summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* glsl: Add a visitor to determine whether a uniform block is ever usedIan Romanick2013-01-253-0/+225
| | | | | Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* glsl: Lower UBO references using link-time data instead of compile-time dataIan Romanick2013-01-251-3/+86
| | | | | | | | Pretty much all of the compile-time, per-compilation unit block data is about to get the axe. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* glsl: Add gl_uniform_buffer_variable::IndexName fieldIan Romanick2013-01-254-2/+22
| | | | | | | | | | | | | glGetUniformIndices requires that the block instance index not be present in the name of queried uniforms. However, gl_uniform_buffer_variable::Name will include the instance index. The IndexName field is added to handle this difference. Note that currently IndexName will always point to the same string as Name. This will change soon. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]>
* glsl: Make the align function available elsewhere in the linkerIan Romanick2013-01-254-31/+20
| | | | | Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]>
* glsl: Calculate link-time uniform block data without using compile-time ↵Ian Romanick2013-01-251-24/+54
| | | | | | | | | | block data Pretty much all of the compile-time, per-compilation unit block data is about to get the axe. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Jordan Justen <[email protected]>
* glsl: Assert that interfaces, like structures, are not seen as leaf typesIan Romanick2013-01-251-0/+4
| | | | | | Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Add new uniform_field_visitor::process variantIan Romanick2013-01-252-0/+39
| | | | | | | | | | This flavor takes a type and a base name. It will be used to handle cases where the block name (instead of the instance name) is used for an interface block. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Recurse into uniform blocks just like uniform structuresIan Romanick2013-01-251-4/+20
| | | | | | | | v2: Inspite of the spell checker, spell recurse correctly. Suggested by Carl Worth. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Handle instance array declarationsIan Romanick2013-01-251-3/+19
| | | | | | | | | v2: Add a comment and an assertion about the array size in the non-instance name case. Suggested by Paul Berry. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Paul Berry <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Track blocks in the symbol table using the glsl_type instead of the ↵Ian Romanick2013-01-253-18/+9
| | | | | | | | | | | | | gl_uniform_block Eventually the gl_uniform_block information won't be calculated until linking. Block names need to be checked for name clashes during compiling, so we have to track it differently. v2: Update the commit message. Suggested by Carl Worth. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Add new uniform_field_visitor::visit_field variantIan Romanick2013-01-252-0/+17
| | | | | | Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Modify uniform_field_visitor::visit_field to take a row_major parameterIan Romanick2013-01-253-7/+19
| | | | | | | | | | Not used yet, but the UBO layout visitor will use this. v2: Remove a spruious hunk. This is moved to the patch "glsl: Remove ir_variable::uniform_block". Suggested by Carl Worth. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Modify uniform_field_visitor::recursion to take a row_major parameterIan Romanick2013-01-252-5/+19
| | | | | | | | | | | Not used yet, but the UBO layout visitor will use this. v2: Add some commentary as to why row_major is always set to false in process. Suggesed by Paul Berry. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Add a predicate to determine whether a variable is an interface blockIan Romanick2013-01-251-0/+27
| | | | | | | | | | | | | | | | | | | | | | | | | For the first declaration below, there will be an ir_variable named "instance" whose type and whose instance_type will be the same glsl_type. For the second declaration, there will be an ir_variable named "f" whose type is float and whose instance_type is B2. "instance" is an interface instance variable, but "f" is not. uniform B1 { float f; } instance; uniform B2 { float f; }; v2: Copy the comment message documentation into the code. Suggested by Paul Berry. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Paul Berry <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Require that indices into uniform block arrays be constantsIan Romanick2013-01-251-0/+9
| | | | | | Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Add ir_variable::interface_type fieldIan Romanick2013-01-253-1/+13
| | | | | | | | | | | | | | | For variables that are in an interface block or are an instance of an interface block, this is the GLSL_TYPE_INTERFACE type for that block. Convert the ir_variable::is_in_uniform_block method added in the previous commit to use this field instead of ir_variable::uniform_block. v2: Fix the place-holder comment on ir_variable::interface_type. Suggested by Paul Berry. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Add ir_variable::is_in_uniform_block predicateIan Romanick2013-01-258-8/+16
| | | | | | | | | The way a variable is tested for this property is about to change, and this makes the code easier to modify. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Generate an interface type for uniform blocksIan Romanick2013-01-251-49/+118
| | | | | | | | | | | | | | If the block has an instance name, add the instance name to the symbol table instead of the individual fields. Fixes the piglit test interface-name-access-without-interface-name.vert for real. v2: Update the comment before the assertion that interface block definitions won't generate instructions. Suggested by Paul Berry. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Paul Berry <[email protected]>
* glsl: Add GLSL_TYPE_INTERFACEIan Romanick2013-01-2511-9/+109
| | | | | | | | | | | | | | | | | | | | | Interfaces are structurally identical to structures from the compiler's point of view. They have some additional restrictions, and generally GPUs use different instructions to access them. Using a different base type should make this a bit easier. This commit also adds the glsl_type::interface_packing fields. For GLSL_TYPE_INTERFACE types, this will track the specified packing mode. It is analogous to gl_uniform_buffer::_Packing. v2: Add serveral missing GLSL_TYPE_INTERFACE cases in switch-statements. v3: Add information about glsl_type::interface_packing. Move row_major checking in glsl_type::record_key_compare from this patch to the previous patch. Both suggested by Paul Berry. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Paul Berry <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Add row_major field to glsl_struct_fieldIan Romanick2013-01-253-37/+42
| | | | | | | | | | | | | For now, this will always be false. In the near future, an "interface" type will be added that shares a lot of infrastructure with structures. v2: Move row_major checking in glsl_type::record_key_compare from the next patch to this patch. Suggested by Paul Berry. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Paul Berry <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Refactor out processing of structure fieldsIan Romanick2013-01-251-13/+41
| | | | | | | | | | | | | This will soon also be used for processing interface block fields. v2: Add a comment explaining the interface of ast_process_structure_or_interface_block. Suggested by Paul Berry. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Paul Berry <[email protected]> Reviewed-by: Chad Versace <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Parse interface array sizeIan Romanick2013-01-252-17/+59
| | | | | | | | | | | | | The size is parsed and stored in the AST, but it is not used yet. Processing of the array size is added in the patch "glsl: Handle instance array declarations" v2: Update the commit message (suggested by Carl Worth). Add a comment to ast_uniform_block::array_size (suggested by Paul Berry). Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Chad Versace <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Parse non-array uniform block instance names in GLSL ES 3.00.Kenneth Graunke2013-01-252-4/+30
| | | | | | | | | | | | | | | | | | | | | | | In GLSL ES 3.00 (and GLSL 1.50), uniform blocks can have an associated "instance name", which essentially namespaces the variables inside. This patch adds basic parsing for this new feature, but doesn't yet hook it up to actually do anything yet. It does not support for arrays of interface blocks; a later commit will take care of that. This change temporarily regresses the piglit test interface-name-access-without-interface-name.vert. This shader failed to compile before (the expected result), but it failed to compile for the wrong reason. This is not a real regression. v2: Add some comments to ast_uniform_block::instance_name. Suggested by Paul Berry. Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Chad Versace <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Refactor uniform block parser rules.Kenneth Graunke2013-01-251-21/+16
| | | | | | | | | | | | | The existing code has a lot of duplication; the only difference between the two cases is whether we merge in an additional layout qualifier. Apparently creating a layout_qualifieropt rule that can be empty causes a lot of conflicts and confusion. However, refactoring out the guts of the ast_uniform_block creation works fine. Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Chad Versace <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* linker: Refactor intra-stage block compatabililty testingIan Romanick2013-01-254-19/+73
| | | | | | | | | | | | | | | | Also slightly change the compatibility test. Instead of comparing the offsets of the block variables, compare the packing mode of the blocks. Ideally we don't want to assign the offsets until a later stage of linking. This is put in a new file called link_uniform_blocks.cpp. Some new functions related to uniform blocks are going to live in that file as well. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Chad Versace <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* mesa: Track the packing mode of a UBO in gl_uniform_bufferIan Romanick2013-01-252-0/+24
| | | | | | | | | This allows the next patch to verify that two uniform blocks match without first calculating the locations of the fields. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Chad Versace <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* glsl: Replace most default cases in switches on GLSL typeIan Romanick2013-01-2510-23/+48
| | | | | | | | | | | | | | | This makes it easier to find switch-statements that need to be updated after a new GLSL_TYPE_* is added because the compiler will generate a warning. Switch-statements that only had a small number of cases (e.g., everything in ir_constant_expression.cpp) were not modified. I may regret that decision when we eventually add support for doubles. Signed-off-by: Ian Romanick <[email protected]> Reviewed-by: Carl Worth <[email protected]> Reviewed-by: Chad Versace <[email protected]> Reviewed-by: Kenneth Graunke <[email protected]>
* i965: Correct gen6+ guardband calculation.Eric Anholt2013-01-252-9/+21
| | | | | | | | | | | Too much attention was paid to the first paragraphs, and not enough to the last little note that "oh, by the way, the rendered things themselves still have to be clipped to just 8192 wide/high". Fixes GTF's clip.c test with 4096 or higher width on ivb, where one of the triangles got the upper half of its pixels dropped. Tested-by: Ian Romanick <[email protected]>
* i965: Use GL_RED for DEPTH_TEXTURE_MODE in ES 3.0 for unsized formats.Kenneth Graunke2013-01-254-7/+21
| | | | | | | | | | | | | | | | | | | | | Khronos has apparently decided that depth textures with sized formats (allowed with ARB_internalformat_query or ES 3.0) should be treated as GL_RED, while unsized formats (an existing feature) should be treated as GL_INTENSITY for compatibility with ES 2.0. Ian is proposing changes to ARB_internalformat_query which will make this actually legal and consistent. A similar problem exists with GL 4.2, but we're going to ignore that for the time being. Tested on Ivybridge: no Piglit regressions; fixes 4 es3conform tests: - depth_texture_fbo - depth_texture_fbo_clear - depth_texture_teximage - depth_texture_texsubimage Reviewed-by: Ian Romanick <[email protected]>
* i965: Bump maximum supported ES2 context version to 3.0Chad Versace2013-01-251-1/+1
| | | | | | | | | | Since patch "i965: Validate requested GLES context version in brwCreateContext", we have been able to create ES 3.0 contexts due to the max version check. So...bump the max version. Reviewed-by: Kenneth Graunke <[email protected]> Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* i965/Gen6+: Enable ARB_ES3_compatibility extensionPaul Berry2013-01-251-0/+1
| | | | Reviewed-by: Ian Romanick <[email protected]>
* mesa/es3: Enable ES 3.0 API and shading language versionIan Romanick2013-01-252-2/+27
| | | | | | | | v2: Add ARB_internalformat_query to the list of required extensions. v3: Add OES_depth_texture_cube_map to the list of required extensions. Signed-off-by: Ian Romanick <[email protected]>
* scons: Add imports.c to builtin_compiler build.Vinson Lee2013-01-241-0/+2
| | | | | | | | Fixes build regression introduced by commit eac030e38e3cdd4ed4534516e3d3a50c8a372719. Signed-off-by: Vinson Lee <[email protected]> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=59835
* i965/fs/gen7: Fix fatal typo in unpackHalf2x16Chad Versace2013-01-241-1/+1
| | | | | | | | | s/src/src_w/ That little typo, which sneaked into v4 of the previous patch, generates incorrect fs code. Signed-off-by: Chad Versace <[email protected]>
* i965/fs/gen7: Emit code for GLSL 3.00 pack/unpack operations (v4)Chad Versace2013-01-245-3/+144
| | | | | | | | | | | | | v2: Remove lewd comment. [for idr] v3: - Optimize away tmp register for packHalf2x16. [for anholt, paul] - Improve comments. [for anholt, paul] - Reduce near-duplicate code by removing vec4_visitor emit_pack/unpack methods. [for chadv] v4: Factor our UD/W register conversion into helper function. [for anholt] Reviewed-by: Eric Anholt <[email protected]> Reviewed-by: Ian Romanick <[email protected]> (v2) Signed-off-by: Chad Versace <[email protected]>
* i965/vs/gen7: Emit code for GLSL ES 3.00 pack/unpack operations (v3)Chad Versace2013-01-243-0/+146
| | | | | | | | | | | | | | FIXME: This patch emits VS code that violates documented hardware restrictions and then relies on undocumented behavior that results from that violation. This patch passes all tests, but should be fixed ASAP to conform to the hardware documentation. v2: Explain undocumented hardware behavior. Improve comments. v3: Use ALU1 helper methods F32TO16() and F16TO32(). [for anholt] Reviewed-by: Eric Anholt <[email protected]> Reviewed-by: Ian Romanick <[email protected]> (v1) Signed-off-by: Chad Versace <[email protected]>
* i965: Quote the PRM on a HorzStride subtletyChad Versace2013-01-241-1/+4
| | | | | Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* i965: Add opcodes for F32TO16 and F16TO32Chad Versace2013-01-244-0/+8
| | | | | | | | | | | | | The GLSL ES 3.00 operations packHalf2x16 and unpackHalf2x16 will emit these opcodes. - Define the opcodes BRW_OPCODE_{F32TO16,F16TO32}. - Add the opcodes to the brw_disasm table. - Define convenience functions brw_{F32TO16,F16TO32}. Reviewed-by: Ian Romanick <[email protected]> Acked-by: Paul Berry <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* i965: Lower the GLSL ES 3.00 pack/unpack operations (v2)Chad Versace2013-01-241-0/+32
| | | | | | | | | | | | | | | | | On gen < 7, we fully lower all operations to arithmetic and bitwise operations. On gen >= 7, we fully lower the Snorm2x16 and Unorm2x16 operations, and partially lower the Half2x16 operations. v2: - Comment that scalarization is needed only for SOA code [for idr]. - Replace switch-statement with if-statement [for idr]. - Remove misplaced hunk from previous patch [found by idr]. Reviewed-by: Ian Romanick <[email protected]> Reviewed-by: Matt Tuner <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* glsl: Add lowering pass for GLSL ES 3.00 pack/unpack operations (v4)Chad Versace2013-01-243-0/+1056
| | | | | | | | | | | | | | | | Lower them to arithmetic and bit manipulation expressions. v2: Rewrite using ir_builder [for idr]. v3: Comment typos. [for mattst88] v4: Fix arithmetic error in comments. Factor out a shift instruction. Don't heap allocate factory.instructions. [for paul] Reviewed-by: Ian Romanick <[email protected]> (v2) Reviewed-by: Matt Tuner <[email protected]> (v3) Reviewed-by: Paul Berry <[email protected]> (v4) Signed-off-by: Chad Versace <[email protected]>
* glsl: Fix type-deduction for and/or/xor expressionsChad Versace2013-01-241-0/+5
| | | | | | | | | In ir_expression's constructor, the cases for {bit,logic}_{and,or,xor} failed to handle the case when both operands were vectors. Note: This is a candidate for the stable branches. Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* glsl: Reformat and/or/xor cases in ir_expression ctorChad Versace2013-01-241-2/+2
| | | | | | | | | | | Replace tabs with spaces. According to docs/devinfo.html, Mesa's indetation style is: indent -br -i3 -npcs --no-tabs infile.c -o outfile.c This patch prevents whitespace weirdness in the next patch. Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* glsl/ir_builder: Add helpers for making if-statementsChad Versace2013-01-242-0/+35
| | | | | | | | | | | Add two overloaded variants of ir_if *if_tree() The new functions allow one to chain together if-trees within a single C++ expression that resembles a real if-statement. Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* glsl/ir_builder: Add `enum writemask`Chad Versace2013-01-241-0/+9
| | | | | | | | Using this enum improves the readibility of calls to assign(), whose third argument is a writemask. Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* glsl/ir_factory: Add helper method for making an ir_constantChad Versace2013-01-241-0/+24
| | | | | | | | Add method ir_factory::constant. This little method constructs an ir_constant using the factory's mem_ctx. Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* glsl/ir_builder: Add more helpers for constructing expressionsChad Versace2013-01-242-0/+156
| | | | | | | | | | | | | | | | | Add the following functions, each of which construct the similarly named ir expression: div, round_even, clamp equal, less, greater, lequal, gequal logic_not, logic_and, logic_or bit_not, bit_or, bit_and, lshift, rshift f2i, i2f, f2u, u2f, i2u, u2i Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* glsl/ir_factory: Initialize members to NULL in constructorChad Versace2013-01-241-0/+7
| | | | | | | This eliminates unexpected behavior due to unitialized values. Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* glsl: Evaluate constant GLSL ES 3.00 pack/unpack expressions (v3)Chad Versace2013-01-241-0/+189
| | | | | | | | | | | | | | | | | | That is, evaluate constant expressions of the following functions: packSnorm2x16 unpackSnorm2x16 packUnorm2x16 unpackUnorm2x16 packHalf2x16 unpackHalf2x16 v2: Reuse _mesa_pack_float_to_half and its inverse to evaluate pack/unpackHalf2x16. [for idr] v3: Whitespace fixes. [for mattst88] Don't cast neg floats directly to uint16; use an intermediate cast to int16. [for paul] Reviewed-by: Ian Romanick <[email protected]> (v2) Reviewed-by: Paul Berry <[email protected]> Reviewed-by: Matt Tuner <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* mesa: Remove rounding bias in _mesa_float_to_half()Chad Versace2013-01-241-29/+37
| | | | | | | | | | | | | | | | | | | | | | | Not all float32 values can be exactly represented as a float16. _mesa_float_to_half() rounded such intermediate float32 values to zero by truncating unrepresentable bits in the mantissa. This patch improves _mesa_float_to_half() by rounding intermediate float32 values to the nearest float16; when the float32 is exactly between two float16 values we round to the one with an even mantissa. This behavior is preferred over the old behavior because: - It has reduced bias relative to the old behavior. - It reproduces the behavior of real hardware: opcode F32TO16 in Intel's GPU ISA. - By reproducing the behavior of the GPU (at least on Intel hardware), compile-time evaluation of constant packHalf2x16 GLSL expressions will result in the same value as if the expression were executed on the GPU. Reviewed-by: Ian Romanick <[email protected]> Reviewed-by: Paul Berry <[email protected]> Signed-off-by: Chad Versace <[email protected]>
* mesa,glsl: Move round_to_even() from glsl to mesa/main (v2)Chad Versace2013-01-246-22/+35
| | | | | | | | | | | | | Move round_to_even's definition to mesa/main so that _mesa_float_to_half() can use it in order to eliminate rounding bias. In additon to moving the fuction definition, prefix its name with "_mesa", just as all other functions in mesa/main are prefixed. v2: Fix Android build. Reviewed-by: Ian Romanick <[email protected]> Signed-off-by: Chad Versace <[email protected]>
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-27 08:14:51 +0000