On Sun, Feb 22, 2009 at 1:48 AM, Frederic Weisbecker <fweisbec@xxxxxxxxx> wrote:
...
>> +static ssize_t xpad_store_rumble_enable(struct device *dev,
>> + struct device_attribute *attr, const char *buf, size_t count)
>> +{
>> + struct usb_xpad *xpad = to_xpad(dev);
>> + int newrumble = xpad->rumble_enable;
>> + if (1 == sscanf(buf, "%d", &newrumble))
>
>
> Oh, that's not wrong but it looks weird, usually, a code reader would
> expect to see if (sscanf(...) == 1)
>
Oops... I changed some stuff around (deleted an unneeded variable) and
didn't change the test form back.
The "backwards" expression is a trick that some of us teach when
teaching C, for the specific case of comparing a variable to a
constant. It allows the compiler to check for an unintentional "="
where a "==" was desired. (foo = 4) is not an error or a warning, but
(4 = foo) is. But in this case, I now effectively have constants on
both sides of the expression, so it just looks funny.
...
>> +
>> +static void xpad_init_controller(struct usb_xpad *xpad)
>> +{
>> + set_stick_limit(XSTICK_LIMIT_DEFAULT, &xpad->left_stick_limit, xpad->left_dead_zone);
>> + set_stick_limit(XSTICK_LIMIT_DEFAULT, &xpad->right_stick_limit, xpad->right_dead_zone);
>> + set_dead_zone(XDEAD_ZONE_DEFAULT, &xpad->left_dead_zone, xpad->left_stick_limit);
>> + set_dead_zone(XDEAD_ZONE_DEFAULT, &xpad->right_dead_zone, xpad->right_stick_limit);
>> +
>> + if (xpad->controller_type == XCONTROLLER_TYPE_GUITAR) {
>> + xpad->rumble_enable = 0;
>> + }
>> + else if (xpad->controller_type == XCONTROLLER_TYPE_DANCE_PAD) {
>> + xpad->rumble_enable = 0;
>> + }
>> + else {
>> + xpad->rumble_enable = 1;
>> + }
>
>
> The brackets are not needed in these checks, may be you should run
> scripts/checkpatch.pl to check obvious coding styles issues.
Will do. This is not yet a "final" patch I'm ready to ask for
inclusion... I want to do some more testing first.
...
>> +#ifdef CONFIG_SYSFS
>> + xpad->sysfs_ok = 1;
>> + error = sysfs_create_group(&input_dev->dev.kobj, &xpad_default_attr_group);
>> + if (error) {
>> + /* Driver will work without the sysfs interface, but parameters
>> + * will not be adjustable, so this failure is a warning. */
>> + printk(KERN_WARNING "xpad: sysfs_create_group failed with error %d\n", error);
>> + xpad->sysfs_ok = 0;
>> + }
>> +#endif
>
>
> No need to check for CONFIG_SYSFS, sysfs_create_group will be a no-op which returns 0
> if not built.
>
Regarding this check and the others, if I remove them, the struct
usb_xpad instances will be slightly larger, and the code will be a bit
bigger than necessary (unless gcc optimizes away the un-called
functions), on systems where CONFIG_SYSFS is not set. In theory, I
don't see a need to have a system that uses this driver but does not
use the sysfs interface. In practice, however, I used to do all kinds
of crazy things when building systems for environmental sensing, so I
guess it could happen.
...
>
>
> Ditto. Actually it's useful for data which consume memory,
> or fields inside structures, but not for types themselves.
>
> But well, these parts are a removing, not new code.
>
> Anyway, looks like a nice work.
>
>
> Frederic.
>
Thanks for the feedback! I will clean up those issues and run the code
through the style checks. Tentatively, I'm hoping to have the final
revision of the patch by around mid-week for maybe one more round of
comments, then I'll post it "for real."
Thanks,
Mike
--
Mike Murphy
Ph.D. Candidate and NSF Graduate Research Fellow
Clemson University School of Computing
120 McAdams Hall
Clemson, SC 29634-0974 USA
Tel: +1 864.656.2838 Fax: +1 864.656.0145
http://cirg.cs.clemson.edu/~mamurph
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
