Re: [PATCHv4 3/8] gpu: host1x: Add channel support

[Mark Zhang <nvmarkzhang@gmail.com>]

On 01/02/2013 05:31 PM, Terje Bergström wrote:
> On 02.01.2013 09:40, Mark Zhang wrote:
>> On 12/21/2012 07:39 PM, Terje Bergstrom wrote:
>>> Add support for host1x client modules, and host1x channels to submit
>>> work to the clients. The work is submitted in GEM CMA buffers, so
>>> this patch adds support for them.
>>>
>>> Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com>
>>> ---
>> [...]
>>> +/*
>>> + * Begin a cdma submit
>>> + */
>>> +int host1x_cdma_begin(struct host1x_cdma *cdma, struct host1x_job *job)
>>> +{
>>> +	struct host1x *host1x = cdma_to_host1x(cdma);
>>> +
>>> +	mutex_lock(&cdma->lock);
>>> +
>>> +	if (job->timeout) {
>>> +		/* init state on first submit with timeout value */
>>> +		if (!cdma->timeout.initialized) {
>>> +			int err;
>>> +			err = host1x->cdma_op.timeout_init(cdma,
>>> +					job->syncpt_id);
>>> +			if (err) {
>>> +				mutex_unlock(&cdma->lock);
>>> +				return err;
>>> +			}
>>> +		}
>>> +	}
>>> +	if (!cdma->running)
>>> +		host1x->cdma_op.start(cdma);
>>> +
>>> +	cdma->slots_free = 0;
>>> +	cdma->slots_used = 0;
>>> +	cdma->first_get = host1x->cdma_pb_op.putptr(&cdma->push_buffer);
>>> +
>>> +	trace_host1x_cdma_begin(job->ch->dev->name);
>>
>> Seems missing "mutex_unlock(&cdma->lock);" here.
> 
> That's intentional. Writing a job to channel is atomic, so lock is taken
> from host1x_cdma_begin() until host1x_cdma_end().
> 

Okay. So can we consider that lock and unlock this mutex in the function
which calls host1x_cdma_begin and host1x_cdma_end, that means, in
current implementation, function "channel_submit"?

Mark
> Terje
>