xorg/driver/xf86-input-fpit - [PATCH] Don't call xf86IsCorePointer.

Peter Hutterer mailinglists at who-t.net
Wed Feb 6 20:10:22 PST 2008


pcpa at mandriva.com.br wrote:
> Quoting Peter Hutterer <mailinglists at who-t.net>:
> 
>> pcpa at mandriva.com.br wrote:
>>> Quoting Peter Hutterer <mailinglists at who-t.net>:>   Thanks for the 
>>> review. When I made the patch I searched for other
>>> modules that had a similar problem, and used the same logic as in
>>> the acecad driver, that has an almost identical code, but that patch
>>> initializes to 1, and then checks the abi version before calling
>>> xf86IsCorePointer().
>> Yes, I know. I fixed this up a few days ago.
>>
>>>   In the case of the acecad driver, I also believe it could be
>>> changed to not call dlopen, and instead be linked agains't libsysfs.
>>>
>>>   I don't know exactly the way the new input code works, so I
>>> don't think I deserve to get credit for this patch, but if you
>>> prefer I can write a new one initializing is_core_pointer to 0,
>>> before the #if :-)
>> yes please. I'm busy enough as it is.
> 
>   Ok, new patch is attached.

pushed, thanks.


>> On another note, do you actually have users that require these drivers?
> 
>   Unfortunately not, at least not a single opened bug report. After
> some googling I found a bug report that may be useful:
> http://www.nabble.com/Bug-457722%3A-xserver-xorg-input-fpit%3A-using-touchscreen-kills-x-td14492063.html

yeah, this is bug #14057 on bugs.fdo.org.
i just noticed that I did attach some patches to this bug that may fix 
things up.

> Since it was calling xf86IsCorePointer in a loop, I did a different
> patch (maybe I just did the compiler optimization by removing the set
> the variable to 0 every iteration...). The loop is a bit weird also,
> as it is basically:
> for (loop = 0; loop < end; loop++) {
>    ...
>    loop += chunk_size - 1;
> }
> but I did not modify it to avoid the need of a bigger patch.

If you have the desire to fix it, feel free to do so. I'll do my best 
reviewing but unless we hear back from an actual user of such a device, 
there isn't all that much we can do.

Cheers,
   Peter



More information about the xorg mailing list