vgem (and our igt tests using vgem) need this. I suspect etnaviv will fare similarly.
Fixes: d5264ed3823a ("drm: Return -ENOTSUPP when called for KMS cap with a non-KMS driver") Cc: Michel Dänzer michel.daenzer@amd.com Cc: Alex Deucher alexander.deucher@amd.com Cc: Chris Wilson chris@chris-wilson.co.uk Signed-off-by: Daniel Vetter daniel.vetter@intel.com Signed-off-by: Daniel Vetter daniel.vetter@intel.com --- drivers/gpu/drm/drm_ioctl.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c index 0a900bd4575d..60bf95644739 100644 --- a/drivers/gpu/drm/drm_ioctl.c +++ b/drivers/gpu/drm/drm_ioctl.c @@ -234,10 +234,15 @@ static int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_
req->value = 0;
- /* Only one cap makes sense with a UMS driver: */ - if (req->capability == DRM_CAP_TIMESTAMP_MONOTONIC) { + /* Only some caps make sense with UMS/render-only drivers. */ + switch (req->capability) ) { + case DRM_CAP_TIMESTAMP_MONOTONIC: req->value = drm_timestamp_monotonic; return 0; + case DRM_CAP_PRIME: + req->value |= dev->driver->prime_fd_to_handle ? DRM_PRIME_CAP_IMPORT : 0; + req->value |= dev->driver->prime_handle_to_fd ? DRM_PRIME_CAP_EXPORT : 0; + return 0; }
/* Other caps only work with KMS drivers */ @@ -258,10 +263,6 @@ static int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_ case DRM_CAP_DUMB_PREFER_SHADOW: req->value = dev->mode_config.prefer_shadow; break; - case DRM_CAP_PRIME: - req->value |= dev->driver->prime_fd_to_handle ? DRM_PRIME_CAP_IMPORT : 0; - req->value |= dev->driver->prime_handle_to_fd ? DRM_PRIME_CAP_EXPORT : 0; - break; case DRM_CAP_ASYNC_PAGE_FLIP: req->value = dev->mode_config.async_page_flip; break;
On Wed, Dec 07, 2016 at 02:13:23PM +0100, Daniel Vetter wrote:
Slightly changes old behaviour but shouldn't we also be checking driver->features?
if (drm_core_check_feature(dev, DRIVER_PRIME)) { req->value |= dev->driver->prime_fd_to_handle ? DRM_PRIME_CAP_IMPORT : 0; req->value |= dev->driver->prime_handle_to_fd ? DRM_PRIME_CAP_EXPORT : 0; }
Paranoia says second patch as a potential change of abi. So,
Reviewed-by: Chris Wilson chris@chris-wilson.co.uk -Chris
vgem (and our igt tests using vgem) need this. I suspect etnaviv will fare similarly.
v2. Make it build. Oops.
Fixes: d5264ed3823a ("drm: Return -ENOTSUPP when called for KMS cap with a non-KMS driver") Cc: Michel Dänzer michel.daenzer@amd.com Cc: Alex Deucher alexander.deucher@amd.com Cc: Chris Wilson chris@chris-wilson.co.uk Reviewed-by: Chris Wilson chris@chris-wilson.co.uk Signed-off-by: Daniel Vetter daniel.vetter@intel.com --- drivers/gpu/drm/drm_ioctl.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c index 0a900bd4575d..a16b6fbd1004 100644 --- a/drivers/gpu/drm/drm_ioctl.c +++ b/drivers/gpu/drm/drm_ioctl.c @@ -234,10 +234,15 @@ static int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_
req->value = 0;
- /* Only one cap makes sense with a UMS driver: */ - if (req->capability == DRM_CAP_TIMESTAMP_MONOTONIC) { + /* Only some caps make sense with UMS/render-only drivers. */ + switch (req->capability) { + case DRM_CAP_TIMESTAMP_MONOTONIC: req->value = drm_timestamp_monotonic; return 0; + case DRM_CAP_PRIME: + req->value |= dev->driver->prime_fd_to_handle ? DRM_PRIME_CAP_IMPORT : 0; + req->value |= dev->driver->prime_handle_to_fd ? DRM_PRIME_CAP_EXPORT : 0; + return 0; }
/* Other caps only work with KMS drivers */ @@ -258,10 +263,6 @@ static int drm_getcap(struct drm_device *dev, void *data, struct drm_file *file_ case DRM_CAP_DUMB_PREFER_SHADOW: req->value = dev->mode_config.prefer_shadow; break; - case DRM_CAP_PRIME: - req->value |= dev->driver->prime_fd_to_handle ? DRM_PRIME_CAP_IMPORT : 0; - req->value |= dev->driver->prime_handle_to_fd ? DRM_PRIME_CAP_EXPORT : 0; - break; case DRM_CAP_ASYNC_PAGE_FLIP: req->value = dev->mode_config.async_page_flip; break;
On 07/12/16 11:49 PM, Daniel Vetter wrote:
Thanks Daniel, and sorry I missed this, guess I was thinking of !MODESET too much in terms of UMS and too little in terms of render-only drivers.
Reviewed-by: Michel Dänzer michel.daenzer@amd.com
On Thu, Dec 08, 2016 at 12:05:48AM +0900, Michel Dänzer wrote:
Thanks for the review, applied to drm-misc. -Daniel
dri-devel@lists.freedesktop.org