dri-devel Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Maxime Ripard <maxime@cerno.tech>
To: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
	Thomas Zimmermann <tzimmermann@suse.de>,
	Maxime Ripard <maxime@cerno.tech>,
	Daniel Vetter <daniel.vetter@intel.com>,
	David Airlie <airlied@linux.ie>
Cc: dri-devel@lists.freedesktop.org
Subject: [PATCH v2 40/68] drm/vc4: hdmi: Switch to device-managed ALSA initialization
Date: Wed, 22 Jun 2022 16:31:41 +0200	[thread overview]
Message-ID: <20220622143209.600298-41-maxime@cerno.tech> (raw)
In-Reply-To: <20220622143209.600298-1-maxime@cerno.tech>

The current code to unregister our ALSA device needs to be undone manually
when we remove the HDMI driver.

Since ALSA doesn't seem to support any mechanism to defer freeing something
until the last user of the ALSA device is gone, we can either use a
device-managed or a DRM-managed action.

The consistent way would be to use a DRM-managed one, just like pretty much
any framework-facing structure should be doing. However, ALSA does a lot of
allocation and registration using device-managed calls. Thus, if we're
going that way, by the time the DRM-managed action would run all of those
allocation would have been freed and we would end up with a use-after-free.

Thus, let's do a device-managed action. It's been tested with KASAN enabled
and doesn't seem to trigger any issue, so it's as good as anything.

Signed-off-by: Maxime Ripard <maxime@cerno.tech>
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 43 ++++++++++++++++++++++++++++------
 1 file changed, 36 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index 2d1f87f89994..2de7532b6c78 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -2022,6 +2022,14 @@ static struct hdmi_codec_pdata vc4_hdmi_codec_pdata = {
 	.i2s = 1,
 };
 
+static void vc4_hdmi_audio_codec_release(void *ptr)
+{
+	struct vc4_hdmi *vc4_hdmi = ptr;
+
+	platform_device_unregister(vc4_hdmi->audio.codec_pdev);
+	vc4_hdmi->audio.codec_pdev = NULL;
+}
+
 static int vc4_hdmi_audio_init(struct vc4_hdmi *vc4_hdmi)
 {
 	const struct vc4_hdmi_register *mai_data =
@@ -2063,6 +2071,30 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *vc4_hdmi)
 	vc4_hdmi->audio.dma_data.addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
 	vc4_hdmi->audio.dma_data.maxburst = 2;
 
+	/*
+	 * NOTE: Strictly speaking, we should probably use a DRM-managed
+	 * registration there to avoid removing all the audio components
+	 * by the time the driver doesn't have any user anymore.
+	 *
+	 * However, the ASoC core uses a number of devm_kzalloc calls
+	 * when registering, even when using non-device-managed
+	 * functions (such as in snd_soc_register_component()).
+	 *
+	 * If we call snd_soc_unregister_component() in a DRM-managed
+	 * action, the device-managed actions have already been executed
+	 * and thus we would access memory that has been freed.
+	 *
+	 * Using device-managed hooks here probably leaves us open to a
+	 * bunch of issues if userspace still has a handle on the ALSA
+	 * device when the device is removed. However, this is mitigated
+	 * by the use of drm_dev_enter()/drm_dev_exit() in the audio
+	 * path to prevent the access to the device resources if it
+	 * isn't there anymore.
+	 *
+	 * Then, the vc4_hdmi structure is DRM-managed and thus only
+	 * freed whenever the last user has closed the DRM device file.
+	 * It should thus outlive ALSA in most situations.
+	 */
 	ret = devm_snd_dmaengine_pcm_register(dev, &pcm_conf, 0);
 	if (ret) {
 		dev_err(dev, "Could not register PCM component: %d\n", ret);
@@ -2086,6 +2118,10 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *vc4_hdmi)
 	}
 	vc4_hdmi->audio.codec_pdev = codec_pdev;
 
+	ret = devm_add_action_or_reset(dev, vc4_hdmi_audio_codec_release, vc4_hdmi);
+	if (ret)
+		return ret;
+
 	dai_link->cpus		= &vc4_hdmi->audio.cpu;
 	dai_link->codecs	= &vc4_hdmi->audio.codec;
 	dai_link->platforms	= &vc4_hdmi->audio.platform;
@@ -2124,12 +2160,6 @@ static int vc4_hdmi_audio_init(struct vc4_hdmi *vc4_hdmi)
 
 }
 
-static void vc4_hdmi_audio_exit(struct vc4_hdmi *vc4_hdmi)
-{
-	platform_device_unregister(vc4_hdmi->audio.codec_pdev);
-	vc4_hdmi->audio.codec_pdev = NULL;
-}
-
 static irqreturn_t vc4_hdmi_hpd_irq_thread(int irq, void *priv)
 {
 	struct vc4_hdmi *vc4_hdmi = priv;
@@ -2993,7 +3023,6 @@ static void vc4_hdmi_unbind(struct device *dev, struct device *master,
 	kfree(vc4_hdmi->hdmi_regset.regs);
 	kfree(vc4_hdmi->hd_regset.regs);
 
-	vc4_hdmi_audio_exit(vc4_hdmi);
 	vc4_hdmi_cec_exit(vc4_hdmi);
 	vc4_hdmi_hotplug_exit(vc4_hdmi);
 
-- 
2.36.1


  parent reply	other threads:[~2022-06-22 14:34 UTC|newest]

Thread overview: 82+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-22 14:31 [PATCH v2 00/68] drm/vc4: Fix hotplug for vc4 Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 01/68] drm/mipi-dsi: Detach devices when removing the host Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 02/68] drm/crtc: Introduce drmm_crtc_init_with_planes Maxime Ripard
2022-06-22 18:37   ` kernel test robot
2022-06-28  8:54   ` Thomas Zimmermann
2022-06-22 14:31 ` [PATCH v2 03/68] drm/encoder: Introduce drmm_encoder_init Maxime Ripard
2022-06-22 19:38   ` kernel test robot
2022-06-28  8:55   ` Thomas Zimmermann
2022-06-22 14:31 ` [PATCH v2 04/68] drm/connector: Reorder headers Maxime Ripard
2022-06-24 19:21   ` Sam Ravnborg
2022-06-22 14:31 ` [PATCH v2 05/68] drm/connector: Mention the cleanup after drm_connector_init Maxime Ripard
2022-06-24 19:25   ` Sam Ravnborg
2022-06-22 14:31 ` [PATCH v2 06/68] drm/connector: Clarify when drm_connector_unregister is needed Maxime Ripard
2022-06-24 19:25   ` Sam Ravnborg
2022-06-22 14:31 ` [PATCH v2 07/68] drm/connector: Introduce drmm_connector_init Maxime Ripard
2022-06-24 19:26   ` Sam Ravnborg
2022-06-22 14:31 ` [PATCH v2 08/68] drm/connector: Introduce drmm_connector_init_with_ddc Maxime Ripard
2022-06-24 19:26   ` Sam Ravnborg
2022-06-28  9:00   ` Thomas Zimmermann
2022-06-28 14:05     ` Jani Nikula
2022-06-22 14:31 ` [PATCH v2 09/68] drm/bridge: panel: Introduce drmm_panel_bridge_add Maxime Ripard
2022-06-24 19:27   ` Sam Ravnborg
2022-06-22 14:31 ` [PATCH v2 10/68] drm/bridge: panel: Introduce drmm_of_get_bridge Maxime Ripard
2022-06-24 19:31   ` Sam Ravnborg
2022-06-22 14:31 ` [PATCH v2 11/68] drm/vc4: drv: Call component_unbind_all() Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 12/68] drm/vc4: drv: Use drm_dev_unplug Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 13/68] drm/vc4: crtc: Create vblank reporting function Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 14/68] drm/vc4: hvs: Protect device resources after removal Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 15/68] drm/vc4: hvs: Remove planes currently allocated before taking down Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 16/68] drm/vc4: plane: Take possible_crtcs as an argument Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 17/68] drm/vc4: crtc: Remove manual plane removal on error Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 18/68] drm/vc4: plane: Switch to drmm_universal_plane_alloc() Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 19/68] drm/vc4: crtc: Move debugfs_name to crtc_data Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 20/68] drm/vc4: crtc: Switch to drmm_kzalloc Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 21/68] drm/vc4: crtc: Switch to DRM-managed CRTC initialization Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 22/68] drm/vc4: dpi: Remove vc4_dev dpi pointer Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 23/68] drm/vc4: dpi: Embed DRM structures into the private structure Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 24/68] drm/vc4: dpi: Switch to drmm_kzalloc Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 25/68] drm/vc4: dpi: Return an error if we can't enable our clock Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 26/68] drm/vc4: dpi: Remove unnecessary drm_of_panel_bridge_remove call Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 27/68] drm/vc4: dpi: Add action to disable the clock Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 28/68] drm/vc4: dpi: Switch to DRM-managed encoder initialization Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 29/68] drm/vc4: dpi: Switch to drmm_of_get_bridge Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 30/68] drm/vc4: dpi: Protect device resources Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 31/68] drm/vc4: dsi: Embed DRM structures into the private structure Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 32/68] drm/vc4: dsi: Switch to DRM-managed encoder initialization Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 33/68] drm/vc4: dsi: Switch to drmm_of_get_bridge Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 34/68] drm/vc4: dsi: Fix the driver structure lifetime Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 35/68] drm/vc4: dsi: Switch to devm_pm_runtime_enable Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 36/68] drm/vc4: hdmi: Switch to drmm_kzalloc Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 37/68] drm/vc4: hdmi: Remove call to drm_connector_unregister() Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 38/68] drm/vc4: hdmi: Switch to DRM-managed encoder initialization Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 39/68] drm/vc4: hdmi: Switch to DRM-managed connector initialization Maxime Ripard
2022-06-22 14:31 ` Maxime Ripard [this message]
2022-06-22 14:31 ` [PATCH v2 41/68] drm/vc4: hdmi: Switch to device-managed CEC initialization Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 42/68] drm/vc4: hdmi: Use a device-managed action for DDC Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 43/68] drm/vc4: hdmi: Switch to DRM-managed kfree to build regsets Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 44/68] drm/vc4: hdmi: Use devm to register hotplug interrupts Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 45/68] drm/vc4: hdmi: Move audio structure offset checks Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 46/68] drm/vc4: hdmi: Protect device resources after removal Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 47/68] drm/vc4: hdmi: Switch to devm_pm_runtime_enable Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 48/68] drm/vc4: txp: Remove vc4_dev txp pointer Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 49/68] drm/vc4: txp: Remove duplicate regset Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 50/68] drm/vc4: txp: Switch to drmm_kzalloc Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 51/68] drm/vc4: txp: Remove call to drm_connector_unregister() Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 52/68] drm/vc4: txp: Protect device resources Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 53/68] drm/vc4: vec: Remove vc4_dev vec pointer Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 54/68] drm/vc4: vec: Embed DRM structures into the private structure Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 55/68] drm/vc4: vec: Switch to drmm_kzalloc Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 56/68] drm/vc4: vec: Remove call to drm_connector_unregister() Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 57/68] drm/vc4: vec: Switch to DRM-managed encoder initialization Maxime Ripard
2022-06-22 14:31 ` [PATCH v2 58/68] drm/vc4: vec: Switch to DRM-managed connector initialization Maxime Ripard
2022-06-22 14:32 ` [PATCH v2 59/68] drm/vc4: vec: Protect device resources after removal Maxime Ripard
2022-06-22 14:32 ` [PATCH v2 60/68] drm/vc4: vec: Switch to devm_pm_runtime_enable Maxime Ripard
2022-06-22 14:32 ` [PATCH v2 61/68] drm/vc4: debugfs: Protect device resources Maxime Ripard
2022-06-22 14:32 ` [PATCH v2 62/68] drm/vc4: debugfs: Return an error on failure Maxime Ripard
2022-06-22 14:32 ` [PATCH v2 63/68] drm/vc4: debugfs: Simplify debugfs registration Maxime Ripard
2022-06-22 14:32 ` [PATCH v2 64/68] drm/vc4: Switch to drmm_mutex_init Maxime Ripard
2022-06-22 14:32 ` [PATCH v2 65/68] drm/vc4: perfmon: Add missing mutex_destroy Maxime Ripard
2022-06-22 14:32 ` [PATCH v2 66/68] drm/vc4: v3d: Stop disabling interrupts Maxime Ripard
2022-06-22 14:32 ` [PATCH v2 67/68] drm/vc4: v3d: Rework the runtime_pm setup Maxime Ripard
2022-06-22 14:32 ` [PATCH v2 68/68] drm/vc4: v3d: Switch to devm_pm_runtime_enable Maxime Ripard

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=20220622143209.600298-41-maxime@cerno.tech \
    --to=maxime@cerno.tech \
    --cc=airlied@linux.ie \
    --cc=daniel.vetter@intel.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=maarten.lankhorst@linux.intel.com \
    --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