dri-devel Archive on lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] drm/appletbdrm: Add suspend and resume support
@ 2026-09-24 19:25 Vasilij Markin via B4 Relay
  2026-09-24 19:33 ` sashiko-bot
                   ` (2 more replies)
  0 siblings, 3 replies; 6+ messages in thread
From: Vasilij Markin via B4 Relay @ 2026-09-24 19:25 UTC (permalink / raw)
  To: Aun-Ali Zaidi, Aditya Garg, Maarten Lankhorst, Maxime Ripard,
	Thomas Zimmermann, David Airlie, Simona Vetter
  Cc: dri-devel, linux-kernel, Kerem Karabay, Atharva Tiwari,
	Vasilij Markin

From: Vasilij Markin <vasilij.kernel@tototo.si>

The driver has no USB power management callbacks, so the USB core
unbinds it before system suspend and binds it again on resume. The DRM
device is unregistered and registered again on every suspend and resume
cycle, userspace driving the Touch Bar through it (such as tiny-dfr)
loses its device, and the Touch Bar stays dark after resume.

Implement suspend and resume with drm_mode_config_helper_suspend() and
drm_mode_config_helper_resume(), so the device stays registered and the
last committed state is restored on resume. If the device was reset
while suspended, repeat the probe-time handshake first.

Fixes: 0670c2f56e45 ("drm/tiny: add driver for Apple Touch Bars in x86 Macs")
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Vasilij Markin <vasilij.kernel@tototo.si>
---
Tested on a MacBookPro16,2 with this change backported to 7.2.7, deep (S3)
suspend by closing the lid, tiny-dfr 0.3.7. Without it the driver is probed
again on every resume and systemd stops tiny-dfr; with it tiny-dfr keeps
running, the driver is not probed again and the Touch Bar shows its content
right after resume. The reset_resume path is untested: the device never came
back reset.

Kernels before commit 3d95f93b30b8 ("drm/appletbdrm: Convert to atomic_create_state")
also need the plane reset hook to free the existing state, otherwise
drm_mode_config_reset() on resume hits WARN_ON(plane->state) in
appletbdrm_primary_plane_reset() and leaks the old state. 7.2.y is such a
kernel; the tested backport carries that change as a separate patch.

Built on drm-misc-next (c4c54ffd6) with W=1 without warnings.
---
 drivers/gpu/drm/tiny/appletbdrm.c | 43 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 43 insertions(+)

diff --git a/drivers/gpu/drm/tiny/appletbdrm.c b/drivers/gpu/drm/tiny/appletbdrm.c
index 4297d9c11..024ad8bf6 100644
--- a/drivers/gpu/drm/tiny/appletbdrm.c
+++ b/drivers/gpu/drm/tiny/appletbdrm.c
@@ -30,6 +30,7 @@
 #include <drm/drm_gem_atomic_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_gem_shmem_helper.h>
+#include <drm/drm_modeset_helper.h>
 #include <drm/drm_plane.h>
 #include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
@@ -821,6 +822,45 @@ static void appletbdrm_shutdown(struct usb_interface *intf)
 	drm_atomic_helper_shutdown(&adev->drm);
 }
 
+static int appletbdrm_suspend(struct usb_interface *intf, pm_message_t message)
+{
+	struct appletbdrm_device *adev = usb_get_intfdata(intf);
+
+	return drm_mode_config_helper_suspend(&adev->drm);
+}
+
+static int appletbdrm_resume(struct usb_interface *intf)
+{
+	struct appletbdrm_device *adev = usb_get_intfdata(intf);
+
+	return drm_mode_config_helper_resume(&adev->drm);
+}
+
+static int appletbdrm_reset_resume(struct usb_interface *intf)
+{
+	struct appletbdrm_device *adev = usb_get_intfdata(intf);
+	struct drm_device *drm = &adev->drm;
+	int ret;
+
+	/*
+	 * A reset puts the device back into its initial state, where it
+	 * expects the same handshake as during probe
+	 */
+	ret = appletbdrm_get_information(adev);
+	if (ret) {
+		drm_err(drm, "Failed to get display information\n");
+		return ret;
+	}
+
+	ret = appletbdrm_signal_readiness(adev);
+	if (ret) {
+		drm_err(drm, "Failed to signal readiness\n");
+		return ret;
+	}
+
+	return drm_mode_config_helper_resume(drm);
+}
+
 static const struct usb_device_id appletbdrm_usb_id_table[] = {
 	{ USB_DEVICE_INTERFACE_CLASS(0x05ac, 0x8302, USB_CLASS_AUDIO_VIDEO) },
 	{}
@@ -832,6 +872,9 @@ static struct usb_driver appletbdrm_usb_driver = {
 	.probe		= appletbdrm_probe,
 	.disconnect	= appletbdrm_disconnect,
 	.shutdown	= appletbdrm_shutdown,
+	.suspend	= appletbdrm_suspend,
+	.resume		= appletbdrm_resume,
+	.reset_resume	= appletbdrm_reset_resume,
 	.id_table	= appletbdrm_usb_id_table,
 };
 module_usb_driver(appletbdrm_usb_driver);

---
base-commit: c4c54ffd65555ac5316f8faf4d7dc8c815877937
change-id: 20260924-appletbdrm-pm-6bd220385beb

Best regards,
-- 
Vasilij Markin <vasilij.kernel@tototo.si>



^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH] drm/appletbdrm: Add suspend and resume support
  2026-09-24 19:25 [PATCH] drm/appletbdrm: Add suspend and resume support Vasilij Markin via B4 Relay
@ 2026-09-24 19:33 ` sashiko-bot
  2026-09-25 11:51 ` Aditya Garg
  2026-10-04 12:06 ` [PATCH v2] " Aditya Garg
  2 siblings, 0 replies; 6+ messages in thread
From: sashiko-bot @ 2026-09-24 19:33 UTC (permalink / raw)
  To: Vasilij Markin; +Cc: dri-devel

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] drm/appletbdrm: Memory leak of suspend_state on reset_resume error
--

commit 60e19da684ddfb4bbf66c6f0c2624ed415d96f12
Author: Vasilij Markin <vasilij.kernel@tototo.si>
Subject: drm/appletbdrm: Add suspend and resume support

This commit introduces suspend and resume support for the appletbdrm driver.
It uses drm_mode_config_helper_suspend() and drm_mode_config_helper_resume()
to keep the device registered across cycles. In case the device was reset while
suspended, it repeats the probe-time handshake.

> diff --git a/drivers/gpu/drm/tiny/appletbdrm.c b/drivers/gpu/drm/tiny/appletbdrm.c
> index 4297d9c1125e4..024ad8bf6e1c4 100644
> --- a/drivers/gpu/drm/tiny/appletbdrm.c
> +++ b/drivers/gpu/drm/tiny/appletbdrm.c
[ ... ]
> @@ -821,6 +822,45 @@ static void appletbdrm_shutdown(struct usb_interface *intf)
>  	drm_atomic_helper_shutdown(&adev->drm);
>  }
>  
> +static int appletbdrm_suspend(struct usb_interface *intf, pm_message_t message)
> +{
> +	struct appletbdrm_device *adev = usb_get_intfdata(intf);
> +
> +	return drm_mode_config_helper_suspend(&adev->drm);
> +}
> +
> +static int appletbdrm_resume(struct usb_interface *intf)
> +{
> +	struct appletbdrm_device *adev = usb_get_intfdata(intf);
> +
> +	return drm_mode_config_helper_resume(&adev->drm);
> +}
> +
> +static int appletbdrm_reset_resume(struct usb_interface *intf)
> +{
> +	struct appletbdrm_device *adev = usb_get_intfdata(intf);
> +	struct drm_device *drm = &adev->drm;
> +	int ret;
> +
> +	/*
> +	 * A reset puts the device back into its initial state, where it
> +	 * expects the same handshake as during probe
> +	 */
> +	ret = appletbdrm_get_information(adev);
> +	if (ret) {
> +		drm_err(drm, "Failed to get display information\n");
> +		return ret;

[Severity: Medium]
During system suspend, drm_mode_config_helper_suspend() duplicates the atomic
commit state and stores it in dev->mode_config.suspend_state.

If appletbdrm_get_information() fails here, we return early and bypass the call
to drm_mode_config_helper_resume(). Does this leak the suspend_state and its
associated DRM object references, since they never get freed or put by
drm_atomic_state_put()?

> +	}
> +
> +	ret = appletbdrm_signal_readiness(adev);
> +	if (ret) {
> +		drm_err(drm, "Failed to signal readiness\n");
> +		return ret;

[Severity: Medium]
Similar to the issue above, if appletbdrm_signal_readiness() fails, does
returning early without calling drm_mode_config_helper_resume() leak the atomic
commit state stored in dev->mode_config.suspend_state?

> +	}
> +
> +	return drm_mode_config_helper_resume(drm);
> +}
> +

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260924-appletbdrm-pm-v1-1-0e5432132161@tototo.si?part=1

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH] drm/appletbdrm: Add suspend and resume support
  2026-09-24 19:25 [PATCH] drm/appletbdrm: Add suspend and resume support Vasilij Markin via B4 Relay
  2026-09-24 19:33 ` sashiko-bot
@ 2026-09-25 11:51 ` Aditya Garg
  2026-10-04 12:06 ` [PATCH v2] " Aditya Garg
  2 siblings, 0 replies; 6+ messages in thread
From: Aditya Garg @ 2026-09-25 11:51 UTC (permalink / raw)
  To: vasilij.kernel, Aun-Ali Zaidi, Maarten Lankhorst, Maxime Ripard,
	Thomas Zimmermann, David Airlie, Simona Vetter
  Cc: dri-devel, linux-kernel, Kerem Karabay, Atharva Tiwari


Hi Vasilij

Thanks for the patch

> From: Vasilij Markin <vasilij.kernel@tototo.si>
> 
> The driver has no USB power management callbacks, so the USB core
> unbinds it before system suspend and binds it again on resume. The DRM
> device is unregistered and registered again on every suspend and resume
> cycle, userspace driving the Touch Bar through it (such as tiny-dfr)
> loses its device, and the Touch Bar stays dark after resume.
> 
> Implement suspend and resume with drm_mode_config_helper_suspend() and
> drm_mode_config_helper_resume(), so the device stays registered and the
> last committed state is restored on resume. If the device was reset
> while suspended, repeat the probe-time handshake first.
> 
> Fixes: 0670c2f56e45 ("drm/tiny: add driver for Apple Touch Bars in x86 Macs")
> Assisted-by: Claude:claude-opus-5-5
> Signed-off-by: Vasilij Markin <vasilij.kernel@tototo.si>
> ---
> Tested on a MacBookPro16,2 with this change backported to 7.2.7, deep (S3)
> suspend by closing the lid, tiny-dfr 0.3.7. Without it the driver is probed
> again on every resume and systemd stops tiny-dfr; with it tiny-dfr keeps
> running, the driver is not probed again and the Touch Bar shows its content
> right after resume. The reset_resume path is untested: the device never came
> back reset.
> 
> Kernels before commit 3d95f93b30b8 ("drm/appletbdrm: Convert to atomic_create_state")
> also need the plane reset hook to free the existing state, otherwise
> drm_mode_config_reset() on resume hits WARN_ON(plane->state) in
> appletbdrm_primary_plane_reset() and leaks the old state. 7.2.y is such a
> kernel; the tested backport carries that change as a separate patch.
> 
> Built on drm-misc-next (c4c54ffd6) with W=1 without warnings.
> ---
>   drivers/gpu/drm/tiny/appletbdrm.c | 43 +++++++++++++++++++++++++++++++++++++++
>   1 file changed, 43 insertions(+)
> 
> diff --git a/drivers/gpu/drm/tiny/appletbdrm.c b/drivers/gpu/drm/tiny/appletbdrm.c
> index 4297d9c11..024ad8bf6 100644
> --- a/drivers/gpu/drm/tiny/appletbdrm.c
> +++ b/drivers/gpu/drm/tiny/appletbdrm.c
> @@ -30,6 +30,7 @@
>   #include <drm/drm_gem_atomic_helper.h>
>   #include <drm/drm_gem_framebuffer_helper.h>
>   #include <drm/drm_gem_shmem_helper.h>
> +#include <drm/drm_modeset_helper.h>
>   #include <drm/drm_plane.h>
>   #include <drm/drm_print.h>
>   #include <drm/drm_probe_helper.h>
> @@ -821,6 +822,45 @@ static void appletbdrm_shutdown(struct usb_interface *intf)
>   	drm_atomic_helper_shutdown(&adev->drm);
>   }
>   
> +static int appletbdrm_suspend(struct usb_interface *intf, pm_message_t message)
> +{
> +	struct appletbdrm_device *adev = usb_get_intfdata(intf);
> +
> +	return drm_mode_config_helper_suspend(&adev->drm);
> +}
> +
> +static int appletbdrm_resume(struct usb_interface *intf)
> +{
> +	struct appletbdrm_device *adev = usb_get_intfdata(intf);
> +
> +	return drm_mode_config_helper_resume(&adev->drm);
> +}
> +
> +static int appletbdrm_reset_resume(struct usb_interface *intf)

We don't need a reset_resume function as the USB device is virtually 
exposed by the t2bce driver and current implementation doesn't reset the 
device in any case. If the need arises, it can be added in a later patch.

> +{
> +	struct appletbdrm_device *adev = usb_get_intfdata(intf);
> +	struct drm_device *drm = &adev->drm;
> +	int ret;
> +
> +	/*
> +	 * A reset puts the device back into its initial state, where it
> +	 * expects the same handshake as during probe
> +	 */
> +	ret = appletbdrm_get_information(adev);
> +	if (ret) {
> +		drm_err(drm, "Failed to get display information\n");
> +		return ret;
> +	}
> +
> +	ret = appletbdrm_signal_readiness(adev);
> +	if (ret) {
> +		drm_err(drm, "Failed to signal readiness\n");
> +		return ret;
> +	}
> +
> +	return drm_mode_config_helper_resume(drm);
> +}
> +
>   static const struct usb_device_id appletbdrm_usb_id_table[] = {
>   	{ USB_DEVICE_INTERFACE_CLASS(0x05ac, 0x8302, USB_CLASS_AUDIO_VIDEO) },
>   	{}
> @@ -832,6 +872,9 @@ static struct usb_driver appletbdrm_usb_driver = {
>   	.probe		= appletbdrm_probe,
>   	.disconnect	= appletbdrm_disconnect,
>   	.shutdown	= appletbdrm_shutdown,
> +	.suspend	= appletbdrm_suspend,
> +	.resume		= appletbdrm_resume,
> +	.reset_resume	= appletbdrm_reset_resume,
>   	.id_table	= appletbdrm_usb_id_table,
>   };
>   module_usb_driver(appletbdrm_usb_driver);
> 
> ---
> base-commit: c4c54ffd65555ac5316f8faf4d7dc8c815877937
> change-id: 20260924-appletbdrm-pm-6bd220385beb
> 
> Best regards,


^ permalink raw reply	[flat|nested] 6+ messages in thread

* [PATCH v2] drm/appletbdrm: Add suspend and resume support
  2026-09-24 19:25 [PATCH] drm/appletbdrm: Add suspend and resume support Vasilij Markin via B4 Relay
  2026-09-24 19:33 ` sashiko-bot
  2026-09-25 11:51 ` Aditya Garg
@ 2026-10-04 12:06 ` Aditya Garg
  2026-10-04 12:19   ` sashiko-bot
  2026-10-07 14:20   ` Aditya Garg
  2 siblings, 2 replies; 6+ messages in thread
From: Aditya Garg @ 2026-10-04 12:06 UTC (permalink / raw)
  To: vasilij.kernel, Aun-Ali Zaidi, Maarten Lankhorst, Maxime Ripard,
	Thomas Zimmermann, David Airlie, Simona Vetter
  Cc: dri-devel, linux-kernel, Kerem Karabay, Atharva Tiwari

From: Vasilij Markin <vasilij.kernel@tototo.si>

The driver has no USB power management callbacks, so the USB core
unbinds it before system suspend and binds it again on resume. The DRM
device is unregistered and registered again on every suspend and resume
cycle, userspace driving the Touch Bar through it (such as tiny-dfr)
loses its device, and the Touch Bar stays dark after resume.

Implement suspend and resume with drm_mode_config_helper_suspend() and
drm_mode_config_helper_resume(), so the device stays registered and the
last committed state is restored on resume. If the device was reset
while suspended, repeat the probe-time handshake first.

Fixes: 0670c2f56e45 ("drm/tiny: add driver for Apple Touch Bars in x86 Macs")
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Vasilij Markin <vasilij.kernel@tototo.si>
Signed-off-by: Aditya Garg <aditya.garg@linux.dev>
---
v2: Remove reset_resume

 drivers/gpu/drm/tiny/appletbdrm.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/gpu/drm/tiny/appletbdrm.c b/drivers/gpu/drm/tiny/appletbdrm.c
index ef71b9957..c94ed1068 100644
--- a/drivers/gpu/drm/tiny/appletbdrm.c
+++ b/drivers/gpu/drm/tiny/appletbdrm.c
@@ -30,6 +30,7 @@
 #include <drm/drm_gem_atomic_helper.h>
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_gem_shmem_helper.h>
+#include <drm/drm_modeset_helper.h>
 #include <drm/drm_plane.h>
 #include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
@@ -821,6 +822,20 @@ static void appletbdrm_shutdown(struct usb_interface *intf)
 	drm_atomic_helper_shutdown(&adev->drm);
 }
 
+static int appletbdrm_suspend(struct usb_interface *intf, pm_message_t message)
+{
+	struct appletbdrm_device *adev = usb_get_intfdata(intf);
+
+	return drm_mode_config_helper_suspend(&adev->drm);
+}
+
+static int appletbdrm_resume(struct usb_interface *intf)
+{
+	struct appletbdrm_device *adev = usb_get_intfdata(intf);
+
+	return drm_mode_config_helper_resume(&adev->drm);
+}
+
 static const struct usb_device_id appletbdrm_usb_id_table[] = {
 	{ USB_DEVICE_INTERFACE_CLASS(0x05ac, 0x8302, USB_CLASS_AUDIO_VIDEO) },
 	{}
@@ -832,6 +847,8 @@ static struct usb_driver appletbdrm_usb_driver = {
 	.probe		= appletbdrm_probe,
 	.disconnect	= appletbdrm_disconnect,
 	.shutdown	= appletbdrm_shutdown,
+	.suspend	= appletbdrm_suspend,
+	.resume		= appletbdrm_resume,
 	.id_table	= appletbdrm_usb_id_table,
 };
 module_usb_driver(appletbdrm_usb_driver);
-- 
2.53.0

^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH v2] drm/appletbdrm: Add suspend and resume support
  2026-10-04 12:06 ` [PATCH v2] " Aditya Garg
@ 2026-10-04 12:19   ` sashiko-bot
  2026-10-07 14:20   ` Aditya Garg
  1 sibling, 0 replies; 6+ messages in thread
From: sashiko-bot @ 2026-10-04 12:19 UTC (permalink / raw)
  To: Aditya Garg; +Cc: dri-devel

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] drm/appletbdrm: memory leak of suspend_state on unbind
- [Medium] drm/appletbdrm: missing reset_resume handler breaks reset recovery
--

commit c3fac1e92efd5498a32a50fb4e9a862ae7b24e77
Author: Vasilij Markin <vasilij.kernel@tototo.si>

drm/appletbdrm: Add suspend and resume support

The driver has no USB power management callbacks, so the USB core
unbinds it before system suspend and binds it again on resume. The DRM
device is unregistered and registered again on every suspend and resume
cycle, userspace driving the Touch Bar through it (such as tiny-dfr)
loses its device, and the Touch Bar stays dark after resume.

> Implement suspend and resume with drm_mode_config_helper_suspend() and
> drm_mode_config_helper_resume(), so the device stays registered and the
> last committed state is restored on resume. If the device was reset
> while suspended, repeat the probe-time handshake first.

[Severity: Medium]
Will the device actually stay registered if it is reset while suspended?

According to the Linux USB core logic, if a device is reset during suspend
and the driver lacks a .reset_resume callback, the USB core sets
needs_binding = 1, which forcefully unbinds and rebinds the driver.
This fallback unbinds the DRM device, destroying the userspace connection
and defeating the stated goal for the reset scenario.

> diff --git a/drivers/gpu/drm/tiny/appletbdrm.c b/drivers/gpu/drm/tiny/appletbdrm.c
> index 4297d9c1125e4..919ca14181af3 100644
> --- a/drivers/gpu/drm/tiny/appletbdrm.c
> +++ b/drivers/gpu/drm/tiny/appletbdrm.c
[ ... ]
> @@ -821,6 +822,20 @@ static void appletbdrm_shutdown(struct usb_interface *intf)
>  	drm_atomic_helper_shutdown(&adev->drm);
>  }
>  
> +static int appletbdrm_suspend(struct usb_interface *intf, pm_message_t message)
> +{
> +	struct appletbdrm_device *adev = usb_get_intfdata(intf);
> +
> +	return drm_mode_config_helper_suspend(&adev->drm);

[Severity: Medium]
Does this code leak adev->drm.mode_config.suspend_state if the device is
unbound while suspended?

drm_mode_config_helper_suspend() allocates an atomic state and stores it in
adev->drm.mode_config.suspend_state. 

If the device is reset during suspend, the USB core forcefully unbinds the
driver because it lacks a reset_resume handler. The unbind path calls
appletbdrm_disconnect() and tears down the DRM device.

Neither appletbdrm_disconnect() nor the DRM core cleanup routines free
suspend_state. Does this cause a memory leak of the atomic commit object
and all its duplicated plane/CRTC states on every suspend/reset cycle?

> +}
> +
> +static int appletbdrm_resume(struct usb_interface *intf)
[ ... ]
> @@ -832,6 +847,8 @@ static struct usb_driver appletbdrm_usb_driver = {
>  	.probe		= appletbdrm_probe,
>  	.disconnect	= appletbdrm_disconnect,
>  	.shutdown	= appletbdrm_shutdown,
> +	.suspend	= appletbdrm_suspend,
> +	.resume		= appletbdrm_resume,
>  	.id_table	= appletbdrm_usb_id_table,
>  };

[Severity: Medium]
Is the .reset_resume callback missing here?

The commit message states: "If the device was reset while suspended, repeat
the probe-time handshake first" to ensure the device stays registered.

However, intentionally omitting the .reset_resume callback causes the USB
core to unbind the driver on reset instead of restoring the DRM state.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/586d0b39-40c8-4f3d-a786-c1085f87b858@linux.dev?part=1

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH v2] drm/appletbdrm: Add suspend and resume support
  2026-10-04 12:06 ` [PATCH v2] " Aditya Garg
  2026-10-04 12:19   ` sashiko-bot
@ 2026-10-07 14:20   ` Aditya Garg
  1 sibling, 0 replies; 6+ messages in thread
From: Aditya Garg @ 2026-10-07 14:20 UTC (permalink / raw)
  To: vasilij.kernel, Aun-Ali Zaidi, Maarten Lankhorst, Maxime Ripard,
	Thomas Zimmermann, David Airlie, Simona Vetter
  Cc: dri-devel, linux-kernel, Kerem Karabay, Atharva Tiwari



> On 4 Oct 2026, at 5:36 PM, Aditya Garg <aditya.garg@linux.dev> wrote:
> 
> From: Vasilij Markin <vasilij.kernel@tototo.si>
> 
> The driver has no USB power management callbacks, so the USB core
> unbinds it before system suspend and binds it again on resume. The DRM
> device is unregistered and registered again on every suspend and resume
> cycle, userspace driving the Touch Bar through it (such as tiny-dfr)
> loses its device, and the Touch Bar stays dark after resume.
> 
> Implement suspend and resume with drm_mode_config_helper_suspend() and
> drm_mode_config_helper_resume(), so the device stays registered and the
> last committed state is restored on resume. If the device was reset
> while suspended, repeat the probe-time handshake first.
> 
> Fixes: 0670c2f56e45 ("drm/tiny: add driver for Apple Touch Bars in x86 Macs")
> Assisted-by: Claude:claude-opus-5-5
> Signed-off-by: Vasilij Markin <vasilij.kernel@tototo.si>
> Signed-off-by: Aditya Garg <aditya.garg@linux.dev>

Could anyone please merge it if they are ok with it? Thanks.
> ---
> v2: Remove reset_resume
> 
> drivers/gpu/drm/tiny/appletbdrm.c | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
> 
> diff --git a/drivers/gpu/drm/tiny/appletbdrm.c b/drivers/gpu/drm/tiny/appletbdrm.c
> index ef71b9957..c94ed1068 100644
> --- a/drivers/gpu/drm/tiny/appletbdrm.c
> +++ b/drivers/gpu/drm/tiny/appletbdrm.c
> @@ -30,6 +30,7 @@
> #include <drm/drm_gem_atomic_helper.h>
> #include <drm/drm_gem_framebuffer_helper.h>
> #include <drm/drm_gem_shmem_helper.h>
> +#include <drm/drm_modeset_helper.h>
> #include <drm/drm_plane.h>
> #include <drm/drm_print.h>
> #include <drm/drm_probe_helper.h>
> @@ -821,6 +822,20 @@ static void appletbdrm_shutdown(struct usb_interface *intf)
>    drm_atomic_helper_shutdown(&adev->drm);
> }
> 
> +static int appletbdrm_suspend(struct usb_interface *intf, pm_message_t message)
> +{
> +    struct appletbdrm_device *adev = usb_get_intfdata(intf);
> +
> +    return drm_mode_config_helper_suspend(&adev->drm);
> +}
> +
> +static int appletbdrm_resume(struct usb_interface *intf)
> +{
> +    struct appletbdrm_device *adev = usb_get_intfdata(intf);
> +
> +    return drm_mode_config_helper_resume(&adev->drm);
> +}
> +
> static const struct usb_device_id appletbdrm_usb_id_table[] = {
>    { USB_DEVICE_INTERFACE_CLASS(0x05ac, 0x8302, USB_CLASS_AUDIO_VIDEO) },
>    {}
> @@ -832,6 +847,8 @@ static struct usb_driver appletbdrm_usb_driver = {
>    .probe        = appletbdrm_probe,
>    .disconnect    = appletbdrm_disconnect,
>    .shutdown    = appletbdrm_shutdown,
> +    .suspend    = appletbdrm_suspend,
> +    .resume        = appletbdrm_resume,
>    .id_table    = appletbdrm_usb_id_table,
> };
> module_usb_driver(appletbdrm_usb_driver);
> --
> 2.53.0

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2026-10-07 14:22 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-09-24 19:25 [PATCH] drm/appletbdrm: Add suspend and resume support Vasilij Markin via B4 Relay
2026-09-24 19:33 ` sashiko-bot
2026-09-25 11:51 ` Aditya Garg
2026-10-04 12:06 ` [PATCH v2] " Aditya Garg
2026-10-04 12:19   ` sashiko-bot
2026-10-07 14:20   ` Aditya Garg

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox