| Age | Commit message (Collapse) | Author |
|---|
|
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Part-of: <https://gitlab.freedesktop.org/xorg/test/x11perf/-/merge_requests/10>
|
|
Also bumps package version to 1.6.99.0 to prepare for 1.7 release.
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Part-of: <https://gitlab.freedesktop.org/xorg/test/x11perf/-/merge_requests/10>
|
|
Xft can no longer support 'render' in font names due to a
change in fontconfig. Work around this by removing that from
the font names and apply it to the pattern manually whenever
the 'bfont' name is not NULL.
Signed-off-by: Keith Packard <keithp@keithp.com>
Part-of: <https://gitlab.freedesktop.org/xorg/test/x11perf/-/merge_requests/1>
|
|
This compares the font family from the loaded font with the requested
family and skips tests where the requested family wasn't available.
It's not perfect, as someone could have a mis-configured font family,
but should catch most incorrect font usages.
I've also switched from Charter to DejaVu Sans as that is more widely
installed.
Signed-off-by: Keith Packard <keithp@keithp.com>
Part-of: <https://gitlab.freedesktop.org/xorg/test/x11perf/-/merge_requests/1>
|
|
Signed-off-by: Keith Packard <keithp@keithp.com>
Part-of: <https://gitlab.freedesktop.org/xorg/test/x11perf/-/merge_requests/1>
|
|
Part-of: <https://gitlab.freedesktop.org/xorg/test/x11perf/-/merge_requests/1>
|
|
Part-of: <https://gitlab.freedesktop.org/xorg/test/x11perf/-/merge_requests/1>
|
|
Signed-off-by: Keith Packard <keithp@keithp.com>
Part-of: <https://gitlab.freedesktop.org/xorg/test/x11perf/-/merge_requests/1>
|
|
This avoids the obvious optimization which catch overlapping
rectangles to avoid re-rendering the same pixels.
Part-of: <https://gitlab.freedesktop.org/xorg/test/x11perf/-/merge_requests/1>
|
|
Some drivers manage to optimize the solid block of segments into a
single square fill operation, which is a nifty optimization, but
doesn't really measure what we're interested here.
Signed-off-by: Keith Packard <keithp@keithp.com>
Part-of: <https://gitlab.freedesktop.org/xorg/test/x11perf/-/merge_requests/1>
|
|
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
|
|
The other minpixwin tests draw to a window 50% the size of the pixmap,
it's clear this one was meant to too.
Signed-off-by: Adam Jackson <ajax@redhat.com>
|
|
Prepend a c to the test names of the "core" versions, to match the
-aa*text and -caa*text tests.
Signed-off-by: Peter Harris <pharris@opentext.com>
|
|
|
|
|
|
Similar to GetImage and ShmPutImage, test the performance of retrieving
pixel data from the X server without the overhead of copying the pixels.
In the upper bound, using XShmGetImage, the performance will be limited
by the RTT latency (x11perf -prop). However, comparing the scaling factors
of XShmGetImage and XGetImage will give some insight into the driver
efficiency.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Adam Jackson <ajax@redhat.com>
|
|
https://bugs.freedesktop.org/show_bug.cgi?id=31640
Fix trapezoid checks to look for both render & xft
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
|
|
Signed-off-by: Adam Jackson <ajax@redhat.com>
|
|
|
|
-compwinwinXX, -comppixwinXX where XX is 10, 100 or 500.
-pop <POP> <POP> ... where <POP> is one of Clear, Src, Dst, Over,
OverReverse, In, InReverse, Out, OutReverse, Atop, AtopReverse, Xor, Add or
Saturate.
-format <FORMAT> <FORMAT> ... where <FORMAT> is one of RGB24, ARGB32, A8,
A4, A1 or NATIVE
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
