From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
To: Matt Ranostay <matt@ranostay.consulting>
Cc: linux-media@vger.kernel.org, Attila Kinali <attila@kinali.ch>,
Marek Vasut <marex@denx.de>, Luca Barbato <lu_zero@gentoo.org>
Subject: Re: [PATCH v5] media: video-i2c: add video-i2c driver
Date: Fri, 13 Jan 2017 13:47 +0200 [thread overview]
Message-ID: <4891099.8VPSOPnGff@avalon> (raw)
In-Reply-To: <1482548666-25272-1-git-send-email-matt@ranostay.consulting>
Hi Matt,
Thank you for the patch.
On Friday 23 Dec 2016 19:04:26 Matt Ranostay wrote:
> There are several thermal sensors that only have a low-speed bus
> interface but output valid video data. This patchset enables support
> for the AMG88xx "Grid-Eye" sensor family.
>
> Cc: Attila Kinali <attila@kinali.ch>
> Cc: Marek Vasut <marex@denx.de>
> Cc: Luca Barbato <lu_zero@gentoo.org>
> Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> Signed-off-by: Matt Ranostay <matt@ranostay.consulting>
> ---
> Changes from v1:
> * correct i2c_polling_remove() operations
> * fixed delay calcuation in buffer_queue()
> * add include linux/slab.h
>
> Changes from v2:
> * fix build error due to typo in include of slab.h
>
> Changes from v3:
> * switch data transport to a kthread to avoid to .buf_queue that can't sleep
> * change naming from i2c-polling to video-i2c
> * make the driver for single chipset under another uses the driver
>
> Changes from v4:
> * fix wraparound issue with jiffies and schedule_timeout_interruptible()
>
> drivers/media/i2c/Kconfig | 9 +
> drivers/media/i2c/Makefile | 1 +
> drivers/media/i2c/video-i2c.c | 569 ++++++++++++++++++++++++++++++++++++++
> 3 files changed, 579 insertions(+)
> create mode 100644 drivers/media/i2c/video-i2c.c
[snip]
> diff --git a/drivers/media/i2c/video-i2c.c b/drivers/media/i2c/video-i2c.c
> new file mode 100644
> index 000000000000..9390560bd117
> --- /dev/null
> +++ b/drivers/media/i2c/video-i2c.c
[snip]
> +#include <linux/delay.h>
> +#include <linux/freezer.h>
> +#include <linux/i2c.h>
> +#include <linux/list.h>
> +#include <linux/module.h>
> +#include <linux/mutex.h>
> +#include <linux/sched.h>
> +#include <linux/slab.h>
> +#include <linux/videodev2.h>
> +#include <media/v4l2-common.h>
> +#include <media/v4l2-ioctl.h>
> +#include <media/v4l2-device.h>
> +#include <media/v4l2-fh.h>
> +#include <media/v4l2-event.h>
Alphabetical order please.
> +#include <media/videobuf2-v4l2.h>
> +#include <media/videobuf2-vmalloc.h>
[snip]
> +static struct v4l2_fmtdesc amg88xx_format = {
> + .description = "12-bit Greyscale",
If I'm not mistaken the V4L2 core fills that for you nowadays, you don't have
to set it.
> + .pixelformat = V4L2_PIX_FMT_Y12,
> +};
> +
> +static struct v4l2_frmsize_discrete amg88xx_size = {
> + .width = 8,
> + .height = 8,
> +};
> +
> +struct video_i2c_chip {
> + /* video dimensions */
> + struct v4l2_fmtdesc *format;
> + struct v4l2_frmsize_discrete *size;
You can make those two pointers (and the variables they point to) const.
> +
> + /* max frames per second */
> + unsigned int max_fps;
> +
> + /* pixel buffer size */
> + unsigned int buffer_size;
> +
> + /* pixel size in bits */
> + unsigned int bpp;
> +
> + /* xfer function */
> + int (*xfer)(struct video_i2c_data *data, char *buf);
> +};
[snip]
> +static int video_i2c_thread_vid_cap(void *priv)
> +{
> + struct video_i2c_data *data = priv;
> + struct video_i2c_buffer *vid_cap_buf = NULL;
> +
> + set_freezable();
> +
> + do {
> + unsigned int start_jiffies = jiffies;
jiffies is an unsigned long.
> + unsigned int delay = msecs_to_jiffies(1000 / data->chip-
>max_fps);
> + int schedule_delay;
> +
> + try_to_freeze();
> +
> + mutex_lock(&data->lock);
Why do you need the mutex here ?
> + spin_lock(&data->slock);
> +
> + if (!list_empty(&data->vid_cap_active)) {
> + vid_cap_buf = list_entry(data->vid_cap_active.next,
> + struct video_i2c_buffer,
list);
> + list_del(&vid_cap_buf->list);
> + }
> +
> + if (vid_cap_buf) {
vid_cap_buf will only be non-NULL in all but the first iteration of the loop,
even if the list is empty. You should declare the variable inside the loop,
not at the function level.
> + struct vb2_buffer *vb2_buf = &vid_cap_buf->vb.vb2_buf;
> + void *vbuf = vb2_plane_vaddr(vb2_buf, 0);
> + int ret = data->chip->xfer(data, vbuf);
> +
> + vb2_buf->timestamp = ktime_get_ns();
> + vb2_buffer_done(vb2_buf, ret ?
> + VB2_BUF_STATE_ERROR :
VB2_BUF_STATE_DONE);
> + }
> +
> + spin_unlock(&data->slock);
The spinlock must be unlocked before data->chip->xfer. You can't hold it and
sleep.
> + mutex_unlock(&data->lock);
> +
> + schedule_delay = delay - (jiffies - start_jiffies);
Does this still work when jiffies wraps around ?
> + if (schedule_delay < 0)
> + schedule_delay = delay;
> +
> + schedule_timeout_interruptible(schedule_delay);
> + } while (!kthread_should_stop());
> +
> + return 0;
> +}
> +
> +static int start_streaming(struct vb2_queue *vq, unsigned int count)
> +{
> + struct video_i2c_data *data = vb2_get_drv_priv(vq);
> +
> + if (data->kthread_vid_cap)
> + return 0;
> +
> + data->kthread_vid_cap = kthread_run(video_i2c_thread_vid_cap, data,
> + "%s-vid-cap", data-
>v4l2_dev.name);
> +
No need for a blank line here.
> + if (IS_ERR(data->kthread_vid_cap)) {
> + struct video_i2c_buffer *buf, *tmp;
> +
> + list_for_each_entry_safe(buf, tmp, &data->vid_cap_active,
list) {
> + list_del(&buf->list);
> + vb2_buffer_done(&buf->vb.vb2_buf,
> + VB2_BUF_STATE_QUEUED);
> + }
Shouldn't you protect the whole loop with the data->slock spinlock ?
> +
> + return PTR_ERR(data->kthread_vid_cap);
> + }
> +
> + return 0;
> +}
> +
> +static void stop_streaming(struct vb2_queue *vq)
> +{
> + struct video_i2c_data *data = vb2_get_drv_priv(vq);
> +
> + if (data->kthread_vid_cap == NULL)
> + return;
> +
> + mutex_lock(&data->lock);
Do you need the mutex here ?
> + spin_lock(&data->slock);
> +
> + while (!list_empty(&data->vid_cap_active)) {
> + struct video_i2c_buffer *buf;
> +
> + buf = list_entry(data->vid_cap_active.next,
> + struct video_i2c_buffer, list);
> + list_del(&buf->list);
> + vb2_buffer_done(&buf->vb.vb2_buf, VB2_BUF_STATE_ERROR);
> + }
> + spin_unlock(&data->slock);
> + mutex_unlock(&data->lock);
> +
> + kthread_stop(data->kthread_vid_cap);
Shouldn't you stop the thread first ?
> + data->kthread_vid_cap = NULL;
> +}
[snip]
> +
> +static int video_i2c_querycap(struct file *file, void *priv,
> + struct v4l2_capability *vcap)
> +{
> + struct video_i2c_data *data = video_drvdata(file);
> + struct i2c_client *client = data->client;
> +
> + strlcpy(vcap->driver, data->v4l2_dev.name, sizeof(vcap->driver));
> + strlcpy(vcap->card, "I2C Transport Video", sizeof(vcap->card));
Shouldn't this include the I2C device name ?
> +
> + sprintf(vcap->bus_info, "I2C:%d-%d", client->adapter->nr, client-
>addr);
> + return 0;
> +}
[snip]
> +static int video_i2c_enum_frameintervals(struct file *file, void *priv,
> + struct v4l2_frmivalenum *fe)
> +{
> + struct video_i2c_data *data = video_drvdata(file);
> + struct v4l2_frmsize_discrete *size = data->chip->size;
> +
> + if (fe->index > 0)
> + return -EINVAL;
> +
> + if ((fe->width != size->width) || (fe->height != size->height))
No need for the inner parentheses.
> + return -EINVAL;
> +
> + fe->type = V4L2_FRMIVAL_TYPE_DISCRETE;
> + fe->discrete.numerator = 1;
> + fe->discrete.denominator = data->chip->max_fps;
> +
> + return 0;
> +}
[snip]
> +static int video_i2c_fmt_vid_cap(struct file *file, void *fh,
> + struct v4l2_format *fmt)
Should you call the function video_i2c_s_fmt_vid_cap ?
> +{
> + struct video_i2c_data *data = video_drvdata(file);
> + int ret = video_i2c_try_fmt_vid_cap(file, fh, fmt);
> +
> + if (ret < 0)
> + return ret;
video_i2c_try_fmt_vid_cap() always returns 0, so you can drop the check (and
the ret variable).
> + if (vb2_is_busy(&data->vb_vidq))
> + return -EBUSY;
> +
> + return 0;
> +}
> +
> +static int video_i2c_g_parm(struct file *filp, void *priv,
> + struct v4l2_streamparm *parm)
> +{
> + struct video_i2c_data *data = video_drvdata(filp);
> +
> + if (parm->type != V4L2_BUF_TYPE_VIDEO_CAPTURE)
> + return -EINVAL;
> +
> + parm->parm.capture.readbuffers = 1;
> + parm->parm.capture.capability = V4L2_CAP_TIMEPERFRAME;
> + parm->parm.capture.timeperframe.numerator = 1;
> + parm->parm.capture.timeperframe.denominator = data->chip->max_fps;
> +
> + return 0;
> +}
> +
> +static int video_i2c_s_parm(struct file *filp, void *priv,
> + struct v4l2_streamparm *parm)
> +{
> + if (parm->type != V4L2_BUF_TYPE_VIDEO_CAPTURE)
> + return -EINVAL;
> +
> + return video_i2c_g_parm(filp, priv, parm);
You can just use the video_i2c_g_parm() function as the .vidioc_s_parm
handler, there's no need to declare a separate function.
> +}
[snip]
> +static int video_i2c_probe(struct i2c_client *client,
> + const struct i2c_device_id *id)
> +{
> + struct video_i2c_data *data;
> + struct v4l2_device *v4l2_dev;
> + struct vb2_queue *queue;
> + int ret;
> +
> + data = kzalloc(sizeof(*data), GFP_KERNEL);
> + if (!data)
> + return -ENOMEM;
> +
> + data->chip = &video_i2c_chip;
> + data->client = client;
> + v4l2_dev = &data->v4l2_dev;
> + strlcpy(v4l2_dev->name, VIDEO_I2C_DRIVER, sizeof(v4l2_dev->name));
> +
> + ret = v4l2_device_register(&client->dev, v4l2_dev);
> + if (ret < 0)
> + goto error_free_device;
> +
> + mutex_init(&data->lock);
> + mutex_init(&data->queue_lock);
> +
> + queue = &data->vb_vidq;
> + queue->type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
> + queue->io_modes = VB2_MMAP | VB2_USERPTR | VB2_READ;
> + queue->timestamp_flags = V4L2_BUF_FLAG_TIMESTAMP_MONOTONIC |
> + V4L2_BUF_FLAG_TSTAMP_SRC_EOF;
> + queue->drv_priv = data;
> + queue->buf_struct_size = sizeof(struct video_i2c_buffer);
> + queue->min_buffers_needed = 1;
> + queue->ops = &video_i2c_video_qops;
> + queue->mem_ops = &vb2_vmalloc_memops;
> +
> + ret = vb2_queue_init(queue);
> + if (ret < 0)
> + goto error_free_device;
This should be error_unregister_device.
> +
> + data->vdev.queue = queue;
> + data->vdev.queue->lock = &data->queue_lock;
> +
> + strlcpy(data->vdev.name, "I2C Transport Video", sizeof(data-
>vdev.name));
Shouldn't this be called after the I2C device name ?
> + data->vdev.v4l2_dev = v4l2_dev;
> + data->vdev.fops = &video_i2c_fops;
> + data->vdev.lock = &data->lock;
> + data->vdev.ioctl_ops = &video_i2c_ioctl_ops;
> + data->vdev.release = video_i2c_release;
> + data->vdev.device_caps = V4L2_CAP_VIDEO_CAPTURE |
> + V4L2_CAP_READWRITE | V4L2_CAP_STREAMING;
> +
> + spin_lock_init(&data->slock);
> + INIT_LIST_HEAD(&data->vid_cap_active);
> +
> + video_set_drvdata(&data->vdev, data);
> + i2c_set_clientdata(client, data);
> +
> + ret = video_register_device(&data->vdev, VFL_TYPE_GRABBER, -1);
> + if (ret < 0)
> + goto error_unregister_device;
> +
> + return 0;
> +
> +error_unregister_device:
> + v4l2_device_unregister(v4l2_dev);
> +
> +error_free_device:
> + kfree(data);
> +
> + return ret;
> +}
> +
> +static int video_i2c_remove(struct i2c_client *client)
> +{
> + struct video_i2c_data *data = i2c_get_clientdata(client);
> +
> + v4l2_device_unregister(&data->v4l2_dev);
> + video_unregister_device(&data->vdev);
I believe you have to swap those two calls, as vdev references v4l2_dev.
> +
> + return 0;
> +}
[snip]
--
Regards,
Laurent Pinchart
next prev parent reply other threads:[~2017-01-13 11:46 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-24 3:04 [PATCH v5] media: video-i2c: add video-i2c driver Matt Ranostay
2017-01-09 5:17 ` Matt Ranostay
2017-01-09 5:33 ` Marek Vasut
2017-01-13 4:45 ` Matt Ranostay
2017-01-13 10:22 ` Laurent Pinchart
2017-01-14 6:27 ` Matt Ranostay
2017-01-13 11:47 ` Laurent Pinchart [this message]
2017-01-14 6:42 ` Matt Ranostay
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=4891099.8VPSOPnGff@avalon \
--to=laurent.pinchart@ideasonboard.com \
--cc=attila@kinali.ch \
--cc=linux-media@vger.kernel.org \
--cc=lu_zero@gentoo.org \
--cc=marex@denx.de \
--cc=matt@ranostay.consulting \
/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