[PATCH v3 08/12] drm/amdgpu: Suspend and resume internal clients with client helpers

Cavitt, Jonathan jonathan.cavitt at intel.com
Tue Oct 8 21:37:04 UTC 2024


-----Original Message-----
From: Intel-xe <intel-xe-bounces at lists.freedesktop.org> On Behalf Of Thomas Zimmermann
Sent: Tuesday, October 8, 2024 4:59 AM
To: simona at ffwll.ch; airlied at gmail.com; javierm at redhat.com; jfalempe at redhat.com
Cc: dri-devel at lists.freedesktop.org; amd-gfx at lists.freedesktop.org; intel-gfx at lists.freedesktop.org; intel-xe at lists.freedesktop.org; Thomas Zimmermann <tzimmermann at suse.de>; Alex Deucher <alexander.deucher at amd.com>; Christian König <christian.koenig at amd.com>; Xinhui Pan <Xinhui.Pan at amd.com>
Subject: [PATCH v3 08/12] drm/amdgpu: Suspend and resume internal clients with client helpers
> 
> Replace calls to drm_fb_helper_set_suspend_unlocked() with calls
> to the client functions drm_client_dev_suspend() and
> drm_client_dev_resume(). Any registered in-kernel client will now
> receive suspend and resume events.
> 
> Signed-off-by: Thomas Zimmermann <tzimmermann at suse.de>
> Cc: Alex Deucher <alexander.deucher at amd.com>
> Cc: "Christian König" <christian.koenig at amd.com>
> Cc: Xinhui Pan <Xinhui.Pan at amd.com>

LGTM, though perhaps it's not my place to say this is okay.

I'd wait for an expert on the amdgpu to respond first before
jumping to push.

Reviewed-by: Jonathan Cavitt <jonathan.cavitt at intel.com>
-Jonathan Cavitt

> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 22 +++++++++++-----------
>  1 file changed, 11 insertions(+), 11 deletions(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index fd853dc843e9..9c40d620a658 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -37,8 +37,8 @@
>  
>  #include <drm/drm_aperture.h>
>  #include <drm/drm_atomic_helper.h>
> +#include <drm/drm_client_event.h>
>  #include <drm/drm_crtc_helper.h>
> -#include <drm/drm_fb_helper.h>
>  #include <drm/drm_probe_helper.h>
>  #include <drm/amdgpu_drm.h>
>  #include <linux/device.h>
> @@ -4711,13 +4711,13 @@ int amdgpu_device_prepare(struct drm_device *dev)
>   * amdgpu_device_suspend - initiate device suspend
>   *
>   * @dev: drm dev pointer
> - * @fbcon : notify the fbdev of suspend
> + * @notify_clients: notify in-kernel DRM clients
>   *
>   * Puts the hw in the suspend state (all asics).
>   * Returns 0 for success or an error on failure.
>   * Called at driver suspend.
>   */
> -int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
> +int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients)
>  {
>  	struct amdgpu_device *adev = drm_to_adev(dev);
>  	int r = 0;
> @@ -4737,8 +4737,8 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
>  	if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D3))
>  		DRM_WARN("smart shift update failed\n");
>  
> -	if (fbcon)
> -		drm_fb_helper_set_suspend_unlocked(adev_to_drm(adev)->fb_helper, true);
> +	if (notify_clients)
> +		drm_client_dev_suspend(adev_to_drm(adev), false);
>  
>  	cancel_delayed_work_sync(&adev->delayed_init_work);
>  
> @@ -4773,13 +4773,13 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon)
>   * amdgpu_device_resume - initiate device resume
>   *
>   * @dev: drm dev pointer
> - * @fbcon : notify the fbdev of resume
> + * @notify_clients: notify in-kernel DRM clients
>   *
>   * Bring the hw back to operating state (all asics).
>   * Returns 0 for success or an error on failure.
>   * Called at driver resume.
>   */
> -int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
> +int amdgpu_device_resume(struct drm_device *dev, bool notify_clients)
>  {
>  	struct amdgpu_device *adev = drm_to_adev(dev);
>  	int r = 0;
> @@ -4835,8 +4835,8 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon)
>  	/* Make sure IB tests flushed */
>  	flush_delayed_work(&adev->delayed_init_work);
>  
> -	if (fbcon)
> -		drm_fb_helper_set_suspend_unlocked(adev_to_drm(adev)->fb_helper, false);
> +	if (notify_clients)
> +		drm_client_dev_resume(adev_to_drm(adev), false);
>  
>  	amdgpu_ras_resume(adev);
>  
> @@ -5448,7 +5448,7 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle,
>  				if (r)
>  					goto out;
>  
> -				drm_fb_helper_set_suspend_unlocked(adev_to_drm(tmp_adev)->fb_helper, false);
> +				drm_client_dev_resume(adev_to_drm(tmp_adev), false);
>  
>  				/*
>  				 * The GPU enters bad state once faulty pages
> @@ -5734,7 +5734,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
>  		 */
>  		amdgpu_unregister_gpu_instance(tmp_adev);
>  
> -		drm_fb_helper_set_suspend_unlocked(adev_to_drm(tmp_adev)->fb_helper, true);
> +		drm_client_dev_suspend(adev_to_drm(tmp_adev), false);
>  
>  		/* disable ras on ALL IPs */
>  		if (!need_emergency_restart &&
> -- 
> 2.46.0
> 
> 


More information about the Intel-gfx mailing list