summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* nv50: create value references with the right typeChristoph Bumiller2010-09-092-26/+29
| | | | | | | | | Since atm our OPs aren't typed but instead values are, we need to take care if they're used as different types (e.g. a load makes a value u32 by default). Maybe this should be changed (also to match TGSI), but it should work as well if done properly.
* nv50: use actual loads/stores if TEMPs are accessed indirectlyChristoph Bumiller2010-09-0910-23/+122
|
* nv50: don't parse again in tgsi_2_ncChristoph Bumiller2010-09-091-15/+3
|
* nv50: prepare for having multiple functionsChristoph Bumiller2010-09-098-52/+171
| | | | | | | | | At some point we'll want to support real subroutines instead of just inlining them into the main shader. Since recursive calls are forbidden, we can just save all used registers to a fixed local memory region and restore them on a return, no need for a stack pointer.
* nv50: save tgsi instructionsChristoph Bumiller2010-09-092-0/+6
|
* nv50: load address register before using it, not afterChristoph Bumiller2010-09-031-5/+7
|
* Merge remote branch 'origin/master' into nv50-compilerChristoph Bumiller2010-09-02303-8515/+18498
|\ | | | | | | | | Conflicts: src/gallium/drivers/nv50/nv50_program.c
| * r600g: fix memory/bo leakJerome Glisse2010-09-024-2/+21
| | | | | | | | Signed-off-by: Jerome Glisse <[email protected]>
| * r600g: fix thinko in shadow code.Dave Airlie2010-09-021-1/+1
| | | | | | | | spotted by taiu on irc
| * r600g: fix logicop, the 3d ROP is the 2D rop shifted twice.Dave Airlie2010-09-021-1/+1
| |
| * r600g: fix depth texture testsDave Airlie2010-09-021-2/+2
| |
| * r600g: add missing vertex fetch formats to the translation table.Dave Airlie2010-09-022-0/+3
| | | | | | | | fixes at least 2 more piglits.
| * r600g: fix binding of same texture to several target slotJerome Glisse2010-09-012-23/+62
| | | | | | | | | | | | | | | | | | One can bind same texture or sampler to different slot, each slot needs it own state. The solution implemented here is not exactly beautifull or optimal need to think to somethings better. Signed-off-by: Jerome Glisse <[email protected]>
| * r600g: fix incorrect state naming in pipe_sampler vs pipe_sampler_viewDave Airlie2010-09-021-1/+1
| | | | | | | | fixes problems in valgrind with uninitialised values.
| * r600g: silence compiler warningJerome Glisse2010-09-011-1/+1
| | | | | | | | Signed-off-by: Jerome Glisse <[email protected]>
| * r600g: refix db/cb stateJerome Glisse2010-09-016-32/+119
| | | | | | | | | | Signed-off-by: Dave Airlie <[email protected]> Signed-off-by: Jerome Glisse <[email protected]>
| * r600g: fix up default state differences between r6xx and r7xxAlex Deucher2010-09-011-6/+16
| | | | | | | | Signed-off-by: Alex Deucher <[email protected]>
| * st/glx: re-order destruction of buffers, visualsBrian Paul2010-09-011-1/+1
| | | | | | | | | | Free the buffers before the visuals. Fixes valgrind warning reported in fd.o bug 29919.
| * r600g: avoid dynamic allocation of statesJerome Glisse2010-09-0117-1237/+523
| | | | | | | | | | | | | | | | | | | | | | Make state statically allocated, this kills a bunch of code and avoid intensive use of malloc/free. There is still a lot of useless duplicate function wrapping that can be kill. This doesn't improve yet performance, needs to avoid memcpy states in radeon_ctx_set_draw and to avoid rebuilding vs_resources, dsa, scissor, cb_cntl, ... states at each draw command. Signed-off-by: Jerome Glisse <[email protected]>
| * Revert "Revert "r600g: precompute some of the hw state""Jerome Glisse2010-09-0111-120/+229
| | | | | | | | | | | | | | | | This reverts commit 1fa7245c348cb7aced81f1672140f64cb6450e2f. Conflicts: src/gallium/drivers/r600/r600_state.c
| * nouveau/nvfx: Remove enforcement of bit depth being same as front bufferPatrice Mandin2010-09-011-17/+0
| | | | | | | | Signed-off-by: Patrice Mandin <[email protected]>
| * r600g: correct cb/zb offset emits.Dave Airlie2010-09-011-2/+2
| | | | | | | | This fixes fbo-3d and fbo-cubemap
| * Revert "r600g: precompute some of the hw state"Dave Airlie2010-09-0111-229/+216
| | | | | | | | | | | | | | | | | | | | This reverts commit de0b76cab22caa9fc7260f80acb8f151ccced6c5, its pre-computes the texture state wrong, you can't just use an array of levels, since you can have FBOs to depth texture slices inside a level as well it would get really messy quickly. Probably need to split commits like this up into pieces for each piece of state, so we can revert bits easier in case of regressions. This also break 5 piglit tests, and valgrind starts to warn about invalid read/writes after this.
| * r600g: fix typo causing segfault.Dave Airlie2010-09-011-1/+1
| | | | | | | | | | | | | | | | fixes warning that r600_blit.c: In function ‘r600_resource_copy_region’: r600_blit.c:136: warning: passing argument 1 of ‘util_resource_copy_region’ from incompatible pointer type and also 7 more piglit tests.
| * r600g: fix glean texCube and shadows.Dave Airlie2010-09-011-5/+149
| | | | | | | | add cube and shadow support to the texture code.
| * gallivm: fix bug in nested conditionalsBrian Paul2010-08-311-2/+4
| | | | | | | | This, plus the previous commit fix fd.o bug 29806.
| * llvmpipe: slightly simplify build_maskKeith Whitwell2010-08-311-6/+2
| |
| * llvmpipe: combine linear mask calculationKeith Whitwell2010-08-312-15/+84
| |
| * llvmpipe: intrinsics versions of build_mask functionsKeith Whitwell2010-08-311-1/+77
| |
| * st/egl: Enable EGL_MESA_drm_display.Chia-I Wu2010-08-311-1/+3
| |
| * r600g: fix up depth write swizzles.Dave Airlie2010-08-311-1/+4
| | | | | | | | | | | | | | | | For some reason r600c, emits extra instructions in the FP to do the depth write output swizzle, I'm not sure this is required, so here I'm doing it in the exports. this fixes the mesa trivial demos tri-depthwrite and tri-depthwrite2, it doesn't fix the glsl1 gl_FragDepth writing test however.
| * r600g: fix fp-fragment-position test.Dave Airlie2010-08-311-0/+1
| |
| * r600g: fix typo in last commitDave Airlie2010-08-311-1/+1
| |
| * r600g: fix position input to fragment shader.Dave Airlie2010-08-311-0/+7
| | | | | | | | this fixes a few if the fs shader tests, 10 more piglits
| * r600g: remove unneeded function call from scsDave Airlie2010-08-311-4/+0
| |
| * r600g: make LIT work properlyDave Airlie2010-08-311-8/+3
| | | | | | | | | | | | | | this is a bit of a workaround, something is wrong with the literal emits here so we just use the trig copy function to copy the immd to a temp at start of op. fix VP/FP LIT tests
| * r600g: fixup trig functions when input is a literalDave Airlie2010-08-311-9/+67
| | | | | | | | | | | | | | | | | | So as the trig functions used up the literal spots for the PI work, if the arg0 was an immediate we'd hit failure, so copy the literal before starting. add some tracking of max temp used to avoid trashing temp regs. 5 more piglits, fp1 COS,SCS,SIN tests
| * r600g: make sure LIT splits constantsDave Airlie2010-08-311-14/+11
| |
| * r600g: fix constant splittingDave Airlie2010-08-311-2/+2
| | | | | | | | constant splitting was broken for multi-constant cases, fixes fp1 CMP+MAD, vp1 CMP.
| * r600g: fix LIT testsDave Airlie2010-08-311-2/+3
| |
| * r600g: add missing literalsDave Airlie2010-08-312-1/+33
| | | | | | | | | | | | | | Also add an error if we hit this problem again, we need to do this better possibly tying the literal addition to the last flag. Signed-off-by: Dave Airlie <[email protected]>
| * r600g: precompute some of the hw stateJerome Glisse2010-08-3011-216/+229
| | | | | | | | | | | | | | | | | | | | Idea is to build hw state at pipe state creation and reuse them while keeping a non PM4 packet interface btw winsys & pipe driver. This commit also force rebuild of pm4 packet on each call to radeon_state_pm4 which in turn slow down everythings, this will be addressed. Signed-off-by: Jerome Glisse <[email protected]>
| * r600g: fix depth buffer decompression after states reworkJerome Glisse2010-08-301-1/+1
| | | | | | | | Signed-off-by: Jerome Glisse <[email protected]>
| * r600g: fixup states generation in winsys.Dave Airlie2010-08-3015-335/+295
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current states code had an unhealthy relationship between that had to somehow magically align themselves, editing either place meant renumbering all states after the one you were on, and it was pretty unapproachable code. This replaces the huge types structures with a simple type + sub type struct, which is keyed on an stype enum in radeon.h. Each stype can have a per-shader type subclassing (4 types supported, PS/VS/GS/FS), and also has a number of states per-subtype. So you have 256 constants per 4 shaders per one CONSTANT stype. The interface from the driver is changed to pass in the tuple, (stype, id, shader_type), and we look for this. If radeon_state_shader ever shows up on profile, it could use a hashtable based on stype/shader_type to speed things up. Signed-off-by: Dave Airlie <[email protected]>
| * gallivm: Compute the 4 texel offsets for linear filtering en ensemble.José Fonseca2010-08-303-126/+280
| |
| * gallivm: Disable LLVM's pretty stack trace dumper.José Fonseca2010-08-301-0/+8
| | | | | | | | | | | | | | | | By default LLVM adds a signal handler to output a pretty stack trace. This signal handler is never removed, causing problems when unloading the shared object where the gallium driver resides. Thanks to Chris Li for finding this.
| * gallivm: Correct copy'n'pasted comments.José Fonseca2010-08-301-4/+4
| |
| * gallivm: Fix lp_build_sum_vector.José Fonseca2010-08-301-6/+4
| | | | | | | | | | | | | | The result is scalar, so when argument is zero/undef we can pass vector zero/undef. Also, support the scalar case.
| * svga: Fix CMP translation for vertex shader targets.Michal Krol2010-08-301-0/+19
| | | | | | | | | | SVGA3DOP_CMP is not supported for vertex shaders; use SLT + LRP instead.
| * svga: Re-emit bound rendertargets and texture samplers at the beginning of ↵José Fonseca2010-08-304-8/+27
| | | | | | | | | | | | | | | | | | every command buffer. Only non null resources. To ensure that relocations are emitted for every resource currently referred.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-27 01:18:10 +0000