Re: [patch/rfc 1/4] GPIO implementation framework

[David Brownell <david-b@pacbell.net>]

On Monday 29 October 2007, David Brownell wrote:
> 
> Provides new implementation infrastructure that platforms may choose to use
> when implementing the GPIO programming interface.  Platforms can update their
> GPIO support to use this.  The downside is slower access to non-inlined GPIOs;
> rarely a problem except when bitbanging some protocol.

I was asked just what that overhead *is* ... and it surprised me.
A summary of the results is appended to this note.

Fortuntely it turns out those problems all go away if the gpiolib
code uses a *raw* spinlock to guard its table lookups.  With a raw
spinlock, any performance impact of gpiolib seems to be well under
a microsecond in this bitbang context (and not objectionable).
Preempt became free; enabling debug options had only a minor cost.

That's as it should be, since the only substantive changes were to
grab and release a lock, do one table lookup a bit differently, and
add one indirection function call ... changes which should not have
any visible performance impact on per-bit codepaths, and one might
expect to cost on the order of one dozen instructions.


So the next version of this code will include a few minor bugfixes,
and will also use a raw spinlock to protect that table.  A raw lock
seems appropriate there in any case, since non-sleeping GPIOs should
be accessible from hardirq contexts even on RT kernels.

If anyone has any strong arguments against using a raw spinlock
to protect that table, it'd be nice to know them sooner rather
than later.

- Dave


SUMMARY:

Using the i2c-gpio driver on a preempt kernel with all the usual
kernel debug options enabled, the per-bit times (*) went up in a
bad way:  from about 6.4 usec/bit (original GPIO code on this board)
up to about 11.2 usec/bit (just switching to gpiolib), which is
well into "objectionable overhead" territory for bit access.

Just enabling preempt shot the time up to 7.4 usec/bit ... which is
also objectionable (it's all-the-time overhead that is clearly
needless), but much less so.

Converting the table lock to be a raw spinlock essentially removed
all non-debug overheads.  It took enabling all those debug options
plus internal gpiolib debugging overhead to get those times up to
the 7.4 usec/bit that previously applied even with just preempt.

(*) Those times being eyeballed medians; I didn't make time to find
    a way to export a few thousand measurements from the tool and
    do the math.  The typical range was +/- one usec.

    The numbers include udelay() calls, so the relevant point is
    the time *delta* attributable only to increased gpiolib costs,
    not the base time (with udelays).  The delta probably reflects
    on the order of four GPIO calls:  set two different bits, clear
    one of them, and read it to make sure it cleared.


> The upside is: 
> 
>   * Providing two features which were "want to have (but OK to defer)" when
>     GPIO interfaces were first discussed in November 2006:
> 
>     -   A "struct gpio_chip" to plug in GPIOs that aren't directly supported
>         by SOC platforms, but come from FPGAs or other multifunction devices
>         (like UCB-1x00 GPIOs).
> 
>     -   Full support for message-based GPIO expanders, needing a gpio_chip
>         hookup; previous support for this part of the programming interface
>         was just stubs.  (One example: the widely used pcf8574 I2C chips,
>         with 8 GPIOs each.)
> 
>   * Including a non-stub implementation of the gpio_{request,free}() calls,
>     which makes those calls much more useful.  The diagnostic labels are
>     also recorded given DEBUG_FS, so /sys/kernel/debug/gpio can show a
>     snapshot of all GPIOs known to this infrastructure.
> 
> The driver programming interfaces introduced in 2.6.21 do not change at all;
> this new infrastructure is entirely below the covers.