mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Thomas Zimmermann <tzimmermann@suse.de>
To: Ruben Wauters <rubenru09@aol.com>,
	Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
	Maxime Ripard <mripard@kernel.org>,
	David Airlie <airlied@gmail.com>, Simona Vetter <simona@ffwll.ch>
Cc: dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] drm/gud: move plane init to gud_pipe.c
Date: Tue, 7 Oct 2025 11:17:49 +0200	[thread overview]
Message-ID: <e0d81b43-22cf-4004-936f-2a1dae9d8741@suse.de> (raw)
In-Reply-To: <20251004175900.15235-2-rubenru09@aol.com>

Hi Ruben,

please see my comments below.

Am 04.10.25 um 19:49 schrieb Ruben Wauters:
> gud_probe() currently is a quite large function that does a lot of
> different things, including USB detection, plane init, and several other
> things.
>
> This patch moves the plane and crtc init into gud_plane_init() in
> gud_pipe.c, which is a more appropriate file for this. Associated
> variables and structs have also been moved to gud_pipe.c
>
> Signed-off-by: Ruben Wauters <rubenru09@aol.com>
> ---
> It was somewhat difficult to determine what exactly should be moved
> over, gud_probe() as a function quite a mess, so I need to figure out
> exactly how to split this one up.

Agreed. The probe function looks really chaotic.

I think that just moving CRTC and plane is a not enough. In ast and udl, 
we have functions that init the whole display pipeline from 
drmm_mode_config_init() to _reset(). See [1] and [2] for examples. That 
would likely be a good model for gud as well, but gud's probe function 
mixes up pipeline init with other code.

[1] 
https://elixir.bootlin.com/linux/v6.17.1/source/drivers/gpu/drm/ast/ast_mode.c#L1005
[2] 
https://elixir.bootlin.com/linux/v6.17.1/source/drivers/gpu/drm/udl/udl_modeset.c#L482


Looking over gud_probe, the following blocks are related to pipeline init:

- lines 466-469 [3]
- lines 486-489
- lines 558-565
- lines 590-599
- lines 610-623
- line 641

[3] 
https://gitlab.freedesktop.org/drm/misc/kernel/-/blob/drm-misc-next/drivers/gpu/drm/gud/gud_drv.c#L466

I'd try to move these lines into a new helper that initializes the full 
modesetting pipeline.

The other code that happens in between is either preparation or clean up 
and should be done before or after creating the pipeline.


>
> As an aside, I noticed that the driver doesn't have a version macro in
> gud_drv.c, and therefore is shown as 1.0.0. I was thinking of
> introducing a version, but I wanted to know how others generally deal
> with driver versions. I'm not 100% sure if it's *necessary* for GUD but
> it might be a good idea.

I wouldn't bother at all about module versions. AFAIK no one cares about 
it anyway.

Best regards
Thomas

> ---
>   drivers/gpu/drm/gud/gud_drv.c      | 48 +-----------------------
>   drivers/gpu/drm/gud/gud_internal.h |  1 +
>   drivers/gpu/drm/gud/gud_pipe.c     | 60 ++++++++++++++++++++++++++++++
>   3 files changed, 62 insertions(+), 47 deletions(-)
>
> diff --git a/drivers/gpu/drm/gud/gud_drv.c b/drivers/gpu/drm/gud/gud_drv.c
> index b7345c8d823d..967c16479b5c 100644
> --- a/drivers/gpu/drm/gud/gud_drv.c
> +++ b/drivers/gpu/drm/gud/gud_drv.c
> @@ -16,7 +16,6 @@
>   #include <drm/clients/drm_client_setup.h>
>   #include <drm/drm_atomic_helper.h>
>   #include <drm/drm_blend.h>
> -#include <drm/drm_crtc_helper.h>
>   #include <drm/drm_damage_helper.h>
>   #include <drm/drm_debugfs.h>
>   #include <drm/drm_drv.h>
> @@ -338,43 +337,12 @@ static int gud_stats_debugfs(struct seq_file *m, void *data)
>   	return 0;
>   }
>   
> -static const struct drm_crtc_helper_funcs gud_crtc_helper_funcs = {
> -	.atomic_check = drm_crtc_helper_atomic_check
> -};
> -
> -static const struct drm_crtc_funcs gud_crtc_funcs = {
> -	.reset = drm_atomic_helper_crtc_reset,
> -	.destroy = drm_crtc_cleanup,
> -	.set_config = drm_atomic_helper_set_config,
> -	.page_flip = drm_atomic_helper_page_flip,
> -	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
> -	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
> -};
> -
> -static const struct drm_plane_helper_funcs gud_plane_helper_funcs = {
> -	DRM_GEM_SHADOW_PLANE_HELPER_FUNCS,
> -	.atomic_check = gud_plane_atomic_check,
> -	.atomic_update = gud_plane_atomic_update,
> -};
> -
> -static const struct drm_plane_funcs gud_plane_funcs = {
> -	.update_plane = drm_atomic_helper_update_plane,
> -	.disable_plane = drm_atomic_helper_disable_plane,
> -	.destroy = drm_plane_cleanup,
> -	DRM_GEM_SHADOW_PLANE_FUNCS,
> -};
> -
>   static const struct drm_mode_config_funcs gud_mode_config_funcs = {
>   	.fb_create = drm_gem_fb_create_with_dirty,
>   	.atomic_check = drm_atomic_helper_check,
>   	.atomic_commit = drm_atomic_helper_commit,
>   };
>   
> -static const u64 gud_plane_modifiers[] = {
> -	DRM_FORMAT_MOD_LINEAR,
> -	DRM_FORMAT_MOD_INVALID
> -};
> -
>   DEFINE_DRM_GEM_FOPS(gud_fops);
>   
>   static const struct drm_driver gud_drm_driver = {
> @@ -587,17 +555,10 @@ static int gud_probe(struct usb_interface *intf, const struct usb_device_id *id)
>   			return -ENOMEM;
>   	}
>   
> -	ret = drm_universal_plane_init(drm, &gdrm->plane, 0,
> -				       &gud_plane_funcs,
> -				       formats, num_formats,
> -				       gud_plane_modifiers,
> -				       DRM_PLANE_TYPE_PRIMARY, NULL);
> +	ret = gud_plane_init(gdrm, formats, num_formats);
>   	if (ret)
>   		return ret;
>   
> -	drm_plane_helper_add(&gdrm->plane, &gud_plane_helper_funcs);
> -	drm_plane_enable_fb_damage_clips(&gdrm->plane);
> -
>   	devm_kfree(dev, formats);
>   	devm_kfree(dev, formats_dev);
>   
> @@ -607,13 +568,6 @@ static int gud_probe(struct usb_interface *intf, const struct usb_device_id *id)
>   		return ret;
>   	}
>   
> -	ret = drm_crtc_init_with_planes(drm, &gdrm->crtc, &gdrm->plane, NULL,
> -					&gud_crtc_funcs, NULL);
> -	if (ret)
> -		return ret;
> -
> -	drm_crtc_helper_add(&gdrm->crtc, &gud_crtc_helper_funcs);
> -
>   	ret = gud_get_connectors(gdrm);
>   	if (ret) {
>   		dev_err(dev, "Failed to get connectors (error=%d)\n", ret);
> diff --git a/drivers/gpu/drm/gud/gud_internal.h b/drivers/gpu/drm/gud/gud_internal.h
> index d27c31648341..4a91aae61e50 100644
> --- a/drivers/gpu/drm/gud/gud_internal.h
> +++ b/drivers/gpu/drm/gud/gud_internal.h
> @@ -69,6 +69,7 @@ void gud_plane_atomic_update(struct drm_plane *plane,
>   int gud_connector_fill_properties(struct drm_connector_state *connector_state,
>   				  struct gud_property_req *properties);
>   int gud_get_connectors(struct gud_device *gdrm);
> +int gud_plane_init(struct gud_device *gdrm, u32 *formats, unsigned int num_formats);
>   
>   /* Driver internal fourcc transfer formats */
>   #define GUD_DRM_FORMAT_R1		0x00000122
> diff --git a/drivers/gpu/drm/gud/gud_pipe.c b/drivers/gpu/drm/gud/gud_pipe.c
> index 3a208e956dff..1f7af86b28fd 100644
> --- a/drivers/gpu/drm/gud/gud_pipe.c
> +++ b/drivers/gpu/drm/gud/gud_pipe.c
> @@ -10,6 +10,7 @@
>   
>   #include <drm/drm_atomic.h>
>   #include <drm/drm_connector.h>
> +#include <drm/drm_crtc_helper.h>
>   #include <drm/drm_damage_helper.h>
>   #include <drm/drm_drv.h>
>   #include <drm/drm_format_helper.h>
> @@ -450,6 +451,65 @@ static void gud_fb_handle_damage(struct gud_device *gdrm, struct drm_framebuffer
>   	gud_flush_damage(gdrm, fb, src, !fb->obj[0]->import_attach, damage);
>   }
>   
> +static const struct drm_plane_funcs gud_plane_funcs = {
> +	.update_plane = drm_atomic_helper_update_plane,
> +	.disable_plane = drm_atomic_helper_disable_plane,
> +	.destroy = drm_plane_cleanup,
> +	DRM_GEM_SHADOW_PLANE_FUNCS,
> +};
> +
> +static const struct drm_plane_helper_funcs gud_plane_helper_funcs = {
> +	DRM_GEM_SHADOW_PLANE_HELPER_FUNCS,
> +	.atomic_check = gud_plane_atomic_check,
> +	.atomic_update = gud_plane_atomic_update,
> +};
> +
> +static const struct drm_crtc_helper_funcs gud_crtc_helper_funcs = {
> +	.atomic_check = drm_crtc_helper_atomic_check
> +};
> +
> +static const struct drm_crtc_funcs gud_crtc_funcs = {
> +	.reset = drm_atomic_helper_crtc_reset,
> +	.destroy = drm_crtc_cleanup,
> +	.set_config = drm_atomic_helper_set_config,
> +	.page_flip = drm_atomic_helper_page_flip,
> +	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
> +	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
> +};
> +
> +static const u64 gud_plane_modifiers[] = {
> +	DRM_FORMAT_MOD_LINEAR,
> +	DRM_FORMAT_MOD_INVALID
> +};
> +
> +int gud_plane_init(struct gud_device *gdrm, u32 *formats, unsigned int num_formats)
> +{
> +	struct drm_device *drm = &gdrm->drm;
> +	struct drm_plane *plane = &gdrm->plane;
> +	struct drm_crtc *crtc = &gdrm->crtc;
> +	int ret;
> +
> +	ret = drm_universal_plane_init(drm, plane, 0,
> +				       &gud_plane_funcs,
> +				       formats, num_formats,
> +				       gud_plane_modifiers,
> +				       DRM_PLANE_TYPE_PRIMARY, NULL);
> +	if (ret)
> +		return ret;
> +
> +	drm_plane_helper_add(plane, &gud_plane_helper_funcs);
> +	drm_plane_enable_fb_damage_clips(plane);
> +
> +	ret = drm_crtc_init_with_planes(drm, crtc, plane, NULL,
> +					&gud_crtc_funcs, NULL);
> +	if (ret)
> +		return ret;
> +
> +	drm_crtc_helper_add(crtc, &gud_crtc_helper_funcs);
> +
> +	return 0;
> +}
> +
>   int gud_plane_atomic_check(struct drm_plane *plane,
>   			   struct drm_atomic_state *state)
>   {

-- 
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstrasse 146, 90461 Nuernberg, Germany
GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman
HRB 36809 (AG Nuernberg)



  reply	other threads:[~2025-10-07  9:17 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20251004175900.15235-2-rubenru09.ref@aol.com>
2025-10-04 17:49 ` Ruben Wauters
2025-10-07  9:17   ` Thomas Zimmermann [this message]
2025-10-07  9:52     ` Ruben Wauters
2025-10-07 10:56       ` Thomas Zimmermann
2025-10-07 11:15         ` Ruben Wauters

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=e0d81b43-22cf-4004-936f-2a1dae9d8741@suse.de \
    --to=tzimmermann@suse.de \
    --cc=airlied@gmail.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=mripard@kernel.org \
    --cc=rubenru09@aol.com \
    --cc=simona@ffwll.ch \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®