[PULL] unreviewed patches Jan 2013

[Peter Hutterer]

On Thu, Feb 07, 2013 at 04:53:55PM -0800, Keith Packard wrote:
> Peter Hutterer <peter.hutterer at who-t.net> writes:
> 
> > Peter Hutterer (14):
> >       test/signal-logging: simplify tests using sprintf
> 
> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       os: silently ignore length modifiers in pnprintf
> 
> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       os: add support for %f to pnprintf
> 
> (this could use some improvements to clean up rounding, but meh)
> 
> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       dix: fix ptraccel debugging printfs
> 
> There are some gratuitous format width changes in this patch which
> may not give the same result with a 'real' printf. I'd revert those
> given that you ignore length modifiers entirely at this point.

done, see patch.

> 
> >       dix: add some more info to a ptraccel debug msg
> 
> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       dix: use BUG_RETURN_VAL for an error message
> 
> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       dix: unify prefix for ptraccel debugging in DebugAccelF macro
> 
> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       dix: pre-scale relative events from abs devices to desktop ratio (#31636)
> 
> So many coordinate systems, so little time.
> 
> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       Xext: pass the current screen to miProcessDeviceEvent() from xtest calls
> 
> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       Xext: if a root window is given in XTestFakeInput, move to that
> 
> The comment before the call to scale_from_screen needs updating to say
> /* valuators are in screen or desktop coords */

done.

> 
> otherwise,
> 
> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       dmx: don't include dmx-config.h from xdmxconfig (#37502)
> 
> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       dix: when shutting down slave devices, shut down xtest devices last
> 
> Would be lovely to have comments before each loop labeling what devices
> they expect to shut down. Otherwise:

done

> Reviewed-by: Keith Packard <keithp at keithp.com>
> 
> >       dix: support the transformation matrix for relative devices.
> 
> I suggest leaving the current transform called 'transform' and creating
> a new transform called 'relative_transform' or some such that has the
> correct values in it (i.e., with the translation components set to zero
> so you don't have to copy each time you use it). That will preserve API
> at least.

fwiw, this isn't API, it's only used internally. did half of what you
suggested, rel matrix is stored without zero mangling now, but abs axis is
renamed, both for obviousness and the compiler telling me where I forgot the
update.

> Also, this needs to bump the ABI version number.

While we're there, how about:
http://patchwork.freedesktop.org/patch/12960/

because if that goes in too, I can skip the minor bump.

Cheers,
   Peter

> 
> >       Merge branch 'ptraccel-fixes' into unreviewed
> >
> >  Xext/xtest.c               |  12 ++-
> >  dix/devices.c              |   8 +-
> >  dix/getevents.c            |  72 ++++++++++++++--
> >  dix/ptrveloc.c             |  40 ++++-----
> >  hw/dmx/config/xdmxconfig.c |   3 -
> >  include/input.h            |   1 +
> >  include/inputstr.h         |   5 +-
> >  include/misc.h             |   1 +
> >  os/log.c                   |  19 ++++-
> >  os/utils.c                 |  32 ++++++++
> >  test/signal-logging.c      | 200 ++++++++++++++++++++++-----------------------
> >  11 files changed, 253 insertions(+), 140 deletions(-)
> 
> -- 
> keith.packard at intel.com