kc3-lang/angle/src/compiler

src/compiler

Log

Author Commit Date CI Message
Olli Etuaho 693c9aa0 2015-04-07T17:50:36 Replace setArray function in TPublicType with functions from TType Having only one way of setting or clearing arrayness of types clarifies code dealing with arrays. TEST=angle_unittests BUG=angleproject:941 Change-Id: I98cb7c44fd66440c9de8b4c6c4a02827e9300db7 Reviewed-on: https://chromium-review.googlesource.com/264361 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 93a90fd0 2015-04-07T18:14:07 Remove half-baked non-sized array support ANGLE used to accept shaders with non-sized arrays in initializer lists. Fix this. Proper support for implicitly sized arrays will be implemented later for ESSL3. TEST=angle_unittests BUG=angleproject:941 Change-Id: I53c5ccf0f7ef09ad30e142f8350812959fc4846c Reviewed-on: https://chromium-review.googlesource.com/264360 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Olli Etuaho 9638c352 2015-04-01T14:34:52 Support assigning array constructors in HLSL output This adds support to the following type of shader code, where an array constructor is used as a source of assignment: float a[3]; a = float[3](0.0, 1.0, 2.0); The assignment gets replaced in the HLSL code by a function call where the array is an out parameter and the constructor parameters are regular parameters. Constructors cannot yet be used in initializers. With this change in place, some of the relevant dEQP tests start passing. TEST=dEQP-GLES3.functional.shaders.arrays.constructor.* BUG=angleproject:941 Change-Id: I13ed603c02a30d9f9950a61c4d562ae9fac058ec Reviewed-on: https://chromium-review.googlesource.com/263403 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho f40319e9 2015-03-10T14:33:00 Add basic support for array constructors Add limited support for parsing array constructors and writing them out as GLSL code. Still missing from this version: HLSL output, array support in initializer lists, arrays with implicit size. BUG=angleproject:941 Change-Id: I7febf80923c4cd0b730399f1f49f9456cf3668e9 Reviewed-on: https://chromium-review.googlesource.com/260572 Reviewed-by: Geoff Lang <geofflang@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Corentin Wallez 711d8663 2015-04-02T19:43:57 Revert "Implement gradient operation AST analysis" This reverts commit 2fc57a2ace58b7ae3106344c56be94948f761b2a. Change-Id: I619297090c8441c1b90099437f8764263cdd68cc Reviewed-on: https://chromium-review.googlesource.com/263728 Reviewed-by: Corentin Wallez <cwallez@chromium.org> Tested-by: Corentin Wallez <cwallez@chromium.org>
Corentin Wallez c457e7bb 2015-04-02T19:43:52 Revert "Implement discontinuous loops AST analysis" This reverts commit f30a9ba96bdfa014fc182060c714a2c75451f4cf. Change-Id: I577cfd1e7d5d6138413c29205cbf4d35b6861226 Reviewed-on: https://chromium-review.googlesource.com/263754 Reviewed-by: Corentin Wallez <cwallez@chromium.org> Tested-by: Corentin Wallez <cwallez@chromium.org>
Geoff Lang 186160b2 2015-04-02T19:34:38 Revert "Implemented a CallDAG to allow for more AST analysis" failures in angle_unittests and compilation failure on clang (see: https://build.chromium.org/p/chromium.gpu.fyi/builders/GPU%20Mac%20Builder%20%28dbg%29/builds/30016/steps/compile/logs/stdio) . This reverts commit b34d1d12969496b1b0fb53934b8ce3200304f900. Change-Id: Ia995fb2db0e891294f3461de01617cb13e5ae381 Reviewed-on: https://chromium-review.googlesource.com/263727 Reviewed-by: Geoff Lang <geofflang@chromium.org> Tested-by: Geoff Lang <geofflang@chromium.org>
Corentin Wallez f30a9ba9 2015-03-19T10:39:13 Implement discontinuous loops AST analysis This will allow narrowing down which usages of [[flatten]] and [[unroll]] are actually useful. BUG=angleproject:937 BUG=395048 Change-Id: Icc37abf51a1407c9e7bc708dc08d3ba4050a496a Reviewed-on: https://chromium-review.googlesource.com/261189 Tested-by: Corentin Wallez <cwallez@chromium.org> Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Corentin Wallez 2fc57a2a 2015-03-18T12:55:45 Implement gradient operation AST analysis This will allow narrowing down which usages of [[flatten]] and [[unroll]] are actually useful. BUG=angleproject:937 BUG=395048 Change-Id: Ic31e434c7b5bb6c32eff2336e5949391bfee25a5 Reviewed-on: https://chromium-review.googlesource.com/260930 Reviewed-by: Nicolas Capens <capn@chromium.org> Tested-by: Corentin Wallez <cwallez@chromium.org> Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Corentin Wallez b34d1d12 2015-02-11T11:15:24 Implemented a CallDAG to allow for more AST analysis The CallDAG preprocesses the AST to construct a DAG of functions that can be used for several analysis. Use it to implement check for recursion and max call depth. It will also be used to limit the usage of [[flatten]] and [[unroll]]. BUG=angleproject:937 BUG=395048 Change-Id: I643e0ed605ad095e4fd7da4977d842be5e6a12e0 Reviewed-on: https://chromium-review.googlesource.com/229352 Reviewed-by: Nicolas Capens <capn@chromium.org> Tested-by: Corentin Wallez <cwallez@chromium.org> Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Olli Etuaho b2d6a9be 2015-03-30T16:00:53 Stub simplifying array assignment for HLSL output Add an AST traverser that rejects shaders with expressions that use the return value of an assignment where an array is assigned to another array. In the future this should be made into a tool that can simplify these expressions so that return values of array assignments are not used. In its current form this code prevents OutputHLSL from producing garbage code when the original code contains expressions like a = (b = c); where a, b, and c are all arrays. BUG=angleproject:960 Change-Id: I11353d7ed7160c853e58a0ef3471ca439cb314c8 Reviewed-on: https://chromium-review.googlesource.com/263070 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Olli Etuaho 1269076c 2015-03-31T12:55:28 Add basic support for assigning arrays in HLSL output Implement support for assignments where the return value of the assignment is not used in another part of the expression. TEST=WebGL conformance tests BUG=angleproject:960 Change-Id: Ibf9d71a75d27d139d2aabb5162ab04a0974321d3 Reviewed-on: https://chromium-review.googlesource.com/263222 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Olli Etuaho f6c694bc 2015-03-26T14:50:53 Assign built-in function return type in promote() This finishes the refactoring of unary math operation handling so that IntermUnary::promote has the complete code for setting the return type of the node. BUG=angleproject:952 TEST=angle_unittests, WebGL conformance tests Change-Id: I19bd8d53029e24f734c9436eceb446b37e7fcf26 Reviewed-on: https://chromium-review.googlesource.com/262416 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho dca3e796 2015-03-26T13:24:04 Refactor unary math operator handling to clarify responsibilities Shuffle the code around so that each part has a clear responsibility: IntermUnary::promote is responsible for setting the return type of the node, Intermediate::addUnaryMath is responsible for creating the node object, and ParseContext::createUnaryMath is responsible for validating the operand type. This removes duplicated bool type check for logical not. BUG=angleproject:952 TEST=angle_unittests, WebGL conformance tests Change-Id: I9f5a0abb6434ad2730441ea9199ec3f5382ebcda Reviewed-on: https://chromium-review.googlesource.com/262415 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 5321c808 2015-04-02T17:08:16 Add unsigned literal support to GLSL output BUG=angleproject:964 Change-Id: I8eb3cfd8d49beded4ac55ead97b50c7691e4980a Reviewed-on: https://chromium-review.googlesource.com/263710 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill f0d10f89 2015-03-31T12:56:52 Replace non-copyable macro with a helper class. This class provides a simpler scheme for blocking default copy and assignment operators. It also reduces the amount of code needed since it's inherited to child classes. This also fixes the conflict between our macro and the same-named macro in Chromium code. BUG=angleproject:956 Change-Id: If0dc72aa3f63fbc7b8fa34907418821c64c39e2f Reviewed-on: https://chromium-review.googlesource.com/263257 Tested-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Brandon Jones <bajones@chromium.org>
Olli Etuaho fc7cfd1c 2015-03-31T14:46:18 Fix array function parameter syntax in HLSL equality function The fact that passing shaders to the platform compiler is deferred until linking is performed allowed this bug to pass undetected in earlier manual testing. TEST=WebGL conformance tests BUG=angleproject:941 Change-Id: Ic12b5d0c5547057551df90086453e897a161d559 Reviewed-on: https://chromium-review.googlesource.com/263221 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Nicolas Capens <capn@chromium.org>
Olli Etuaho 69c11b5d 2015-03-26T12:59:00 Move validation from Intermediate::addUnaryMath to ParseContext Intermediate should only have logic for creating node objects, validation of parameter types belongs in ParseContext. BUG=angleproject:952 TEST=angle_unittests, WebGL conformance tests Change-Id: Ie90697641fabb2a837ccc4571a93616d63ea64e6 Reviewed-on: https://chromium-review.googlesource.com/262414 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Nicolas Capens <capn@chromium.org>
Olli Etuaho 5e5c826c 2015-03-26T14:04:54 Fix float-int conversion return type precision tracking Float-int conversion functions are defined to always return highp values. BUG=angleproject:865 TEST=angle_unittests Change-Id: Idf243b483f7b5edfcb54de2755af11e17c3756d3 Reviewed-on: https://chromium-review.googlesource.com/262413 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Nicolas Capens <capn@chromium.org>
Olli Etuaho ff699005 2015-03-23T14:38:42 Disallow operations on structs containing samplers ESSL 1.00 spec section 5.9 says that equality operators don't operate on structs containing sampler types. Section 5.7 also suggests this. ESSL 3.00 doesn't have a similar restriction. ESSL 1.00 spec section 4.1.7 says that structs containing samplers can't be used as l-values. This is interpreted to apply also in the case of ESSL 3.00, which similarly disallows samplers as l-values, but doesn't explicitly mention structs. BUG=angleproject:954 TEST=angle_unittests, WebGL conformance tests Change-Id: I73f74962a192e8d9449990ffa5f3d8c851491601 Reviewed-on: https://chromium-review.googlesource.com/261822 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Olli Etuaho ae37a5c8 2015-03-20T16:50:15 Support equality between structs containing arrays in HLSL output This requires sorting all equality functions together so that struct equality functions can have dependencies on array equality functions, but support for array equality functions that have dependencies on struct equality functions is also maintained. There's no automated test coverage for this specifically. The change was tested by manually inspecting shader output and ensuring that there were no test regressions in tests listed below. TEST=dEQP-GLES3.functional.shaders.* BUG=angleproject:954 Change-Id: If7199ab2446804afae50f103bb625101172882b9 Reviewed-on: https://chromium-review.googlesource.com/261550 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 7fb4955d 2015-03-18T17:27:44 Support array equality operator in HLSL output This requires adding functions to the shader source that can do the comparison for a specific array size. There's no automated test coverage specifically for this functionality, since all deqp tests that cover this also require array constructors to be supported. The change has been tested by manually inspecting shader output. No regressions were seen in automated tests listed below. TEST=dEQP-GLES3.functional.shaders.*, angle_unittests BUG=angleproject:941 Change-Id: Ie2ca7c016a3f0bcb3392a96d6d20d6f803d28bf0 Reviewed-on: https://chromium-review.googlesource.com/261530 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Nicolas Capens <capn@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 9dd217bc 2015-03-20T14:24:31 Disallow operations on structures containing arrays in ESSL1 Comparing structures that contain arrays to each other and assigning structures that contain arrays is "not defined" in ESSL 1.00 (section 5.7). Sections 5.8 and 5.9 further suggest that these operations are not allowed. Additionally some platform drivers on Linux seem to reject shaders produced by ANGLE which compare structures containing arrays. This might require changing the output GLSL version for ESSL 3.00. BUG=angleproject:954 TEST=angle_unittests Change-Id: I5f3a016f360f940f2fc1ec1ff8e60d13a977eb69 Reviewed-on: https://chromium-review.googlesource.com/261531 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 47fd36a7 2015-03-19T14:22:24 Move some validation from IntermBinary::promote to ParseContext This makes the role of promote() in the system clearer and helps to make the code more understandable, since more of the checks are in the same logical place. BUG=angleproject:952 TEST=angle_unittests, WebGL conformance tests Change-Id: Idb2de927d872e46210d71cf6de06a6f8c1fc5da1 Reviewed-on: https://chromium-review.googlesource.com/260803 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Austin Kinross c8ef69d2 2015-03-18T16:43:22 Fix C4702 issues (unreachable code) in ANGLE, excluding <xtree> Change-Id: Ia7603139af266fd7b14efc8c3465225738456e67 Reviewed-on: https://chromium-review.googlesource.com/261038 Tested-by: Austin Kinross <aukinros@microsoft.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho e79904c3 2015-03-18T16:56:42 Accept equality and assignment for arrays in parsing This is enough to support the operations in GLSL output. HLSL output will likely require additional work to support this. BUG=angleproject:941 Change-Id: I728d511ab07af94bc3382dc2796c1e9ac79d1442 Reviewed-on: https://chromium-review.googlesource.com/260801 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 1ade24db 2015-03-19T16:29:32 Add #version 300 es to ESSL output when needed BUG=angleproject:953 Change-Id: Ib9e45fab1d35f8c78abd0ac2209ac4f63afe4465 Reviewed-on: https://chromium-review.googlesource.com/261170 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Olli Etuaho d6b14287 2015-03-17T14:31:35 Move binary op array check to ParseContext Also piece together an addAssign function in ParseContext that uses the binary op array check. This will make it easier to change the array-related checks in the future to use shaderVersion. Moving validation out from IntermBinary::promote also makes the architecture clearer, promote()'s role should be mainly to determine the type of the return value of the binary operation, not to do validation. BUG=angleproject:941 TEST=angle_unittests, WebGL conformance tests Change-Id: If1de33ea250893527be7f0d7930d4678a0864684 Reviewed-on: https://chromium-review.googlesource.com/260571 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho b2983c95 2015-03-18T14:02:46 Correctly insert unmangled function names to symbol table This fixes detecting some cases of function parameter mismatch (previously a mangled function name corresponding to no function parameters was added to the symbol table for each user-defined function, and this was returned when doing function lookups with no parameters). Also fixes detection of reusing a function name as a variable/struct name. New unit tests are added to ensure that these fixes don't regress. BUG=angleproject:936 TEST=angle_unittests, WebGL conformance tests Change-Id: I2dadde9dcc01c7a4a653c1982c36377b89e6d437 Reviewed-on: https://chromium-review.googlesource.com/260800 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho fc1806e1 2015-03-17T13:03:11 Move most of addBinaryMath from Intermediate to ParseContext Some type checks for binary math will be different based on the shading language version, which is easily accessible in ParseContext. Because of this and also for architectural simplicity it makes more sense to have the checks in ParseContext. BUG=angle:941 TEST=angle_unittests, WebGL conformance tests Change-Id: I92a499f47e1cbc6a7b6391ce0fa04284803e7140 Reviewed-on: https://chromium-review.googlesource.com/260570 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Arun Patole 6341939f 2015-03-13T11:51:07 Fix code duplication in OutputHLSL While debugging, I came across the piece of code in OutputHLSL::OutputHLSL which is unnecessarily duplicated, this change just fixes it. Change-Id: I0549d58696417ccf651bacfa19f489b290d4ba6e Reviewed-on: https://chromium-review.googlesource.com/259960 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 2728eda4 2015-03-16T10:23:50 Revert "A special state to track field selection is not necessary." This reverts commit f8dc4fb66d2ee473b6ba37626f02756422863f6c. The special state to track field selection is actually necessary to avoid reduce/reduce conflicts when array constructors are added to the grammar. BUG=angleproject:941 TEST=WebGL conformance tests, angle_unittests Change-Id: I55476483c9e83241e8978cd58f05ef303c7c8680 Reviewed-on: https://chromium-review.googlesource.com/260260 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Nicolas Capens <capn@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 4f52da4a 2015-03-11T10:48:41 Remove no-op functionality from IdentifyBuiltIns and refactor Adding array built-ins like gl_FragData and gl_LastFragData to the symbol table with a name that included their size in brackets was incorrect and unnecessary. Remove this. The array built-ins should only be added with their proper name and with an array type. Also refactor the code to reduce unnecessary repetition of conditions. The previous version attempted to split the built-ins that had dependencies on resources into their own part of the code, but this split was not clearly defined and only made the code harder to follow and prone to issues. TEST=WebGL conformance tests Change-Id: I7a77865a594864a22f3b566f9d3da1d0ead46466 Reviewed-on: https://chromium-review.googlesource.com/258751 Reviewed-by: Nicolas Capens <capn@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 61ffec88 2015-03-11T10:35:01 Fix formatting of IdentifyBuiltIns Fix indentation of case labels, fix too long lines, put braces on their own lines and remove extraneous spacing and comments that didn't contribute to the readability of the code. TEST=WebGL conformance tests Change-Id: Ie33d0a7c4412c448081f5d879fa9f64baaf8d537 Reviewed-on: https://chromium-review.googlesource.com/258750 Reviewed-by: Nicolas Capens <capn@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Arun Patole 44efa0b8 2015-03-04T17:11:05 Conditionally enable IEEE strictness for isnan() ANGLE's ESSL3 built-in function isnan() implementation just uses HLSL intrinsic function isnan(). For HLSL intrinsic function isnan() to work properly, IEEE strictness needs to be enabled for D3D compiler. This change detects use of isnan() in shaders and passes compiler flag D3DCOMPILE_IEEE_STRICTNESS whenever isnan is used in shaders. This change also moves existing workarounds in D3DWorkaroundType to D3DCompilerWorkarounds. BUG=angle:927 TEST= dEQP tests dEQP-GLES3.functional.shaders.builtin_functions.common.isnan.* Change-Id: I1ce5b1a7a825fdd720a37dc9aeb71320e55162d9 Reviewed-on: https://chromium-review.googlesource.com/255834 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Jamie Madill <jmadill@chromium.org>
Nico Weber 092e1a12 2015-03-09T07:40:11 Initialize fragmentPrecisionHigh in constructor. This was the only field not initialized by the constructor, which looks like an oversight. Change-Id: If61424d574c663113a4c6d31e94b6ec13b412884 Reviewed-on: https://chromium-review.googlesource.com/257360 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Geoff Lang <geofflang@chromium.org>
Olli Etuaho fc7fab72 2015-03-06T12:03:18 Fix matrix compound multiplication Matrix compound multiplication was lacking transposes. Add them so that results match normal multiplication. BUG=angleproject:938 TEST=dEQP-GLES3.functional.shaders.matrix.mul_assign.* Change-Id: I4c5159de0727afb4ee2831586f04fa817619de7a Reviewed-on: https://chromium-review.googlesource.com/256366 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Olli Etuaho 51585037 2015-03-05T19:40:42 Remove unnecessary ifdef APPLE from built-in emulation The emulation is already toggled by a compilation flag, so having it behind an ifdef only adds unnecessary complexity, particularly when testing changes to the emulation code. BUG=angleproject:865 Change-Id: Idf7854c2b5323609880b527ed806440dcc8ac091 Reviewed-on: https://chromium-review.googlesource.com/256365 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Olli Etuaho 8efc5ad5 2015-03-03T17:21:10 Initialize BuiltInFunctionEmulator outside Compiler This moves GLSL output specific code from the Compiler class to the GLSL/ESSL translators. BUG=angleproject:865 Change-Id: I2d552e9cdb41f7d8ddfee7b0249a99d629a6d7d7 Reviewed-on: https://chromium-review.googlesource.com/255471 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 72ba85bf 2015-03-04T14:23:26 Fix regression in addFunctionCallOrMethod Fix a crash when parsing a shader with incorrect constructor arguments. BUG=angleproject:911 TEST=WebGL conformance tests Change-Id: I3459612f36185db861a361fa916cc6b671795037 Reviewed-on: https://chromium-review.googlesource.com/255870 Reviewed-by: Zhenyao Mo <zmo@chromium.org> Tested-by: Zhenyao Mo <zmo@chromium.org>
Olli Etuaho 95cd3c68 2015-03-03T16:45:32 Move some output out of BuiltInFunctionEmulator subclasses This paves the way for getting rid of the BuiltInFunctionEmulator subclasses in favor of initializing the BuiltInFunctionEmulator dynamically in Compiler subclasses. The eventual goal is getting rid of GLSL output specific functionality in Compiler, which should be language-agnostic. BUG=angle:865 Change-Id: Ibb114b905785c7343b2726c97699268c982536a0 Reviewed-on: https://chromium-review.googlesource.com/255470 Reviewed-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho c4ba3bef 2015-03-02T14:42:24 Refactor addFunctionCallOrMethod out of glslang.y TEST=angle_unittests BUG=angle:911 Change-Id: I6d4c0f0bfbf9e6252e241ae5bb1b6acdb463fce6 Reviewed-on: https://chromium-review.googlesource.com/255451 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Qingqing Deng 5dbece59 2015-02-27T20:35:38 Implement trunc, round, roundEven (Emulated roundEven) BUG=angle:934 Change-Id: If33256bf3707f3f0202147276c71f08c65518205 Reviewed-on: https://chromium-review.googlesource.com/254842 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 2cd7a0ed 2015-02-27T13:57:32 Remove switch fall-through on HLSL Remove fall-through from non-empty labels in switch statements. Tested with dEQP-GLES3.functional.shaders.*switch*. All pass except for tests that require dFdx/dFdy, which fail for an unrelated reason. Additional test coverage could still be useful for cases where a label is terminated by a continue or return statement. BUG=angle:921 Change-Id: I4741867789a9308d66d0adeabdaf83907106e2d2 Reviewed-on: https://chromium-review.googlesource.com/254550 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Olli Etuaho 687ef22c 2015-03-03T10:58:05 Fix derivatives being related to an extension on ESSL3 dFdx, dFdy and fwidth are parts of the core language in ESSL3. This fixes a regression from the patch "Set symbol extension at insertion". BUG=angle:926 Change-Id: I6c0bbce3afad1f064ff115a16e23ea79bd584e49 Reviewed-on: https://chromium-review.googlesource.com/255450 Reviewed-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Zhenyao Mo 05b6b7fc 2015-03-02T17:08:09 Add an SH_GLSL_CORE_OUTPUT profile. So we could generate shaders for Apple using core GL profile. By switching to core profile, we still pass most WebGL conformance tests 1.0.2 on Linux, but not all, so apparently more work is needed. However, I think it's OK to check this CL in because this output profile will be only used behind a chromium switch. BUG=angleproject:933 TEST=webgl conformance tests Change-Id: Iad70e1aebf82349d3fc5f4116c1d6bc4448193fd Reviewed-on: https://chromium-review.googlesource.com/255282 Tested-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Arun Patole 7749005a 2015-03-03T18:36:05 Fix min built-in function support One of the recent change that uses generic types to compact the symbol table initialization has introduced a bug in 'min' function implementation, just fixing it here. TEST= dEQP tests: dEQP-GLES3.functional.shaders.operator.common_functions.min.lowp_uvec* Change-Id: I0e3d386c89bc07c3b656ec5029b8ce57f663cf46 Reviewed-on: https://chromium-review.googlesource.com/255369 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Nicolas Capens c9d9b30a 2015-02-20T23:02:15 Set symbol extension at insertion. BUG=angle:926 Change-Id: I4c96ae89c79d720d99bde803541c3158ba4629ab Reviewed-on: https://chromium-review.googlesource.com/251662 Tested-by: Nicolas Capens <capn@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 05ae50dc 2015-02-20T10:16:47 Add basic support for HLSL output of switch and case This patch outputs switch and case statements in HLSL, but ignores fall-through. Switch statements that include fall-through cause the platform HLSL compiler to generate errors. BUG=angle:921 Change-Id: I10efbfb45076cf4fd79840698abafd8111e597d4 Reviewed-on: https://chromium-review.googlesource.com/251527 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 01cd8afa 2015-02-20T10:39:20 Add GLSL output of switch and case BUG=angle:921 Change-Id: I0d752440ce6ffdfcc005f1a6123694bdfeb1b067 Reviewed-on: https://chromium-review.googlesource.com/251526 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho c8716df9 2015-02-26T17:17:22 Fix an issue with handling an error case inside ValidateSwitch In case a condition with a wrong type is given to a case statement, it generates an error but the compiler recovers. This caused ValidateSwitch to assert. Fix this. BUG=angle:921 Change-Id: I7949798cab923c2b168817471896470c6c611878 Reviewed-on: https://chromium-review.googlesource.com/254080 Reviewed-by: Zhenyao Mo <zmo@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Nicolas Capens 482907eb 2015-02-23T16:56:33 Set symbol operations at insertion. BUG=angle:926 Change-Id: Ifc8f62ebe1277214a4d3fbd5c56eb5faeff8e389 Reviewed-on: https://chromium-review.googlesource.com/251650 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Nicolas Capens <capn@chromium.org> Reviewed-by: Olli Etuaho <oetuaho@nvidia.com>
Jamie Madill e9cc469f 2015-02-19T16:00:13 Accept mismatching auxiliary interpolations. The ES3.1 spec, and discussion on Khronos.org, confirm that dEQP is correct in accepting mismatching centroid specifiers in shader linkage. Mismatching flat/smooth is still a link error. Fixes: shaders.linkage.varying.rules.differing_interpolation_2 Change-Id: I3016f4147e7c1b16b02371ee95866c8daf826212 Reviewed-on: https://chromium-review.googlesource.com/251205 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org> Tested-by: Jamie Madill <jmadill@chromium.org>
Gregoire Payen de La Garanderie 5cc9ac85 2015-02-20T11:27:56 Fix shadow samplers with textureProj and vertex shaders. Fixes: dEQP-GLES3.functional.shaders.texture_functions.textureproj.sampler2dshadow_fragment dEQP-GLES3.functional.shaders.texture_functions.textureproj.sampler2dshadow_bias_fragment dEQP-GLES3.functional.shaders.texture_functions.textureprojoffset.sampler2dshadow_fragment dEQP-GLES3.functional.shaders.texture_functions.textureprojoffset.sampler2dshadow_bias_fragment and some of: dEQP-GLES3.functional.shaders.texture_functions.texture*.sampler2dshadow_vertex BUG=angle:925 Change-Id: I080e379ded95469f0486ae9c8bb1756842118a2d Reviewed-on: https://chromium-review.googlesource.com/251530 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Nicolas Capens <capn@chromium.org> Tested-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho ac5274df 2015-02-20T10:19:08 Add validation for the structure of switch statements Implement a traverser to check for the following errors: -More than one default or replicated constant expression -No statement between a label and the end of a switch statement -Statements in a switch statement before the first case statement -Mismatch between the type of init-expression and type of a case label -Case or default label nested inside other control flow nested within the corresponding switch Tested by manually disabling shading language version checks for switch and by manually enabling case statements in a Chromium build and checking that the expected errors are generated. BUG=angle:921 Change-Id: I99c49c17c8b520849adbe4d8521e46cb10e20e41 Reviewed-on: https://chromium-review.googlesource.com/251524 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 3c1dfb5a 2015-02-20T11:34:03 Add stubs for switch and case output BUG=angle:921 Change-Id: I58bd645a8d53ef5bad9b680e54c8948d50932fca Reviewed-on: https://chromium-review.googlesource.com/251525 Reviewed-by: Zhenyao Mo <zmo@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Daniel Bratell 73941deb 2015-02-25T14:34:49 Disable the HLSL code in Angle if it is not being used. We're only using the HLSL code in Windows so it's not necessary to compile and distribute it on other platforms. This adds a defined ANGLE_ENABLE_HLSL that can be checked in files that are used by non-HLSL code as well. Mostly the HLSL code is just not include by the build system. Details of the space savings (heavily truncated) Total change: -165717 bytes =========================== -606 - Source: angle/src/common/utilities.cpp -627 - Source: angle/src/compiler/translator/FlagStd140Structs.cpp -695 - Source: /usr/include/c++/4.8/bits/stl_algo.h -710 - Source: angle/src/compiler/translator/TranslatorHLSL.cpp -713 - Source: angle/src/compiler/translator/IntermNode.h -863 - Source: /usr/include/c++/4.8/bits/stl_map.h -935 - Source: angle/src/compiler/translator/blocklayout.cpp -1515 - Source: angle/src/compiler/translator/BuiltInFunctionEmulator.cpp -1655 - Source: angle/src/compiler/translator/UnfoldShortCircuit.cpp -2375 - Source: /usr/include/c++/4.8/bits/vector.tcc -3135 - Source: angle/src/compiler/translator/RewriteElseBlocks.cpp -4656 - Source: angle/src/compiler/translator/UtilsHLSL.cpp -5265 - Source: angle/src/compiler/translator/BuiltInFunctionEmulatorHLSL.cpp -6505 - Source: /usr/include/c++/4.8/bits/stl_tree.h -11480 - Source: angle/src/compiler/translator/UniformHLSL.cpp -13580 - Source: angle/src/compiler/translator/StructureHLSL.cpp -18964 - Source: ?? (constant strings and a few vtbls) -89332 - Source: angle/src/compiler/translator/OutputHLSL.cpp Change-Id: I23ccc98abd0a21f847dd34f9482800b3ba679d56 Reviewed-on: https://chromium-review.googlesource.com/251528 Tested-by: bratell at Opera <bratell@opera.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho 53f076fa 2015-02-20T10:55:14 Add basic error checks for switch and case Check that the case labels have constant scalar integers, and that switch statements are initialized with scalar integers. Also check that case and default labels do not exist outside switch statements. Allow break statements inside switch statements. Tested by manually disabling shading language version checks for switch in a Chromium build and checking that the expected errors are generated. BUG=angle:921 Change-Id: Ibe83d0db52958c493ded5640d4babf670dc02d55 Reviewed-on: https://chromium-review.googlesource.com/251523 Reviewed-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho a3a36664 2015-02-17T13:46:51 Implement parsing switch statements Put in some groundwork for parsing switch statements and case labels in the parser, including definitions for IntermNode classes. Intermediate functions for adding the statements are stubbed to only generate errors for now. Tested by manually disabling shading language version checks for switch in a Chromium build and checking that the expected errors are generated. BUG=angle:921 Change-Id: I064b3e0c4c1b724a083cf5bc78eebfdd3794eb1b Reviewed-on: https://chromium-review.googlesource.com/250380 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 49300865 2015-02-20T14:54:49 Refactor validating jump statements out of glslang.y Move the logic to ParseContext. This will simplify changing the logic for the break statement that is needed to implement switch. BUG=angle:921, angle:911 TEST=WebGL conformance tests Change-Id: Ib8bbc7075c05d345e90377202b5446fb3307f94c Reviewed-on: https://chromium-review.googlesource.com/251522 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Nicolas Capens f3cc4aef 2015-02-23T13:51:25 Use generic types to compact the symbol table initialization. Just like in the spec text, use symbolic types that represent multiple concrete types. The few signatures that get generated more than once are only added to the symbol table once because it ignores duplicates. BUG=angle:926 Change-Id: I216f03d22502b724dbefc87c5a021d6021c3a434 Reviewed-on: https://chromium-review.googlesource.com/251620 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Nicolas Capens <capn@chromium.org>
Arun Patole 6b19d765 2015-02-19T09:40:39 Implement missing variants of abs/sign This change adds ANGLE support for abs/sign variants introduced in ESSL3. BUG=angle:923 TEST=Unit tests, dEQP tests Unit Tests: TypeTrackingTest.BuiltInAbsSignFunctionFloatResultTypeAndPrecision TypeTrackingTest.BuiltInAbsSignFunctionIntResultTypeAndPrecision dEQP tests passing 100% because of this change: dEQP-GLES3.functional.shaders.builtin_functions.common.abs dEQP-GLES3.functional.shaders.builtin_functions.common.sign Change-Id: I2a3b028611f0eaaac3c031f8926d34a0e146663d Reviewed-on: https://chromium-review.googlesource.com/251491 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Daniel Bratell 29190088 2015-02-20T16:42:54 Make Angle code 40 KB smaller by using string literals directly. The implicit conversion of hundreds of string literals to TString generated a lot of machine code. By keeping them as string literals all the way the code will be smaller and faster. This is the change with clang for x64 (note VisitUnary in particular): Total change: -41392 bytes ========================== 2 added, totalling +469 bytes across 1 sources 2 removed, totalling -472 bytes across 1 sources 5 shrunk, for a net change of -41389 bytes (54126 bytes before, 12737 bytes after) across 1 sources 279692 unchanged, totalling 51433327 bytes ------------------------------------------------------------------------------------------------------------------------------------ -41392 - Source: /home/bratell/src/chromium/src/third_party/angle/src/compiler/translator/OutputHLSL.cpp - (gained 469, lost 41861) ------------------------------------------------------------------------------------------------------------------------------------ New symbols: +328: sh::OutputHLSL::outputConstructor(Visit, TType const&, char const*, TVector<TIntermNode*> const*) type=t, size=328 bytes +141: sh::OutputHLSL::outputTriplet(Visit, char const*, char const*, char const*) type=t, size=141 bytes Removed symbols: -133: sh::OutputHLSL::outputTriplet(Visit, std::basic_string<char, std::char_traits<char>, pool_allocator<char> > const&, std::basic_string<char, std::char_traits<char>, pool_allocator<char> > const&, std::basic_string<char, std::char_traits<char>, pool_allocator<char> > const&) type=t, size=133 bytes -339: sh::OutputHLSL::outputConstructor(Visit, TType const&, std::basic_string<char, std::char_traits<char>, pool_allocator<char> > const&, TVector<TIntermNode*> const*) type=t, size=339 bytes Shrunk symbols: -388: sh::OutputHLSL::writeEmulatedFunctionTriplet(Visit, char const*) type=t, (was 628 bytes, now 240 bytes) -714: sh::OutputHLSL::visitBranch(Visit, TIntermBranch*) type=t, (was 1017 bytes, now 303 bytes) -9738: sh::OutputHLSL::visitAggregate(Visit, TIntermAggregate*) type=t, (was 17609 bytes, now 7871 bytes) -14132: sh::OutputHLSL::visitBinary(Visit, TIntermBinary*) type=t, (was 17627 bytes, now 3495 bytes) -16417: sh::OutputHLSL::visitUnary(Visit, TIntermUnary*) type=t, (was 17245 bytes, now 828 bytes) Change-Id: Id0f87d72f6d7f1ab7b543f0d28d5a8b7c7db9ec7 Reviewed-on: https://chromium-review.googlesource.com/251090 Reviewed-by: Geoff Lang <geofflang@chromium.org> Tested-by: bratell at Opera <bratell@opera.com>
Jamie Madill 80d934bb 2015-02-19T10:16:12 Enable MSVS warning 4512. Fix up the missing DISALLOW_COPY_AND_ASSIGN macros and various small problems preventing us from turning on this warning. This should ensure we more often use the DISALLOW macro going forward. Change-Id: I2e1a9d23a31a51279a577fad8dffb8c1530e2ee8 Reviewed-on: https://chromium-review.googlesource.com/251100 Tested-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Cooper Partin <coopp@microsoft.com> Tested-by: Cooper Partin <coopp@microsoft.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill 7c1f6f19 2015-02-19T18:56:09 Revert "Implement missing variants of abs/sign" This for some reason is failing angle_unittests: http://build.chromium.org/p/chromium.gpu.fyi/builders/Win7%20Release%20%28NVIDIA%29/builds/13296 Happens across platforms. This reverts commit 3e142e47cc5b40564bcdba4252fd326bc28ca343. Change-Id: Ideb20457546c0dec10c9584182cc8d5ead42309d Reviewed-on: https://chromium-review.googlesource.com/251160 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho b6e07a6a 2015-02-16T12:22:10 Implement ESSL3 modf This is the first built-in function that has an out parameter, so l-value checks are added for built-ins. BUG=angle:918 Change-Id: Ifd5befe955224f706f864e25107879c9cdce9e9f Reviewed-on: https://chromium-review.googlesource.com/250780 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Arun Patole 3e142e47 2015-02-19T09:40:39 Implement missing variants of abs/sign This change adds ANGLE support for abs/sign variants introduced in ESSL3. BUG=angle:923 TEST=dEQP tests Tests passing 100% because of this change: dEQP-GLES3.functional.shaders.builtin_functions.common.abs dEQP-GLES3.functional.shaders.builtin_functions.common.sign Change-Id: If22032be2c1ed08451275262e311ef5ac613d45e Reviewed-on: https://chromium-review.googlesource.com/251060 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho 3eab00a0 2015-02-12T16:02:39 Fix constant folding integer modulus Constant folding % used to generate results from division instead of results from modulus. Fix this and introduce a test for this and other integer constant folding, that checks that the right constant values are found in the IntermNode tree. BUG=angle:916 TEST=angle_unittests Change-Id: I75e59b7734eb206378a1b2957af0a5437a90097c Reviewed-on: https://chromium-review.googlesource.com/249691 Reviewed-by: Gregoire Payen de La Garanderie <Gregory.Payen@imgtec.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Nicolas Capens <capn@chromium.org>
Arun Patole 0c1726e6 2015-02-18T14:35:02 Implement isnan/isinf This change adds ANGLE support for isnan/isinf. BUG=angle:922 TEST=dEQP tests: dEQP-GLES3.functional.shaders.builtin_functions.common.isnan.* (high precision tests still fail) Tests passing 100% because of this change: dEQP-GLES3.functional.shaders.builtin_functions.common.isinf.* Reduced number of errors in tests: EQP-GLES3.functional.shaders.constant_expressions.builtin_functions.common.isnan.* dEQP-GLES3.functional.shaders.constant_expressions.builtin_functions.common.isinf_* Change-Id: I52bf18b59365ea7031746ae840550162d7e6b96e Reviewed-on: https://chromium-review.googlesource.com/250761 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Nicolas Capens <capn@chromium.org>
Arun Patole 2b45573b 2015-02-17T20:24:59 Implement missing vector relational functions This change adds ANGLE support for following vector relational function variants introduced in GLES3: bvec lessThan(uvec x, uvec y) bvec lessThanEqual(uvec x, uvec y) bvec greaterThan(uvec x, uvec y) bvec greaterThanEqual(uvec x, uvec y) bvec equal(uvec x, uvec y) bvec notEqual(uvec x, uvec y) BUG=angle:920 TEST=dEQP tests Reduced number of errors in tests: dEQP-GLES3.functional.shaders.constant_expressions.builtin_functions.vector_relational.* Change-Id: Id45ce77b35f39403e587a577622271d8e9815225 Reviewed-on: https://chromium-review.googlesource.com/250377 Reviewed-by: Nicolas Capens <capn@chromium.org> Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho a3a5cc6a 2015-02-13T13:12:22 Expose the IntermNode tree generated in the compiler for testing This refactoring makes it possible for tests to access the IntermNode tree produced by compilation by calling compileTree(). Removing ParseContext usage from OutputHLSL has the additional benefit of better separation between parsing and output. BUG=angle:916 Change-Id: Ib40954832316328772a5c1dcbbe6b46b238e4e65 Reviewed-on: https://chromium-review.googlesource.com/249723 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Arun Patole 42077bb1 2015-02-16T15:32:37 Implement missing variants of min, max and clamp BUG=angle:914 TEST=dEQP tests Tests passing 100% because of this change: dEQP-GLES3.functional.shaders.operator.common_functions.min.* dEQP-GLES3.functional.shaders.operator.common_functions.max.* dEQP-GLES3.functional.shaders.operator.common_functions.clamp.* Reduced number of errors in tests: EQP-GLES3.functional.shaders.constant_expressions.builtin_functions.common.* These tests are not yet passing because of constant expressions bug (angle bug:913) Change-Id: Ibccee47ed78e994e40a0effe9afa9cf9692492b0 Reviewed-on: https://chromium-review.googlesource.com/250330 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Nicolas Capens <capn@chromium.org> Tested-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho ff805cc3 2015-02-13T10:59:34 Separate integer and float modulus Integer and float modulus are separated into Mod and IMod operator types to make the code clearer and to make GetOperatorString return the correct value for both. Change-Id: Ibfbca2c558bf919b0eab6404d7349f87fe47a18c Reviewed-on: https://chromium-review.googlesource.com/249692 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Olli Etuaho 09b22474 2015-02-11T11:47:26 Add addBinaryMath and addUnaryMath helpers to ParseContext This refactoring reduces code duplication and moves functionality that is not immediately related to the language grammar out of glslang.y. BUG=angle:911 Change-Id: If5e225461890ed542dee01905829df1c9a6f5e27 Reviewed-on: https://chromium-review.googlesource.com/248570 Reviewed-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho 1033e1d8 2015-02-12T12:03:13 Move operator definitions to their own file This generalizes GetOperatorString so that it can be used outside ValidateLimitations. BUG=angle:911 Change-Id: I43a160b3d7d60388ad529a72f1720d3171f67315 Reviewed-on: https://chromium-review.googlesource.com/249073 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Jamie Madill 9b82084f 2015-02-12T10:40:10 Fix HLSL varying struct linking. We would always expand a struct name to always include its symbol ID. This fixes the workaround to only affect scoped structs. We can leave global structs, which are by definition uniquely named, without decorating them with a symbol ID. This fixes several tests in dEQP's shader.linkage.varying.struct. Re-land with GLSL translator bug fixed. BUG=angle:910 Change-Id: I23a932bd1dadea5e9aafabde697e6a2af9a43f2b Reviewed-on: https://chromium-review.googlesource.com/249134 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org> Tested-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill b960cc4a 2015-02-12T15:33:20 Revert "Allow varying structs to match between stages." This patch has a bug which breaks the GLSL to GLSL translator. See broken GPU FYI bot builds on the Chromium waterfall: http://build.chromium.org/p/chromium.gpu.fyi/builders/Win7%20Release%20%28NVIDIA%29/builds/13074 This reverts commit 489ffc4d69656c2dbfdbe07e3fa6e38baeb3e072. Change-Id: I8241c6d658f4f31af91b89af1141d3d778ff7c70 Reviewed-on: https://chromium-review.googlesource.com/249131 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill f4780c1c 2015-02-11T16:33:11 Use correct non-square register count in structs. For struct varyings with nested structs, we would count registers incorrectly, and produce D3D link errors. This fixes tests in the dEQP's shaders.linkage.varying.struct. BUG=angle:910 Change-Id: I47ea6dba36bf57cf94a7e7f30997c6c9096c2b40 Reviewed-on: https://chromium-review.googlesource.com/247243 Tested-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Jamie Madill 489ffc4d 2015-02-11T16:33:07 Allow varying structs to match between stages. We would always expand a struct name to always include its symbol ID. This workaround fixed scoped structs with the same name. It would also block any possiblity of linking the structs successfully. Instead we can use the workaround only on inner-scoped structs, and leave global structs, which are by definition uniquely named, as they are. This fixes several tests in dEQP's shader.linkage.varying.struct. BUG=angle:910 Change-Id: I81b8dadc7ea493152aff0c44d607114eaaabb142 Reviewed-on: https://chromium-review.googlesource.com/247242 Reviewed-by: Geoff Lang <geofflang@chromium.org> Tested-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 2857f489 2015-02-09T15:35:29 Fix struct uniform packing. Our current code would not adapt properly for members that did not fit in the previous register. Instead, use the correct adjusted value as returned from the block layout encoder. This fixes a dEQP WebGL test: gles2/shaders/linkage. BUG=angle:910 Change-Id: Id77d0c76ce767665b3db97cb4d14e8008254fd9f Reviewed-on: https://chromium-review.googlesource.com/247241 Tested-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill 55e79e09 2015-02-09T15:35:00 Implement equality ops for nested structs. This fixes the WebGL test glsl_misc_struct_equals as well as several dEQP tests in functional.shaders.struct. BUG=391957 BUG=angle:910 Change-Id: I09f3cd3f51bbc3541b64dbcfddfe01884ddba6f5 Reviewed-on: https://chromium-review.googlesource.com/247083 Tested-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Gregoire Payen de La Garanderie b3dced29 2015-01-12T14:54:55 Implementing gl_InstanceID in the HLSL compiler. Fixes: dEQP-GLES3.functional.instanced.draw_arrays_instanced.instance_id dEQP-GLES3.functional.instanced.draw_arrays_instanced.mixed dEQP-GLES3.functional.instanced.draw_elements_instanced.instance_id dEQP-GLES3.functional.instanced.draw_elements_instanced.mixed BUG=angle:601 Change-Id: I6e120eebc90d00e025fc58f096064e6ed1da826b Reviewed-on: https://chromium-review.googlesource.com/246911 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Gregoire Payen de La Garanderie <Gregory.Payen@imgtec.com> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
David Yen 0fbd128c 2015-02-02T14:46:09 Extension behavior is now reset between each shader translation unit. The extension behavior was being shared between translation units, this was causing states to be cached between shader compilers. This has been fixed now by adding a new ResetExtensionBehavior() function. A unit test has also been added for testing extensions when compiling shaders. A test has been included which tests that the internal state of the extension behavior is being reset properly. BUG=453543 Change-Id: Icb2a07019b5db972dc75cdbbdece4b7e9757c682 Reviewed-on: https://chromium-review.googlesource.com/245522 Tested-by: David Yen <dyen@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Jamie Madill 37997145 2015-01-28T10:06:34 Defer dependent HLSL global var inits. Some global initializers depend on other globals, for instance a varying or attribute value. Since we use a static proxy variable for these varyings, we need to initialize the global static after we initialize the proxy in the shader preamble. This fixes a long- standing compiler bug. We should also add a WebGL test for this. BUG=angle:878 Change-Id: I71db103a6b8c24fb862e0d8b32293da9bc2e8103 Reviewed-on: https://chromium-review.googlesource.com/243581 Tested-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Jamie Madill 32aab01d 2015-01-27T14:12:26 Use a stack for OutputHLSL info log output. Previously we would always reference mBody in several intermediate output methods. This made using these traversals from within the header, or for utility methods, very difficult. Instead, use a stack where we write to the top InfoLog, and can push/pop from the stack. This gives us more flexibility. BUG=angle:878 Change-Id: I8a6c0382bad18b44d75158274c701db13d4d4e65 Reviewed-on: https://chromium-review.googlesource.com/243580 Tested-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Austin Kinross 3ae6465f 2015-01-26T15:51:39 Fix lots of variable shadowing in ANGLE BUG=angle:877 Change-Id: I3df0fffb19f5ecbe439fbc2a8d6d239a5dc6b638 Reviewed-on: https://chromium-review.googlesource.com/243334 Tested-by: Austin Kinross <aukinros@microsoft.com> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Olli Etuaho 1be8870b 2015-01-19T16:56:44 Avoid precision emulation overhead on unused values Avoid rounding intermediate values when the intermediate value is not consumed. For example, this avoids rounding the return value of assignment, so "b = a;" becomes "b = frm(a);" instead of "frm(b = frm(a))". BUG=angle:874 TEST=compiler_tests Change-Id: Ifcdb53fb1d07a2cf24e429cc237c2d0262dc32f8 Reviewed-on: https://chromium-review.googlesource.com/241852 Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Geoff Lang 74cafab1 2015-01-23T23:17:32 Revert "Fix lots of variable shadowing in ANGLE" Caused WebGL CTS failures on the texture-npot test: https://www.khronos.org/registry/webgl/sdk/tests/conformance/textures/texture-npot.html This reverts commit c67e6e9fade44ef8938724e82db11db725e9c8e5. Change-Id: I089e99859231e0d657084ac3647257c650a9da92 Reviewed-on: https://chromium-review.googlesource.com/243041 Reviewed-by: Geoff Lang <geofflang@chromium.org> Tested-by: Geoff Lang <geofflang@chromium.org>
Austin Kinross c67e6e9f 2015-01-21T16:01:07 Fix lots of variable shadowing in ANGLE BUG=angle:877 Change-Id: I15168ae32605b26aee08274464ffe68adb5a7e87 Reviewed-on: https://chromium-review.googlesource.com/242351 Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Austin Kinross <aukinros@microsoft.com>
Cooper Partin e6664f06 2015-01-09T16:22:24 Added PointSprites Support for renderers that do not support Geometry Shaders Change-Id: Iae9ac5f8fbba68dba5e49ccda7bb7eebb05c8e9a Reviewed-on: https://chromium-review.googlesource.com/240450 Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Tested-by: Geoff Lang <geofflang@chromium.org>
Olli Etuaho 31b5fc62 2015-01-16T12:13:36 Add support for bitwise operations in ESSL3 Add support for <<, >>, &, |, ^, and their compound assignment variants <<=, >>=, &=, |=, ^=. Also add support for bitwise not (~). BUG=angle:870 Change-Id: I5e6a835409589556d5d58d58078fdf505cfd8da5 Reviewed-on: https://chromium-review.googlesource.com/241850 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org>
Olli Etuaho 7700ff65 2015-01-15T12:16:29 Add basic support for ESSL3 vec2 (un)pack functions The functions are emulated on HLSL, and use the native functions on OpenGL 4.2+ and GLES3.0. Emulation for OpenGL versions <= 4.1 is not yet implemented. BUG=angle:865 Change-Id: I6803a1767dacdb3dca12f13924651fd38fcacb75 Reviewed-on: https://chromium-review.googlesource.com/240961 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Geoff Lang <geofflang@chromium.org> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho abf6dadd 2015-01-14T14:45:16 Implement matrix inverse built-in Inverse is emulated in HLSL by calculating the cofactor matrix and dividing that with the determinant. This results in the transpose of the inverse as is required. Better performing options might exist especially for 4x4 matrices, but this is enough for a working implementation. BUG=angle:859 Change-Id: I5185797cc1ed86865f5f4342707abdc2977a186b Reviewed-on: https://chromium-review.googlesource.com/240331 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Geoff Lang <geofflang@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho d68157fc 2015-01-16T12:24:17 Fix issues related to modulus operator Make sure that "scalar %= vector" is not approved by the compiler, and that % and %= are only exposed on GLES3. BUG=angle:854 Change-Id: I3ac7eb607bd86a6caa905d2bef5d397688ad3fd3 Reviewed-on: https://chromium-review.googlesource.com/241410 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Gregoire Payen de La Garanderie <Gregory.Payen@imgtec.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho 80a5a6c6 2015-01-12T15:35:27 Clean up BuiltInFunctionEmulator public interface This removes all language-specific bits from BuiltInFunctionEmulator parent class, and makes the public interface of BuiltInFunctionEmulator minimal. Writing comments around emulated function definitions is removed from OutputHLSL, they are not necessary as the emulated function definitions are just another part of the shader header. Change-Id: I9abf57d86f4e37b0674d7dfafe653298f205dd27 Reviewed-on: https://chromium-review.googlesource.com/240230 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Nicolas Capens <capn@chromium.org> Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho e39706d0 2014-12-30T16:40:36 Implement simple ESSL3 matrix functions Add support for built-in functions outerProduct, determinant, transpose and the variations of matrixCompMult that were previously unsupported. BUG=angle:859 Change-Id: Ie2b9cf83cd80c5a886c6d4eb190c7ce25a32d0a4 Reviewed-on: https://chromium-review.googlesource.com/239873 Reviewed-by: Nicolas Capens <capn@chromium.org> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho e17e3197 2015-01-02T12:47:59 Use BuiltInFunctionEmulatorHLSL for all emulated functions Implementation of missing built-in functions is a separate concern from outputting the intermediate tree itself as HLSL, so it makes sense to have all of the built-in emulation in a class that is separate from OutputHLSL. Being able to reuse the same logic for different emulated functions also makes the code more compact. Change-Id: Id503dc3a5c5e743ec65722add56d6ba216a03a7f Reviewed-on: https://chromium-review.googlesource.com/239872 Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Nicolas Capens <capn@chromium.org> Reviewed-by: Olli Etuaho <oetuaho@nvidia.com> Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Olli Etuaho 2aafa18b 2015-01-08T19:04:48 Fix null check after addUnaryMath is called for a built-in function The intermTypedNode pointer null check should be done before the pointer is used, not after. In practice the pointer should never be null if IntermUnary::promote() is consistent with the built-in function table, but it's useful to have the null check as a fail-safe when developing. Change-Id: I24e0d7399ef9d628f474e99fbca08dbd67fdd8c3 Reviewed-on: https://chromium-review.googlesource.com/240411 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org>
Jamie Madill 9e0478f6 2015-01-13T11:13:54 Move the block layout code to translator. This code is easily accessible from the translator, with the proper export calls. This facilitates adding a common static library, since this code calls some methods in translator. BUG=angle:773 Change-Id: I0c50098ec3f67c2df7749b3c2518be0a9fd939e2 Reviewed-on: https://chromium-review.googlesource.com/240093 Reviewed-by: Zhenyao Mo <zmo@chromium.org> Tested-by: Jamie Madill <jmadill@chromium.org>
Olli Etuaho f6bdb312 2015-01-09T15:21:38 Refactor defining emulated functions in BuiltInFunctionEmulator This removes the fixed enumeration of emulated functions in favor of a dynamic map. This makes the code more compact and flexible. The main benefit is that maintaining the list emulated functions is a lot simpler now. Change-Id: Ic9951a496b9f021c76ad5b4c3daccd89af5ac093 Reviewed-on: https://chromium-review.googlesource.com/239871 Tested-by: Olli Etuaho <oetuaho@nvidia.com> Reviewed-by: Jamie Madill <jmadill@chromium.org> Reviewed-by: Nicolas Capens <capn@chromium.org>
kmx.io     mail     discord kmxgit v0.4.0