linux-kernel - Re: [PATCH v5 3/6] drm/bridge: cadence: cdns-mhdp8546-core: Set the mhdp connector earlier in atomic

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <7c0957a9-a970-4332-9df9-26ba432a0aec@ti.com>
Date: Tue, 2 Sep 2025 14:57:29 +0530
From: Harikrishna Shenoy <h-shenoy@...com>
To: Tomi Valkeinen <tomi.valkeinen@...asonboard.com>
CC: <jonas@...boo.se>, <jernej.skrabec@...il.com>,
        <maarten.lankhorst@...ux.intel.com>, <tzimmermann@...e.de>,
        <airlied@...il.com>, <simona@...ll.ch>, <lyude@...hat.com>,
        <luca.ceresoli@...tlin.com>, <viro@...iv.linux.org.uk>,
        <andy.yan@...k-chips.com>, <linux@...blig.org>, <javierm@...hat.com>,
        <linux-kernel@...r.kernel.org>, <devarsht@...com>,
        <j-choudhary@...com>, <u-kumar1@...com>, <s-jain1@...com>,
        <andrzej.hajda@...el.com>, <neil.armstrong@...aro.org>,
        <rfoss@...nel.org>, <Laurent.pinchart@...asonboard.com>,
        <mripard@...nel.org>, <lumag@...nel.org>, <dianders@...omium.org>,
        <dri-devel@...ts.freedesktop.org>
Subject: Re: [PATCH v5 3/6] drm/bridge: cadence: cdns-mhdp8546-core: Set the
 mhdp connector earlier in atomic_enable()


On 9/1/25 15:35, Tomi Valkeinen wrote:
> Hi,
>
> On 11/08/2025 10:59, Harikrishna Shenoy wrote:
>> From: Jayesh Choudhary <j-choudhary@...com>
>>
>> In case if we get errors in cdns_mhdp_link_up() or cdns_mhdp_reg_read()
>> in atomic_enable, we will go to cdns_mhdp_modeset_retry_fn() and will hit
>> NULL pointer while trying to access the mutex. We need the connector to
>> be set before that. Unlike in legacy !(DBANC) cases, we do not have
>> connector initialised in bridge_attach(). So set the mhdp->connector
>> in atomic_enable() earlier to avoid possible NULL pointer.
>>
>> Fixes: c932ced6b585 ("drm/tidss: Update encoder/bridge chain connect model")
>> Signed-off-by: Jayesh Choudhary <j-choudhary@...com>
>> ---
>>   .../drm/bridge/cadence/cdns-mhdp8546-core.c   | 20 +++++++++----------
>>   1 file changed, 10 insertions(+), 10 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
>> index c2ce3d6e5a88..b2f5a48cac2d 100644
>> --- a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
>> +++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c
>> @@ -1759,12 +1759,21 @@ static void cdns_mhdp_atomic_enable(struct drm_bridge *bridge,
>>   	struct drm_bridge_state *new_state;
>>   	const struct drm_display_mode *mode;
>>   	u32 resp;
>> -	int ret;
>> +	int ret = 0;
>>   
>>   	dev_dbg(mhdp->dev, "bridge enable\n");
>>   
>>   	mutex_lock(&mhdp->link_mutex);
>>   
>> +	mhdp->connector = drm_atomic_get_new_connector_for_encoder(state,
>> +								   bridge->encoder);
>> +	if (WARN_ON(!mhdp->connector))
>> +		goto out;
>> +
>> +	conn_state = drm_atomic_get_new_connector_state(state, mhdp->connector);
>> +	if (WARN_ON(!conn_state))
>> +		goto out;
> You are just moving code here, but... Shouldn't these be errors? If I
> read this right, ret is 0 here, and thus if we hit either of those
> issues above, we'll return 0.
>
>   Tomi
>
This is a void function, no need to return
>> +
>>   	if (mhdp->plugged && !mhdp->link_up) {
>>   		ret = cdns_mhdp_link_up(mhdp);
>>   		if (ret < 0)
>> @@ -1784,15 +1793,6 @@ static void cdns_mhdp_atomic_enable(struct drm_bridge *bridge,
>>   	cdns_mhdp_reg_write(mhdp, CDNS_DPTX_CAR,
>>   			    resp | CDNS_VIF_CLK_EN | CDNS_VIF_CLK_RSTN);
>>   
>> -	mhdp->connector = drm_atomic_get_new_connector_for_encoder(state,
>> -								   bridge->encoder);
>> -	if (WARN_ON(!mhdp->connector))
>> -		goto out;
>> -
>> -	conn_state = drm_atomic_get_new_connector_state(state, mhdp->connector);
>> -	if (WARN_ON(!conn_state))
>> -		goto out;
>> -
>>   	if (mhdp->hdcp_supported &&
>>   	    mhdp->hw_state == MHDP_HW_READY &&
>>   	    conn_state->content_protection ==