| Age | Commit message (Collapse) | Author |
|---|
|
Bugzilla #6734 <https://bugs.freedesktop.org/show_bug.cgi=6734> Apply patch
from Philip Langdale which causes us to make sure that O_NONBLOCK is
set on the inotify device.
|
|
|
|
evdevAxesRec: Make btnMap an int array instead of a Card8 array.
Make abs support and non-core stuff actually work.
Relative emulation for abs mode is still a bit broken, but that's far less
critical.
How many buttons we've registered is configuration information, not an
error message.
|
|
Actually, well, _work_. Note, abs support still segfaults, I'll debug it
after I've slept. But I should get it tomorrow.
|
|
mostly work for older servers again.)
Close the device properly on read errer.
Minimal inotify support, not ideal yet, but...
Minimal inotify support, not ideal yet, but...
Minimal inotify support, not ideal yet, but...
|
|
Change the calls to refer to EvdevAxes instead of Abs and Rel. Include
mipointer.h. (Fixes a compile error.) Remove an unused variable from
EvdevReadInput.
Removed evdev_abs.c and evdev_rel.c, added evdev_axes.c. Added count to rel
and abs. Moved btnMap from rel to the new axes. Shuffle some includes,
and include stdarg.h.
Removed.
New, a mix of rel and abs that should actually work in the rel case.
Rearrange include order. btnMap was moved from being part of rel to being
part of axes.
Removed.
Rearrange include order.
Rearrange include order.
NOTE: This commit has been compile tested, not run tested.
I have some other changes to make before I can do more heavy testing.
|
|
|
|
Disable the undocumented xkb guts options in the key driver. (Requested by
Daniel Stone.)
|
|
Detect keys numbered higher then buttons. Remove the (depreciated since
long before xkb support was added to evdev) XkbKeymap option.
|
|
|
|
getting unplugged, that sort of stuff.)
|
|
Work properly if we're defining entirely by capability maps.
|
|
gets it.
Bugzilla #5943 <https://bugs.freedesktop.org/show_bug.cgi=5943> Make sure
we include errno.h.
Reduce EVDEV_MAXBUTTONS to 96.
Split up evdevStateRec into a struct with pointers to new structs for btn,
abs, rel, and key.
New structure type for handling the device capability bitmaps.
Add device bits and struct input_id to evdevDeviceRec.
Add matching device bits, struct input_id, and pass number to
evdevDriverRec.
Prototype for evdevGetBits from evdev_brain.c.
Conversion for the evdevStateRec split.
Remove the errno.h include, it's in evdev.h for now.
Move the bit getting from the drivers to here, into evdevDeviceRec.
Fix a rare case of fd leakage.
Add several new (and somewhat ugly) device matching options: <map>Bits:
Where map is one of ev, key, rel, abs, msc, led, snd, or ff. In the
format of '+0 +3 -1-2 ~5-10', requires bits 0 and 3 be set, bits 1 and
2 to not be set, and at least one bit in the range of 5 to
10 be set. bustype, vendor, product, and version: Simple integer options
for matching the struct device_id fields, must be 0 (the default) or
the exact value you wish to match against. pass: Bounded to 0-3,
devices are matched to the first matching entry found, order for
multiple matching entries in the same pass is undefined, but it starts
with pass 0 and goes to pass 3.
Adaptation for the evdevStateRec split and the change in capability bitmap
handling.
Add evdevGetBits to fill the new evdevBitsRec struct type.
Lots of somewhat ugly code for matching by capability bits.
Split out of evdevRescanDevices to smaller handling functions. The new
design should be better if I decide to handle arbitrary Device fields
again.
Adaptation for the evdevStateRec split and the change in capability bitmap
handling.
Handle all buttons, no button compression at this time, however we reorder
things so that BTN_MOUSE comes before BTN_MISC, somewhat evily.
Support for the new btn->state[] array of int pointers.
Adaptation for the evdevStateRec split and the change in capability bitmap
handling.
Adaptation for the evdevStateRec split and the change in capability bitmap
handling.
I really hope I didn't miss any changes.
|
|
|
|
(maybe) fix for bug #5950. Though, I'm not convinced and still don't
see how this bug could be happening, especially if this doesn't do it.
|
|
|
|
little bug, properly handle wheels on 3 button mice.
|
|
updated version of the patch listed. Basicly a rewrite of the driver,
with a few pieces of the old. XKB support, proper device matching,
basic absolute pointer support. Lots more, will require some user
config changes.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
update to fix bug #5167 (Linux prefers *.1x man pages in man1 subdir)
|
|
|
|
|
|
|
|
|
|
- Only define EXTRA_DIST when there are actually man pages to be processed
- Add missing man pages to keyboard and ati drivers EXTRA_DIST
|
|
substitutions to work better with BSD make
|
|
of indirectly
|
|
|
|
|
|
|
|
drivers
|
|
- Verify and update package version numbers as needed
- Implement versioning scheme
- Change bug address to point to bugzilla bug entry form
- Disable loadable i18n in libX11 by default (use --enable-loadable-i18n to
reenable it)
- Fix makedepend to use pkgconfig and pass distcheck
- Update build script to build macros first
- Update modular Xorg version
|
|
|
|
|
|
|
|
|
|
|
|
|
|
installed. Allow the use of evdev for mice on 2.4 kernels.
|
|
|
|
issues when input goes to multiple places. Restrict building the evdev
driver to 2.6 kernels since EVIOCGRAB didn't exist in 2.4.
|
|
don't overlap the wheel buttons. Patch from Alan Swanson (#968).
|
|
|
