From: Sasha Levin <sashal@kernel.org>
To: stable@vger.kernel.org
Cc: Thomas Zimmermann <tzimmermann@suse.de>,
Javier Martinez Canillas <javierm@redhat.com>,
Alex Deucher <alexander.deucher@amd.com>,
dri-devel@lists.freedesktop.org, nouveau@lists.freedesktop.org,
amd-gfx@lists.freedesktop.org, linux-fbdev@vger.kernel.org,
Sasha Levin <sashal@kernel.org>
Subject: [PATCH 6.6.y] drm, fbcon, vga_switcheroo: Avoid race condition in fbcon setup
Date: Mon, 1 Dec 2025 17:51:23 -0500 [thread overview]
Message-ID: <20251201225123.1298682-1-sashal@kernel.org> (raw)
In-Reply-To: <2025120119-quake-universal-d896@gregkh>
From: Thomas Zimmermann <tzimmermann@suse.de>
[ Upstream commit eb76d0f5553575599561010f24c277cc5b31d003 ]
Protect vga_switcheroo_client_fb_set() with console lock. Avoids OOB
access in fbcon_remap_all(). Without holding the console lock the call
races with switching outputs.
VGA switcheroo calls fbcon_remap_all() when switching clients. The fbcon
function uses struct fb_info.node, which is set by register_framebuffer().
As the fb-helper code currently sets up VGA switcheroo before registering
the framebuffer, the value of node is -1 and therefore not a legal value.
For example, fbcon uses the value within set_con2fb_map() [1] as an index
into an array.
Moving vga_switcheroo_client_fb_set() after register_framebuffer() can
result in VGA switching that does not switch fbcon correctly.
Therefore move vga_switcheroo_client_fb_set() under fbcon_fb_registered(),
which already holds the console lock. Fbdev calls fbcon_fb_registered()
from within register_framebuffer(). Serializes the helper with VGA
switcheroo's call to fbcon_remap_all().
Although vga_switcheroo_client_fb_set() takes an instance of struct fb_info
as parameter, it really only needs the contained fbcon state. Moving the
call to fbcon initialization is therefore cleaner than before. Only amdgpu,
i915, nouveau and radeon support vga_switcheroo. For all other drivers,
this change does nothing.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://elixir.bootlin.com/linux/v6.17/source/drivers/video/fbdev/core/fbcon.c#L2942 # [1]
Fixes: 6a9ee8af344e ("vga_switcheroo: initial implementation (v15)")
Acked-by: Javier Martinez Canillas <javierm@redhat.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Cc: dri-devel@lists.freedesktop.org
Cc: nouveau@lists.freedesktop.org
Cc: amd-gfx@lists.freedesktop.org
Cc: linux-fbdev@vger.kernel.org
Cc: <stable@vger.kernel.org> # v2.6.34+
Link: https://patch.msgid.link/20251105161549.98836-1-tzimmermann@suse.de
[ drm_fb_helper_unregister_info() lacks vga_switcheroo code ]
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/gpu/drm/drm_fb_helper.c | 7 -------
drivers/video/fbdev/core/fbcon.c | 9 +++++++++
2 files changed, 9 insertions(+), 7 deletions(-)
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index b507c1c008a3e..3891837a78414 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -30,9 +30,7 @@
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/console.h>
-#include <linux/pci.h>
#include <linux/sysrq.h>
-#include <linux/vga_switcheroo.h>
#include <drm/drm_atomic.h>
#include <drm/drm_drv.h>
@@ -1668,7 +1666,6 @@ static int drm_fb_helper_find_sizes(struct drm_fb_helper *fb_helper,
static int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper)
{
struct drm_client_dev *client = &fb_helper->client;
- struct drm_device *dev = fb_helper->dev;
struct drm_fb_helper_surface_size sizes;
int ret;
@@ -1687,10 +1684,6 @@ static int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper)
strcpy(fb_helper->fb->comm, "[fbcon]");
- /* Set the fb info for vgaswitcheroo clients. Does nothing otherwise. */
- if (dev_is_pci(dev->dev))
- vga_switcheroo_client_fb_set(to_pci_dev(dev->dev), fb_helper->info);
-
return 0;
}
diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c
index 78a5b22c8d150..8b2c3065c0c26 100644
--- a/drivers/video/fbdev/core/fbcon.c
+++ b/drivers/video/fbdev/core/fbcon.c
@@ -64,6 +64,7 @@
#include <linux/console.h>
#include <linux/string.h>
#include <linux/kd.h>
+#include <linux/pci.h>
#include <linux/slab.h>
#include <linux/fb.h>
#include <linux/fbcon.h>
@@ -75,6 +76,7 @@
#include <linux/interrupt.h>
#include <linux/crc32.h> /* For counting font checksums */
#include <linux/uaccess.h>
+#include <linux/vga_switcheroo.h>
#include <asm/irq.h>
#include "fbcon.h"
@@ -2913,6 +2915,9 @@ void fbcon_fb_unregistered(struct fb_info *info)
console_lock();
+ if (info->device && dev_is_pci(info->device))
+ vga_switcheroo_client_fb_set(to_pci_dev(info->device), NULL);
+
fbcon_registered_fb[info->node] = NULL;
fbcon_num_registered_fb--;
@@ -3046,6 +3051,10 @@ static int do_fb_registered(struct fb_info *info)
}
}
+ /* Set the fb info for vga_switcheroo clients. Does nothing otherwise. */
+ if (info->device && dev_is_pci(info->device))
+ vga_switcheroo_client_fb_set(to_pci_dev(info->device), info);
+
return ret;
}
--
2.51.0
next parent reply other threads:[~2025-12-01 22:51 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <2025120119-quake-universal-d896@gregkh>
2025-12-01 22:51 ` Sasha Levin [this message]
2025-12-02 7:42 ` [PATCH 6.6.y] drm, fbcon, vga_switcheroo: Avoid race condition in fbcon setup Thomas Zimmermann
2025-12-02 15:53 ` Greg KH
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=20251201225123.1298682-1-sashal@kernel.org \
--to=sashal@kernel.org \
--cc=alexander.deucher@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=javierm@redhat.com \
--cc=linux-fbdev@vger.kernel.org \
--cc=nouveau@lists.freedesktop.org \
--cc=stable@vger.kernel.org \
--cc=tzimmermann@suse.de \
/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).