On 03/30/2018 07:33 AM, Andy Shevchenko wrote:
On Wed, Mar 28, 2018 at 9:18 PM, Laura Abbott <labb...@redhat.com> wrote:
The new challenge is to remove VLAs from the kernel
(see https://lkml.org/lkml/2018/3/7/621) to eventually
turn on -Wvla.
Using a kmalloc array is the easy way to fix this but kmalloc is still
more expensive than stack allocation. Introduce a fast path with a
fixed size stack array to cover most chip with gpios below some fixed
amount. The slow path dynamically allocates an array to cover those
chips with a large number of gpios.
+ ret = gpiod_set_array_value_complex(false,
true,
lh->numdescs,
lh->descs,
vals);
+ if (ret)
+ return ret;
+
return 0;
Can't we
return gpiod_set_array_value_complex(); ?
Yeah I'll clean that up for v4.
+ slowpath = kcalloc(2 * BITS_TO_LONGS(chip->ngpio),
+ sizeof(*slowpath),
+ can_sleep ? GFP_KERNEL : GFP_ATOMIC);
+ if (slowpath)
+ kfree(slowpath);
+ if (slowpath)
+ kfree(slowpath);
Since slowpath is a pointer, conditionals above are redundant.
+ slowpath = kcalloc(2 * BITS_TO_LONGS(chip->ngpio),
+ sizeof(*slowpath),
+ can_sleep ? GFP_KERNEL : GFP_ATOMIC);
+ if (slowpath)
+ kfree(slowpath);
Ditto.
This was caught by a coccinelle script via 0-day but I think the request
was to not do it. I'll add a comment explaining why we are going against
style.
Thanks,
Laura
