Age | Commit message (Collapse) | Author | |
---|---|---|---|
2010-09-07 | i965: Don't bother with RNDZ for f2i. | Eric Anholt | |
The default type conversion for MOV should be fine, and RNDZ actually requires two instructions. | |||
2010-09-07 | i965: Add some validation on BRW_OPCODE_MUL and ADD's arguments. | Eric Anholt | |
Now that we're playing with other types in brw_fs.cpp, it's easy to trip over issues like these. | |||
2010-09-07 | i965: Add assertion for another requirement about types. | Eric Anholt | |
This catches a failure in the FS backend. | |||
2010-09-07 | i965: Align the start of attribute interp coefficients in FS to use PLN. | Eric Anholt | |
2010-09-07 | i965: Just assert when we flagged a compile error in the FS for now. | Eric Anholt | |
Dumping back to potentially 16-wide dispatch doesn't really work out at the moment, and hopefully I'll just be able to resolve all the failures so we never have to do this at all. | |||
2010-09-07 | i965: Clean up fs_reg setup by using a helper for constructors. | Eric Anholt | |
2010-09-07 | i965: When using the new FS backend, don't validate the Mesa IR version. | Eric Anholt | |
2010-09-07 | i965: Add a bit of validation for some ISA restrictions in the docs. | Eric Anholt | |
2010-09-07 | i965: set the source width/stride when handling reladdr dests in the VS. | Eric Anholt | |
This is a requirement specified in the docs. No behavior change in glsl-vs-varying-array.shader_test that violated these requirements. | |||
2010-09-07 | Make XF86VIDMODE extension optional | Jon TURNEY | |
Code in glx/glxcmds.c which uses the XF86VIDMODE extension is already guarded. Also use that guard to control inclusion of the xf86vmode.h header, and only enable that guard if the XF86VIDMODE extension is found by pkgconfig. This changes the behaviour on platforms which XF86VIDMODE exists, in that XF86VIDMODE used to be mandatory, but is now optional. Presumably other build systems are already arranging for -DXF86VIDMODE to be supplied to the complier when glxcmds.c is compiled, so are not affected by this change Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk> | |||
2010-09-07 | glx: Drop unused dri2proto.h include | Kristian Høgsberg | |
2010-09-07 | glx: Move dpy and scr fields out of direct rendering conditional | Kristian Høgsberg | |
Nothing direct rendering specific about these fields. Moving them out makes no-direct-rendering compilation work again. Signed-off-by: Kristian Høgsberg <krh@bitplanet.net> | |||
2010-09-07 | Some fixes for GLX_INDIRECT_RENDERING only build | Jon TURNEY | |
This fixes some of the build issues with GLX_INDIRECT_RENDERING but !GLX_DIRECT_RENDERING due to recent changes. Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk> Signed-off-by: Kristian Høgsberg <krh@bitplanet.net> | |||
2010-09-07 | dri: Make it a little clearer that we're not dereferencing a NULL pointer | nobled | |
2010-09-07 | dri: Use the right type for the API token | nobled | |
Pass mesa_api to CreateContext, and abort early if the requested API isn't recognized. | |||
2010-09-07 | libgl-xlib: Fix --enable-gallium-llvm build. | Chia-I Wu | |
Check MESA_LLVM and link to LLVM as other targets do. | |||
2010-09-07 | llvmpipe: Add lp_rast_debug.c to Makefile. | Chia-I Wu | |
2010-09-07 | glx: Only clear the stored context tag when the context has been unbound | Jon TURNEY | |
The calling order of ->bind and ->unbind changed and then ->unbind would clear the currentContextTag of the old context before ->bind could reuse it in the make current request, in the indirect case. Instead, clear the old currentContextTag if and only if we send a request to the server to actually unbind it or reassign it to another context. https://bugs.freedesktop.org/show_bug.cgi?id=29977 Signed-off-by: Jon TURNEY <jon.turney@dronecode.org.uk> Signed-off-by: Kristian Høgsberg <krh@bitplanet.net> | |||
2010-09-07 | llvmpipe: Refactor lp_scene_add_resource_reference | José Fonseca | |
Less goto spaghetti. | |||
2010-09-07 | llvmpipe: Fix negated logic in lp_scene_add_resource_reference(). | José Fonseca | |
Fixes performance regression. | |||
2010-09-07 | llvmpipe: move more coef setup into lp_setup_coef.c | Keith Whitwell | |
2010-09-07 | llvmpipe: declare fence handle struct | Keith Whitwell | |
2010-09-07 | llvmpipe: rasterization debugging helpers | Keith Whitwell | |
2010-09-07 | llvmpipe: use opcodes instead of function pointers in bins | Keith Whitwell | |
Also, move some state from rasterizer struct to the scene. | |||
2010-09-07 | llvmpipe: rearrange queries | Keith Whitwell | |
2010-09-07 | llvmpipe: enforce fixed memory limit on scenes | Keith Whitwell | |
2010-09-07 | llvmpipe: clean up deferred zstencil clears | Keith Whitwell | |
2010-09-07 | llvmpipe: rework fences and queries | Keith Whitwell | |
2010-09-07 | llvmpipe: move whole-tile emit into a function | Keith Whitwell | |
2010-09-07 | llvmpipe: put fs variant dumping in a function | Keith Whitwell | |
2010-09-07 | Replace reference to tgsi-instruction-set.txt. | Tilman Sauerbeck | |
That file has been replaced by tgsi.rst. Signed-off-by: José Fonseca <jfonseca@vmware.com> | |||
2010-09-07 | nouveau: restore nouveau_class.h for nv04-nv20 | Luca Barbieri | |
Sorry, I deleted the Gallium copy without realizing that the DRI one was just a symlink to it. | |||
2010-09-06 | nvfx: fix return in vp main | Luca Barbieri | |
2010-09-06 | mesa: Define C99's __func__ macro on MSVC. | José Fonseca | |
2010-09-06 | r600g: add error print for no literals for r700s as well | Dave Airlie | |
2010-09-06 | r600g: fixup r700 assembler for clamp/relative addressing | Dave Airlie | |
2010-09-06 | r600g: add script to generate header file with offsets into state objects. | Dave Airlie | |
This was inherently fragile as any changes to r600_states.h would also need manual updating of all of the bits in radeon.h. Just add a simple python script to do the conversion, its not hooked up to make at all. This also will make adding evergreen a bit easier. | |||
2010-09-06 | r600g: fix up surface references | Dave Airlie | |
We end up referencing the new surf and derefing the old surface which is a copy of the pointer to the new surf. So just bump the ref count directly. | |||
2010-09-06 | r600g: search for sampler views in context on removal. | Dave Airlie | |
Need to remove from context as well. | |||
2010-09-06 | r600g: add missing printf operand | Dave Airlie | |
2010-09-05 | nvfx: support nv30 simulation on nv40 | Luca Barbieri | |
2010-09-05 | radeon: Add radeon_buffer_objects.c. | Henri Verbeet | |
2010-09-05 | r200: Add radeon_buffer_objects.c. | Henri Verbeet | |
2010-09-05 | radeon: Make do_blit_readpixels() into a PBO work. | Henri Verbeet | |
2010-09-05 | mesa: don't smash the stack in _mesa_find_used_registers | Luca Barbieri | |
At some point this actually triggered, not sure if it still does. Give a meaningful assert and refuse to smash the stack anyway. | |||
2010-09-05 | mesa: don't expose unsupported GL_ARB_geometry_shader4 for now | Luca Barbieri | |
The new GLSL compiler doesn't support it. Advertising it prevents Unigine Heaven from working, since it attempts to use it. | |||
2010-09-05 | nouveau: delete nouveau_class.h, move nv50 regs to nv50_reg.h | Luca Barbieri | |
nv50 should switch to rules-ng-ng too at some point. The classic Mesa Nouveau driver also includes a copy of nouveau_class.h, and should convert to rules-ng-ng too and remove it. | |||
2010-09-05 | nvfx: move nv04_2d to rules-ng-ng | Luca Barbieri | |
2010-09-05 | nvfx: switch to rules-ng-ng register headers | Luca Barbieri | |
This is the new register generation toolkit in use by nouveau. As far as I know, this is the best register description toolkit in existence, and you should use it too for your hardware :) Thanks to Marcin Kościelnicki for inventing it and performing invaluable reverse engineering work of nVidia chips. | |||
2010-09-05 | nvfx: remove remaining BEGIN_RING/eng3d uses | Luca Barbieri | |