summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Expand)AuthorFilesLines
2015-04-01i915g: Implement EGL_EXT_image_dma_buf_importStéphane Marchesin5-7/+38
2015-04-01i965/fs: Relax type check in cmod propagation.Matt Turner1-1/+3
2015-04-01nir: Remove useless ftrunc inside f2i/f2u.Matt Turner1-0/+4
2015-04-01nir: Recognize (a < b || a < c) as a < max(b, c).Matt Turner1-0/+2
2015-04-01nir: Add addition/multiplication identities of exp/log.Matt Turner1-0/+6
2015-04-01nir: Add identities for the log function.Matt Turner1-0/+8
2015-04-01nir: Add identities for the exponential function.Matt Turner1-0/+6
2015-04-01nir: Recognize another open coded lrp.Matt Turner1-0/+1
2015-04-01nir: Recognize open coded lrp.Matt Turner1-0/+1
2015-04-01nir: Use _mesa_flsll(InputsRead) in prog->nir.Kenneth Graunke1-2/+2
2015-04-01mesa: Implement _mesa_flsll().Kenneth Graunke1-0/+24
2015-04-01nir: In prog->nir, don't wrap dot products with ptn_channel(..., X).Kenneth Graunke1-4/+4
2015-04-01i965: Use the same nir options for all gensJason Ekstrand1-10/+2
2015-04-01i965/nir: Run DCE again before going out of SSAJason Ekstrand1-0/+2
2015-04-01i965/nir: Run the ffma peephole after the rest of the optimizationsJason Ekstrand2-2/+11
2015-04-01nir/peephole_ffma: Be less agressive about fusing multiply-addsJason Ekstrand1-0/+41
2015-04-01nir: Add a dedicated ffma peephole optimizationJason Ekstrand3-0/+223
2015-04-01nir: Move the compare-with-zero optimizations to the late sectionJason Ekstrand1-4/+4
2015-04-01nir/algebraic: Add a seperate section for "late" optimizationsJason Ekstrand3-0/+13
2015-04-01nir/algebraic: Remove a duplicate optimizationJason Ekstrand1-3/+0
2015-04-01nir/algebraic: #define around structure definitionsJason Ekstrand1-6/+11
2015-04-01nir/print: Don't print extra swizzzle componentsJason Ekstrand1-7/+19
2015-04-01configure: nuke --with-max-{width,height}Emil Velikov2-2/+2
2015-04-01gallium: ship tgsi_to_nir.h in the tarballEmil Velikov1-1/+2
2015-04-01glsl: Make sure not to dereference NULL.Matt Turner1-0/+2
2015-04-01main: create_buffers unlocks mutex when throwing OUT_OF_MEMORY.Laura Ekstrand1-0/+1
2015-04-01automake,scons: Put NIR source files in a separate var to fix SCons build.Jose Fonseca2-1/+4
2015-04-01automake: Fix out-of-source builds.Jose Fonseca1-0/+1
2015-04-01mesa: don't include colormac.h in format codeBrian Paul3-2/+2
2015-04-01mesa: remove unneeded #include of colormac.hBrian Paul13-13/+5
2015-04-01tnl: remove unneeded #include of colormac.hBrian Paul11-11/+1
2015-04-01swrast: remove unneeded #include of colormac.hBrian Paul19-19/+4
2015-04-01mesa: remove unused macros from colormac.hBrian Paul1-45/+0
2015-04-01nir: Recognize a pattern of bool frobbing from TGSI KILL_IF.Eric Anholt1-0/+2
2015-04-01nir: Recognize a pattern for doing b2f without the opcode.Eric Anholt1-0/+1
2015-04-01vc4: Add shader-db dumping of NIR instruction count.Eric Anholt1-0/+29
2015-04-01vc4: Convert to consuming NIR.Eric Anholt5-720/+707
2015-04-01gallium: Add tgsi_to_nir to get a nir_shader for a TGSI shader.Eric Anholt3-0/+1454
2015-04-01vc4: Tell shader-db how big our UBOs are, if present.Eric Anholt1-0/+6
2015-04-01mesa: Make a shared header for 3D pipeline enum / #defines.Eric Anholt3-142/+173
2015-04-01nir: add nir_builder.h to the tarballEmil Velikov1-0/+1
2015-04-01xmlpool: remove the clean targetEmil Velikov1-6/+4
2015-04-01xmlpool: don't forget to ship the MOSEmil Velikov1-1/+8
2015-04-01osmesa: don't try to bundle osmesa.def SConscriptEmil Velikov1-2/+0
2015-04-01i965: Handle scratch accesses where reladdr also points to scratch spaceIago Toral Quiroga2-26/+76
2015-04-01gallivm: (trivial) fix the logic deciding if function call should be used...Roland Scheidegger1-3/+1
2015-04-01mesa/fbo: lock ctx->Shared->Mutex when allocating renderbuffersMartin Peres1-0/+2
2015-04-01mesa/fbo: do not assign a value that is never read later onMartin Peres1-6/+3
2015-04-01egl: add initial EGL_MESA_image_dma_buf_export v2.4Dave Airlie7-4/+179
2015-03-31i965/state: Remove brw->state.dirtyJordan Justen2-7/+0