android-x86-mesa.git - Androïd/x86 port of Mesa drivers

Age	Commit message (Collapse)	Author
2009-01-18	nouveau: Rename buffer on map if discardable, busy, and write-only.	Younes Manton

2009-01-18	gallium: Add PIPE_BUFFER_USAGE_DISCARD.	Younes Manton
	When passed to map() signals that the buffer's previous contents are not required, allowing the driver to allocate a new buffer if the current buffer can not be mapped immediately.
2009-01-18	g3dvl: Return BadAlloc if we can't create an XvMC surface.	Younes Manton

2009-01-19	i915: Build gem and egl winsys by default	Jakob Bornecrantz

2009-01-19	i915: Make gem submit commands	Jakob Bornecrantz

2009-01-19	egl: Fix swap and creation of front buffer	Jakob Bornecrantz

2009-01-18	i915: Use new egl state_tracker	Jakob Bornecrantz

2009-01-18	egl: Add a egl state_tracker that use Gallium	Jakob Bornecrantz
	This works on top Gallium and KMS. The only thing that does not work currently is swap buffers for shown mesa screens. So the only fun thing this will produce is a white screen. The driver wishing to us the state_tracker needs to implement the intrace as define in drm_api.h located in gallium/include/state_tracker. And also have a working KMS implementation.
2009-01-18	i915: Update gem backend a bit	Jakob Bornecrantz

2009-01-14	gallium: Disable memory debugging for Windows OGL.	José Fonseca
	Unfortunately both Mesa and Gallium use the same defines for memory allocation (MALLOC, FREE, etc), and worse, some times memory is allocated with one set and freed with the other set, causing the homegrown memory debugger to trip on itself. In the future mesa and gallium should use different names, but for now, memory debugging on Windows will have to be carried with different tools..
2009-01-14	intel: Add a none working GEM backend for intel	Jakob Bornecrantz

2009-01-14	cell: Specify constant as float for CEILF().	Jonathan Adamczewski
	Without the f, the constant is treated as a double, resulting in slower arithmetic and libgcc conversion calls each time CEILF() is used.
2009-01-13	nv50: fix progs/tests/manytex	Ben Skeggs
	Previously all squares were textured with the same texture.. not quite what the demo was supposed to look like!
2009-01-13	nv50: get glxgears showing all 3 gears instead of 1!!	Ben Skeggs
	This fixes a lot of other things where not all the geometry got drawn also.
2009-01-13	nv50: change some magic reg, makes more things work	Ben Skeggs
	No real idea what this does.. but a lot of things that misrendered and made the GPU throw a DATA_ERROR now work.. I'm wondering what side-effects we'll see from this :)
2009-01-13	nv50: add DXTn formats	Ben Skeggs

2009-01-13	nv50: shadow mapping	Ben Skeggs

2009-01-13	nv50: aniso	Ben Skeggs

2009-01-13	nv50: occlusion queries	Ben Skeggs
	Not quite working, but the general idea is right I think.
2009-01-13	nv50: implement KIL enough for progs/fp/kil to work	Ben Skeggs

2009-01-13	cell: Add missing suffix to SHUFFLE macro	Jonathan Adamczewski

2009-01-12	cell: allocate batch buffers w/ 16-byte alignment	Jonathan Adamczewski
	Replace cell_batch{align,alloc)*() with cell_batch_alloc16(), allocating multiples of 16 bytes that are 16 byte aligned. Opcodes are stored in preferred slot of SPU machine word. Various structures are explicitly padded to 16 byte multiples. Added STATIC_ASSERT().
2009-01-12	gallivm: Make sure the bitcode buffer is followed by a 0 byte.	Michel Dänzer
	May fail to parse otherwise.
2009-01-12	gallivm: Print error message from ParseBitcodeFile() in case it fails.	Michel Dänzer

2009-01-12	gallivm: Adapt to header file move in LLVM 2.4.	Michel Dänzer

2009-01-12	nv50: make rtt work again	Ben Skeggs

2009-01-12	nv50: fix handling of depth textures	Ben Skeggs

2009-01-12	nv50: another typo..	Ben Skeggs

2009-01-12	nouveau: fix warning	Ben Skeggs

2009-01-12	nv50: fix assertion failure	Ben Skeggs

2009-01-12	nv50: remove previous hack to manage tiled surfaces	Ben Skeggs

2009-01-12	nv50: fix a typo and a thinko	Ben Skeggs

2009-01-12	nv50: enable npot textures	Ben Skeggs

2009-01-12	nv50: disable shader debug	Ben Skeggs

2009-01-12	nv50: any cpu access to a texture is done on its backing images	Ben Skeggs
	Still a little dodgy: - RTT will hit an assertion (hopefully!) and fail - 3D textures with depth >= 32 will cause bad things to happen
2009-01-12	nv50: create buffers for each image that makes up a texture	Ben Skeggs

2009-01-12	nouveau: return buffer map to something sane.	Ben Skeggs
	Sorry, but no, we're not doing this.. Correctness always takes precedence over speed. Implement this higher up where you know it's safe to do so, and doesn't break other things in the process.
2009-01-12	nouveau: use usage, not uninitialised flags value...	Ben Skeggs

2009-01-11	cell: optimize unpack_colors() function, saving 12 cycles	Brian Paul

2009-01-11	cell: move color unpacking code into separate function	Brian Paul

2009-01-11	cell: re-order the z/stencil fetch/extract/convert instructions for better perf	Brian Paul
	The new instruction order is 10 cycles faster.
2009-01-11	cell: datatype clean-ups in SPE rtasm	Brian Paul

2009-01-11	cell: simplify the 'optional register' code	Brian Paul

2009-01-11	cell: asst datatype clean-ups	Brian Paul

2009-01-11	cell: move depth/stencil code into separate function	Brian Paul

2009-01-11	cell: clean-up, re-indent, comments	Brian Paul

2009-01-10	cell: use tgsi_dump_instruction() instead of spe_comment()	Brian Paul

2009-01-10	gallium: use tgsi_dump_instruction() instead of ppc_comment()	Brian Paul

2009-01-10	gallium: remove unused struct type	Brian Paul

2009-01-10	gallium: fix register clobber bug in TGSI->PPC codegen	Brian Paul
	When negating a src vector that's stored in a altivec register, need to put negated value into a new register so we don't upset the original value. This solves the dark colors in the mandelbrot GLSL demo. Also, use new predicate functions to check if a TGSI temp is stored in an altivec register.