Message-ID: <20210205124401.00003dd0@Huawei.com>
Date:   Fri, 5 Feb 2021 12:44:01 +0000
From:   Jonathan Cameron <Jonathan.Cameron@...wei.com>
To:     Alexandru Ardelean <ardeleanalex@...il.com>
CC:     Alexandru Ardelean <alexandru.ardelean@...log.com>,
        LKML <linux-kernel@...r.kernel.org>,
        linux-iio <linux-iio@...r.kernel.org>,
        Lars-Peter Clausen <lars@...afoo.de>,
        "Hennerich, Michael" <Michael.Hennerich@...log.com>,
        Jonathan Cameron <jic23@...nel.org>,
        Nuno Sá <nuno.sa@...log.com>,
        "Bogdan, Dragos" <dragos.bogdan@...log.com>,
        "Rafael J. Wysocki" <rafael@...nel.org>,
        "Greg Kroah-Hartman" <gregkh@...uxfoundation.org>
Subject: Re: [PATCH v3 11/11] iio: buffer: add ioctl() to support opening
 extra buffers for IIO device

On Fri, 5 Feb 2021 11:51:13 +0200
Alexandru Ardelean <ardeleanalex@...il.com> wrote:

> On Thu, Feb 4, 2021 at 9:06 PM Jonathan Cameron
> <Jonathan.Cameron@...wei.com> wrote:
> >
> > On Mon, 1 Feb 2021 16:51:05 +0200
> > Alexandru Ardelean <alexandru.ardelean@...log.com> wrote:
> >  
> > > With this change, an ioctl() call is added to open a character device for a
> > > buffer. The ioctl() number is 'i' 0x91, which follows the
> > > IIO_GET_EVENT_FD_IOCTL ioctl.
> > >
> > > The ioctl() will return a 0 FD for the first buffer, as that FD for buffer0
> > > is the same FD as the one used for this ioctl().  
> >
> > That sounds dangerous as code might just use it without checking properly.
> > Perhaps take a leaf out of open() and return negative?  
> 
> So, initially this returned -EBUSY.
> But that is also confusing.
> And returning 0 isn't great either.
> 
> I've been thinking about this for the last few days.
> Would it be too bad if return another FD for the same buffer0 ?
> Admittedly, you'd be able to access the same buffer0 via 2 FDs, but
> coming back to the idea of creating a new API and having to live with
> the old one as well, it feels like these 2 FDs for buffer0 are falling
> under the same principle.

It would be in the crazy category if anyone actually tried to read from
both, unless we did something to spin up a clone of the data and that's
getting complex.  I guess we could do something like return errors on
all the original FD activities related to the new one (i.e. poll, read)
That might work. What do you think?

> 
> >
> > What would happen if we just returned an error in this path?  
> 
> We need to find a good error code [if we return an error].
> -EBUSY would sound the closest to something correct.
range error maybe?  or just -EINVAL if we do go this way.

Jonathan

> 
> >  
> > >
> > > For any other extra buffer, this ioctl() will return an anon inode FD that
> > > would access any extra buffer.
> > >
> > > Right now, there doesn't seem to be (or I couldn't find) a way for this
> > > ioctl() to return the FD for buffer0 (i.e. to return the same FD as used
> > > for the ioctl()).
> > > So, usespace would need to know that  ioctl(fd,
> > > IIO_GET_EVENT_FD_IOCTL, 0) will return FD 0.
> > > We could also return another FD for buffer 0, but duplicating FDs for the
> > > same IIO buffer sounds problematic.
> > >
> > > Also, there is no IIO_BUFFER_GET_BUFFER_COUNT ioctl() implemented, as the
> > > index for each buffer (and the count) can be deduced from the
> > > '/sys/bus/iio/devices/iio:deviceX/bufferY' folders (i.e the number of
> > > bufferY folders).
> > >
> > > Used following C code to test this:
> > > -------------------------------------------------------------------
> > >
> > >  #include <stdio.h>
> > >  #include <stdlib.h>
> > >  #include <unistd.h>
> > >  #include <sys/ioctl.h>
> > >  #include <fcntl.h"
> > >  #include <errno.h>
> > >
> > >  #define IIO_BUFFER_GET_FD_IOCTL      _IOWR('i', 0x91, int)
> > >
> > > int main(int argc, char *argv[])
> > > {
> > >         int fd;
> > >         int fd1;
> > >         int ret;
> > >
> > >         if ((fd = open("/dev/iio:device0", O_RDWR))<0) {
> > >                 fprintf(stderr, "Error open() %d errno %d\n",fd, errno);
> > >                 return -1;
> > >         }
> > >
> > >         fprintf(stderr, "Using FD %d\n", fd);
> > >
> > >         fd1 = atoi(argv[1]);
> > >
> > >         ret = ioctl(fd, IIO_BUFFER_GET_FD_IOCTL, &fd1);
> > >         if (ret < 0) {
> > >                 fprintf(stderr, "Error for buffer %d ioctl() %d errno %d\n", fd1, ret, errno);
> > >                 close(fd);
> > >                 return -1;
> > >         }
> > >
> > >         fprintf(stderr, "Got FD %d\n", fd1);
> > >
> > >         close(fd1);
> > >         close(fd);
> > >
> > >         return 0;
> > > }
> > > -------------------------------------------------------------------
> > >
> > > Results are:
> > > -------------------------------------------------------------------
> > >  # ./test 0
> > >  Using FD 3
> > >  Got FD 0
> > >
> > >  # ./test 1
> > >  Using FD 3
> > >  Got FD 4
> > >
> > >  # ./test 2
> > >  Using FD 3
> > >  Got FD 4
> > >
> > >  # ./test 3
> > >  Using FD 3
> > >  Got FD 4
> > >
> > >  # ls /sys/bus/iio/devices/iio\:device0
> > >  buffer  buffer0  buffer1  buffer2  buffer3  dev
> > >  in_voltage_sampling_frequency  in_voltage_scale
> > >  in_voltage_scale_available
> > >  name  of_node  power  scan_elements  subsystem  uevent
> > > -------------------------------------------------------------------
> > >
> > > iio:device0 has some fake kfifo buffers attached to an IIO device.  
> > If you get a chance to add a parameter to the tools/iio/ buffer
> > program that does much the same as above that would be great.  
> 
> will take a look
> 
> >  
> > >
> > > Signed-off-by: Alexandru Ardelean <alexandru.ardelean@...log.com>  
> >
> > Otherwise a few trivials inline.
> >
> > Good work on bringing this new approach together so quickly.
> >
> > It's a lot simpler which is good and still gives us a reasonable interface.
> >
> > If you can update the example code in tree and perhaps add something to
> > the main IIO docs that would be great.
> > Documentation/driver-api/iio/buffers.rst I think.
> >
> > We'll also  (at somepoint) want to do a scrub of the ABI docs to
> > either include the new paths or make them less specific to incorporate
> > both.
> >
> > Thanks,
> >
> > Jonathan
> >  
> > > ---
> > >  drivers/iio/industrialio-buffer.c | 118 ++++++++++++++++++++++++++++++
> > >  drivers/iio/industrialio-core.c   |   8 ++
> > >  include/linux/iio/buffer_impl.h   |   5 ++
> > >  include/linux/iio/iio-opaque.h    |   2 +
> > >  include/uapi/linux/iio/buffer.h   |  10 +++
> > >  5 files changed, 143 insertions(+)
> > >  create mode 100644 include/uapi/linux/iio/buffer.h
> > >
> > > diff --git a/drivers/iio/industrialio-buffer.c b/drivers/iio/industrialio-buffer.c
> > > index a69bb705d173..119c2b5aa863 100644
> > > --- a/drivers/iio/industrialio-buffer.c
> > > +++ b/drivers/iio/industrialio-buffer.c
> > > @@ -9,6 +9,7 @@
> > >   * - Better memory allocation techniques?
> > >   * - Alternative access techniques?
> > >   */
> > > +#include <linux/anon_inodes.h>
> > >  #include <linux/kernel.h>
> > >  #include <linux/export.h>
> > >  #include <linux/device.h>
> > > @@ -1333,6 +1334,106 @@ void iio_buffer_unregister_legacy_sysfs_groups(struct iio_dev *indio_dev)
> > >       kfree(iio_dev_opaque->legacy_scan_el_group.attrs);
> > >  }
> > >
> > > +static int iio_buffer_chrdev_release(struct inode *inode, struct file *filep)
> > > +{
> > > +     struct iio_dev_buffer_pair *ib = filep->private_data;
> > > +     struct iio_dev *indio_dev = ib->indio_dev;
> > > +     struct iio_buffer *buffer = ib->buffer;
> > > +
> > > +     clear_bit(IIO_BUSY_BIT_POS, &buffer->flags);
> > > +     iio_device_put(indio_dev);
> > > +     kfree(ib);
> > > +
> > > +     return 0;
> > > +}
> > > +
> > > +static const struct file_operations iio_buffer_chrdev_fileops = {
> > > +     .owner = THIS_MODULE,
> > > +     .llseek = noop_llseek,
> > > +     .read = iio_buffer_read_outer_addr,
> > > +     .poll = iio_buffer_poll_addr,
> > > +     .compat_ioctl = compat_ptr_ioctl,
> > > +     .release = iio_buffer_chrdev_release,
> > > +};
> > > +
> > > +static long iio_device_buffer_getfd(struct iio_dev *indio_dev, unsigned long arg)
> > > +{
> > > +     struct iio_dev_opaque *iio_dev_opaque = to_iio_dev_opaque(indio_dev);
> > > +     int __user *ival = (int __user *)arg;
> > > +     char buf_name[sizeof("iio:buffer:xxx")];
> > > +     struct iio_dev_buffer_pair *ib;
> > > +     struct iio_buffer *buffer;
> > > +     int fd, idx;
> > > +
> > > +     if (copy_from_user(&idx, ival, sizeof(idx)))
> > > +             return -EFAULT;
> > > +
> > > +     if (idx == 0) {
> > > +             fd = 0;
> > > +             if (copy_to_user(ival, &fd, sizeof(fd)))
> > > +                     return -EFAULT;
> > > +             return 0;
> > > +     }
> > > +
> > > +     if (idx >= iio_dev_opaque->attached_buffers_cnt)
> > > +             return -ENOENT;
> > > +
> > > +     fd = mutex_lock_interruptible(&indio_dev->mlock);  
> >
> > Why mlock?  Might be fine, but this doesn't feel like it is in the normal scope
> > of that lock.  So perhaps a comment on why it makes sense here.  
> 
> Copy+paste from iio_event_getfd()
> It does the same thing.
> So, I'm a bit vague on this mlock being used here.
> Will take a look.

The Event one could well be wrong as well :)

> 
> >  
> > > +     if (fd)
> > > +             return fd;
> > > +
> > > +     buffer = iio_dev_opaque->attached_buffers[idx];
> > > +
> > > +     if (test_and_set_bit(IIO_BUSY_BIT_POS, &buffer->flags)) {
> > > +             fd = -EBUSY;
> > > +             goto error_unlock;
> > > +     }
> > > +
> > > +     iio_device_get(indio_dev);
> > > +
> > > +     ib = kzalloc(sizeof(*ib), GFP_KERNEL);
> > > +     if (!ib) {
> > > +             fd = -ENOMEM;
> > > +             goto error_iio_dev_put;
> > > +     }
> > > +
> > > +     ib->indio_dev = indio_dev;
> > > +     ib->buffer = buffer;
> > > +
> > > +     fd = anon_inode_getfd(buf_name, &iio_buffer_chrdev_fileops,
> > > +                           ib, O_RDWR | O_CLOEXEC);
> > > +     if (fd < 0)
> > > +             goto error_free_ib;
> > > +
> > > +     if (copy_to_user(ival, &fd, sizeof(fd))) {
> > > +             fd = -EFAULT;
> > > +             goto error_free_ib;
> > > +     }
> > > +
> > > +     mutex_unlock(&indio_dev->mlock);
> > > +     return fd;
> > > +
> > > +error_free_ib:
> > > +     kfree(ib);
> > > +error_iio_dev_put:
> > > +     iio_device_put(indio_dev);
> > > +     clear_bit(IIO_BUSY_BIT_POS, &buffer->flags);
> > > +error_unlock:
> > > +     mutex_unlock(&indio_dev->mlock);
> > > +     return fd;
> > > +}
> > > +
> > > +static long iio_device_buffer_ioctl(struct iio_dev *indio_dev, struct file *filp,
> > > +                                 unsigned int cmd, unsigned long arg)
> > > +{
> > > +     switch (cmd) {
> > > +     case IIO_BUFFER_GET_FD_IOCTL:
> > > +             return iio_device_buffer_getfd(indio_dev, arg);
> > > +     default:
> > > +             return IIO_IOCTL_UNHANDLED;
> > > +     }
> > > +}
> > > +
> > >  static int __iio_buffer_alloc_sysfs_and_mask(struct iio_buffer *buffer,
> > >                                            struct iio_dev *indio_dev,
> > >                                            int index)
> > > @@ -1460,6 +1561,7 @@ int iio_buffer_alloc_sysfs_and_mask(struct iio_dev *indio_dev)
> > >       const struct iio_chan_spec *channels;
> > >       int unwind_idx;
> > >       int ret, i;
> > > +     size_t sz;
> > >
> > >       channels = indio_dev->channels;
> > >       if (channels) {
> > > @@ -1481,6 +1583,18 @@ int iio_buffer_alloc_sysfs_and_mask(struct iio_dev *indio_dev)
> > >                       goto error_unwind_sysfs_and_mask;
> > >               }
> > >       }
> > > +     unwind_idx = iio_dev_opaque->attached_buffers_cnt - 1;
> > > +
> > > +     sz = sizeof(*(iio_dev_opaque->buffer_ioctl_handler));
> > > +     iio_dev_opaque->buffer_ioctl_handler = kzalloc(sz, GFP_KERNEL);  
> >  
> > > +     if (!iio_dev_opaque->buffer_ioctl_handler) {
> > > +             ret = -ENOMEM;
> > > +             goto error_unwind_sysfs_and_mask;
> > > +     }
> > > +
> > > +     iio_dev_opaque->buffer_ioctl_handler->ioctl = iio_device_buffer_ioctl;
> > > +     iio_device_ioctl_handler_register(indio_dev,
> > > +                                       iio_dev_opaque->buffer_ioctl_handler);
> > >
> > >       return 0;
> > >
> > > @@ -1503,6 +1617,10 @@ void iio_buffer_free_sysfs_and_mask(struct iio_dev *indio_dev)
> > >       if (!buffer)
> > >               return;
> > >
> > > +     iio_device_ioctl_handler_unregister(iio_dev_opaque->buffer_ioctl_handler);
> > > +     kfree(iio_dev_opaque->buffer_ioctl_handler);
> > > +     iio_dev_opaque->buffer_ioctl_handler = NULL;
> > > +
> > >       iio_buffer_unregister_legacy_sysfs_groups(indio_dev);
> > >
> > >       for (i = iio_dev_opaque->attached_buffers_cnt - 1; i >= 0; i--) {
> > > diff --git a/drivers/iio/industrialio-core.c b/drivers/iio/industrialio-core.c
> > > index 8af85838d1c2..78807b62ff52 100644
> > > --- a/drivers/iio/industrialio-core.c
> > > +++ b/drivers/iio/industrialio-core.c
> > > @@ -1722,6 +1722,9 @@ static int iio_chrdev_open(struct inode *inode, struct file *filp)
> > >       ib->indio_dev = indio_dev;
> > >       ib->buffer = indio_dev->buffer;
> > >
> > > +     if (indio_dev->buffer)
> > > +             test_and_set_bit(IIO_BUSY_BIT_POS, &indio_dev->buffer->flags);
> > > +  
> >
> > What is this test preventing?  I guess it's keeping the flag for the first
> > buffer in sync with what other buffers will look at, but does anything check those?
> > We'll not be able to open this one twice anyway due to the check above.
> >
> > Perhaps a comment on why this is here?  
> 
> Hmm, I admit this is a bit sloppy.
> This should technically be an impossible condition, so maybe just
> set_bit() would have been enough.
> It's making sure that the ioctl() on buffer 0 would have return
> -EBUSY; but that was on an older version.
> 
> >  
> > >       filp->private_data = ib;
> > >
> > >       return 0;
> > > @@ -1739,6 +1742,11 @@ static int iio_chrdev_release(struct inode *inode, struct file *filp)
> > >       struct iio_dev_buffer_pair *ib = filp->private_data;
> > >       struct iio_dev *indio_dev = container_of(inode->i_cdev,
> > >                                               struct iio_dev, chrdev);
> > > +     struct iio_buffer *buffer = ib->buffer;
> > > +
> > > +     if (buffer)
> > > +             clear_bit(IIO_BUSY_BIT_POS, &buffer->flags);
> > > +
> > >       clear_bit(IIO_BUSY_BIT_POS, &indio_dev->flags);
> > >       iio_device_put(indio_dev);
> > >       kfree(ib);
> > > diff --git a/include/linux/iio/buffer_impl.h b/include/linux/iio/buffer_impl.h
> > > index 768b90c64412..245b32918ae1 100644
> > > --- a/include/linux/iio/buffer_impl.h
> > > +++ b/include/linux/iio/buffer_impl.h
> > > @@ -6,6 +6,8 @@
> > >
> > >  #ifdef CONFIG_IIO_BUFFER
> > >
> > > +#include <uapi/linux/iio/buffer.h>
> > > +
> > >  struct iio_dev;
> > >  struct iio_buffer;
> > >
> > > @@ -72,6 +74,9 @@ struct iio_buffer {
> > >       /** @length: Number of datums in buffer. */
> > >       unsigned int length;
> > >
> > > +     /** @flags: File ops flags including busy flag. */
> > > +     unsigned long flags;
> > > +
> > >       /**  @bytes_per_datum: Size of individual datum including timestamp. */
> > >       size_t bytes_per_datum;
> > >
> > > diff --git a/include/linux/iio/iio-opaque.h b/include/linux/iio/iio-opaque.h
> > > index c909835b6247..2c3374d465da 100644
> > > --- a/include/linux/iio/iio-opaque.h
> > > +++ b/include/linux/iio/iio-opaque.h
> > > @@ -9,6 +9,7 @@
> > >   * @event_interface:         event chrdevs associated with interrupt lines
> > >   * @attached_buffers:                array of buffers statically attached by the driver
> > >   * @attached_buffers_cnt:    number of buffers in the array of statically attached buffers
> > > + * @buffer_ioctl_handler:    ioctl() handler for this IIO device's buffer interface
> > >   * @buffer_list:             list of all buffers currently attached
> > >   * @channel_attr_list:               keep track of automatically created channel
> > >   *                           attributes
> > > @@ -28,6 +29,7 @@ struct iio_dev_opaque {
> > >       struct iio_event_interface      *event_interface;
> > >       struct iio_buffer               **attached_buffers;
> > >       unsigned int                    attached_buffers_cnt;
> > > +     struct iio_ioctl_handler        *buffer_ioctl_handler;
> > >       struct list_head                buffer_list;
> > >       struct list_head                channel_attr_list;
> > >       struct attribute_group          chan_attr_group;
> > > diff --git a/include/uapi/linux/iio/buffer.h b/include/uapi/linux/iio/buffer.h
> > > new file mode 100644
> > > index 000000000000..de571c83c9f2
> > > --- /dev/null
> > > +++ b/include/uapi/linux/iio/buffer.h
> > > @@ -0,0 +1,10 @@
> > > +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
> > > +/* industrial I/O buffer definitions needed both in and out of kernel
> > > + */
> > > +
> > > +#ifndef _UAPI_IIO_BUFFER_H_
> > > +#define _UAPI_IIO_BUFFER_H_
> > > +
> > > +#define IIO_BUFFER_GET_FD_IOCTL              _IOWR('i', 0x91, int)
> > > +
> > > +#endif /* _UAPI_IIO_BUFFER_H_ */  
> >  

Powered by blists - more mailing lists