media: staging: rkisp1: cap: fix return values from pm functions

If no errors occurs, pm functions return usage counters, so they can
return positive numbers.
This happens when streaming from multiple capture devices (mainpath and
selfpath).

Fix simultaneous streaming from mainpath and selfpath by not failing
when pm usage counters returns a positive number.

Signed-off-by: Helen Koike <helen.koike@collabora.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
This commit is contained in:
Helen Koike 2020-03-16 22:00:43 +01:00 committed by Mauro Carvalho Chehab
parent b2b1046f1d
commit 60e915354c

View File

@ -892,7 +892,7 @@ static void rkisp1_vb2_stop_streaming(struct vb2_queue *queue)
v4l2_pipeline_pm_put(&node->vdev.entity);
ret = pm_runtime_put(rkisp1->dev);
if (ret)
if (ret < 0)
dev_err(rkisp1->dev, "power down failed error:%d\n", ret);
rkisp1_dummy_buf_destroy(cap);
@ -945,7 +945,7 @@ rkisp1_vb2_start_streaming(struct vb2_queue *queue, unsigned int count)
goto err_ret_buffers;
ret = pm_runtime_get_sync(cap->rkisp1->dev);
if (ret) {
if (ret < 0) {
dev_err(cap->rkisp1->dev, "power up failed %d\n", ret);
goto err_destroy_dummy;
}