Re: [PATCH v3 12/13] gpiolib: add new ioctl() for monitoring changes in line info

[Bartosz Golaszewski]

czw., 19 gru 2019 o 19:17 Andy Shevchenko <andy.shevchenko@xxxxxxxxx>
napisaÅ(a):
>
> On Thu, Dec 19, 2019 at 7:17 PM Bartosz Golaszewski <brgl@xxxxxxxx> wrote:
> >
> > From: Bartosz Golaszewski <bgolaszewski@xxxxxxxxxxxx>
> >
> > Currently there is no way for user-space to be informed about changes
> > in status of GPIO lines e.g. when someone else requests the line or its
> > config changes. We can only periodically re-read the line-info. This
> > is fine for simple one-off user-space tools, but any daemon that provides
> > a centralized access to GPIO chips would benefit hugely from an event
> > driven line info synchronization.
> >
> > This patch adds a new ioctl() that allows user-space processes to reuse
> > the file descriptor associated with the character device for watching
> > any changes in line properties. Every such event contains the updated
> > line information.
> >
> > Currently the events are generated on three types of status changes: when
> > a line is requested, when it's released and when its config is changed.
> > The first two are self-explanatory. For the third one: this will only
> > happen when another user-space process calls the new SET_CONFIG ioctl()
> > as any changes that can happen from within the kernel (i.e.
> > set_transitory() or set_debounce()) are of no interest to user-space.
>
> > -       } else if (cmd == GPIO_GET_LINEINFO_IOCTL) {
> > +       } else if (cmd == GPIO_GET_LINEINFO_IOCTL ||
> > +                  cmd == GPIO_GET_LINEINFO_WATCH_IOCTL) {
>
> Wouldn't be better for maintenance to have them separated from the day 1?
>

Sure. Will fix in v4.

> ...
>
> > +       if (test_bit(desc_to_gpio(desc), priv->watched_lines)) {
>
> if (!test_bit(...))
>   return NOTIFY_DONE;
>
> ?
>
> > +                       pr_debug_ratelimited(
> > +                               "%s: lineinfo event FIFO is full - event dropped\n",
>
> > +                               __func__);
>
> This is in 99.9% cases redundant in *_dbg() calls.
>
> > +               ret = NOTIFY_OK;
> > +       }
> > +
> > +       return ret;
>
> return NOTIFY_OK;

Yeah, makes sense.

> ?
>
> > +}
>
> ...
>
> > @@ -3111,6 +3285,7 @@ static int gpio_set_bias(struct gpio_chip *chip, struct gpio_desc *desc)
> >                 if (ret != -ENOTSUPP)
> >                         return ret;
> >         }
> > +
> >         return 0;
> >  }
> >
>
> This hunk doesn't belong to this patch.
>

Ha! Of course it doesn't. :)

> ...
>
> > +/**
> > + * struct gpioline_info_changed - Information about a change in status
> > + * of a GPIO line
> > + * @info: updated line information
> > + * @timestamp: estimate of time of status change occurrence, in nanoseconds
> > + * and GPIOLINE_CHANGED_CONFIG
> > + * @event_type: one of GPIOLINE_CHANGED_REQUESTED, GPIOLINE_CHANGED_RELEASED
> > + */
> > +struct gpioline_info_changed {
>
> > +       struct gpioline_info info;
>
> Is this guaranteed to be always 8 byte aligned?
> I'm expecting to see some comments there and / or here about it.
>

struct gpioline_info alone is 32-bit aligned but its size is 72 bytes
which works for 64-bit alignment. This new structure's biggest element
in 64-bit, so it's 64-bit aligned on 64-bit arch. We have 72 bytes of
gpioline_info, 8 bytes of timestamp, 32 bytes of event type and 5 * 32
bytes of padding. Should be fine, but I'll add comments to the header.

Bart

> > +       __u64 timestamp;
> > +       __u32 event_type;
> > +       __u32 padding[5]; /* for future use */
> > +};
>
> --
> With Best Regards,
> Andy Shevchenko