Chris Wilson
bfc027ac6d
test/clip-image: Free image after use.
2010-03-23 08:16:57 +00:00
Chris Wilson
e214f09d63
xcb: Consume the implicit reference for the cached connection on finish
2010-03-22 22:20:58 +00:00
Chris Wilson
c235543bdf
xcb: Destroy reference to local source picture in show-glyphs.
2010-03-22 22:15:44 +00:00
Chris Wilson
c42cdd2c9d
xcb: Gracefully destroy any lingering fallback during finish.
...
A fallback should never persist beyond the flush into the finish, but
yet one remains in test/clip-shapes-unaligned-rectangles. For the time
been, simply clean up the rogue surface.
2010-03-22 20:14:09 +00:00
Chris Wilson
fd96aa3de2
gl: Hook in glyph cache finalisation.
2010-03-22 19:07:31 +00:00
Chris Wilson
c7511fa7ad
drm/intel: Revoke glyph caches on device finish
2010-03-22 17:21:38 +00:00
Chris Wilson
9ba73bdfc4
gl: Decouple cache from scaled-font on context destruction
2010-03-22 17:17:08 +00:00
Chris Wilson
7812d095ab
drm/intel: Replace open-coding of _cairo_rtree_node_remove()
2010-03-22 16:46:58 +00:00
Chris Wilson
48d8471625
recording: Cleanup the clip on error.
...
Whilst recording the commands, remember to cleanup the local clip after
detecting an error.
2010-03-22 13:05:23 +00:00
Chris Wilson
c87737c2bf
ps: Skip surface emission if in error state.
...
During finish, if we encountered an error during the construction of the
surface, simply proceed to cleanup rather than emit the incomplete and
potentially subtly broken output.
2010-03-22 13:04:19 +00:00
Chris Wilson
e961cdf568
image: Special case wholly unbounded fixups.
...
In the event of an empty bounded rectangle, the computation of the
unbounded - bounded rectangles leads to negative areas, integer overflow
and death.
[And similarly for the derived surfaces.]
2010-03-22 10:54:08 +00:00
Chris Wilson
844d8ea57d
surface: Check pattern for error status.
...
Ensure that the error status from patterns is propagated when used by
doing so in the surface layer. Similarly check that a surface pattern
has not been finished.
2010-03-22 10:37:18 +00:00
Chris Wilson
213093f437
ps: Free page on error path.
2010-03-22 10:36:47 +00:00
Chris Wilson
c8a8e57d6a
script: Use a stack to push/pop recursed line numbers.
...
Still not entirely helpful in the event of recursive parsing without a
reference to the file as well as the line number in the event of an
exception.
2010-03-21 20:42:34 +00:00
Chris Wilson
4b4de940ee
script: Remove the version check.
...
When compiling we can depend on whatever version of cairo we need, but
we should be wary of checking for runtime compatibility when building
standalone.
2010-03-21 20:41:18 +00:00
Chris Wilson
a7cb314b06
test/any2ppm: Check for errors after executing script.
2010-03-21 20:40:54 +00:00
Chris Wilson
b73a33b89e
script: Make the test suite run again!
2010-03-21 20:40:19 +00:00
Chris Wilson
2a59f0af6a
wrapper: Apply device transform inverse as appropriate.
...
Fixes many failures of the paginated surfaces as they replayed through
the recording surfaces.
2010-03-21 20:38:27 +00:00
Chris Wilson
5720fff827
ps: Destroy the local surface on error.
...
Minor leak of the intermediate surface when converting to an opaque
source following an error.
2010-03-21 20:37:38 +00:00
Chris Wilson
35f318a59c
paginated: Use common is_clear.
...
As equivalent processing to the page_is_blank optimisation done inside
the paginated surface is applied by the gstate, remove the redundant
code.
2010-03-21 20:36:26 +00:00
Chris Wilson
b101c7dab8
gl: Add EGL interface
...
Enable the EGL backend for GL.
2010-03-11 01:49:18 +00:00
Chris Wilson
ca6e4e1f91
spans: Initialise rects->is_bounded for polygon compat.
2010-03-11 01:48:08 +00:00
Chris Wilson
6754f1db17
gstate: Account for undropped non-transformed glyphs.
...
In the simplest case of pass-through glyphs we made a mistake in
propagating the number of glyphs.
Reported-by: Christophe de Dinechin <christophe@taodyne.com>
2010-03-10 15:13:00 +00:00
Chris Wilson
5fb36fe5ab
drm/i965: Acquire device around commit.
2010-03-08 15:37:17 +00:00
Chris Wilson
4083f40fbd
glew: don't pull in GL/glu.h
2010-03-07 16:59:57 +00:00
Chris Wilson
473e006ad6
test: Add a1-mask-sample
...
Test sampling positions with a NEAREST mask.
2010-03-06 18:12:36 +00:00
Chris Wilson
f79b2ceea4
drm/i915: Discard redundant solitary clips during fill
2010-03-04 19:52:58 +00:00
Chris Wilson
7cea9ae290
drm/i915: Apply more micro-optimisations when targetting CONTENT_ALPHA
2010-03-04 17:23:55 +00:00
Chris Wilson
dce8b028cd
drm/i915: Adjust clip matrix for clip origin.
2010-03-04 17:13:53 +00:00
Chris Wilson
8e702044b3
drm/i915: Handle clip/combine coords for spans.
2010-03-04 17:13:02 +00:00
Chris Wilson
e1f0c2e73f
perf: Check for and include unistd.h
...
isatty() and access() require unistd.h, so include it!
2010-03-04 17:11:58 +00:00
Chris Wilson
271752b944
gl: Pad the gradient texture.
...
Fixes test/linear-gradient-subset
2010-03-04 08:24:08 +00:00
Chris Wilson
afe6f4f051
drm/i915: Use a preserved temporary to store the radial texcoord.
2010-03-04 00:04:30 +00:00
Chris Wilson
4b2e047882
drm/i915: Eliminate redundant radial vertex shader.
...
It is always equivalent to a simple texture transformation, so reduce it
to the texture shader.
2010-03-04 00:04:30 +00:00
Chris Wilson
5161051798
drm/i915: Fix encoding of negated operand channels
...
Long ago when converting the pixel shader structs into macros and
reducing the code size by ~100k (the inlines were too depth for constant
propagation and CSE), I broke the encoding of negated channels. So
instead use a single bit to indicate a negation rather than 2s
complement (with sign extension into neighbouring channels, oops). The
disadvantage is that expressing the negated channel is a little more
ugly.
2010-03-04 00:04:30 +00:00
Chris Wilson
5b973c6708
drm/i915: Fix crash with linear-step-function
...
Avoid the potential divide by zero by falling back to pixman rendered
gradient textures for such troublesome input.
2010-03-04 00:04:30 +00:00
Chris Wilson
404ead7eeb
drm/i915: Ensure simple alpha is written to green on alpha-only surfaces.
...
The intricacies of h/w make convoluted s/w.
2010-03-04 00:04:30 +00:00
Chris Wilson
1f2daf6524
drm/i915: Avoid double multiplication of linear alpha.
2010-03-04 00:04:21 +00:00
Chris Wilson
cdaef58c19
drm/i915: Acquire device before shader commit.
2010-03-03 23:57:36 +00:00
Chris Wilson
dd6f96dd4a
drm/i965: Ignore flushing on device finish.
2010-03-03 23:57:36 +00:00
Chris Wilson
3692597567
drm/i965: Merge the constants into the surface stream.
...
As we can use an offset from the surface base for the constant buffers,
we need to fold the constants into the surface stream.
2010-03-03 23:57:35 +00:00
Chris Wilson
cc58594757
drm/intel: Set the 1D gradient texture to be untiled.
2010-03-03 23:57:35 +00:00
Chris Wilson
543cf07364
drm/intel: Set default extend mode for rendering gradient subsets.
2010-03-03 23:57:35 +00:00
Chris Wilson
521d1c1d2c
surface: Simplify is_clear logic.
...
As we pre-check whether the operation is a no-op on an already cleared
surface, then for all but a paint with CLEAR and no clip, the result is
a non-clear surface.
2010-03-03 23:57:35 +00:00
Chris Wilson
017db95664
gstate: Use _cairo_pattern_clear for CLEAR.
2010-03-03 23:57:35 +00:00
Chris Wilson
70412bba0e
test: Avoid derefencing a NULL xcb connection
2010-03-03 23:57:35 +00:00
Chris Wilson
5bfea8447b
test: Add references for rounded-rectangle-{fill,stroke}
2010-03-03 23:57:35 +00:00
Chris Wilson
4126d580d8
surface-fallback: Free traps on composite_trapezoids() error
2010-03-02 14:38:36 +00:00
Chris Wilson
8a59522bbd
surface-fallback: Propagate NOTHING_TO_DO
...
NOTHING_TO_DO is converted to SUCCESS by the surface layer, so clean up
the code slightly by reducing the number of checks and conversions.
2010-03-02 14:36:16 +00:00
Chris Wilson
f071958606
xlib: Handle a1 image uploads through converter
...
Fixes test/large-source [xlib]
2010-03-02 13:58:18 +00:00
