Also ensure that the number of texture coordinates is exactly
the right number required for the given texture dimension.
I think SPIR-V is looser in this respect.
Assumes coordinates are floating point.
Bug: tint:349
Change-Id: I4512c333fada3647c66f13ef31897b2d73b46cf0
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/33982
Reviewed-by: Ben Clayton <bclayton@google.com>
Commit-Queue: David Neto <dneto@google.com>
Auto-Submit: David Neto <dneto@google.com>
- Emit (non-depth) sampler variables
- Emit sampled texture variables
- Test emission of textureSample, textureBias, textureLevel
TODO: convert unsigned offset parameter to signed. crbug.com/tint/348
TODO: support arrayed access, where we have to split out the array index
into a separate operand. crbug.com/tint/349
TODO: for explicit-lod sampling, we may have to convert coordinates to
floating point. crbug.com/tint/346
Bug: tint:109
Change-Id: I12558f99473ca234ce0d09a87fc0c2f4730497bc
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/33342
Reviewed-by: Ryan Harrison <rharrison@chromium.org>
Commit-Queue: Ryan Harrison <rharrison@chromium.org>
Auto-Submit: David Neto <dneto@google.com>
Migrate all uses to use this and the new `unique_type<T>()` and `types()` methods.
Remove the `type_mgr()` accessor. `TypeManager` is now an implementation detail of the module, allowing us to unify the allocation of types and nodes (if we so wish).
Fixes: tint:337
Bug: tint:307
Change-Id: I233fa9dc73d60515dd721f02ea7ba089ef7d374f
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/33667
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: David Neto <dneto@google.com>
Remove all redundant std::move()s. I've also removed calls to
std::move() in tests, even if they act as an optimization. This is for
two reasons:
(a) Performance is not important for testing, and this helps with
readability.
(b) A whole bunch tests were relying on std::move() clearing vectors so
they can be repopulated and used again. This is undefined behavior:
> Objects of types defined in the C++ standard library may be moved from
> (12.8). Move operations may be explicitly specified or implicitly
> generated. Unless otherwise specified, such moved-from objects shall
> be placed in a valid but unspecified state.
All of these UB cases have been fixed.
Removed all duplicate variables left over from:
`auto* foo_ptr = foo.get()`
which became:
`auto* foo_ptr = foo`
Bug: tint:322
Change-Id: Ibd08a2379671382320fd4d8da296ccc6a378b8af
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32900
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
This is a minimal effort to fix up the code. There's substantial code
cleanup which can now be done, which is done in the next change.
Bug: tint:322
Change-Id: Iafcf5e814837d9534889e8c21333de4931a19cfa
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32864
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: I72558482c4b6aff7a655087ee010b3e16b006192
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32860
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: I36a1cf55c31a1dabccde272b2be415f98c16b18f
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32677
Commit-Queue: Ben Clayton <bclayton@google.com>
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: David Neto <dneto@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: I2a571d0a4727d6dc3d6c38e8b6602e131292f49c
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/32676
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Commit-Queue: 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>
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>
Translate OpSpecConstantTrue, OpSpecConstantFalse, and OpSpecConstant.
The latter only can be used with integer or float scalars.
If the constant has a SpecId decoration, then generate a module-scope
decorated constant. Otherwise generate a module-scope constant without
decorations.
Register the ID so we know to use the declared const identifier in
expressions later in the module.
Bug: tint:156
Change-Id: Icd6e9b60225ced7ee99963c4f85cec1eb0e3ae6b
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31541
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
This is the first step in improving the error messages produced while parsing.
The `line` and `column` information of `Source` has been moved to `Source::Location`.
`Source::Range` has been added that contains a `Location` interval - allowing error messages to highlight the full region of the error.
The `File` information provides an optional file path, and pre-splits the content into lines. These lines can be used to print the full line containing an error.
This CL contains a few temporary changes that help split up this work, and to ease integration with Tint.
Bug: tint:282
Change-Id: I7aa501b0a9631f286e8e93fd7396bdbe38175727
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/31420
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Reviewed-by: David Neto <dneto@google.com>
Commit-Queue: David Neto <dneto@google.com>
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 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 adds the emission of StageDecoration to entry point functions.
EntryPoint nodes are still emitted. We duplicate the function emission
if there are multiple entry points pointing to the same function.
Change-Id: Icb48a063f5c6a30948bbe2c37c7fce7431af5864
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/28665
Reviewed-by: David Neto <dneto@google.com>
Reviewed-by: Sarah Mashayekhi <sarahmashay@google.com>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
Set source location on function variable declarations, const
declarations, and most normal instructions.
Bug: tint:3
Change-Id: I469afcdf1b7d8f6d1e64617189a6fa329056737f
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/27700
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
This is required so we actually emit values that are only used
in phis.
Bug: tint:3, tint:215
Change-Id: I9d957a697839b8d09246905c3f28064f0bc01731
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/27701
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
If a definition occurs in the first block of an IfSelection or
SwitchSelection, then it should count as belonging to the scope of the
parent of that if or switch selection. This prevents some bad hoisting
decisions.
Bug: tint:3, tint:213
Change-Id: I89df5e8cbc163577e19e78c2bf393eb1eec4a0aa
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/27660
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Commit-Queue: David Neto <dneto@google.com>
Follow the actual SPIR-V type when computing a composite extract
instad of the canonicalized view in the optimizer's type manager.
Do this so we can generate the correct member name for a struct,
rather than using the member name for the other representative
struct type. The optimizer's type canonicalizer is insensitive to
struct member names.
Prompted by tint:213, for which the original case was an
access chain.
Bug: tint:3, tint:213
Change-Id: I8705c7ee655fe47c8b7a3658db524fe18833efdb
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/27603
Commit-Queue: David Neto <dneto@google.com>
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Follow the actual SPIR-V type when computing an access chain expression,
instead of the canonicalized view in the optimizer's type manager.
Do this so we can generate the correct member name for a struct,
rather than using the member name for the other representative
struct type. The optimizer's type canonicalizer is insensitive to
struct member names.
Fixes tint:213
Bug: tint:3, tint:213
Change-Id: I88ec42a4cb049b011a59d5522e4cb39bc181a4fb
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/27602
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Commit-Queue: dan sinclair <dsinclair@chromium.org>
This CL updates the SPIR-V Reader to create BlockStatements instead of
StatementLists.
Bug: tint:136
Change-Id: I957019446ca00306187de701f86ae3e0dd5c5eb8
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/25740
Reviewed-by: Sarah Mashayekhi <sarahmashay@google.com>
This CL removes the KillStatement from the AST and replaces all test
usage with DiscardStatement.
Bug: tint:169
Change-Id: Ie68dd3cdd54056f144d10506f05cc1f6903d1cda
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/25605
Reviewed-by: David Neto <dneto@google.com>
This CL changes the SPIRV-Reader to generate DiscardStatements instead
of KillStatements for an OpKill.
Bug: tint:168
Change-Id: Ic4223b614c822c6ad449f61aedaddc9605c42535
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/25642
Reviewed-by: David Neto <dneto@google.com>
Still TODO: OpSelect over arrays and structures, as permitted in SPIR-V 1.4
Bug: tint:3, tint:99
Change-Id: I70f6c8a43ea3339cd715813c6eb0128d66ff0df8
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/25301
Reviewed-by: dan sinclair <dsinclair@chromium.org>
This CL adds support for handling uniform data. Currently the uniform is
added to a buffer where the number is the binding value. This will need
to be updated to accept the correct mapping from the embedder.
Bug: tint:8
Change-Id: Icccccbe599a9555defa6136e384745f4093df020
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/25104
Reviewed-by: David Neto <dneto@google.com>
Old way:
- struct decorated with BufferBlock
- Uniform storage class
New way
- struct decorated with Block
- StorageBuffer storage class
Also fixes the result type for an access chain.
Bug: tint:99
Change-Id: I2324ba94bb19b369d206313de798bdfec6099fe0
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24605
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Emits it as a module-level variable. Deconstruct and throw away
the gl_PerVertex struct.
Not handled: unusual patterns that are technically valid but
which don't occur in practice:
- loading, storing, or producing intermediate values of the whole structure.
- multiple definitions of the per-vertex structure (e.g. if someone had
put both a vertex shader and a tessellation shader in the same
module.)
Bug: tint:3, tint:99
Change-Id: I3ad9ff6ab780a002367f01f385bfa7d6ddba6db9
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24880
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Take the type of a const-definition (or a write to a hoisted variable)
from the generated expression rather than the directly-converted type
from the SPIR-V ID.
This prepares the way for remapping storage buffers from
Uniform/BufferBlock to StorageBuffer/Block representation.
Bug: tint:99
Change-Id: I6104d8ad96312053562468a3cdb247932c1cb6cb
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24604
Reviewed-by: dan sinclair <dsinclair@chromium.org>
Use the fact that in WGSL the scope corresponding to the loop construct
encloses the scope for its associated continue construct.
In our construct data structure, the two are adjacent but not
overlapping.
This improvement means that when a definition is in a loop construct,
but used only in the loop or associated continue construct, then no
hoisting is required.
Bug: tint:3
Change-Id: I8d33b8f76303ab2868306847e846b4c26899e746
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24420
Reviewed-by: dan sinclair <dsinclair@google.com>
When we hoist a variable out of a continue construct, put it
in associated loop construct, if it exists. This reduces its
lifetime in WGSL, and easier to understand as a code reader.
Change-Id: I8f0cc37640bfe67874cbc27b55029e79e9a8992c
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24321
Reviewed-by: dan sinclair <dsinclair@google.com>
The special case is not "single-block-loop" but rather the
case where a continue target is also its own loop header.
This can occur for single-block loops and multi-block loops.
Bug: tint:3
Change-Id: I4af1410793caf8b26a1f781e221fc0b395f07aa3
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24320
Reviewed-by: dan sinclair <dsinclair@google.com>
For each OpPhi, make a variable to carry values from predecessor blocks
to the OpPhi. Declare the variable at the smallest scope enclosing all
the predecessor blocks (where we write to it), and the OpPhi (where we
read from it).
Bug: tint:3
Change-Id: I7898b4b903d9ee1a25a7466e3c5aaf6840550e2d
Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/24181
Reviewed-by: dan sinclair <dsinclair@google.com>
