_dev)
> dev_dbg(&idxd->pdev->dev,
> "User type WQ cannot be enabled without SVA.\n");
>
> - return -EOPNOTSUPP;
> - }
> -
> - mutex_lock(&wq->wq_lock);
> -
> - if (!idxd_wq_driver_name_match(wq, dev)) {
> - idxd->cmd_status = IDXD_SCMD_WQ_NO_DRV_NAME;
> - rc = -ENODEV;
> + rc = -EOPNOTSUPP;
> goto wq_err;
> }
>
