intel-gfx.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: Jani Nikula <jani.nikula@linux.intel.com>
To: Imre Deak <imre.deak@intel.com>, intel-gfx@lists.freedesktop.org
Cc: dri-devel@lists.freedesktop.org, Simona Vetter <simona.vetter@ffwll.ch>
Subject: Re: [PATCH v3 02/11] drm/connector: Add FIXME for GETRESOURCES ioctl wrt. uninited connectors
Date: Thu, 12 Dec 2024 12:06:12 +0200	[thread overview]
Message-ID: <87h679xlkb.fsf@intel.com> (raw)
In-Reply-To: <20241211230328.4012496-3-imre.deak@intel.com>

On Thu, 12 Dec 2024, Imre Deak <imre.deak@intel.com> wrote:
> The connectors enumerated by the GETRESOURCES ioctl may not be fully
> initialized yet wrt. to the state set up during connector registration
> (for instance the connector's debugfs/sysfs interfaces may not exist
> yet). This can happen in two ways:
>
> 1. Connectors initialized and added to the
>    drm_mode_config::connector_list during driver loading will be visible
>    to the GETRESOURCES ioctl caller once the driver is registered via
>    drm_dev_register()->drm_minor_register(DRM_MINOR_PRIMARY) and before
>    the connectors are registered via drm_dev_register()->
>    drm_modeset_register_all().
>
> 2. Dynamic connectors (MST) - after being initialized - may be added to
>    the connector_list after the driver is loaded and registered and before
>    the connector's userspace interfaces (debugfs, sysfs etc.) are added
>    in drm_connector_dynamic_register().
>
> A solution for 1. would be to register the driver only after the
> connectors are registered, for 2. to add the connector to connector_list
> only after the userspace interfaces are registered.
>
> The fix would require a bigger change, not the topic of this patchset,
> so adding only a FIXME comment here.

References to "this patchset" become meaningless after this has been
committed.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>

>
> Suggested-by: Simona Vetter <simona.vetter@ffwll.ch>
> Signed-off-by: Imre Deak <imre.deak@intel.com>
> ---
>  drivers/gpu/drm/drm_mode_config.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
>
> diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c
> index 37d2e0a4ef4be..8642a2fb25a90 100644
> --- a/drivers/gpu/drm/drm_mode_config.c
> +++ b/drivers/gpu/drm/drm_mode_config.c
> @@ -150,6 +150,15 @@ int drm_mode_getresources(struct drm_device *dev, void *data,
>  	drm_connector_list_iter_begin(dev, &conn_iter);
>  	count = 0;
>  	connector_id = u64_to_user_ptr(card_res->connector_id_ptr);
> +	/*
> +	 * FIXME: the connectors on the list may not be fully initialized yet,
> +	 * if the ioctl is called before the connectors are registered. (See
> +	 * drm_dev_register()->drm_modeset_register_all() for static and
> +	 * drm_connector_dynamic_register() for dynamic connectors.)
> +	 * The driver should only get registered after static connectors are
> +	 * fully initialized and dynamic connectors should be added to the
> +	 * connector list only after fully initializing them.
> +	 */
>  	drm_for_each_connector_iter(connector, &conn_iter) {
>  		/* only expose writeback connectors if userspace understands them */
>  		if (!file_priv->writeback_connectors &&

-- 
Jani Nikula, Intel

  reply	other threads:[~2024-12-12 10:06 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-11 23:03 [PATCH v3 0/7] drm/connector: Expose only a properly inited connector Imre Deak
2024-12-11 23:03 ` [PATCH v3 01/11] drm/connector: Add a way to init/add a connector in separate steps Imre Deak
2024-12-12  0:33   ` Lyude Paul
2024-12-12 10:16     ` Simona Vetter
2024-12-12 10:04   ` Jani Nikula
2024-12-12 11:45     ` Imre Deak
2024-12-16 12:22   ` Maxime Ripard
2024-12-11 23:03 ` [PATCH v3 02/11] drm/connector: Add FIXME for GETRESOURCES ioctl wrt. uninited connectors Imre Deak
2024-12-12 10:06   ` Jani Nikula [this message]
2024-12-13 12:02     ` Imre Deak
2024-12-11 23:03 ` [PATCH v3 03/11] drm/connector: Add deprication notes for drm_connector_register/unregister Imre Deak
2024-12-12 10:10   ` Jani Nikula
2024-12-12 12:02     ` Imre Deak
2024-12-16 14:37       ` Simona Vetter
2024-12-11 23:03 ` [PATCH v3 04/11] drm/dp_mst: Register connectors via drm_connector_dynamic_register() Imre Deak
2024-12-12 10:12   ` Jani Nikula
2024-12-13 12:06     ` Imre Deak
2024-12-16 11:03       ` Jani Nikula
2024-12-16 12:14         ` Imre Deak
2024-12-16 12:23   ` Imre Deak
2024-12-17  8:02     ` Lin, Wayne
2024-12-11 23:03 ` [PATCH v3 05/11] drm/i915/dp_mst: Expose a connector to kernel users after it's properly initialized Imre Deak
2024-12-12 10:13   ` Jani Nikula
2024-12-11 23:03 ` [PATCH v3 06/11] drm/amd/dp_mst: " Imre Deak
2024-12-12 10:14   ` Jani Nikula
2024-12-16 12:45   ` Imre Deak
2024-12-16 18:26     ` Alex Deucher
2024-12-11 23:03 ` [PATCH v3 07/11] drm/nouveau/dp_mst: " Imre Deak
2024-12-12 10:14   ` Jani Nikula
2024-12-11 23:03 ` [PATCH v3 08/11] drm/connector: Warn if a connector is registered/added incorrectly Imre Deak
2024-12-12 10:15   ` Jani Nikula
2024-12-11 23:03 ` [PATCH v3 09/11] drm/tests: Add tests for drm_connector_dynamic_init()/register() Imre Deak
2024-12-16 12:22   ` Maxime Ripard
2024-12-11 23:03 ` [PATCH v3 10/11] drm/i915/dp_mst: Fix error handling while adding a connector Imre Deak
2024-12-12 10:16   ` Jani Nikula
2024-12-11 23:03 ` [PATCH v3 11/11] drm/i915/dp_mst: Use intel_connector vs. drm_connector pointer in intel_dp_mst.c Imre Deak
2024-12-12 10:25   ` Jani Nikula
2024-12-12 12:23     ` Imre Deak
2024-12-11 23:33 ` ✗ Fi.CI.CHECKPATCH: warning for drm/connector: Expose only a properly inited connector Patchwork
2024-12-11 23:33 ` ✗ Fi.CI.SPARSE: " Patchwork
2024-12-11 23:57 ` ✓ i915.CI.BAT: success " Patchwork
2024-12-12  7:33 ` ✗ i915.CI.Full: failure " Patchwork
2024-12-13 15:41   ` Imre Deak
2024-12-16 12:22 ` [PATCH v3 0/7] " Imre Deak
2024-12-16 12:39   ` Maxime Ripard
2024-12-17 14:47 ` Imre Deak

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=87h679xlkb.fsf@intel.com \
    --to=jani.nikula@linux.intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=imre.deak@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=simona.vetter@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;
as well as URLs for NNTP newsgroup(s).