summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* radeonsi: make LLVM IR dumping less messyMarek Olšák2016-02-093-9/+15
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: move a few r600_can_dump_shader calls to where they're neededMarek Olšák2016-02-091-5/+5
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: remove useless code that handles dx10_clamp_modeMarek Olšák2016-02-093-14/+6
| | | | | | | "enable-no-nans-fp-math" is a wrong string and there was a disagreement about fixing it. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: dump SPI_PS_INPUT values along with shader statsMarek Olšák2016-02-091-0/+7
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: read SPI_PS_INPUT_ADDR from LLVM if it returns itMarek Olšák2016-02-093-2/+7
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: don't force gl_SampleMaskIn to 1 for smoothingMarek Olšák2016-02-091-7/+4
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: split PS input interpolation code into its own functionMarek Olšák2016-02-091-56/+71
| | | | | | This will be used by the fragment shader prolog. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: implement forcing per-sample_interpolation using the shader key onlyMarek Olšák2016-02-096-152/+55
| | | | | | | | | | | It was partly a state and partly emulated by shader code, but since we want to do this in a fragment shader prolog, we need to put it into the shader key, which will be used to generate the prolog. This also removes the spi_ps_input states and moves the registers to the PS state. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: remove si_shader::ps_input_interpolateMarek Olšák2016-02-092-6/+3
| | | | | | tgsi_shader_info has this too. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: move BCOLOR PS input locations after all other inputsMarek Olšák2016-02-093-29/+50
| | | | | | | | | | | | | | | | BCOLOR inputs were immediately after COLOR inputs. Thus, all following inputs were offset by 1 if color_two_side was enabled, and not offset if it was not enabled, which is a variation that's problematic if we want to have 1 variant per shader and the variant doesn't care about color_two_side (that should be handled by other bytecode attached at the beginning). Instead, move BCOLOR inputs after all other inputs, so BCOLOR0 is at location "num_inputs" if it's present. BCOLOR1 is next. This also allows removing si_shader::nparam and si_shader::ps_input_param_offset, which are useless now. Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: move SPI_PS_INPUT_CNTL value computation to a separate functionMarek Olšák2016-02-091-34/+40
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: generate a color_two_side variant only if the shader reads colorsMarek Olšák2016-02-091-1/+1
| | | | Reviewed-by: Nicolai Hähnle <[email protected]>
* radeonsi: move si_shader_context initialization into a separate functionMarek Olšák2016-02-091-43/+60
| | | | | | This will be re-used later. Reviewed-by: Nicolai Hähnle <[email protected]>
* st/mesa: remove st_is_program_nativeMarek Olšák2016-02-091-13/+0
| | | | | | The default scenario sets GL_TRUE too. Reviewed-by: Edward O'Callaghan <[email protected]>
* st/mesa: unify destroy_program_variants cases for TCS, TES, GSMarek Olšák2016-02-091-50/+16
| | | | Reviewed-by: Edward O'Callaghan <[email protected]>
* st/mesa: unify get_variant functions for TCS, TES, GSMarek Olšák2016-02-093-176/+31
| | | | Reviewed-by: Edward O'Callaghan <[email protected]>
* st/mesa: unify variants and delete functions for TCS, TES, GSMarek Olšák2016-02-095-214/+108
| | | | | | no difference between those Reviewed-by: Edward O'Callaghan <[email protected]>
* mesa: fix incorrect viewport position when GL_CLIP_ORIGIN = GL_LOWER_LEFTBrian Paul2016-02-091-2/+2
| | | | | | | | | Ilia Mirkin found/fixed the mistake. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=93813 Cc: "11.1" <[email protected]> Reviewed-by: Jose Fonseca <[email protected]> Reviewed-by: Roland Scheidegger <[email protected]>
* mesa: rewrite save_CallLists() codeBrian Paul2016-02-091-26/+35
| | | | | | | | When glCallLists() is compiled into a display list, preserve the call as a single glCallLists rather than 'n' glCallList calls. This will matter for an upcoming display list optimization project. Reviewed-by: Ian Romanick <[email protected]>
* mesa: add missing error check in _mesa_CallLists()Brian Paul2016-02-091-0/+8
| | | | | | | | Generate GL_INVALID_VALUE if n < 0. Return early if n==0 or lists==NULL. v2: fix formatting, also check for lists==NULL. Reviewed-by: Ian Romanick <[email protected]>
* mesa: whitespace clean-ups in dlist.hBrian Paul2016-02-091-16/+31
| | | | And remove 'extern' qualifiers.
* st/mesa: don't allocate bitmap drawing state until neededBrian Paul2016-02-093-72/+77
| | | | | | | | | Most apps don't use glBitmap so don't allocate the bitmap cache or gallium state objects/shaders/etc until the first call to st_Bitmap(). v2: simplify a conditional, per Gustaw Smolarczyk. Reviewed-by: Nicolai Hähnle <[email protected]>
* st/mesa: move the setup_bitmap_vertex_data() code into draw_bitmap_quad()Brian Paul2016-02-091-90/+78
| | | | | | Now all the code to setup the vertex data and draw it is in one place. Reviewed-by: Nicolai Hähnle <[email protected]>
* st/mesa: refactor some bitmap drawing codeBrian Paul2016-02-091-33/+57
| | | | | | | Move setup/restoration of rendering state into helper functions. This makes the draw_bitmap_quad() function much more concise. Reviewed-by: Nicolai Hähnle <[email protected]>
* mesa: remove hack to fix up GL_ANY_SAMPLES_PASSED resultsIlia Mirkin2016-02-091-5/+0
| | | | | | | | | | Both st/mesa and i965 should return a true/false result now, and the only other driver implementing queries (radeon) doesn't support ARB_occlusion_query2 which added that pname. Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Marek Olšák <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* st/mesa: make use of the occlusion predicate queryIlia Mirkin2016-02-091-2/+10
| | | | | | Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Marek Olšák <[email protected]> Reviewed-by: Nicolai Hähnle <[email protected]>
* nv50: add PIPE_QUERY_OCCLUSION_PREDICATE supportIlia Mirkin2016-02-091-0/+6
| | | | | Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Samuel Pitoiset <[email protected]>
* nv30: add PIPE_QUERY_OCCLUSION_PREDICATE supportIlia Mirkin2016-02-091-2/+5
| | | | Signed-off-by: Ilia Mirkin <[email protected]>
* ilo: add PIPE_QUERY_OCCLUSION_PREDICATE supportIlia Mirkin2016-02-093-1/+12
| | | | | Signed-off-by: Ilia Mirkin <[email protected]> Reviewed-by: Chia-I Wu <[email protected]>
* draw: use util_pstipple_* function for stipple pattern textures and samplersNicolai Hähnle2016-02-093-112/+18
| | | | | | | | This reduces code duplication. Suggested-by: Jose Fonseca <[email protected]> Reviewed-by: Brian Paul <[email protected]> Reviewed-by: Jose Fonseca <[email protected]>
* draw: use util_pstipple_create_fragment_shaderNicolai Hähnle2016-02-091-197/+12
| | | | | | | | | This reduces code duplication. It also adds support for drivers where the fragment position is a system value. Suggested-by: Jose Fonseca <[email protected]> Reviewed-by: Brian Paul <[email protected]> Reviewed-by: Jose Fonseca <[email protected]>
* winsys/radeon: fix a wrong NUM_TILE_PIPES value from the kernelMarek Olšák2016-02-091-0/+6
| | | | | | | Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=94019 Tested-by: Nick Sarnie <[email protected]> Reviewed-by: Michel Dänzer <[email protected]>
* nir: remove unused nir_variable fieldsTimothy Arceri2016-02-092-20/+0
| | | | | | | These are used in GLSL IR to removed unused varyings and match transform feedback variables. There is no need to use these in NIR. Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: remove unrequired forward declarationTimothy Arceri2016-02-091-2/+0
| | | | | | | This was added in 2548092ad80156a4 although I don't see why as it was already in the linker.h header. Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: clean up and fix bug in varying linking rulesTimothy Arceri2016-02-091-74/+63
| | | | | | | | | | | | | | | | The existing code was very hard to follow and has been the source of at least 3 bugs in the past year. The existing code also has a bug for SSO where if we have a multi-stage SSO for example a tes -> gs program, if we try to use transform feedback with gs the existing code would look for the transform feedback varyings in the tes stage and fail as it can't find them. V2: Add more code comments, always try to remove unused inputs to the first stage. Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: simplify ES Vertex/Fragment shader requirementsTimothy Arceri2016-02-091-28/+28
| | | | | | | | | | | | We really just needed to skip the existing ES < 3.1 check if we have a compute shader, all other scenarios are already covered. * No shaders is a link error. * Geom or Tess without Vertex is a link error which means we always require a Vertex shader and hence a Fragment shader. * Finally a Compute shader linked with any other stage is a link error. Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: simplify required stages for linking rulesTimothy Arceri2016-02-091-43/+41
| | | | Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: small tidy up now that link_shaders() exits early with 0 shadersTimothy Arceri2016-02-091-6/+4
| | | | Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: don't attempt to link empty programTimothy Arceri2016-02-091-23/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously an empty program would go through the entire link_shaders() function and we would have to be careful not to cause a segfault. In core profile also now set link_status to false by generating an error, it was previously set to true. From Section 7.3 (PROGRAM OBJECTS) of the OpenGL 4.5 spec: "Linking can fail for a variety of reasons as specified in the OpenGL Shading Language Specification, as well as any of the following reasons: - No shader objects are attached to program." V2: Only generate an error in core profile and add spec quote (Ian) V3: generate error in ES too, remove previous check which was only applying the rule to GL 4.5/ES 3.1 and above. My understand is that this spec change is clarifying previously undefined behaviour and therefore should be applied retrospectively. The ES CTS tests for this are in ES 2 I suspect it was passing because it would have generated an error for not having both a vertex and fragment shader. Reviewed-by: Kenneth Graunke <[email protected]>
* nir: Recognize open-coded bitfield_reverse.Matt Turner2016-02-081-0/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | Helps 11 shaders in UnrealEngine4 demos. I seriously hope they would have given us bitfieldReverse() if we exposed GL 4.0 (but we do expose ARB_gpu_shader5, so why not use that anyway?). instructions in affected programs: 4875 -> 4633 (-4.96%) cycles in affected programs: 270516 -> 244516 (-9.61%) I suspect there's a *lot* of room to improve nir_search/opt_algebraic's handling of this. We'd actually like to match, e.g., step2 by matching step1 once and then doing a pointer comparison for the second instance of step1, but unfortunately we generate an enormous tuple for instead. The .text size increases by 6.5% and the .data by 17.5%. text data bss dec hex filename 22957 45224 0 68181 10a55 nir_libnir_la-nir_opt_algebraic.o 24461 53160 0 77621 12f35 nir_libnir_la-nir_opt_algebraic.o I'd be happy to remove this if Unreal4 uses bitfieldReverse() if it is in a GL 4.0 context once we expose GL 4.0. Reviewed-by: Jason Ekstrand <[email protected]>
* nir: Handle large unsigned values in opt_algebraic.Matt Turner2016-02-081-4/+1
| | | | | | | | | | | | | | | The next patch adds an algebraic rule that uses the constant 0xff00ff00. Without this change, the build fails with return hex(struct.unpack('I', struct.pack('i', self.value))[0]) struct.error: 'i' format requires -2147483648 <= number <= 2147483647 The hex() function handles integers of any size, and assigning a negative value to an unsigned does what we want in C. The pack/unpack is unnecessary (and as we see, buggy). Reviewed-by: Dylan Baker <[email protected]>
* nir: Do opt_algebraic in reverse order.Matt Turner2016-02-081-2/+2
| | | | | | | | | | | | | | | | | | | | | | Walking the SSA definitions in order means that we consider the smallest algebraic optimizations before larger optimizations. So if a smaller rule is part of a larger rule, the smaller one will happen first, preventing the larger one from happening. instructions in affected programs: 32721 -> 32611 (-0.34%) helped: 106 In programs whose nir_optimize loop count changes (129 of them): before: 1164 optimization loops after: 1071 optimization loops Of the 129 affected, 16 programs' optimization loop counts increased. Prevents regressions and annoyances in the next commits. Reviewed-by: Eduardo Lima Mitev <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* nir: Recognize product of open-coded pow()s.Matt Turner2016-02-081-0/+2
| | | | | | Prevents regressions in the next commit. Reviewed-by: Jason Ekstrand <[email protected]>
* nir: Add opt_algebraic rules for xor with zero.Matt Turner2016-02-081-0/+2
| | | | | | | | instructions in affected programs: 668 -> 664 (-0.60%) helped: 4 Reviewed-by: Eduardo Lima Mitev <[email protected]> Reviewed-by: Jason Ekstrand <[email protected]>
* glsl: validate arrays of arrays on empty type delclarationsTimothy Arceri2016-02-091-25/+38
| | | | | | | | Fixes: dEQP-GLES31.functional.shaders.arrays_of_arrays.invalid.empty_declaration_without_var_name_fragment dEQP-GLES31.functional.shaders.arrays_of_arrays.invalid.empty_declaration_without_var_name_vertex Reviewed-by: Dave Airlie <[email protected]>
* i965: Use nir_lower_load_const_to_scalar().Kenneth Graunke2016-02-081-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I don't know why, but we never hooked up this pass Eric wrote. Otherwise, you can end up with stupid scalarized code such as: vec4 ssa_7 = load_const (0.0, 0.0, 0.0, 0.0) vec4 ssa_8 = ... vec1 ssa_9 = feq ssa_8, ssa_7 vec1 ssa_10 = feq ssa_8.y, ssa_7.y vec1 ssa_11 = feq ssa_8, ssa_7.z vec1 ssa_12 = feq ssa_8.y, ssa_7.w ssa_8.xyxy == <0, 0, 0, 0> should only take two feq instructions. shader-db on Skylake: total instructions in shared programs: 9121153 -> 9120749 (-0.00%) instructions in affected programs: 32421 -> 32017 (-1.25%) helped: 277 HURT: 69 total cycles in shared programs: 69003364 -> 69000912 (-0.00%) cycles in affected programs: 899186 -> 896734 (-0.27%) helped: 313 HURT: 403 This also prevents regressions when disabling channel expressions. v2: Don't call opt_cse afterwards (requested by Matt). It should happen in the optimization loop below anyway. Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Eduardo Lima Mitev <[email protected]> Reviewed-by: Matt Turner <[email protected]>
* mesa: remove now unused sampler index handing codeTimothy Arceri2016-02-094-172/+0
| | | | Reviewed-by: Dave Airlie <[email protected]>
* mesa: compute sampler index in ir_to_mesa rather than using UniformHashTimothy Arceri2016-02-091-3/+78
| | | | | | | | The aim of this is to work towards removing UniformHash from the program struct so that we don't need to hold onto it in memory and pass it around outside the linker. Reviewed-by: Dave Airlie <[email protected]>
* i965: Don't add barrier deps for FB write messages.Kenneth Graunke2016-02-081-3/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are never render target reads, so there are no scheduling hazards. Giving the extra flexibility to the scheduler makes it possible to do FB writes as soon as their sources are available, reducing register pressure. It also makes it possible to do the payload setup for more than one FB write message at a time, which could better hide latency. shader-db results on Skylake: total instructions in shared programs: 9110254 -> 9110211 (-0.00%) instructions in affected programs: 2898 -> 2855 (-1.48%) helped: 3 HURT: 0 LOST: 0 GAINED: 1 A reduction in instruction counts is surprising, but legitimate: the three shaders helped were spilling, and reducing register pressure allowed us to issue fewer spills/fills. total cycles in shared programs: 69035108 -> 68928820 (-0.15%) cycles in affected programs: 4412402 -> 4306114 (-2.41%) helped: 4457 HURT: 213 Signed-off-by: Kenneth Graunke <[email protected]> Reviewed-by: Matt Turner <[email protected]> Reviewed-by: Connor Abbott <[email protected]>
* st/mesa: enable AoA for gallium drivers reporting GLSL 1.30Dave Airlie2016-02-091-0/+1
| | | | | | Acked-by: Ilia Mirkin <[email protected]> Reviewed-by: Timothy Arceri <[email protected]> Signed-off-by: Dave Airlie <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-22 20:50:25 +0000