[Intel-gfx] [PATCH 05/59] drm/prime: Make DRIVER_PRIME a no-op

Emil Velikov emil.l.velikov at gmail.com
Mon Jun 17 17:53:21 UTC 2019


On 2019/06/14, Daniel Vetter wrote:
> Drivers must fill out the handle_to_fd and fd_to_handle hooks to
> enable export/import prime functionality already. The additional
> DRIVER_PRIME flag doesn't serve any real purpose, since the overall
> flag doesn't even tell you whether import or export or maybe even both
> is supported.
> 
> Ditch it.
> 
> This patch just makes it defunct, subsequent patches will remove it
> from all the drivers.
> 
> Signed-off-by: Daniel Vetter <daniel.vetter at intel.com>
> ---
>  drivers/gpu/drm/drm_prime.c | 9 +--------
>  1 file changed, 1 insertion(+), 8 deletions(-)
> 
> diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
> index f08159a8b03a..78f6f10b2060 100644
> --- a/drivers/gpu/drm/drm_prime.c
> +++ b/drivers/gpu/drm/drm_prime.c
> @@ -47,8 +47,7 @@
>   * between applications, they can't be guessed like the globally unique GEM
>   * names.
>   *
> - * Drivers that support the PRIME API must set the DRIVER_PRIME bit in the
> - * &drm_driver.driver_features field, and implement the
> + * Drivers that support the PRIME API implement the
>   * &drm_driver.prime_handle_to_fd and &drm_driver.prime_fd_to_handle operations.
>   * GEM based drivers must use drm_gem_prime_handle_to_fd() an
>   * drm_gem_prime_fd_to_handle() to implement these. For GEM based drivers the
> @@ -361,9 +360,6 @@ int drm_prime_fd_to_handle_ioctl(struct drm_device *dev, void *data,
>  {
>  	struct drm_prime_handle *args = data;
>  
> -	if (!drm_core_check_feature(dev, DRIVER_PRIME))
> -		return -EOPNOTSUPP;
> -
>  	if (!dev->driver->prime_fd_to_handle)
>  		return -ENOSYS;
>  
Userspace will see a trivial change here EOPNOTSUPP vs ENOSYS. Although
I seriously doubt they had such fine-grained error checking and fallback
paths.

Please mention the changed return value in the commit message.

Thanks
Emil


More information about the Intel-gfx mailing list