Re: [PATCH v4 1/3] drm/bridge: ti-sn65dsi83: Fix error handling in sn65dsi83_reset_work()

[Esben Haabendal]

"Herve Codina" <herve.codina@xxxxxxxxxxx> writes:

> Hi Esben,
>
> On Mon, 10 Aug 2026 09:19:40 +0200
> Esben Haabendal <esben@xxxxxxxxxx> wrote:
>
>> The error handling of sn65dsi83_reset_pipe() failure in
>> sn65dsi83_reset_work() has been a bit strange all the time, missing both
>> enable_irq() and drm_bridge_exit() in case of failure. But since
>> sn65dsi83_reset_pipe() have never been failing (always returning 0), it has
>> not caused any problems.
>>
>> As we are going to change sn65dsi83_reset_pipe() to return error values in
>> case of failure, we need to fix this.
>>
>> Fixes: ad5c6ecef27e ("drm: bridge: ti-sn65dsi83: Add error recovery mechanism")
>> Cc: stable@xxxxxxxxxxxxxxx
>> Signed-off-by: Esben Haabendal <esben@xxxxxxxxxx>
>> ---
>>  drivers/gpu/drm/bridge/ti-sn65dsi83.c | 5 ++---
>>  1 file changed, 2 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi83.c b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
>> index 42b451432bbb..0ccfff401b2a 100644
>> --- a/drivers/gpu/drm/bridge/ti-sn65dsi83.c
>> +++ b/drivers/gpu/drm/bridge/ti-sn65dsi83.c
>> @@ -417,10 +417,9 @@ static void sn65dsi83_reset_work(struct work_struct *ws)
>>
>>  	/* Reset the pipe */
>>  	ret = sn65dsi83_reset_pipe(ctx);
>> -	if (ret) {
>> +	if (ret)
>>  		dev_err(ctx->dev, "reset pipe failed %pe\n", ERR_PTR(ret));
>> -		return;
>> -	}
>> +
>>  	if (ctx->irq)
>>  		enable_irq(ctx->irq);
>>
>>
>
> In this v4, enable_irq() is called even if we cannot reset the pipe. I know that
> sn65dsi83_reset_pipe() always returns 0 (and this will be change) but we have to
> take care here.
>
> If the pipe cannot be reset, the sn65dsi83 component could be in a state where
> it can force the interrupt line. This was the reason the irq is disabled
>   https://elixir.bootlin.com/linux/v7.2/source/drivers/gpu/drm/bridge/ti-sn65dsi83.c#L455

Ok. But maybe take it in two steps.

The change here, together with the change that makes
sn65dsi83_reset_pipe() return non-zero on error, should ensure that
behavior is unchanged. As long as sn65dsi83_reset_pipe() was always
returning 0, we have been calling enable_irq() even if the pipe cannot
be reset.

If that is a problem, I think it is fair to do that as a separate
change. As you say, we have to take care here.

If we leave the irq disabled when sn65dsi83_reset_pipe() fails, do we
simply want to leave the device with irq permanently disabled, or should
we add some additional logic to sn65dsi83_handle_errors() so that we
will try to recover periodically via sn65dsi83_monitor_work()?

I think this is tricky enough to warrant a new patch series.

/Esben