Recent probe-input endianess patch seems to be flawed

[Martin Pitt]

Hi Johannes, hi hal masters,

Johannes, you recently posted a patch to fix endianess issues in
probe-input [1] which was subsequently applied upstream [2].

The patch worked fine on my amd64 desktop and indeed fixed some
properties on my powerpc laptop (iBook G4), but it caused crashes for
a lot of users (see [3] and the list of duplicates), so I backed it
out again in Ubuntu for now. One duplicate [4] has a hald debug
output, in case it is useful.

I quickly googled for some other usages of the EVIOCGBIT ioctl (I
found some in X.org [5] and the kernel [6]), and they use char arrays,
not longs. Unfortunately I did not find a documentation of the ioctl.

So it seems that there is a problem in the patch. Do you have any idea
about that?

Thanks,

Martin


[1] http://lists.freedesktop.org/archives/hal/2006-April/005080.html
[2] http://webcvs.freedesktop.org/hal/hal/hald/linux2/probing/probe-input.c?r1=1.14&r2=1.15
[3] https://launchpad.net/bugs/43962
[4] https://launchpad.net/bugs/44091
[5] http://lists.freedesktop.org/archives/xorg/2005-September/010144.html
[6] http://svn.icculus.org/*checkout*/manymouse/trunk/linux_evdev.c?rev=31

-- 
Martin Pitt        http://www.piware.de
Ubuntu Developer   http://www.ubuntu.com
Debian Developer   http://www.debian.org

In a world without walls and fences, who needs Windows and Gates?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 191 bytes
Desc: Digital signature
Url : http://lists.freedesktop.org/archives/hal/attachments/20060512/09cd8232/attachment.pgp