Re: [PATCH] gpio: remove gpio_descs global array

On Tue, Dec 2, 2014 at 7:32 PM, Geert Uytterhoeven [off-list ref] wrote:

Hi Linus, Alexandre,

On Fri, Nov 28, 2014 at 11:29 AM, Linus Walleij
[off-list ref] wrote:

quoted

On Wed, Nov 19, 2014 at 8:51 AM, Alexandre Courbot [off-list ref] wrote:

quoted

Replace the ARCH_NR_GPIOS-sized static array of GPIO descriptors by
dynamically-allocated arrays for each GPIO chip.

This change makes gpio_to_desc() perform in O(n) (where n is the number
of GPIO chips registered) instead of O(1), however since n is rarely
bigger than 1 or 2 no noticeable performance issue is expected.
Besides this provides more incentive for GPIO consumers to move to the
gpiod interface. One could use a O(log(n)) structure to link the GPIO
chips together, but considering the low limit of n the hypothetical
performance benefit is probably not worth the added complexity.

This patch uses kcalloc() in gpiochip_add(), which removes the ability
to add a chip before kcalloc() can operate. I am not aware of such
cases, but if someone bisects up to this patch then I will be proven
wrong...

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>

OK patch applied. Let's see if the world explodes.

This patch changes a return value from -EPROBE_DEFER to -EINVAL in
regulator-gpio when a GPIO cannot be found yet, causing probe failures
on r8a7791/koelsch:

Hi Geert,

Thanks for signaling this. I think I understand what is going wrong
and will send a fixup patch in a few minutes.

Cheers,
Alex.