Atomic modesetting doesn't use struct drm_connector_funcs.dpms and the set function, drm_helper_connector_dpms(), wouldn't support it anyway. So keep the pointer to NULL.
Signed-off-by: Thomas Zimmermann tzimmermann@suse.de --- drivers/gpu/drm/bochs/bochs_kms.c | 1 - 1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/bochs/bochs_kms.c b/drivers/gpu/drm/bochs/bochs_kms.c index cc93ff74fbd8..7410b0ea4623 100644 --- a/drivers/gpu/drm/bochs/bochs_kms.c +++ b/drivers/gpu/drm/bochs/bochs_kms.c @@ -88,7 +88,6 @@ static const struct drm_connector_helper_funcs bochs_connector_connector_helper_ };
static const struct drm_connector_funcs bochs_connector_connector_funcs = { - .dpms = drm_helper_connector_dpms, .fill_modes = drm_helper_probe_single_connector_modes, .destroy = drm_connector_cleanup, .reset = drm_atomic_helper_connector_reset,
Atomic modesetting doesn't use struct drm_connector_funcs.dpms and the set function, drm_helper_connector_dpms(), wouldn't support it anyway. So keep the pointer to NULL.
Signed-off-by: Thomas Zimmermann tzimmermann@suse.de --- drivers/gpu/drm/udl/udl_connector.c | 1 - 1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/udl/udl_connector.c b/drivers/gpu/drm/udl/udl_connector.c index e9671d38b4a0..0afdfb0d1fe1 100644 --- a/drivers/gpu/drm/udl/udl_connector.c +++ b/drivers/gpu/drm/udl/udl_connector.c @@ -109,7 +109,6 @@ static const struct drm_connector_helper_funcs udl_connector_helper_funcs = { };
static const struct drm_connector_funcs udl_connector_funcs = { - .dpms = drm_helper_connector_dpms, .reset = drm_atomic_helper_connector_reset, .detect = udl_detect, .fill_modes = drm_helper_probe_single_connector_modes,
On Fri, Feb 07, 2020 at 03:16:02PM +0100, Thomas Zimmermann wrote:
Atomic modesetting doesn't use struct drm_connector_funcs.dpms and the set function, drm_helper_connector_dpms(), wouldn't support it anyway. So keep the pointer to NULL.
Signed-off-by: Thomas Zimmermann tzimmermann@suse.de
On both patches:
Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
I just suggested a drm_mode_config_validate() to Ville in his patch series, which runs in drm_dev_register(). Maybe we could add a check for that for atomic drivers? It's a bit a game of whack-a-mole otherwise :-) -Daniel
drivers/gpu/drm/udl/udl_connector.c | 1 - 1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/udl/udl_connector.c b/drivers/gpu/drm/udl/udl_connector.c index e9671d38b4a0..0afdfb0d1fe1 100644 --- a/drivers/gpu/drm/udl/udl_connector.c +++ b/drivers/gpu/drm/udl/udl_connector.c @@ -109,7 +109,6 @@ static const struct drm_connector_helper_funcs udl_connector_helper_funcs = { };
static const struct drm_connector_funcs udl_connector_funcs = {
- .dpms = drm_helper_connector_dpms, .reset = drm_atomic_helper_connector_reset, .detect = udl_detect, .fill_modes = drm_helper_probe_single_connector_modes,
-- 2.25.0
Hi
Am 07.02.20 um 17:41 schrieb Daniel Vetter:
On Fri, Feb 07, 2020 at 03:16:02PM +0100, Thomas Zimmermann wrote:
Atomic modesetting doesn't use struct drm_connector_funcs.dpms and the set function, drm_helper_connector_dpms(), wouldn't support it anyway. So keep the pointer to NULL.
Signed-off-by: Thomas Zimmermann tzimmermann@suse.de
On both patches:
Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
I just suggested a drm_mode_config_validate() to Ville in his patch series, which runs in drm_dev_register(). Maybe we could add a check for that for atomic drivers? It's a bit a game of whack-a-mole otherwise :-)
From the 'git grep' I'd say those where the only atomic drivers to set the callback. The others where non-atomic ones. OTOH, I recently thought on several occasions that this or that behaviour should be enforced at least by WARN_ON.
Best regards Thomas
-Daniel
drivers/gpu/drm/udl/udl_connector.c | 1 - 1 file changed, 1 deletion(-)
diff --git a/drivers/gpu/drm/udl/udl_connector.c b/drivers/gpu/drm/udl/udl_connector.c index e9671d38b4a0..0afdfb0d1fe1 100644 --- a/drivers/gpu/drm/udl/udl_connector.c +++ b/drivers/gpu/drm/udl/udl_connector.c @@ -109,7 +109,6 @@ static const struct drm_connector_helper_funcs udl_connector_helper_funcs = { };
static const struct drm_connector_funcs udl_connector_funcs = {
- .dpms = drm_helper_connector_dpms, .reset = drm_atomic_helper_connector_reset, .detect = udl_detect, .fill_modes = drm_helper_probe_single_connector_modes,
-- 2.25.0
On Mon, 10 Feb 2020 at 08:10, Thomas Zimmermann tzimmermann@suse.de wrote:
Hi
Am 07.02.20 um 17:41 schrieb Daniel Vetter:
On Fri, Feb 07, 2020 at 03:16:02PM +0100, Thomas Zimmermann wrote:
Atomic modesetting doesn't use struct drm_connector_funcs.dpms and the set function, drm_helper_connector_dpms(), wouldn't support it anyway. So keep the pointer to NULL.
Signed-off-by: Thomas Zimmermann tzimmermann@suse.de
On both patches:
Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
I just suggested a drm_mode_config_validate() to Ville in his patch series, which runs in drm_dev_register(). Maybe we could add a check for that for atomic drivers? It's a bit a game of whack-a-mole otherwise :-)
From the 'git grep' I'd say those where the only atomic drivers to set the callback. The others where non-atomic ones. OTOH, I recently thought on several occasions that this or that behaviour should be enforced at least by WARN_ON.
Having a WARN_ON was my first idea as I saw these patches fly-by.
Sure it will add some dmesg noise in the very odd corner-case. At the same time, it will high-light existing and future bugs so that we can address them promptly.
Emil
On Fri, Feb 07, 2020 at 03:16:01PM +0100, Thomas Zimmermann wrote:
Atomic modesetting doesn't use struct drm_connector_funcs.dpms and the set function, drm_helper_connector_dpms(), wouldn't support it anyway. So keep the pointer to NULL.
Signed-off-by: Thomas Zimmermann tzimmermann@suse.de
Acked-by: Gerd Hoffmann kraxel@redhat.com
dri-devel@lists.freedesktop.org
-
Daniel Vetter -
Emil Velikov -
Gerd Hoffmann -
Thomas Zimmermann