create() is currently just a simple forwarder to std::make_unique<>, but
will be later replaced with a function that returns a raw pointer,
and owned by the context.
Bug: tint:322
Change-Id: I9d85e925538789d9b58f32c2bba32a05e22aea1c
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32863
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
When extracting values out of constant arrays we sometimes generate a
temporary variable. That generation was over eager in the creation of
the variable and was also creating it for constant vector extractions
where it was providing the wrong souce type.
Bug: tint:318
Change-Id: I8d16182fd1fcf7d7aba0b0e1b7d947137efc136b
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32801
Reviewed-by: Sarah Mashayekhi <sarahmashay@google.com>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Auto-Submit: dan sinclair <dsinclair@chromium.org>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
When building a const variable from a pointer we need to make sure to
load the value and use the loaded ID as the constant Id.
Bug: tint:310
Change-Id: Ia544fd69f3d2ae13e9ff9a983935ddc332d8d6ff
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32800
Reviewed-by: Sarah Mashayekhi <sarahmashay@google.com>
Reviewed-by: David Neto <dneto@google.com>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: David Neto <dneto@google.com>
Auto-Submit: dan sinclair <dsinclair@chromium.org>
In a near-future change, AST nodes, such as ast::BlockStatement will no longer
be std::unique_ptrs, and will have to be constructed and owned by an external
class. This means AST nodes can no longer allocate default child nodes.
Bug: tint:322
Change-Id: I3db9b3c037896f07b84b14b7b8d4da0f066b69b0
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32679
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
In a near-future change, AST nodes, such as ast::BlockStatement will no longer
be std::unique_ptrs, and will have to be constructed and owned by an external
class. This means AST nodes can no longer allocate default child nodes.
Bug: tint:322
Change-Id: Iddb5605b9bc0de80ad2710ced0e429f89410af2f
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32675
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
create() is currently just a simple forwarder to std::make_unique<>, but
will be later replaced with a function that returns a raw pointer,
and owned by the context.
Bug: tint:322
Change-Id: I5321553847b6a7d47ac211ba093d219c7f3bb9bd
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32673
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
This CL adds the missing load of conditional values.
Bug: tint:327
Change-Id: I836ecfacb3a237a54886ebc7625c9449ba33fdc7
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32700
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: David Neto <dneto@google.com>
This CL moves the namer into the context object and makes it a parameter
to the various generators. The old constructor is maintained until we've
updated downstream repos.
Bug: tint:273
Change-Id: I49b2519c4250be21fb73374b16e7c702b727078f
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32580
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
This Cl adds an ast::Builder class which provides a series of helper
methods to make creating AST nodes simpler.
Change-Id: Ife57f27e598d575681f7192d65fab968191699b1
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32560
Reviewed-by: David Neto <dneto@google.com>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
This CL rolls the entry point name change back for the spirv-backend to
allow Dawn to continue to work.
We'll need to roll this together with the inspector change, MSL and HLSL
changes.
Bug: tint:273
Change-Id: I54e7e05a4c277d6c55bd6808f50fb0c7903bc418
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32500
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Add helper methods for constructing `CallExpressions`, which is tediously verbose.
Change-Id: Idb6fbf25e79f07f4108df224fedb88f3fe3fb392
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32401
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Move common Context, Module, TypeDeterminer, Builder and common types into a test base class.
Add helpers for making common types.
Rename `BuilderTest` to `IntrinsicBuilderTest`, otherwise it will collide with tests in another .cc.
Change-Id: I2badbbb381336ddd52aef72811da8adaebdeaf31
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32400
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
This Cl hashes the OpName, OpEntryPoint and OpMemberName strings so we
are no longer passing user provided strings through into the resulting
SPIR-V binary.
Bug: tint:273
Change-Id: I0ca2c65d0cd2800c54d867ab698c7751c341778c
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32061
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
This CL updates Tint to use the GLSL450 memory model.
Bug: tint:63
Change-Id: Ia32851fbf8ee71233977a00386af5a91cb4b4b58
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31601
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
This CL fixes the issue with duplicate globals being emitted in HLSL if
used in multiple entry points. Tests are added for the other backends to
verify the issue does not exist there.
Bug: tint:297
Change-Id: I16d7504e8458fd375c6e1896758fe180ad963871
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31880
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Once a `Decoration` has been parsed, it'll be placed into a `DecorationList` and validated later in the parse. In order to create error diagnostics that refer back to the decoration, we need to know its source.
Bug: tint:282
Bug: tint:291
Change-Id: I38de708adbd041601b61d7e0a4d0402e9a2fe526
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31722
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
This is the first step in unifying the way decorations are parsed - i.e. instead of parsing decorations in different ways based on the predicted grammar that follows, we can parse decorations blocks in a unified way, then later verify what we have is as expected.
`StructDecoration` has been transformed from an `enum class` to a proper class so it can derive from `Decoration`.
Bug: tint:282
Bug: tint:291
Change-Id: Iaf12d266068d03edf695acdf2cd21e6cc3ea8eb3
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31663
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Commit-Queue: Ben Clayton <bclayton@google.com>
This CL fixes up the various lint errors.
Change-Id: If4d3077b55aadec33980452c43917194d803fac6
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31680
Reviewed-by: Ben Clayton <bclayton@google.com>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
When generating a unary operation we need to make sure we correctly
generate the load otherwise the resulting SPIR-V will be invalid. This
CL adds the required GenerateLoadIfNeeded call into the unary
generation.
Bug: tint292
Change-Id: Ia04314726afdda8f63a78e8e52f996681373db6e
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31620
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: David Neto <dneto@google.com>
This CL adds emission of the decorations for the access control flags.
Bug: tint:208 tint:108
Change-Id: I3286132dad8edd2586228dc6e87749ad49451739
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31082
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
This CL merges the StorageAccess enum with the AccessControl enum. The
enum is moved up to src/ast and placed in its own file for clarity.
Change-Id: I95a905a399b5d2e046ea1ea429b35f2064510c2d
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31242
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: David Neto <dneto@google.com>
With the addition of the AccessControlType we want to look through the
access control as well as the aliases as we work through the type tree.
This CL renames UnwrapAliasesIfNeeded to be UnwrapIfNeeded and
UnwrapAliasPtrAlias to UnwrapAll.
Change-Id: I5b027919c3143a89be24c4d87b8106f70358c03b
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31104
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: Sarah Mashayekhi <sarahmashay@google.com>
Reviewed-by: David Neto <dneto@google.com>
spirv-writer: treat function parameter as const
Bug: tint:274
Change-Id: Iba4d90ea10a70ce1f551b7b8da6efcc01fe97412
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/30924
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Both sampler types generate an `OpTypeSampler` in SPIR-V so we need to
make sure they're considered duplicates in the SPIR-V backend. This CL
registers the generated ID with for both sampler type_names when either
one is emitted.
Bug: tint:272
Change-Id: If459bbb34fe8670a7e29d101686b70bf83b184c2
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/30662
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: David Neto <dneto@google.com>
This CL adds the parsing of structs with names. The parsing of type
aliased structs remains to allow for migration to the new system. The
named struct format is always emitted.
Bug: tint:175
Change-Id: Ic0579dedbd2dd0edc7dfd30bc2ec02972091e718
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/30341
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
This CL moves the StructType name into the constructor of the struct
type instead of receiving through an accessor. The |set_name| accessor
is removed as it should not be needed anymore. All call sites have been
updated.
The vertex pulling transform was fixed to correctly register an alias
for the structure being created so it will be emitted.
Bug: tint:175
Change-Id: I8802931d9bdbc6c2f12982eea9032931939d195c
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/30280
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
This Cl updates the identifier type determination check to fail if the
identifier is not found and is not an intrinsic method.
Bug: tint:139
Change-Id: I332dd7fb42dae62bdee459c4a8819bdb5685c903
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/30081
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: Sarah Mashayekhi <sarahmashay@google.com>
The change from `cast` to type constructor casts causes our current
determination if a constructor is constant to no longer be correct.
This Cl updates the determination to match the current spec and adds a
bunch of unit tests to verify the behaviour..
Bug: tint:270
Change-Id: I8ce74eb7c3f849ce62815868313449d8ca2de6be
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/30020
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Sarah Mashayekhi <sarahmashay@google.com>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Currently, if we access an array of non-scalar items we'll incorrectly
emit an OpVectorExtractDynamic which will fail as the result is not
scalar.
This CL updates the array accessor code such that if the base array is
an array of non-scalars we'll do load of the array and then access chain
into the loaded variable.
Bug: tint:267
Change-Id: Ia4d7052b57d8b31b835714b7b922c7859e3dce1f
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/29844
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: David Neto <dneto@google.com>
The push_preamble method was dealing with multiple sections of the
SPIR-V binary layout. As we changed the way things write (like
extensions getting written later) the preamble section was ending up in
incorrect order.
This CL replaces push_preamble with push methods for each of the
sections at the start of the SPIR-V module which should fixup the
ordering issue.
Bug: tint:267
Change-Id: Ib73a66d0fdb2c67dd6e80582289dd18475fad9f9
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/29841
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
This CL updates the WGSL parser to allow array decorations to accept
multiple blocks. The stride decoration on arrays was turned into a
proper decoration object instead of just storing the stride directly.
Bug: tint:240
Change-Id: I6cdc7400d8847e3e043b846ea5c9f86cb795cf86
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/29780
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: Sarah Mashayekhi <sarahmashay@google.com>
Reviewed-by: David Neto <dneto@google.com>
This CL updates to match the spec change allowing multiple struct
decorations.
Bug: tint:240
Change-Id: Id859c6a331c67c46597fc3c70de06d6cc0f486ec
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/29260
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
This CL adds support for retrieving the array length of a Runtime Array
in the SPIR-V backend.
Bug: tint:252
Change-Id: Ic13c4a99da5760738d57702c45f52c6a194a172d
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/29220
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: David Neto <dneto@google.com>
Currently if a constructor contains constructors we consider it const.
This falls down with the new type constructor syntax if the types don't
match. In the case they don't match we no longer consider the
constructor const as we'll generate OpBitcast and OpCopyObject
instructions which we need to build the composite from.
Bug: tint: 263
Change-Id: Ic85f58c8410e862a2ec30c7d93c9b87a61822f6d
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/29523
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
All of the OpEntryPoint declarations must come before OpExecutionMode.
Currently if you have multiple fragment shaders we'll interleave the
OpEntryPoint and OpExeutionMode which will fail to validate.
Bug: tint:263
Change-Id: I7c925cf6b5345c03bfaf1aa15115caa1bdb9af4c
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/29522
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
In WGSL you can provide a constant_id variable without a constructor. In
SPIR-V we must synthesize a constant to attach the SpecId too. This CL
adds that variable creation.
Bug: tint:254
Change-Id: I2f25fdc3cb7e2c9c0f9e2129885865bd24298416
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/29200
Commit-Queue: Sarah Mashayekhi <sarahmashay@google.com>
Reviewed-by: Sarah Mashayekhi <sarahmashay@google.com>
This CL adds OpSpec constants to the SPIRV-Writer.
Bug: tint:151
Change-Id: I309013ca0b4cb514edd92fab3dab2e4faa15969a
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/29101
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: David Neto <dneto@google.com>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
This CL adds a test for struct constructors in the SPIR-V backend.
Change-Id: I9ae82ce86cde159b87cce2ee95d0b16910aadc7f
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28901
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: David Neto <dneto@google.com>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
This CL emits array type constructors in the SPIR-V backend.
Change-Id: I796e81964df1af39ad1aacdd4ab8181852f661fa
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28900
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
This CL adds the matrix constructors to the SPIR-V output.
Change-Id: I7471700c262b499f843ffa4e3c49bcac0b0c7565
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28881
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
This CL removes the cast operator and converts the tests over to using
type constructors.
Bug: tint:241
Change-Id: I2526acb61f5624b2e1c068612a2ddcc748c92aed
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28860
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
This CL updates the SPIR-V writer to emit the arithmetic right shift if
the LHS of the expression is a signed scalar or vector.
Bug: tint:84
Change-Id: I4ca33a31783e1954515db5f12b2cf1d364aedee4
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28940
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
The ! operator should emit an OpLogicalNot instead of an OpNot.
Bug: tint:248
Change-Id: I6e404c3286db7819e0692603e3cf9a87a1b37ed9
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28841
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
This CL updates Tint to use 'bitcast' instead of 'as' for the OpBitcast
conversions.
Bug: tint:241
Change-Id: I53a80de10421b2d9cc009527eebe5ff07e1285c2
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28801
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: David Neto <dneto@google.com>
This CL removes the import mechanism, the identifier paths and updates
all of the standard library methods to be builtins.
Bug: tint:242
Change-Id: If09b98a155ae49ced3986ba2c9b517a060693006
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28720
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
This CL removes the EntryPoint node and transitions everything to the
stage decoration.
Change-Id: Ib2840155905c8fa60ff35870f0c4b6705efb73ff
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28705
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
Reviewed-by: Sarah Mashayekhi <sarahmashay@google.com>
This CL updates the SPIRV-Writer to emit entry point information based
on the function stage as well as EntryPoint nodes.
Change-Id: I1fa937cbb2159b31516b0189216d679e03f0384d
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28702
Reviewed-by: David Neto <dneto@google.com>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
This CL adds a test for multisampled textureLoad and fixes up a few type
determiner issues for multisampled textures.
Bug: tint:237
Change-Id: I5c33797b197b6f092842b22aa93d2076b0779766
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28582
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
This CL removes the complete coverage of the possible input formats
infavor of simpler testing of a few of the inputs. This should be
sufficient coverage and makes the tests easier to read.
The incorrect OpConstantNull for the samplers and textures has also been
removed.
Change-Id: I9cd7d6299f0454bb8e2019968e0422221e4ec183
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28581
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
This CL removes the instantiated test suites, instead it tests each of
the required formats with one texture type and then uses the rest of the
tests to verify the texture type generated is correct.
This makes the tests a lot easier to read as the string generation for
results is removed.
Change-Id: I2c54471a2afc8bf2602ec8e4b50b27ad078dc4ee
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28580
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Ben Clayton
dan sinclair
dan sinclair
David Neto