caskbench - no desc

	Commit message (Collapse)	Author	Age
...
*	Merge pull request #42 from SRA-SiliconValley/Issue_30	bryceharrington	2014-05-22
\|\ \| \| \| \|	Issue #30. Initialize device config structure using memset.
\| *	Issue #30. Initialize device config structure using memset.	suyambulingamrm	2014-05-19
\| \|
* \|	Initialization fixes to avert crashes	chevalam	2014-05-21
\|/ \| \| \|	ref: https://github.com/SRA-SiliconValley/caskbench/issues/29
*	Drop legacy skia support. Remove ENABLE_LEGACY_SKIA_SRA	Bryce Harrington	2014-05-08
\|
*	Track the image path as part of the cairo context, not a shape default	Bryce Harrington	2014-05-08
\| \| \| \| \|	We aren't randomizing the image or varying it from shape to shape, so it doesn't really belong in the shapes structure.
*	Use shape_type_t in shapes structure	Bryce Harrington	2014-05-08
\|
*	Break out a shape_defaults_init() routine	Bryce Harrington	2014-05-08
\|
*	Switch shapes' fill_type from a char* to a fill_type_t	Bryce Harrington	2014-05-08
\| \| \| \|	This eliminates strcmp in active testing, which should boost performance
*	Rename cairo_context structure member shape_args to shape_defaults	Bryce Harrington	2014-05-08
\|
*	Draw circles offset by radius	Bryce Harrington	2014-05-08
\| \| \| \| \| \|	This makes the circle center_x, center_y positioning more consistent with the other shapes, and indeed we're not tracking the center point any more, so rename the variable to be less misleading.
*	tests: Fix multishape by working around triangle anti-aliasing bug	Bryce Harrington	2014-05-08
\|
*	Shorten existing enum names	Bryce Harrington	2014-05-06
\|
*	Rename Shapes enum to shape_type_t and fix enum names	Bryce Harrington	2014-05-06
\| \| \| \| \| \| \| \| \| \| \| \|	We already have a 'shapes' data structure, so to avoid confusion don't name our shape type enums 'Shapes'. Instead, name it shape_type_t, which is more consistent with the code style of other enums in the codebase. Also, rename Circle / Triangle / etc. to CB_SHAPE_CIRCLE / CB_SHAPE_TRIANGLE / etc. The former could be confused as class names; the latter are less likely to get confused this way, and follow the style of other enums in the codebase.
*	boilerplate: Add copyright and license statement on all code files	Bryce Harrington	2014-05-05
\|
*	boilerplate: Add emacs/vi local variables	Bryce Harrington	2014-05-05
\|
*	Use string names for specifying shapes externally instead of ID numbers	Bryce Harrington	2014-05-05
\|
*	Rename centre to center	Bryce Harrington	2014-05-05
\|
*	util: Move skia and cairo image export functionality to utils modules	Bryce Harrington	2014-05-05
\| \| \| \| \|	This cleans up some of the cairo/skia-specific code in caskbench.cpp, moving it out to *-util.cpp files that can be conditionally compiled.
*	cmake: Make skia optional (run only cairo tests)	Bryce Harrington	2014-05-01
\|
*	caskbench: Reword -x and -y option descriptions	Bryce Harrington	2014-04-25
\|
*	caskbench: Disable line style options for the release	Bryce Harrington	2014-04-24
\| \| \| \| \|	None of these appear to have an effect on tests other than stroke, and stroke is disabled.
*	caskbench: Disable multi-shapes option for the release	Bryce Harrington	2014-04-24
\| \| \| \|	It just leads to crashing behavior in one of the tests
*	caskbench: Use - rather than _ in option names	Bryce Harrington	2014-04-24
\|
*	caskbench: Disable the clip test, which simply fails	Bryce Harrington	2014-04-24
\|
*	caskbench: Disable RGBA options for the release	Bryce Harrington	2014-04-24
\| \| \| \| \|	These only appear to affect the fill test, so are of limited utility, and is not the interface we want to be supporting.
*	caskbench: Disable the multishape test for release	Bryce Harrington	2014-04-24
\| \| \| \|	This now causes a crash when used with the EGL backend on PC
*	caskbench: Disable stroke test for the release	Bryce Harrington	2014-04-24
\| \| \| \| \|	This no longer displays output to the screen when using the image backend.
*	Add cairoRandomizeColor and skiaRandomizeColor	Bryce Harrington	2014-04-24
\|
*	Centralize the skia atomics path workaround to config.h	Bryce Harrington	2014-04-23
\|
*	skia: Define location for SK_ATOMICS_PLATFORM_H and SK_MUTEX_PLATFORM_H	Bryce Harrington	2014-04-19
\| \| \| \| \|	The generic location for these is set to "../../src/ports", but src is an inappropriate place for headers to link from.
*	Whitespace cleanup	Bryce Harrington	2014-04-16
\|
*	Addition of user options	chevalam	2014-04-16
\| \| \| \|	Signed-off-by: Bryce Harrington <b.harrington@samsung.com>
*	Set up bitmaps for skia-samsung using the info structure	Bryce Harrington	2014-04-16
\|
*	caskbench: Current Skia needs app to define gPrintInstCount for ref counting	Bryce Harrington	2014-04-15
\|
*	caskbench: Only define SkToS32 if it's undefined	Bryce Harrington	2014-04-14
\|
*	cmake: Make GLX/EGL compilation optional (with EGL as default)	Bryce Harrington	2014-04-14
\| \| \| \| \| \| \| \| \| \| \| \| \|	GLX and GLES2/3 are incompatible as apparently the have overlapping API's so when both are linked in an executable one lib's functions are going to subvert the others. This patch provides options to configure GLX and EGL on or off. Longer term, this will probably need to be revisited. Perhaps there should be a GL_FLAVOR, that can be set to "glx", "egl", etc. so only one is defined at a time. Or, perhaps we generate multiple binaries, one for each, ala caskbench_egl, caskbench_glx, etc.
*	Don't require GL if we don't absolutely require it	Bryce Harrington	2014-03-29
\|
*	json: Fix invalid json syntax (damn commas!)	Bryce Harrington	2014-03-27
\|
*	Reduce default number of iterations.	Bryce Harrington	2014-03-27
\| \| \| \| \|	During development or when kicking the tires, it's nice if it runs swiftly. Let's allow the -i option to exist for a reason. :-)
*	egl: Make EGL sample buffers a runtime option	Bryce Harrington	2014-03-27
\| \| \| \| \| \| \|	Run caskbench with --enable-egl-sample-buffers to turn this feature on. This implementation is sort of hacky but gives us a starting point. I'll generalize better once there are additional device options.
*	Add device_config_t	Bryce Harrington	2014-03-27
\| \| \| \| \| \|	This structure will be used for communicating various configuration settings to the image, egl, and glx setup routines. For now just use it for passing width/height and the surface type.
*	Fix skia image writing so it'll work with EGL.	Bryce Harrington	2014-03-27
\| \| \| \|	Fixes warnings printed about incorrect stride.
*	Cleanup debug printfs	Bryce Harrington	2014-03-27
\|
*	Add a context_update() virtual function	Bryce Harrington	2014-03-20
\| \| \| \|	This permits us to flush and swap buffers when needed.
*	Virtualize the context destructor as well	Bryce Harrington	2014-03-20
\|
*	Split context setup for cairo and skia, and only set up the one actually needed	Bryce Harrington	2014-03-20
\|
*	Split cairo and skia setup into separate routines	Bryce Harrington	2014-03-20
\|
*	Check for surface error status	Bryce Harrington	2014-03-14
\|
*	Improve error handling for GLX.	Bryce Harrington	2014-03-12
\|
*	Increase Cairo's tessellation tolerance	Bryce Harrington	2014-03-10
\| \| \| \| \| \| \| \| \| \| \| \| \|	Suggestion from Henry Song: """ Hi, Bryce Looks like cairo bentley-ottmann tessellation is quite slow. In fact, the default tolerance of 0.001 for pixel-based render target does not make much sense. one way to speed up tessellation is cairo_set_tolerance (0.25). """