Message-ID: <CAHp75VfoOCJDmpDJ6wTEKTDUO1zFRZ6MRSN7dy3cAdepv0s2rQ@mail.gmail.com>
Date:   Thu, 24 Sep 2020 11:39:03 +0300
From:   Andy Shevchenko <andy.shevchenko@...il.com>
To:     Kent Gibson <warthog618@...il.com>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        "open list:GPIO SUBSYSTEM" <linux-gpio@...r.kernel.org>,
        Bartosz Golaszewski <bgolaszewski@...libre.com>,
        Linus Walleij <linus.walleij@...aro.org>,
        Arnd Bergmann <arnd@...db.de>
Subject: Re: [PATCH v9 08/20] gpiolib: cdev: support GPIO_V2_GET_LINEINFO_IOCTL
 and GPIO_V2_GET_LINEINFO_WATCH_IOCTL

On Thu, Sep 24, 2020 at 5:39 AM Kent Gibson <warthog618@...il.com> wrote:
> On Wed, Sep 23, 2020 at 06:41:45PM +0300, Andy Shevchenko wrote:
> > On Tue, Sep 22, 2020 at 5:35 AM Kent Gibson <warthog618@...il.com> wrote:

...

> > > +       memcpy(info_v1->consumer, info_v2->consumer,
> > > +              sizeof(info_v1->consumer));
> >
> > One line?
> >
>
> It can be now the line length limit has been raised - it just breaks the
> old 80 character limit.

I really wouldn't care about this if it's only for a couple of characters.

...

> > > +static int lineinfo_ensure_abi_version(struct gpio_chardev_data *cdata,
> > > +                                      unsigned int version)
> > > +{
> >
> > > +       int abiv = atomic_read(&cdata->watch_abi_version);
> > > +
> > > +       if (abiv == 0) {
> >
> > > +               atomic_cmpxchg(&cdata->watch_abi_version, 0, version);
> > > +               abiv = atomic_read(&cdata->watch_abi_version);
> >
> > atomic_cmpxchng() returns a value.
>
> Yep, it returns the old value - which we don't care about - see below.

Then what's the point to read back?..

> > Also there are no barriers here...
> >
>
> No barriers required - the atomic_cmpxchg() is sufficient.
>
> > > +       }
> > > +       if (abiv != version)
> > > +               return -EPERM;
> >
> > I'm not sure I understand why this is atomic.
> >
>
> The algorithm requires some level of protection and atomic is
> sufficient.
>
> > Also this seems to be racy if cdata changed in background.
> >
>
> Can you provide a case?

CPU0:                CPU1:
 xchg()               ...
 ...                      xchg()
 ...                      read() -> OK
read() ->NOK

> The atomic_cmpxchg() ensures cdata->watch_abi_version is only set
> once - first in wins.  The atomic_read() is so we can check that
> the set version matches what the caller wants.
> Note that multiple callers may request the same version - and all
> should succeed.

So, that's basically what you need when using _old_ value.

0 means you were first, right?
Anything else you simply compare and bail out if it's not the same as
what has been asked.

>
> > Shouldn't be rather
> >
> > if (atomic_cmpxchg() == 0) {
> >   if (atomic_read() != version)
> >     return ...;
> > }
> >
>
> My algorithm allows for multiple callers requesting the same version
> to all succeed.  Yours would fail the first conditional for all but
> the first, and you haven't provided an else for that case...
>
> ... but it would probably look the same so the conditional is pointless,
> or it would reject the request - which would be wrong.
>
> > But here is still the question: why do you expect the version to be
> > changed on background? And what about barriers?
> >
>
> While it is unlikely that userspace will be attempting to use both ABI
> versions simultaneously on the same chip request, it is a possiblity and
> so needs to be protected against. And better to have the watch request
> fail than the read fail or worse - return the wrong struct version.
>
> The atomic_cmpxchg() is sufficient for this algorithm - no barriers
> required.  It could also be written with a spinlock but I was trying to
> avoid locks unless they were absolutely necessary.  A spinlock version
> may arguably be more readable, but it would certainly be more verbose,
> larger and slower.
>
> I'm happy to add some documentation to the function if that would help.

Yes, I guess documentation is what is eagerly needed here.

> > > +       return 0;
> > > +}
> > > +#endif
> > > +
> > > +static int lineinfo_get(struct gpio_chardev_data *cdev, void __user *ip,
> > > +                       bool watch)
> > > +{
> > > +       struct gpio_desc *desc;
> > > +       struct gpio_v2_line_info lineinfo;
> > > +
> > > +       if (copy_from_user(&lineinfo, ip, sizeof(lineinfo)))
> > > +               return -EFAULT;
> > > +
> > > +       if (memchr_inv(lineinfo.padding, 0, sizeof(lineinfo.padding)))
> > > +               return -EINVAL;
> > > +
> > > +       desc = gpiochip_get_desc(cdev->gdev->chip, lineinfo.offset);
> > > +       if (IS_ERR(desc))
> > > +               return PTR_ERR(desc);
> > > +
> > > +       if (watch) {
> > > +#ifdef CONFIG_GPIO_CDEV_V1
> >
> > > +               if (lineinfo_ensure_abi_version(cdev, 2))
> > > +                       return -EPERM;
> >
> > Can't you propagate error code from the function?
> >
>
> You mean:
> +               ret = lineinfo_ensure_abi_version(cdev, 2)
> +               if (ret)
> +                       return ret;
>
> That seems more verbose and less clear.  And I'd need to conditionally
> declare a ret - as this test is compiled out if CDEV_V1 is not defined.
>
> I did flip-flop on what lineinfo_ensure_abi_version() should return -
> either a bool or an error code.
>
> If a bool then the code would include the dreaded negative conditional
> ;-(:
>
> +               if (!lineinfo_is_abi_version(cdev, 2))
> +                       return -EPERM;
>
> so I eventually settled for the error code.  But I'm on the fence on
> this one and happy to change it if you think the bool form is clearer.
>
> Cheers,
> Kent.



-- 
With Best Regards,
Andy Shevchenko

Powered by blists - more mailing lists