xf86-input-libinput: Changes to 'master'

[Peter Hutterer]

 src/xf86libinput.c |  103 +++++++++++++++++++++++++++++++++++++++++------------
 test/test-bezier.c |    7 +++
 2 files changed, 87 insertions(+), 23 deletions(-)

New commits:
commit 72fb6d304eec6eeeac6b42963c2729134d56de57
Author: Peter Hutterer <peter.hutterer at who-t.net>
Date:   Tue Feb 28 14:45:29 2017 +1000

    test: fix a test failure on ppc64(le) and aarch64
    
    Caused by different results in -O0 vs -O2. The resulting array differs only
    slightly but the initial sequence has one extra zero. That triggers our
    assert, no other compiler flag seem to be affecting this.
    
    Compiled with -O0:
    Breakpoint 1, test_nonzero_x_linear () at test-bezier.c:157
    157			assert(bezier[x] > bezier[x-1]);
    (gdb) p bezier
    $6 = {0 <repeats 409 times>, 1, 2, 4, 5, 7, 9, 10, 12, 14, 15, 17, 19, 21, 22,
    
    Compiled with -O2:
    (gdb) p bezier
    $1 = {0 <repeats 410 times>, 1, 3, 5, 7, 9, 10, 12, 14, 15, 17, 19, 20, 22,
    
    Printing of the temporary numbers in the decasteljau function shows that a few
    of them are off by one, e.g.
        408.530612/0.836735 with O0, but
        409.510204/0.836735 with O2
    Note: these are not rounding errors caused by the code, the cast to int
    happens afterwards.
    
    Hack around this by allowing for one extra zero before we check that the rest
    of the curve is ascending again.
    
    https://bugs.freedesktop.org/show_bug.cgi?id=99992
    
    Signed-off-by: Peter Hutterer <peter.hutterer at who-t.net>

commit aae2c8ad9a9f1712149c93d50284ddb5f37e4cbd
Author: Peter Hutterer <peter.hutterer at who-t.net>
Date:   Fri Feb 24 12:56:41 2017 +1000

    Open sysfs files directly instead of going through the server
    
    Only use-case here are pad mode LEDs that now live in /sys/class/leds. Asking
    the server to open them is pointless, the server only knows how to open Option
    "Device". And since the LEDs are in sysfs we should have access to them
    anyway, so no need for jumping through or hula-ing hoops.
    
    xf86CloseSerial() works as intended as it's a slim wrapper around close(), so
    we only have to worry about the open() path here.
    
    Signed-off-by: Peter Hutterer <peter.hutterer at who-t.net>

commit dafc296f2df587a1bb5feb37697c50608db4f246
Author: Peter Hutterer <peter.hutterer at who-t.net>
Date:   Fri Feb 24 12:34:10 2017 +1000

    Add streq() macro, replace strcmp instances with it
    
    And why isn't this a thing in glibc yet
    
    Signed-off-by: Peter Hutterer <peter.hutterer at who-t.net>
    Reviewed-by: Eric Engestrom <eric.engestrom at imgtec.com>

commit 7c90f06d569b1b14d84075e7cea22bce06b925e6
Author: Peter Hutterer <peter.hutterer at who-t.net>
Date:   Fri Feb 24 12:27:37 2017 +1000

    Update pad modes in a workproc, not during the input thread
    
    Updating the property directly causes us to send events from the input thread
    which has some "interesting" side effects like messing up the reply order or
    just crashing the server.
    
    Schedule a work proc instead and update it whenever the server is back in the
    main thread.
    
    Signed-off-by: Peter Hutterer <peter.hutterer at who-t.net>