All of lore.kernel.org
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Louis Chauvet" <louis.chauvet@bootlin.com>
Cc: dri-devel@lists.freedesktop.org
Subject: Re: [PATCH v5 38/38] drm/vkms: Add ConfigFS interface for connector parent and port_id
Date: Sat, 27 Jun 2026 03:45:48 +0000	[thread overview]
Message-ID: <20260627034549.5163B1F000E9@smtp.kernel.org> (raw)
In-Reply-To: <20260627-vkms-all-config-v5-38-854aa0840926@bootlin.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] The function `connector_parent_connector_allow_link` returns an uninitialized variable `ret` on success, which can cause state desynchronization and a Use-After-Free vulnerability.
--

commit bf8b7f51f1321ae67ea645909d25e75e04da5354
Author: Louis Chauvet <louis.chauvet@bootlin.com>

drm/vkms: Add ConfigFS interface for connector parent and port_id

This commit adds configuration interfaces for a connector's parent and port_id
using configfs. It allows port_id to hold any u8 value and permits the parent
to contain at most one symlink to another connector.

> diff --git a/drivers/gpu/drm/vkms/vkms_configfs.c b/drivers/gpu/drm/vkms/vkms_configfs.c
> index dd2a9e450150b..4f4ec99448042 100644
> --- a/drivers/gpu/drm/vkms/vkms_configfs.c
> +++ b/drivers/gpu/drm/vkms/vkms_configfs.c

[ ... ]

> @@ -1512,6 +1555,63 @@ static const struct config_item_type connector_possible_encoders_group_type = {
>  	.ct_owner	= THIS_MODULE,
>  };
>  
> +static int connector_parent_connector_allow_link(struct config_item *src,
> +						 struct config_item *target)
> +{
> +	struct vkms_configfs_connector *connector;
> +	struct vkms_config_connector *connector_cfg;
> +	struct vkms_configfs_connector *parent;
> +	struct vkms_config_connector *parent_cfg;
> +	int ret;
> +
> +	if (target->ci_type != &connector_item_type)
> +		return -EINVAL;
> +
> +	connector = connector_parent_item_to_vkms_configfs_connector(src);
> +	connector_cfg = connector->config;
> +	parent = connector_item_to_vkms_configfs_connector(target);
> +	parent_cfg = parent->config;
> +
> +	if (connector->dev != parent->dev)
> +		return -EINVAL;
> +
> +	scoped_guard(mutex, &connector->dev->lock) {
> +		if (vkms_config_connector_get_parent(connector_cfg))
> +			return -EMLINK;
> +
> +		if (connector->dev->enabled && connector_cfg->enabled)
> +			return -EBUSY;
> +
> +		vkms_config_connector_attach_parent(connector_cfg,
> +						    parent_cfg);
> +	}
> +
> +	return ret;

[Severity: High]
Could this return an uninitialized value on the success path? 

In connector_parent_connector_allow_link(), ret is declared but never
initialized before being returned at the end of the function. 

If ret happens to contain a non-zero value from the stack, configfs will 
treat the link creation as a failure and abort creating the symlink. 
However, the internal VKMS state has already been modified to point to the 
target connector via vkms_config_connector_attach_parent(). 

If the user subsequently removes the target connector using rmdir, does 
this leave a dangling pointer in the VKMS configuration that could result 
in a use-after-free later on?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260627-vkms-all-config-v5-0-854aa0840926@bootlin.com?part=38

      reply	other threads:[~2026-06-27  3:45 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-06-27  3:30 [PATCH v5 00/38] VKMS: Introduce multiple configFS attributes Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 01/38] drm/drm_mode_config: Add helper to get plane type name Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 02/38] drm/blend: Get a rotation name from it's bitfield Louis Chauvet
2026-06-27  3:41   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 03/38] drm/drm_color_mgmt: Expose drm_get_color_encoding_name Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 04/38] drm/drm_color_mgmt: Expose drm_get_color_range_name Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 05/38] drm/connector: Export drm_get_colorspace_name Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 06/38] drm/drm_atomic_state_helper: Properly load default value for rotation Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 07/38] Documentation: ABI: vkms: Add current VKMS ABI documentation Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 08/38] drm/vkms: Add error handling in plane config creation Louis Chauvet
2026-06-27  3:41   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 09/38] drm/vkms: Simplify plane_release code Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 10/38] drm/vkms: Explicitly display plane type Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 11/38] drm/vkms: Use enabled/disabled instead of 1/0 for debug Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 12/38] drm/vkms: Explicitly display connector status Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 13/38] drm/vkms: Introduce config for plane name Louis Chauvet
2026-06-27  3:46   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 14/38] drm/vkms: Use plane folder name as " Louis Chauvet
2026-06-27  3:43   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 15/38] drm/vkms: Introduce config for plane rotation Louis Chauvet
2026-06-27  3:42   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 16/38] drm/vkms: Use DRM_ROTATION_FMT macros for rotation display Louis Chauvet
2026-06-27  3:39   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 17/38] drm/vkms: Introduce configfs for plane rotation Louis Chauvet
2026-06-27  3:46   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 18/38] drm/vkms: Introduce config for plane color encoding Louis Chauvet
2026-06-27  3:43   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 19/38] drm/vkms: Introduce configfs " Louis Chauvet
2026-06-27  3:49   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 20/38] drm/vkms: Introduce config for plane color range Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 21/38] drm/vkms: Introduce configfs " Louis Chauvet
2026-06-27  3:45   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 22/38] drm/vkms: Introduce config for plane format Louis Chauvet
2026-06-27  3:46   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 23/38] drm/vkms: Introduce configfs " Louis Chauvet
2026-06-27  3:45   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 24/38] drm/vkms: Properly render plane using their zpos Louis Chauvet
2026-06-27  3:44   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 25/38] drm/vkms: Introduce config for plane zpos property Louis Chauvet
2026-06-27  3:41   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 26/38] drm/vkms: Introduce configfs " Louis Chauvet
2026-06-27  3:46   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 27/38] drm/vkms: Introduce config for connector type Louis Chauvet
2026-06-27  3:45   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 28/38] drm/vkms: Introduce configfs " Louis Chauvet
2026-06-27  3:45   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 29/38] drm/vkms: Rename vkms_connector_init to vkms_connector_init_static Louis Chauvet
2026-06-27  3:30 ` [PATCH v5 30/38] drm/vkms: Introduce config for connector supported colorspace Louis Chauvet
2026-06-27  3:58   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 31/38] drm/vkms: Introduce configfs " Louis Chauvet
2026-06-27  3:48   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 32/38] drm/vkms: Introduce config for connector EDID Louis Chauvet
2026-06-27  3:48   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 33/38] drm/vkms: Introduce configfs " Louis Chauvet
2026-06-27  3:50   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 34/38] drm/vkms: Store the enabled/disabled status for connector Louis Chauvet
2026-06-27  3:48   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 35/38] drm/vkms: Allow to hot-add connectors Louis Chauvet
2026-06-27  3:50   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 36/38] drm/vkms: Introduce configfs for dynamic connector creation Louis Chauvet
2026-06-27  3:55   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 37/38] drm/vkms: Add connector parent configuration in vkms_config Louis Chauvet
2026-06-27  3:57   ` sashiko-bot
2026-06-27  3:30 ` [PATCH v5 38/38] drm/vkms: Add ConfigFS interface for connector parent and port_id Louis Chauvet
2026-06-27  3:45   ` sashiko-bot [this message]

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=20260627034549.5163B1F000E9@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=louis.chauvet@bootlin.com \
    --cc=sashiko-reviews@lists.linux.dev \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.