|
74da73fe
|
2017-02-01T15:37:48
|
|
Add ESSL 3.10 ldexp/frexp builtins
This adds new built-ins found in ESSL 3.10 section 8.3 Common
Functions.
This includes constant folding support for ldexp and support for both
GLSL and HLSL output. In HLSL these functions need to be emulated.
BUG=angleproject:1730
TEST=angle_unittests
Change-Id: I1330e69978b0cf53efbc3416150194764414e96c
Reviewed-on: https://chromium-review.googlesource.com/435342
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
25aef453
|
2017-01-29T16:15:44
|
|
Add new ESSL 3.10 pack/unpack builtins
This adds new built-ins found in ESSL 3.10 section 8.4 Floating-Point
Pack and Unpack Functions.
This includes constant folding support and support for both GLSL and
HLSL output. In HLSL all of these functions need to be emulated.
BUG=angleproject:1730
TEST=angle_unittests
TEST=dEQP-GLES31.functional.shaders.*pack*norm4x8*
Change-Id: Ibed60286a366cd35c4faafd405e79af562a02a06
Reviewed-on: https://chromium-review.googlesource.com/434170
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
1ecd14b8
|
2017-01-26T13:54:15
|
|
Fold user-definedness of function nodes into TOperator
Whether a function call is user-defined is not orthogonal to TOperator
associated with the call node - other ops than function calls can't be
user-defined. Because of this it makes sense to store the user-
definedness by having different TOperator enums for different types of
calls.
This patch also tags internal helper functions that have a raw
definition outside the AST with a separate TOperator enum. This way
they can be handled with logic that is easy to understand. Before this,
function calls like this left the user-defined bit unset, despite not
really being built-ins either. The EmulatePrecision traverser uses
this. This is also something that could be used to clean up built-in
emulation in the future.
BUG=angleproject:1490
TEST=angle_unittests
Change-Id: I597fcd9789d0cc22b689ef3ce5a0cc3f621d4859
Reviewed-on: https://chromium-review.googlesource.com/433443
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
9250cb24
|
2017-01-21T10:51:27
|
|
Add ESSL 3.10 integer math built-ins
This adds built-ins found in ESSL 3.10 section 8.8 Integer functions.
This includes constant folding support for functions that may be
constant folded, and support for both GLSL and HLSL output. In HLSL
several of the functions need to be emulated.
The precision qualification for the return value of some of these
functions is determined by special rules, that are now part of type
promotion for TIntermUnary nodes and determining the type of
TIntermAggregate nodes.
BUG=angleproject:1730
TEST=angle_unittests
TEST=dEQP-GLES31.functional.shaders.builtin_functions.integer.*
Change-Id: Ib0056c17671c42b6496c2f0ef059b99f8f25c122
Reviewed-on: https://chromium-review.googlesource.com/431310
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
dfa75e87
|
2017-01-23T09:43:06
|
|
Add support for 4-parameter functions to BuiltInFunctionEmulator
New entry points are needed to support built-ins with more parameters.
Also, now that ops that are not function calls don't use the
TIntermAggregate class any more, it's easier to exclude nodes that are
not candidates for built-in emulation using a simple blacklist rather
than to use a whitelist.
Also includes function name style cleanup in BuiltInFunctionEmulator.
This will make it possible to add necessary emulation for built-ins
from ESSL 3.10.
BUG=angleproject:1730
TEST=angle_unittests
Change-Id: If267fc68f5cb9b2ee6703cbcbbe4d157da44a7e0
Reviewed-on: https://chromium-review.googlesource.com/431297
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
1d9dcc24
|
2017-01-19T11:25:32
|
|
Make AST path always include the current node being traversed
AST traversers tend to sometimes call traverse() functions manually
during PreVisit. Change TIntermTraverser so that even if this happens,
all the nodes are automatically added to the traversal path, instead
of having to add them manually in each individual AST traverser.
This also makes calling getParentNode() return the correct node during
InVisit.
This does cause the same node being added to the traversal path twice
in some cases, where nodes are repeatedly traversed, like in
OutputHLSL, but this should not have adverse side effects. The more
common case is that the traverse() function is called on the children
of the node being currently traversed.
This fixes a bug in OVR_multiview validation, which did not previously
call incrementDepth and decrementDepth when it should have.
BUG=angleproject:1725
TEST=angle_unittests, angle_end2end_tests
Change-Id: I6ae762eef760509ebe853eefa37dac28c16e7a9b
Reviewed-on: https://chromium-review.googlesource.com/430732
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
b123938d
|
2016-12-13T15:07:05
|
|
D3D11: Add support to compile and link compute shaders
This is a reland of 2cd9d7e032fb412b539a907c58342060340387a1.
BUG=angleproject:1442
TEST=angle_end2end_tests
Change-Id: I5be0032b97617c31cdd4c66a823e8eb3b518867a
Reviewed-on: https://chromium-review.googlesource.com/430199
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
8ad9e757
|
2017-01-16T19:55:20
|
|
Always store function headers in TIntermFunctionPrototype nodes
TIntermFunctionDefinition nodes now have a TIntermFunctionPrototype
child that stores the function signature, instead of having a separate
type and an aggregate child that stores the parameters.
This makes parsing functions simpler, and paves the way for further
simplifications of function parsing, like reducing conversions between
symbol table structures and AST structures.
TIntermAggregate is now only used for function calls.
BUG=angleproject:1490
TEST=angle_unittests, angle_end2end_tests
Change-Id: Ib56a77b5ef5123b142963a18499690bf37fed987
Reviewed-on: https://chromium-review.googlesource.com/427945
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
16c745a3
|
2017-01-16T17:02:27
|
|
Split TIntermFunctionPrototype from TIntermAggregate
Function prototypes now have their own class TIntermFunctionPrototype.
It's only used for prototypes, not function parameter lists.
TIntermAggregate is still used for parameter lists and function calls.
BUGS=angleproject:1490
TEST=angle_unittests
Change-Id: I6e246ad00a29c2335bd2ab7f61cf73fe463b74bb
Reviewed-on: https://chromium-review.googlesource.com/427944
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
417df92f
|
2017-01-12T09:23:07
|
|
Revert "D3D11: Add support to compile and link compute shaders."
Fails https://build.chromium.org/p/chromium.gpu.fyi/builders/Linux%20Debug%20%28New%20Intel%29/builds/5769
BUG=angleproject:1442
This reverts commit 2cd9d7e032fb412b539a907c58342060340387a1.
Change-Id: Ic1610d20ba0449b423528fa9840aa951c012cf84
Reviewed-on: https://chromium-review.googlesource.com/427229
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
2cd9d7e0
|
2016-12-13T15:07:05
|
|
D3D11: Add support to compile and link compute shaders.
BUG=angleproject:1442
Change-Id: I13240e931e6f121d175d2cd6b41324d38bb39a5c
Reviewed-on: https://chromium-review.googlesource.com/405831
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
d68924e5
|
2017-01-02T17:34:40
|
|
Use GetOperatorString when writing GLSL unary built-in calls
GetOperatorString is now used when writing GLSL for built-in calls
that fall under TIntermUnary. Component-wise not TOperator enum is
renamed for consistency.
This also cleans up some unnecessary creation of string objects when
writing built-in functions.
BUG=angleproject:1682
TEST=angle_unittests, angle_end2end_tests, WebGL conformance tests
Change-Id: I89b2ef222bf5af479d4977417f320789b58ace85
Reviewed-on: https://chromium-review.googlesource.com/424552
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
e180559f
|
2017-01-02T16:41:20
|
|
Use GetOperatorString when writing GLSL built-in function calls
GetOperatorString is now used when writing GLSL for built-in calls
that fall under TIntermAggregate. Component wise and not component
wise TOperator enums are disambiguated from each other.
BUG=angleproject:1682
TEST=angle_unittests, angle_end2end_tests, WebGL conformance tests
Change-Id: I861f1e94eb695eb712592df99705848b442ef07b
Reviewed-on: https://chromium-review.googlesource.com/424532
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
77ba408a
|
2016-12-16T12:01:18
|
|
Unify Diagnostics interface
Use the same kind of interface for reporting preprocessor errors as
for reporting regular compiler errors, and make global errors like
having too many uniforms also go through Diagnostics. Also don't
create std::string objects unnecessarily.
Includes cleanups of some dead code related to reporting errors.
BUG=angleproject:1670
TEST=angle_unittests
Change-Id: I3ee794d32ddeec1826bdf1b76b558f35259f82c0
Reviewed-on: https://chromium-review.googlesource.com/421527
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
bf4e1b73
|
2016-12-09T11:30:15
|
|
Split TIntermInvariantDeclaration from TIntermAggregate
This change is pure refactoring and doesn't fix bugs related to
invariant declarations. Invariant declarations are supposed to accept
a list of identifiers, but this refactoring keeps the current behavior
of only accepting a single identifier in an invariant declaration.
When the bug will be fixed, the new TIntermInvariantDeclaration class
that now has only a single child node can be changed so that it may
have multiple children.
TIntermAggregate is still used for function calls, function
prototypes and function parameter lists.
BUG=angleproject:1490
TEST=angle_unittests
Change-Id: I3e22092c87e1c06445fd7e123d9922c2fcb59428
Reviewed-on: https://chromium-review.googlesource.com/419415
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
d7b1ab58
|
2016-12-12T14:42:19
|
|
Fix up translator style.
Using git cl format.
BUG=angleproject:650
Change-Id: I7d3f98d2b0dcfb0a8de6c35327db74e55c28d761
Reviewed-on: https://chromium-review.googlesource.com/419059
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
56a2f95f
|
2016-12-08T12:16:27
|
|
Output infinity and NaN literals correctly in shaders
Previously infinity and NaN resulting from constant folding would be
clamped to finite 32-bit float range when they were written in shader
output. Now they are written as a bit pattern in case the shader
version allows it.
This does not guarantee that NaNs work, but this is fine, since ESSL
3.00.6 spec has very loose requirements when it comes to NaNs.
BUG=angleproject:1654
TEST=angle_end2end_tests
Change-Id: I9997000beeaa8ed22523c22d5cf6929cdfc93f60
Reviewed-on: https://chromium-review.googlesource.com/417301
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
45bcc784
|
2016-11-07T13:58:48
|
|
translator: Scope all classes with "sh".
I was seeing an odd problem with our PoolAlloc conflicting with the
glslang/Vulkan TIntermNode, so the fix was to move everything to a
separate namespace.
The bison grammars are also regenerated. No functional changes.
BUG=angleproject:1576
Change-Id: I959c7afe4c092f0d458432c07b4dcee4d39513f3
Reviewed-on: https://chromium-review.googlesource.com/408267
Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
60e6edfa
|
2016-10-31T12:17:19
|
|
Make ASSERT reference the conditional expression.
This should prevent further unexpected bot breakage due to
unreferenced variables in the ASSERT expression.
Also remove the no longer needed variable referencing macro.
BUG=angleproject:1586
Change-Id: I127695165bdfe39c51fe8d17e00daf6bf2fa8252
Reviewed-on: https://chromium-review.googlesource.com/404948
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
|
|
|
13389b66
|
2016-10-16T11:48:18
|
|
Split TIntermDeclaration from TIntermAggregate
The new class TIntermDeclaration is now used for struct, interface
block and variable declarations. TIntermDeclaration nodes do not have
a type - rather the type is stored in each child node. The types may
differ in case the declaration is a series of array declarators with
mismatching sizes.
TIntermAggregate is still used for function calls, function
prototypes, function parameter lists and invariant declarations.
BUG=angleproject:1490
TEST=angle_unittests
Change-Id: I0457188f354481470855f61ac1c878fc2579b1d1
Reviewed-on: https://chromium-review.googlesource.com/400023
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
4db7ded5
|
2016-10-13T12:23:11
|
|
Change comma nodes to TIntermBinary
Comma nodes always have just two parameters. If there's an expression
with several commas in the middle, it's parsed as a tree of comma
operations. It makes more sense to represent it as a binary node
rather than an aggregate node.
After this patch, TIntermAggregate is still used for function
prototypes, function parameter lists, function calls, and variable and
invariant declarations.
BUG=angleproject:1490
TEST=angle_unittests, angle_end2end_tests
Change-Id: I66be10624bf27bcf25987b4d93958d4a07600771
Reviewed-on: https://chromium-review.googlesource.com/397320
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
336b1470
|
2016-10-05T16:37:55
|
|
Split TIntermFunctionDefinition from TIntermAggregate
This makes the code easier to understand. Function definition nodes
always have just two children, the parameters node and the function
body node, so there was no proper reason why they should be aggregate
nodes.
As a part of this change, intermediate output is modified to print
symbol table ids of functions so that debugging function id related
functionality will be easier in the future.
After this patch, TIntermAggregate is still used for function
prototypes, function parameter lists, function calls, variable and
invariant declarations and the comma (sequence) operator.
BUG=angleproject:1490
TEST=angle_unittests, angle_end2end_tests
Change-Id: Ib88b4ca5d21abd5f126836ca5900d0baecabd19e
Reviewed-on: https://chromium-review.googlesource.com/394707
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
bd674557
|
2016-10-06T13:28:42
|
|
Separate function info from TIntermAggregate
This change will make it easier to split types of TIntermAggregate
nodes representing functions and function calls into different node
classes.
BUG=angleproject:1490
TEST=angle_unittests
Change-Id: I730aa7858fe31fda86218fc685980c6ad486f5e0
Reviewed-on: https://chromium-review.googlesource.com/394706
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
6d40bbdd
|
2016-09-30T13:49:38
|
|
Split TIntermBlock from TIntermAggregate
The new TIntermBlock node class replaces TIntermAggregate nodes with
the EOpSequence op. It represents the root node of the tree which is
a list of declarations and function definitions, and any code blocks
that can be denoted by curly braces. These include function and loop
bodies, and if-else branches.
This change enables a bunch of more compile-time type checking, and
makes the AST code easier to understand and less error-prone.
The PostProcess step that used to be done to ensure that the root node
is TIntermAggregate is removed in favor of making sure that the root
node is a TIntermBlock in the glslang.y parsing code.
Intermediate output formatting is improved to print the EOpNull error
in a clearer way.
After this patch, TIntermAggregate is still used for function
definitions, function prototypes, function parameter lists, function
calls, variable and invariant declarations and the comma (sequence)
operator.
BUG=angleproject:1490
TEST=angle_unittests, angle_end2end_tests
Change-Id: I04044affff979a11577bc1fe75d747e538b799c8
Reviewed-on: https://chromium-review.googlesource.com/393726
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
5878f832
|
2016-10-07T10:14:58
|
|
Fix formatting of OutputHLSL::visitAggregate
Subsequent refactoring of this code will be easier to review if there
won't be unrelated style changes that git cl format insists on doing.
BUG=angleproject:1490
TEST=angle_unittests
Change-Id: I102fd73bd92317ab438e1676422212f644d2859b
Reviewed-on: https://chromium-review.googlesource.com/394649
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
32db19b7
|
2016-10-04T14:43:16
|
|
Ensure that if-else branches are always sequence nodes
This mainly affects RewriteElseBlocks, which was the only piece of
code still adding TIntermIfElse nodes directly as children of other
TIntermIfElse nodes.
BUG=angleproject:1490
TEST=angle_unittests
Change-Id: I5b25c2fb9c642424417cd6c29e37c20482c6ffaf
Reviewed-on: https://chromium-review.googlesource.com/392847
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
f51fdd2e
|
2016-10-03T10:03:40
|
|
Ensure that all functions have a body node in the AST
Some traversers that insert code to main() assume that the main()
function has a non-null body node in place. This assumption was
previously wrong, since functions could be missing the body node
in case the function body was empty.
Fix possible invalid dereferencing of missing function body nodes by
always adding an empty sequence node to represent the body of
functions that have an empty body in the ESSL source. This also
enables simplifying some tree traversers that used to take the
possibility of missing function body nodes into account.
Also fix AddDefaultReturnStatements to check the last statement inside
the function body for a return statement, instead of checking the
first statement.
BUG=angleproject:1539
TEST=angle_unittests, angle_end2end_tests
Change-Id: I2fbd18c78653fa2f1a96dbd9a619accc4874030d
Reviewed-on: https://chromium-review.googlesource.com/392046
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
b6fa043d
|
2016-09-28T16:28:05
|
|
Split vector swizzle AST nodes into a different node class
This avoids creating a weird aggregate node with a sequence of
constant union nodes to store the offsets. They're stored neatly
inside a vector instead. This makes code that needs to iterate
over the swizzle offsets much simpler.
BUG=angleproject:1490
TEST=angle_unittests
Change-Id: I156b95723529ee05a94d30295ffb6d0952a98564
Reviewed-on: https://chromium-review.googlesource.com/390832
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
6f0a0dca
|
2016-09-27T13:51:29
|
|
Workaround isnan() on Intel drivers
On some Intel drivers, calling function isnan() on highp float will get wrong
answer. This patch work arounds this bug by using an expression to emulate
this function.
BUG=chromium:650547
Change-Id: I5bc5e0352c434f42cd2c55103a74f9f7ba51a72c
Reviewed-on: https://chromium-review.googlesource.com/389834
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
5796127e
|
2016-09-14T13:57:46
|
|
Rename TIntermSelection to TIntermIfElse
Now that ternary nodes are not represented by TIntermSelection any
more, TIntermIfElse is an easier name to understand for newcomers to
the code.
BUG=angleproject:1490
TEST=angle_unittests
Change-Id: Ia1e04e356ab93409400245092a84533d7dfd129d
Reviewed-on: https://chromium-review.googlesource.com/385416
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
7ebb97fc
|
2016-09-08T18:01:50
|
|
Use 64-bits compile options
BUG=chromium:645071
Change-Id: I31825123bf4cb45fb37a93f538e8936487beb5ff
Reviewed-on: https://chromium-review.googlesource.com/382712
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
d0bad2c7
|
2016-09-09T18:01:16
|
|
Split ternary node class from TIntermSelection
Ternary operator nodes are typed parts of expressions, they always
have two children and the children are also guaranteed to be
TIntermTyped. "If" selection nodes can't be a part of an expression,
they can have either one or two children and the children are code
blocks. Due to all of these differences it makes sense to store these
using two different AST node classes.
BUG=angleproject:1490
TEST=angle_unittests
Change-Id: I913ab1d806e3cdb5c21106f078cc9c0b6c72ac54
Reviewed-on: https://chromium-review.googlesource.com/384512
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
|
|
1fe74c7e
|
2016-08-25T13:23:01
|
|
Manually compute the mipmap size for the textureSize builtin.
There were two issues with the current implementation:
* The GetDimensions function already takes into account the base level of the
SRV.
* The GetDimensions function returns doesn't return valid sizes for levels
that don't exist in the SRV. Instead, manually do the lod offset.
BUG=angleproject:931
BUG=angleproject:1316
TEST=dEQP-GLES3.functional.shaders.texture_functions.texturesize.sampler2d_fixed_vertex
Change-Id: I63259b563a42b93b73949e0ef7ac412099a42f13
Reviewed-on: https://chromium-review.googlesource.com/376099
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
856c497e
|
2016-08-08T11:38:39
|
|
Clarify error checking function names in the GLSL parser
Most error checking functions in ParseContext used to follow a format
like <property>ErrorCheck. Sometimes this function would check that
the node/value would have <property>, sometimes it would check that
the node/value would not have it, which was confusing. Change most of
these functions to use a lot more descriptive names, which clearly
communicate what they are checking for.
Also includes a bit of refactoring in constructorErrorCheck(), so that
the function only checks for errors rather than also setting the type
of the constructor node.
Also make TType::arraySize unsigned, and return a sanitized size from
checkIsValidArraySize() instead of using an output parameter.
BUG=angleproject:911
TEST=angle_unittests
Change-Id: Id9767b8c79594ad3f782f801ea68eb96df721a31
Reviewed-on: https://chromium-review.googlesource.com/367070
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
b741c761
|
2016-06-29T15:49:22
|
|
Support precision emulation on HLSL
Re-submit with missing virtual destructor and angle::NonCopyable
added.
Add precision emulation support to HLSL 4.1 output. This makes it
possible for developers to test their shaders for precision issues
easily on Chrome on Windows without having to use the GL backend. The
patch has been verified with Chrome on Windows to reproduce some
precision bugs in real-world WebGL content, including old versions of
the babylon.js library.
The EmulatePrecision AST transformation still relies on writing out
raw shader code for the rounding functions, with raw HLSL code added
alongside pre-existing GLSL and ESSL code. In some ways it would be
nicer to do the EmulatePrecision step as a pure AST transformation,
but on the other hand the raw code is much more readable and easier
to optimize.
To better support multiple output languages in EmulatePrecision, add a
RoundingHelperWriter class that has different subclasses for writing
the rounding functions in different languages.
The unit tests are expanded to cover the HLSL output of precision
emulation. The parts of the tests that require the HLSL output are
only active on Windows where ANGLE_ENABLE_HLSL define is added to the
unit tests. Putting the HLSL tests in an entirely separate file is a
worse alternative, since it would require either a lot of code
duplication or add a lot of boilerplate to the individual tests.
BUG=angleproject:1437
TEST=angle_unittests
Change-Id: I47d501037c206f4bd8b976d3acab9b21c717084c
Reviewed-on: https://chromium-review.googlesource.com/360152
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
3fdaf6f2
|
2016-07-13T15:07:41
|
|
Revert "Support precision emulation on HLSL"
Forgot to add virtual destructors.
BUG=angleproject:1437
This reverts commit a42e8b2cb9d0857f53c0490b5be3bf25b4e1f827.
Change-Id: If33fecfeca9947deedf4668c64dbadf25a5dc5eb
Reviewed-on: https://chromium-review.googlesource.com/360122
Reviewed-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
a42e8b2c
|
2016-06-29T15:49:22
|
|
Support precision emulation on HLSL
Add precision emulation support to HLSL 4.1 output. This makes it
possible for developers to test their shaders for precision issues
easily on Chrome on Windows without having to use the GL backend. The
patch has been verified with Chrome on Windows to reproduce some
precision bugs in real-world WebGL content, including old versions of
the babylon.js library.
The EmulatePrecision AST transformation still relies on writing out
raw shader code for the rounding functions, with raw HLSL code added
alongside pre-existing GLSL and ESSL code. In some ways it would be
nicer to do the EmulatePrecision step as a pure AST transformation,
but on the other hand the raw code is much more readable and easier
to optimize.
To better support multiple output languages in EmulatePrecision, add a
RoundingHelperWriter class that has different subclasses for writing
the rounding functions in different languages.
The unit tests are expanded to cover the HLSL output of precision
emulation. The parts of the tests that require the HLSL output are
only active on Windows where ANGLE_ENABLE_HLSL define is added to the
unit tests. Putting the HLSL tests in an entirely separate file is a
worse alternative, since it would require either a lot of code
duplication or add a lot of boilerplate to the individual tests.
BUG=angleproject:1437
TEST=angle_unittests
Change-Id: Ia4ba0374cd415908f16f34752321af1cb93525a5
Reviewed-on: https://chromium-review.googlesource.com/358473
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
d4f4c11b
|
2016-04-15T15:11:24
|
|
Fix deferring global array initialization
The initial implementation of DeferGlobalInitializers did not take
HLSL corner cases into account. In particular, in case there was a
const-qualified array variable with an initializer that contained
elements that weren't constant folded, initialization would not be
deferred and the global scope of HLSL output would contain a call to
angle_construct_into_*().
On the other hand, deferring global initializers was also done in
cases where it wasn't necessary. Initializers of non-const qualified
array variables that could be written as HLSL literals by HLSL output
were unnecessarily deferred.
This patch fixes both of these issues: Now all global initializers are
potential candidates for deferral instead of just those where the
symbol has the EvqGlobal qualifier, and initializers that are
constructors taking only constant unions as parameters are not
unnecessarily deferred.
BUG=angleproject:1205
BUG=541551
TEST=angle_end2end_tests
Change-Id: I4027059e0e5f39c8a5a48b5c97a3fceaac6b6f8a
Reviewed-on: https://chromium-review.googlesource.com/339201
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
5858f7e3
|
2016-04-08T13:08:46
|
|
Re-land "Refactor texture function handling in OutputHLSL"
This change is pure refactoring, it does not introduce any functional
changes.
Separate texture function output into a helper class and further into
different helper functions to make the code more maintainable.
Some of the logic is simplified slightly by eliminating duplicate
cases and limiting the scope of variables where possible, but care
has been taken to preserve the exact same functionality as before.
Re-land with a fix to typo in include guard.
BUG=angleproject:1349
TEST=dEQP-GLES3.functional.shaders.texture_functions.* (no regression)
dEQP-GLES3.texture.* (no regression)
Change-Id: I57c1ec1950fa05bd16275ca578eb5ee99b34a5ae
Reviewed-on: https://chromium-review.googlesource.com/339180
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
c2ed9380
|
2016-04-15T13:29:25
|
|
Revert "Refactor texture function handling in OutputHLSL"
It triggered an include guard warning on Windows Clang
This reverts commit 6f6c5580553d1f3c584df692823c2f5640e23d88.
Change-Id: Ibd4f2851f311a494f16376d8eed38f3119594761
Reviewed-on: https://chromium-review.googlesource.com/338933
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Corentin Wallez <cwallez@chromium.org>
|
|
|
6f6c5580
|
2016-04-08T13:08:46
|
|
Refactor texture function handling in OutputHLSL
This change is pure refactoring, it does not introduce any functional
changes.
Separate texture function output into a helper class and further into
different helper functions to make the code more maintainable.
Some of the logic is simplified slightly by eliminating duplicate
cases and limiting the scope of variables where possible, but care
has been taken to preserve the exact same functionality as before.
BUG=angleproject:1349
TEST=dEQP-GLES3.functional.shaders.texture_functions.* (no regression)
dEQP-GLES3.texture.* (no regression)
Change-Id: I5d81b842d693c0055890d5724eae6c105e454cd8
Reviewed-on: https://chromium-review.googlesource.com/337931
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
3d932d83
|
2016-04-12T11:10:30
|
|
Defer global initializers when necessary
Move global variable initializers that are not constant expressions to
a function that gets called at the start of main(). This is done
with an AST transformation. This needs to be done because global
variable initializers must be constant in native GL, but ANGLE is more
lenient with what can be put into ESSL 1.00 global initializers to
remain compatible with legacy WebGL content.
Non-constant global variable initializers also caused issues in HLSL
output, since in HLSL output some types of expressions get unfolded
into multiple statements. These include short-circuiting operators and
array initialization. To make sure that these cases are covered, any
initializers that can't be constant folded are deferred, even if they
have the const qualifier.
The old deferring mechanism in OutputHLSL is removed in favor of this
new AST transformation based approach.
BUG=angleproject:819
BUG=angleproject:1205
BUG=angleproject:1350
BUG=596616
TEST=WebGL conformance test
conformance/glsl/misc/global-variable-init.html
Change-Id: I039cc05d6b8c284baeefbdf7f10062cae4bc5716
Reviewed-on: https://chromium-review.googlesource.com/338291
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
b079c7af
|
2016-04-01T12:32:52
|
|
Wrap integer textures with correct wrap mode in HLSL
The wrap mode information for all three dimensions is packed to a
single integer in order to conserve sampler metadata space. Only one
int4 vector is used for the metadata for a single sampler.
The sampler metadata is now packed into a struct instead of an array
of integers in order to make the code more readable and maintainable.
The internalFormatBits field is not removed in this patch. It's better
to remove it in a separate patch, so restoring it is easier in case it
will be used for optimizing some of the texture sampling functions.
The wrap mode passed in sampler metadata is used to wrap the texture
coordinates in the code generated to implement ESSL 3.00 integer
texture sampling built-ins.
Those dEQP-GLES3.functional.texture.units.* tests that sample from
integer cube maps still fail on Intel D3D after this change,
presumably due to driver issues.
BUG=angleproject:1244
BUG=angleproject:1095
BUG=angleproject:1092
TEST=dEQP-GLES3.functional.texture.units.* (all pass on NVIDIA),
dEQP-GLES3.functional.shaders.texture_functions.* (no regressions)
Change-Id: I4e31e5796086f9cc290c6f1f8c4380a768758d71
Reviewed-on: https://chromium-review.googlesource.com/336638
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
ced87057
|
2016-04-04T16:34:27
|
|
Fix integer texture sampling with explicit gradients
The lod level should be selected according to the explicit gradients,
and the gradients are relative to the normalized texture coordinates,
so they need to be multiplied by the base level dimensions to get the
correct gradients to use in the lod level formula.
In the case of sampling integer cube maps, the derivatives of the
texture coordinates on the cube map face need to be calculated based
on the derivatives of the cube map direction vector components.
Also includes fix for sampling integer cube maps with explicit LOD.
BUG=angleproject:1092
TEST=dEQP-GLES3.functional.shaders.texture_functions.*grad*
Change-Id: Iadd358e713fa9695e755e98db8f368e8c512ac45
Reviewed-on: https://chromium-review.googlesource.com/337100
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
d4102f09
|
2016-01-22T14:54:04
|
|
Refactor HLSL texture coordinate output to prepare for wrap modes
Wrap modes for integer textures need to be implemented in shaders in
HLSL. This requires more complex transformations on the texture
coordinates, so store the texture coordinate code for each texture
coordinate in a string variable.
BUG=angleproject:1244
TEST=dEQP-GLES3.functional.shaders.texture_functions.* (no regression),
dEQP-GLES3.functional.texture.units.* (no regression)
Change-Id: Iadd9617a7f906aa1d56dabfba6639f932152e6f1
Reviewed-on: https://chromium-review.googlesource.com/336637
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
9696316d
|
2016-03-21T11:54:33
|
|
Support ESSL structs containing samplers on D3D
Since HLSL can't natively handle samplers in structs, samplers need to
be extracted out of structs into separate variables in the translated
shader code. In HLSL 4.1, samplers that were in structs go into the
normal sampler arrays and are identified by index constants. In other
HLSL versions, samplers that were in structs are translated as uniform
variables.
These transformations are done inside the HLSL output classes, not as
tree transformations. This helps to keep the uniform API provided by
the shader translator intact.
Wherever a struct containing samplers is passed into a user-defined
function, the translated HLSL code passes the separate sampler
variables alongside a struct where the samplers have been removed.
The D3D backend in libANGLE queries the uniform registers of any
samplers that were in uniform structs, and adds them to the register
maps, so that correct sampler state gets assigned to them.
The extracted sampler variables are prefixed with "angle_" instead of
the usual "_" to prevent any name conflicts between them and regular
variables.
BUG=angleproject:504
TEST=angle_end2end_tests,
dEQP-GLES*.functional.shaders.struct.uniform.* (all pass),
dEQP-GLES*.functional.uniform_api.* (most now pass)
Change-Id: Ib79cba2fa0ff8257a973d70dfd917a64f0ca1efb
Reviewed-on: https://chromium-review.googlesource.com/333743
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
b076adde
|
2016-01-11T16:45:46
|
|
Implement gl_VertexID
BUG=angleproject:1217
Change-Id: Ibb9423d7de4966bce231734925a804b6340b5059
Reviewed-on: https://chromium-review.googlesource.com/321420
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
be59c2fb
|
2016-03-07T11:32:34
|
|
Fix ambiguous function call issues in HLSL output
D3D compiler can't resolve between these overloaded functions:
float4 vec4(float2x2 x0);
float4 vec4(float4 x0);
Include the parameter types in the function name to disambiguate
between overloaded user-defined functions and constructors, like this:
float4 vec4_float2x2(float2x2 x0);
float4 vec4_float4(float4 x0);
This is only done for float2x2 and float4 parameters, other parameter
types like float2x3 vs. float3x2 don't need this.
BUG=angleproject:1099
BUG=angleproject:1030
TEST=angle_end2end_tests,
dEQP-GLES3.functional.attribute_location.* (10 more tests pass),
dEQP-GLES2.functional.attribute_location.*
Change-Id: Ief047d41b0adbc238393c3c13cb29771cbb83d58
Reviewed-on: https://chromium-review.googlesource.com/329882
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
f8bf583b
|
2016-02-11T16:21:49
|
|
Pass integer texture format bit count to shaders on D3D11
This will be needed in the future when integer texture wrap mode support
will be added by sampling integer textures through FLOAT/UNORM/SNORM
SRVs.
The bit count needs to be passed for 8-, 10- and 16-bit textures. 32-bit
integer textures are the ones left over. Only passing the bit counts for
the absolute minimum number of formats avoids unnecessary driver constant
buffer updates.
BUG=angleproject:1244
BUG=angleproject:1095
TEST=angle_end2end_tests
Change-Id: I28a84588842b2eb9a1661454437d21c22ce794b7
Reviewed-on: https://chromium-review.googlesource.com/326944
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
bce743ab
|
2016-01-15T17:18:28
|
|
Use texture base level to implement textureSize on D3D11 HLSL
HLSL GetDimensions call doesn't take the texture base level into account,
so ANGLE needs to use the texture base level passed in uniforms to
emulate ESSL textureSize() which does take it into account.
After this change the relevant dEQP tests pass on NVIDIA, Intel is still
suffering from an issue where a wrong value is returned when the lod
is > 0 (tested on Intel HD Graphics 4600). AMD is also suffering from an
unknown issue.
BUG=angleproject:596
TEST=dEQP-GLES3.functional.shaders.texture_functions.texturesize.*
(all pass on NVIDIA now), angle_end2end_tests
Change-Id: I13e33d126008ecdf2b89461a3fb5040949cf19e2
Reviewed-on: https://chromium-review.googlesource.com/322123
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
618bebcd
|
2016-01-15T16:40:00
|
|
Pass texture base level to shaders on D3D11
The base level is passed to shaders in an array included in the driver
uniform block. This is done on feature levels above 9_3, which treat
samplers as indices to sampler arrays in shaders.
A separate uniform block couldn't be used for the sampler metadata,
since that would bring the number of available uniform blocks down to
below minimum level defined by GLES 3.0.
BUG=angleproject:596
TEST=angle_end2end_tests
Change-Id: Ie040521402f9996d51a978aeeba9222e9dd761ce
Reviewed-on: https://chromium-review.googlesource.com/326290
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
2a63b3f8
|
2016-02-08T12:29:08
|
|
Re-land "Implement EGL_experimental_present_path_angle"
- Re-land with clang fix.
This allows ANGLE to render directly onto a D3D swapchain in the correct
orientation when using the D3D11 renderer.
The trick is to add an extra uniform to each shader which takes either
the value +1.0 or -1.0. When rendering to a texture, ANGLE sets this
value to -1.0. When rendering to the default framebuffer, ANGLE sets
this value to +1.0. ANGLE multiplies vertex positions by this value in
the VS to invert rendering when appropriate. It also corrects other
state (e.g. viewport/scissor rect) and shader built-in values
(e.g. gl_FragCoord).
This saves a substantial amount of GPU time and lowers power
consumption. For example, the old method (where ANGLE renders all
content onto an offscreen texture, and then copies/inverts this onto the
swapchain at eglSwapBuffers() time) uses about 20% of the GPU each frame
on a Lumia 630.
Verification:
+ dEQP GL ES2 tests pass when "present path fast" is enabled
+ all ANGLE_end2end_tests pass when "present path fast" is enabled
BUG=angleproject:1219
Change-Id: I56b339897828753a616d7bae837a2f354dba9c63
Reviewed-on: https://chromium-review.googlesource.com/326730
Tryjob-Request: Austin Kinross <aukinros@microsoft.com>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
d1c46228
|
2016-02-08T14:51:18
|
|
Revert "Implement EGL_experimental_present_path_angle"
Compile failure on Clang/Win:
The reason for reverting is: FAILED: ninja -t msvc -e environment.x86 --
"..\..\third_party/llvm-build/Release+Asserts/bin/clang-cl" -m32 /nologo
/showIncludes /FC
@obj\third_party\angle\src\tests\egl_tests\angle_end2end_tests.EGLPresentPathD3D11Test.obj.rsp
/c ..\..\third_party\angle\src\tests\egl_tests\EGLPresentPathD3D11Test.cpp
/Foobj\third_party\angle\src\tests\egl_tests\angle_end2end_tests.EGLPresentPathD3D11Test.obj
/Fdobj\gpu\angle_end2end_tests.cc.pdb
In file included from
..\..\third_party\angle\src\tests\egl_tests\EGLPresentPathD3D11Test.cpp:7:
In file included from
..\..\third_party\angle\src\tests\test_utils/ANGLETest.h:13:
..\..\testing\gtest\include\gtest/gtest.h(1392,16) : error: comparison of
integers of different signs: 'const int' and 'const unsigned int'
[-Werror,-Wsign-compare]
if (expected == actual) {
~~~~~~~~ ^ ~~~~~~
..\..\testing\gtest\include\gtest/gtest.h(1422,12) : note: in instantiation of
function template specialization 'testing::internal::CmpHelperEQ<int, unsigned
int>' requested here
return CmpHelperEQ(expected_expression, actual_expression, expected,
^
..\..\third_party\angle\src\tests\egl_tests\EGLPresentPathD3D11Test.cpp(281,9) :
note: in instantiation of function template specialization
'testing::internal::EqHelper<false>::Compare<int, unsigned int>' requested here
ASSERT_EQ(mWindowWidth * 4, mappedSubresource.RowPitch);
^
..\..\testing\gtest\include\gtest/gtest.h(1960,32) : note: expanded from macro
'ASSERT_EQ'
# define ASSERT_EQ(val1, val2) GTEST_ASSERT_EQ(val1, val2)
^
..\..\testing\gtest\include\gtest/gtest.h(1943,67) : note: expanded from macro
'GTEST_ASSERT_EQ'
EqHelper<GTEST_IS_NULL_LITERAL_(expected)>::Compare, \
^
BUG=angleproject:1219
This reverts commit 6b3c1db5170450bbc4946d8f18ba0d8619da43a0.
Change-Id: Ia67ab82dd13295dc03235d57fa417c73f20a49e6
Reviewed-on: https://chromium-review.googlesource.com/326680
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
6b3c1db5
|
2015-12-18T14:01:46
|
|
Implement EGL_experimental_present_path_angle
This allows ANGLE to render directly onto a D3D swapchain in the correct
orientation when using the D3D11 renderer.
The trick is to add an extra uniform to each shader which takes either
the value +1.0 or -1.0. When rendering to a texture, ANGLE sets this
value to -1.0. When rendering to the default framebuffer, ANGLE sets
this value to +1.0. ANGLE multiplies vertex positions by this value in
the VS to invert rendering when appropriate. It also corrects other
state (e.g. viewport/scissor rect) and shader built-in values
(e.g. gl_FragCoord).
This saves a substantial amount of GPU time and lowers power
consumption. For example, the old method (where ANGLE renders all
content onto an offscreen texture, and then copies/inverts this onto the
swapchain at eglSwapBuffers() time) uses about 20% of the GPU each frame
on a Lumia 630.
Verification:
+ dEQP GL ES2 tests pass when "present path fast" is enabled
+ all ANGLE_end2end_tests pass when "present path fast" is enabled
BUG=angleproject:1219
Change-Id: Ib6eeea46bafa6ebce4adada0ae9db3a433b8fc4c
Reviewed-on: https://chromium-review.googlesource.com/321360
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tryjob-Request: Austin Kinross <aukinros@microsoft.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
|
|
|
00140f4a
|
2016-02-03T18:47:33
|
|
Revert "Pass texture base level to shaders on D3D11"
This reverts commit 3026829e155bd89b5ca5b7b7c5267699b9192557.
Change-Id: I0b4c3f5b1453b993b149423bb1ce407f4918cf54
Reviewed-on: https://chromium-review.googlesource.com/325435
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
3026829e
|
2016-01-15T16:40:00
|
|
Pass texture base level to shaders on D3D11
The base level is passed to shaders in a uniform block created
specifically for passing sampler metadata. This is done on feature levels
above 9_3, which treat samplers as indices to sampler arrays in shaders.
BUG=angleproject:596
TEST=angle_end2end_tests
Change-Id: I846f2fc195ab1fd884052824ffd3c1d65083c0fb
Reviewed-on: https://chromium-review.googlesource.com/322122
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
9b4e8626
|
2015-12-22T15:53:22
|
|
Redesign samplers in shaders on D3D11
Translation of samplers to HLSL on D3D11 is changed as follows:
Instead of passing around HLSL sampler and HLSL texture references in
shaders, all references to ESSL samplers are converted to constant
indices within the shader body. Each ESSL sampler is identified by an
unique index. In the code generated to implement ESSL texture functions,
these indices are used to index arrays of HLSL samplers and HLSL
textures to get the sampler and texture to use.
HLSL textures and samplers are grouped into arrays by their types. Each
unique combination of a HLSL texture type + HLSL sampler type gets its
own array. To convert a unique sampler index to an index to one of these
arrays, a constant offset is applied. In the most common case of a 2D
texture and a regular (non-comparison) sampler, the index offset is
always zero and is omitted.
The end goal of this refactoring is to make adding extra metadata for
samplers easier. The unique sampler index can be used in follow-up
changes to index an array of metadata passed in uniforms, which can
contain such things as the base level of the texture.
This does not solve the issues with samplers in structs.
The interface from the point of view of libANGLE is still exactly the
same, the only thing that changes is how samplers are handled inside the
shader.
On feature level 9_3, the D3D compiler has a bug where it can report that
the maximum sampler index is exceeded when in fact it is not. This can
happen when an array of samplers is declared in the shader. Because of
this the new approach can't be used on D3D11 feature level 9_3, but it
will continue using the old approach instead.
BUG=angleproject:1261
TEST=angle_end2end_tests,
dEQP-GLES3.functional.shaders.texture_functions.* (no regressions)
dEQP-GLES3.functional.texture.units.* (no regressions)
Change-Id: I5fbb0c4280000202dc2795a628b56bd8194ef96f
Reviewed-on: https://chromium-review.googlesource.com/320571
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Commit-Queue: Olli Etuaho <oetuaho@nvidia.com>
Tryjob-Request: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
6c9b2ae9
|
2015-12-08T13:43:11
|
|
Revert "D3D11: Use clamp wrap mode for Integer textures."
dEQP-GLES3 texture failures:
http://build.chromium.org/p/chromium.gpu.fyi/builders/Win7%20Release%20dEQP%20%28NVIDIA%29/builds/4510/steps/angle_deqp_gles3_tests/logs/stdio
dEQP-GLES3.functional.shaders.texture_functions.texture.isampler*
dEQP-GLES3.functional.shaders.texture_functions.texture.usampler*
BUG=angleproject:1244
This reverts commit 7a6a1ffeb275a8b565701305c8b42857ff2bf0b0.
Change-Id: I0e25e7aed0e0d78015d5b8f5a7b9a81e0a5fca4e
Reviewed-on: https://chromium-review.googlesource.com/316641
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
8c46ab11
|
2015-12-07T16:39:19
|
|
OutputHLSL: Avoid using info sink stack as much as possible.
Due to how the traversers work, it might not be trivial to avoid the
info sink stack entirely, but minimize its use to keep the code as
functional as possible.
BUG=angleproject:958
Change-Id: I354313f10f496feea7c6a6f167b0617aeadbe5b0
Reviewed-on: https://chromium-review.googlesource.com/316412
Tryjob-Request: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
7a6a1ffe
|
2015-12-07T16:32:58
|
|
D3D11: Use clamp wrap mode for Integer textures.
Until we can support all wrap modes for int textures, use a correct
clamping scheme. This fixes several dEQP GLES3 FBO tests.
This shuffles some of the tests in functional.texture.units since
it hard-codes a different behaviour for all wrap modes, ignoring
the sampler setting.
BUG=angleproject:1244
TEST=dEQP-GLES3.functional.texture.*
Change-Id: Ic7e89a111728dfb18821534996bf5b9a5ad172b6
Reviewed-on: https://chromium-review.googlesource.com/313997
Tryjob-Request: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
8d9f35f1
|
2015-11-24T16:10:20
|
|
Re-land "D3D11: Fix Integer Texture Cube mip mapping."
We were missing both the correct SRV parameter, as well as the full
computation of the mip level in the HLSL.
Re-land makes the mip computation only happen with implicit sampling.
Before it would confuse the LOD0 computation.
BUG=angleproject:1208
TEST=dEQP-GLES3.functional.texture.*
Change-Id: I4b579033afe5cd1aca1f2d017e48a74c7fc324cc
Reviewed-on: https://chromium-review.googlesource.com/314330
Tryjob-Request: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
ee322723
|
2015-11-24T20:42:48
|
|
Revert "D3D11: Fix Integer Texture Cube mip mapping."
2 tests failed:
dEQP_GLES3.Default/functional_shaders_texture_functions_texture_isamplercube_vertex (c:\b\build\slave\gpu_win_builder\build\src\third_party\angle\src\tests\deqp_support\angle_deqp_gtest.cpp:234)
dEQP_GLES3.Default/functional_shaders_texture_functions_texture_usamplercube_vertex (c:\b\build\slave\gpu_win_builder\build\src\third_party\angle\src\tests\deqp_support\angle_deqp_gtest.cpp:234)
BUG=angleproject:1208
This reverts commit 5778557fe99e300f245a7bfc2200f6b0179e3fbf.
Change-Id: I6f7441bdbfbf937680b0d572645edee8ece649a1
Reviewed-on: https://chromium-review.googlesource.com/314301
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
5778557f
|
2015-11-23T09:37:44
|
|
D3D11: Fix Integer Texture Cube mip mapping.
We were missing both the correct SRV parameter, as well as the full
computation of the mip level in the HLSL.
BUG=angleproject:1208
Change-Id: I345b1e0895495654de3863c4ebcc395d9b7371b9
Reviewed-on: https://chromium-review.googlesource.com/312060
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tryjob-Request: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
e8e4debc
|
2015-11-18T17:15:38
|
|
Fix invalid generated sampling functions in HLSL output
In some generated shaders, "levels" would be read before it is assigned
by GetDimensions when clamping mip level to valid range. Fix these cases.
BUG=angleproject:1092
TEST=dEQP-GLES3.functional.shaders.texture_functions.*
(12 tests start passing)
Change-Id: I9fce8d378606738e5172673d222bce7968e26789
Reviewed-on: https://chromium-review.googlesource.com/312022
Tryjob-Request: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
18b9deb4
|
2015-11-05T12:14:50
|
|
Support writing initializers using HLSL literal syntax
Instead of using constructor functions to initialize variables, it is
better to use literal initializer syntax provided by HLSL when it is
possible. This way shader complexity is reduced and constant array
initialization doesn't have to go through as many AST transformations.
Before this patch, vec4 initialization would result in the following
kind of HLSL:
float4 f = float4(1.0, 2.0, 3.0, 4.0);
After this patch, it will be:
float4 f = {1.0, 2.0, 3.0, 4.0};
Before this patch, vec2 array initialization would result in the
following kind of HLSL:
float2 f[2] = {0, 0, 0, 0};
angle_construct_into_2_float2(f, float2(1.0, 2.0), float2(3.0, 4.0));
After this patch, it will be:
float2 f[2] = {1.0, 2.0, 3.0, 4.0};
BUG=angleproject:1094
BUG=541551
TEST=WebGL conformance tests
Change-Id: I9816a8d95a2cba3964922f6b561862d478da6145
Reviewed-on: https://chromium-review.googlesource.com/311160
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
b1edc4f5
|
2015-11-02T17:20:03
|
|
Accept const array initialization in shader parsing
Array constructors are not folded, unlike all other constant expressions.
Change initializer parsing path so that it accepts constant initializers
whether they are folded or not.
Some parts need to be adapted to work with expressions that are qualified
as constant but that are not necessarily folded:
1. Identifier parsing
2. Indexing parsing
3. Field selection parsing
4. HLSL output for variable declarations
5. Determining unary operator result type
6. Determining binary operator result type
7. Determining built-in function call result type
8. Determining ternary operator result type
Corner cases that are not supported yet:
1. Using array constructors inside case labels
2. Using array constructors inside array size expressions
3. Detecting when a negative constant expression containing an array
constructor is used to index an array
In these cases being able to constant fold the expression is essential to
validating that the code is correct, so they require a more sophisticated
solution. For now we keep the old code that rejects the shader if ANGLE
hasn't been able to constant fold the case label or array size. In case of
indexing an array with a negative constant expression containing an array
constructor, ANGLE will simply treat it as a non-constant expression.
BUG=541551
BUG=angleproject:1094
TEST=dEQP-GLES3.functional.shaders.constant_expressions.* (all pass),
angle_unittests
Change-Id: I0cbc47afd1651a4dece3d68acf7ec72a01fdf047
Reviewed-on: https://chromium-review.googlesource.com/310231
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
5f0246ca
|
2015-07-22T10:30:35
|
|
Implement gl_FragDepth for GLES SL 3.0
Makes it an error to access gl_FragDepthEXT in #version 300 es shader.
TODO:
Lacks the feature to make "#extension GL_EXT_frag_depth : require" an
error for #version 300 es.
Reland of: https://chromium-review.googlesource.com/#/c/287570
BUG=angleproject:1102
TEST=angle_unittest
Change-Id: I064d918d65f37539cb1e14f12173ca5591a4ea3f
Reviewed-on: https://chromium-review.googlesource.com/301711
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>
|
|
|
477b243b
|
2015-08-31T10:41:16
|
|
Change the FLATTEN heuristic to "ifs with a loop with a gradient"
This heuristic makes more sense than the previous "ifs with a
discontinuous loop" as the reason we need to flatten is that we need
gradients to be in branchless code.
Change the UnrollFlatten test accordingly.
Tested with:
- the WebGL CTS
- dev.miaumiau.cat/rayTracer "Skull Demo"
- THe turbulenz engine GPU particle demo
- Lots of ShaderToy Samples (inc. Volcanic, Metropolis and Hierarchical
Voronoi)
- Google Maps Earth mode
- Lots of Chrome experiments
- madebyevan.com/webgl-water
BUG=524297
Change-Id: Iaa727036fffcfde3952716a1ef33b6ee0546b69d
Reviewed-on: https://chromium-review.googlesource.com/296442
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
59f9a641
|
2015-08-06T20:38:26
|
|
Remove EOpInternalFunctionCall
It's cleaner to mark internal functions by using the TName class,
similarly to TIntermSymbol.
TEST=angle_unittests
BUG=angleproject:1116
Change-Id: I12a03a3dea42b3fc571fa25a1b11d0161f24de72
Reviewed-on: https://chromium-review.googlesource.com/291621
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
f5cfc8df
|
2015-08-06T16:36:39
|
|
Track whether a name is internal to ANGLE in a separate class
The AST contains identifiers in a few different places: besides symbols,
there are also function names, which show up in function signatures
and function calls. Any of these can be coming either from the original
shader or from inside ANGLE. A class that encapsulates a string and its
internalness will be useful for implementing a unified way of handling
all names in shader translation. Start implementing this by splitting
the functionality out of TSymbol.
TEST=angle_unittests
BUG=angleproject:1116
Change-Id: I0a1b5936dcccd0d5fc1c0c13c712102fbfff2a79
Reviewed-on: https://chromium-review.googlesource.com/291280
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
e2bfe2cd
|
2015-07-23T21:25:45
|
|
Revert "Implement gl_FragDepth for GLES SL 3.0"
assertion failures in WebGL2 CTS.
This reverts commit 544809610a131fe7040f2212789c62e212bbaf24.
Change-Id: I2e0c7045c5b6ef9031a6e6c5916504fe68f51077
Reviewed-on: https://chromium-review.googlesource.com/287910
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>
|
|
|
54480961
|
2015-07-22T10:30:35
|
|
Implement gl_FragDepth for GLES SL 3.0
Makes it an error to access gl_FragDepthEXT in #version 300 es shader.
TODO:
Lacks the feature to make "#extension GL_EXT_frag_depth : require" an
error for #version 300 es.
BUG=angleproject:1102
TEST=angle_unittest
Change-Id: Ic313a0e1ed6369550d53885290f300fab1aaf304
Reviewed-on: https://chromium-review.googlesource.com/287570
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
07e57df7
|
2015-06-16T16:55:52
|
|
Added non square matrix construction
Added new construction operations for
non square matrices, along with the
required changes to the related
translation functions.
Change-Id: I04ae7d4b2d1bb363b35d088cea45c0e7c4bc8a13
Reviewed-on: https://chromium-review.googlesource.com/277729
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>
|
|
5a0edc62
|
2015-06-30T12:36:07
|
|
Change intializer list order to match field order.
BUG=505304
Change-Id: I38149e72732f0b4eb7d398f90b5d11660e2a7fbc
Reviewed-on: https://chromium-review.googlesource.com/282465
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Geoff Lang <geofflang@chromium.org>
|
|
|
4785fec9
|
2015-05-18T16:09:37
|
|
Clean up most common extra semicolons and braces in HLSL output
Clean up extra semicolons and braces from loops and conditional statements
and extra braces from function definitions.
This makes the HLSL output considerably easier to read.
TEST=angle_unittests, angle_end2end_tests, dEQP-GLES3.functional.shaders.*,
WebGL conformance tests
BUG=angleproject:1013
Change-Id: I8180bab7b3d4bda1cdb8e4fb51cf9d8e384dd797
Reviewed-on: https://chromium-review.googlesource.com/273607
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
4d59f3c6
|
2015-05-28T17:33:53
|
|
Make false blocks produced by RewriteElseBlocks sequence nodes
All child nodes of selection should be sequence nodes, so that they will
output braces and extra braces can be removed from HLSL output.
Also make RewriteElseBlocks to reuse common IntermTraverser functionality
to simplify the code.
TEST=WebGL conformance tests on D3D9
BUG=angleproject:1013
Change-Id: Iafdc05468b22d110abcd020cf52c646dd98fb4a0
Reviewed-on: https://chromium-review.googlesource.com/273608
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
|
|
d94f6647
|
2015-05-18T16:25:12
|
|
Add ANGLE support for ESSL3 variant of mix()
This change adds support for ESSL3 variant of built-in function mix()
that takes last argument as a boolean vector. lerp() - HLSL equivalent
of mix() doesn't look to be supporting a variant that takes last
argument as a boolean vector so emulated it for HLSL.
BUG=angleproject:1006
TESTS=dEQP tests
Fixes "no matching overloaded function found " errors in below tests:
dEQP-GLES3.functional.shaders.constant_expressions.builtin_functions.common.mix_*_b*
(Note: These tests still fail because of constant expression issues)
Change-Id: I79b353933cb450516b8678b1fdaeabe60417e9a7
Reviewed-on: https://chromium-review.googlesource.com/271751
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
fd3b9be6
|
2015-05-18T17:07:36
|
|
Set mUsesDiscardRewriting more conservatively
This fixes a regression from recent refactoring of unfolding short-circuiting
operators. mUsesDiscardRewriting used to be set based on detecting patterns
that the AST traverser now overwrites before the code reaches OutputHLSL. With
this patch mUsesDiscardRewriting is set more conservatively, which ensures that
using discard works correctly also in case the input shader contains the same
kind of pattern as the short-circuiting operator unfolding creates.
TEST=WebGL conformance tests on D3D9
BUG=chromium:488552
Change-Id: Iea0103822a268079593b0913f0b6fc54c745c6e3
Reviewed-on: https://chromium-review.googlesource.com/271761
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
d81ed841
|
2015-05-12T12:46:35
|
|
Defer executing if statements in the global scope
Unfolding of short-circuiting operators (ternary and logical operators) may
create if statements in the global scope, which is not valid HLSL. Use existing
deferred global initialization function to defer execution of if statements in
the global scope.
TEST=WebGL conformance tests
BUG=angleproject:819
Change-Id: I2b0afcc6824dab6bb87eb6abed609e75b1384dab
Reviewed-on: https://chromium-review.googlesource.com/270461
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
a6f22096
|
2015-05-08T18:31:10
|
|
Make UnfoldShortCircuit to change AST instead of writing output
This is needed to make way for further AST transformations to handle array
expressions that need to work correctly together with unfolding short-
circuiting operators. This also improves the maintainability of HLSL output
by isolating the unfolding into a separate compilation step.
The new version of UnfoldShortCircuit traverser will traverse the tree until
an expression that needs to be unfolded is encountered. It then unfolds it and
gets reset. The traverser will be run repeatedly until no more operations to
unfold are found. This helps with keeping the traverser's design relatively
simple.
All declarations are separated to single declarations before short-circuit
unfolding is run. Previously OutputHLSL already output every declaration
separately.
BUG=angleproject:960
TEST=WebGL conformance tests, angle_unittests, angle_end2end_tests
Change-Id: Id769be396adbd4c0223e418980dc464dd855f019
Reviewed-on: https://chromium-review.googlesource.com/270460
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
|
|
|
6ba6eadc
|
2015-05-04T14:21:21
|
|
Rename ConstantUnion to TConstantUnion.
This clarified that we're using the Pool allocator/deallocator for
this type.
BUG=angleproject:993
Change-Id: If8c95f6054d07291e7014be0d4e35766ba2e943b
Reviewed-on: https://chromium-review.googlesource.com/269131
Tested-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
|
|
|
2e295e23
|
2015-04-29T10:41:33
|
|
hlsl: Fix struct specifiers in uniforms.
We would miss the definition for structs specfied in uniforms. Fix
this by always checking to add the constructor. Fixes the WebGL
test 'glsl/misc/struct-specifiers-in-uniforms'.
BUG=angleproject:818
BUG=433412
Change-Id: I411e4a4477f7ef34fceb9faa77489f77d8efdce8
Reviewed-on: https://chromium-review.googlesource.com/267797
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
a1884f2b
|
2015-04-29T10:15:16
|
|
Fixes for the tagging of discontinuous loops
The first fix was for all loops being considered discontinuous in
OutputHLSL because of a typo that produced a tautology.
The error was not detected by the unit tests because as an optimization we
do not generate Lod0 calls when they are not needed for the callee
function (which was correctly detected by the analysis in this case).
Fixed the unit tests by adding a call to a builtin gradient operation.
The second fix was for discard not being taken into account in the
analyses of the AST, which caused a WebGL test regression after the first
fix for conformance/glsl/bugs/conditional-discard-in-loop
BUG=angleproject:982
Change-Id: I1315eac1ad36f726be52d7fda5facf3104341b1f
Reviewed-on: https://chromium-review.googlesource.com/267814
Tested-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Geoff Lang <geofflang@chromium.org>
|
|
|
d57e0db3
|
2015-04-24T15:05:08
|
|
Remove separate compilerdebug.h in favor of debug.h
This unifies the behavior across the compiler and rest of ANGLE - for
example, one can use #define ANGLE_TEST_CONFIG to disable UNIMPLEMENTED
asserts in both the compiler and the rest of ANGLE. Compiler traces from
asserts also go to the same TRACE_OUTPUT_FILE as other traces instead of
being directed through ParseContext.
The compiler build already includes the common sources, so no changes to
build config are needed.
The original version of this change was reverted due to release mode
build issues. This version adds UNUSED_ASSERTION_VARIABLE where needed.
TEST=angle_unittests, angle_end2end_tests,
dEQP-GLES3.functional.shaders.*
BUG=angleproject:983
Change-Id: I36929020a04251b8bc834fbb3c069e10128c3082
Reviewed-on: https://chromium-review.googlesource.com/267411
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Olli Etuaho <oetuaho@nvidia.com>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
a656490d
|
2015-04-27T14:30:32
|
|
Revert "Remove separate compilerdebug.h in favor of debug.h"
Causing compile warnings in Release:
1>compiler\translator\CallDAG.cpp(238): error C2220: warning treated as error - no 'object' file generated
1>compiler\translator\CallDAG.cpp(238): warning C4189: 'op' : local variable is initialized but not referenced
1>compiler\translator\IntermNode.cpp(1495): error C2220: warning treated as error - no 'object' file generated
1>compiler\translator\IntermNode.cpp(1495): warning C4189: 'replaced' : local variable is initialized but not referenced
1>compiler\translator\IntermNode.cpp(1517): warning C4189: 'replaced' : local variable is initialized but not referenced
This reverts commit 5271025865b34685da71d0309131c5aff2e32f71.
Change-Id: Icdf1c37eef22a13d083767609ab0b0285d3dc517
Reviewed-on: https://chromium-review.googlesource.com/267359
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Jamie Madill <jmadill@chromium.org>
|
|
|
52710258
|
2015-04-24T15:05:08
|
|
Remove separate compilerdebug.h in favor of debug.h
This unifies the behavior across the compiler and rest of ANGLE - for
example, one can use #define ANGLE_TEST_CONFIG to disable UNIMPLEMENTED
asserts in both the compiler and the rest of ANGLE. Compiler traces from
asserts also go to the same TRACE_OUTPUT_FILE as other traces instead of
being directed through ParseContext.
The compiler build already includes the common sources, so no changes to
build config are needed.
TEST=angle_unittests, angle_end2end_tests,
dEQP-GLES3.functional.shaders.*
BUG=angleproject:983
Change-Id: Ifca4d16f667b1e5cf9c2e7cc4139940091917a1c
Reviewed-on: https://chromium-review.googlesource.com/266993
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
a8c414ba
|
2015-04-16T15:51:03
|
|
Add basic support for arrays as return values in HLSL output
In HLSL output, user-defined functions that have an array as their return value
get changed so that they have the array as an out parameter instead.
Still missing: support for calling a function that has array as a return value
without assigning the array. Also support for assignments inside complex
expressions.
TEST=dEQP-GLES3.functional.shaders.arrays.return.*
BUG=angleproject:941
Change-Id: I79f5170139116a3dcfb2be2df5f0f79a3d955ca8
Reviewed-on: https://chromium-review.googlesource.com/266003
Reviewed-by: Olli Etuaho <oetuaho@nvidia.com>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
78174db7
|
2015-04-21T16:14:00
|
|
Replace EvqInternal with a separate flag to make it more flexible
The internal flag disables decorating a given symbol in output, effectively
placing it to a different namespace than user-defined symbols. This enables the
compiler to insert symbols to the tree when transforming it to be suitable for
HLSL output without running into name conflicts. In this patch the flag is
separated from the qualifiers since sometimes different qualifiers need to be
used with these internal symbols.
TEST=angle_unittests, angle_end2end_tests, WebGL conformance tests
BUG=angleproject:941
Change-Id: I7036bed98fdb1478a383bb959ca03b42c3cb8100
Reviewed-on: https://chromium-review.googlesource.com/266690
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Tested-by: Olli Etuaho <oetuaho@nvidia.com>
|
|
|
1239ee94
|
2015-03-19T14:38:02
|
|
Use the AST analyses to narrow the usage of [[loop]] and [[unroll]]
These attributes are now used exactly in the loops and ifs
that require them, limiting the number of failed compilations
due to excessive unrolling and flattening.
Also output Lod0 functions only when needed.
Adds unit tests for LOOP, FLATTEN and Lod0 generation.
The patch was tested against the WebGL CTS 1.0.4 for which all the
failures existed prior to this patch and seem to be unrelated to this
change. It also works correctly on the following sites that had trouble
with [[loop]] and [[unroll]]:
* dev.miaumiau.cat/rayTracer "Skull Demo"
* The turbulenz engine particle demo
* Lots of ShaderToy samples (including "Volcanic" and "Metropolis")
* Google Maps Earth mode
* Lots of Chrome Experiments
* Lagoa
* madebyevan.com/webgl-water
* SketchFab
* Unit Tests
BUG=angleproject:937
BUG=395048
Change-Id: I856de9025f10b79781929ec212dbffc2064a940e
Reviewed-on: https://chromium-review.googlesource.com/264791
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
938f0029
|
2015-04-08T19:35:40
|
|
Revert "Use the AST analyses to narrow the usage of [[loop]] and [[unroll]]"
Caused linking failures on mac because the unit test asks for TranslatorHLSL which is not compiled.
This reverts commit 3342e01f2a29343ea95961f0194f9d4f422cb840.
Change-Id: I02b2f54ca5b90611f11b7a549e75bf2e8310639d
Reviewed-on: https://chromium-review.googlesource.com/264790
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
3342e01f
|
2015-03-19T14:38:02
|
|
Use the AST analyses to narrow the usage of [[loop]] and [[unroll]]
These attributes are now used exactly in the loops and ifs
that require them, limiting the number of failed compilations
due to excessive unrolling and flattening.
Also output Lod0 functions only when needed.
Adds unit tests for LOOP, FLATTEN and Lod0 generation.
The patch was tested against the WebGL CTS 1.0.4 for which all the
failures existed prior to this patch and seem to be unrelated to this
change. It also works correctly on the following sites that had trouble
with [[loop]] and [[unroll]]:
* dev.miaumiau.cat/rayTracer "Skull Demo"
* The turbulenz engine particle demo
* Lots of ShaderToy samples (including "Volcanic" and "Metropolis")
* Google Maps Earth mode
* Lots of Chrome Experiments
* Lagoa
* madebyevan.com/webgl-water
* SketchFab
* Unit Tests
BUG=angleproject:937
BUG=395048
Change-Id: If7baddae1cdae0b3a414aa49e5a4c4babedbfe50
Reviewed-on: https://chromium-review.googlesource.com/261263
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
f4eab3b9
|
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: Ib8d7b98431b8cd3563e1eff8ecc9ed5df1a9b7d6
Reviewed-on: https://chromium-review.googlesource.com/263775
Reviewed-by: Jamie Madill <jmadill@chromium.org>
Tested-by: Corentin Wallez <cwallez@chromium.org>
|
|
|
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>
|
|
|
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>
|
|
|
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>
|
|
|
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>
|
|
|
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>
|
|
|
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>
|
|
|
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>
|
|
|
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>
|