89 Commits

Author SHA1 Message Date
Lei Zhang
8656af1e82 [spirv] Use LLVM graph traversal utility for PrettyBlockOrderVisitor
This removes a bunch of special tailored DFS code in favor of the common
LLVM utility. Besides, we avoid recursion with system stack given that
llvm::depth_first_ext is iterator based and maintains its own stack.
PiperOrigin-RevId: 277272961
2019-10-29 07:04:00 -07:00
Lei Zhang
ca2538e9a7 [spirv] Support OpPhi using block arguments
This CL adds another control flow instruction in SPIR-V: OpPhi.
It is modelled as block arguments to be idiomatic with MLIR.
See the rationale.md doc for "Block Arguments vs PHI nodes".
Serialization and deserialization is updated to convert between
block arguments and SPIR-V OpPhi instructions.

PiperOrigin-RevId: 277161545
2019-10-28 15:58:42 -07:00
Kazuaki Ishizaki
8bfedb3ca5 Fix minor spelling tweaks (NFC)
Closes tensorflow/mlir#177

PiperOrigin-RevId: 275692653
2019-10-20 00:11:34 -07:00
Lei Zhang
c5b9fefddc NFC: Rename SPIR-V serializer find*ID() to get*ID() to be consistent
We use get*() in deserizer and other places across the codebase.

PiperOrigin-RevId: 275582390
2019-10-18 18:16:05 -07:00
Lei Zhang
e03e151983 [spirv] Add support for SpecId decoration on spv.specConstant
The SpecId decoration is the handle for providing external specialization.
Similar to descriptor set and binding on global variables, we directly
bake it into assembly parsing and printing.

PiperOrigin-RevId: 274893879
2019-10-15 14:53:30 -07:00
Mahesh Ravishankar
9e9c3a009a Update UndefOp (de)serialization to generate OpUndef at module level.
The SPIR-V spec recommends all OpUndef instructions be generated at
module level. For the SPIR-V dialect its better for UndefOp to produce
an SSA value for use with other instructions. If UndefOp is to be used
at module level, it cannot produce an SSA value (use of this SSA value
within FuncOp would need implicit capture). To satisfy needs of the
SPIR-V spec while making it simpler to represent UndefOp in the SPIR-V
dialect, the serialization is updated to create OpUndef instruction
at module scope.

PiperOrigin-RevId: 273355526
2019-10-07 12:56:38 -07:00
Lei Zhang
ebf584b813 [spirv] Fix function entry block erase after moving to spv.selection
The structured selection/loop's entry block does not have arguments.
If the function's header block is also part of the structured control
flow, we cannot just simply erase it because it may contain arguments
matching the function signature and used by the cloned blocks. Instead,
turn it into a block only containing a spv.Branch op.

Also, we can directly emit instructions for the spv.selection header
block to the block containing the spv.selection op. This eliminates
unnecessary branches in the SPIR-V blob.

Added a test for nested spv.loop.

PiperOrigin-RevId: 273351424
2019-10-07 12:37:13 -07:00
Lei Zhang
f294e0e513 [spirv] Add support for spv.selection
Similar to spv.loop, spv.selection is another op for modelling
SPIR-V structured control flow. It covers both OpBranchConditional
and OpSwitch with OpSelectionMerge.

Instead of having a `spv.SelectionMerge` op to directly model
selection merge instruction for indicating the merge target,
we use regions to delimit the boundary of the selection: the
merge target is the next op following the `spv.selection` op.
This way it's easier to discover all blocks belonging to
the selection and it plays nicer with the MLIR system.

PiperOrigin-RevId: 272475006
2019-10-02 11:01:57 -07:00
Mahesh Ravishankar
75906bd565 Handle OpMemberName instruction in SPIR-V deserializer.
Sdd support in deserializer for OpMemberName instruction. For now
the name is just processed and not associated with the
spirv::StructType being built. That needs an enhancement to
spirv::StructTypes itself.
Add tests to check for errors reported during deserialization with
some refactoring to common out some utility functions.
PiperOrigin-RevId: 270794524
2019-09-23 17:11:18 -07:00
Christian Sigg
c900d4994e Fix a number of Clang-Tidy warnings.
PiperOrigin-RevId: 270632324
2019-09-23 02:34:27 -07:00
Denis Khalikov
2ec8e2be1f [spirv] Add OpControlBarrier and OpMemoryBarrier.
Add OpControlBarrier and OpMemoryBarrier (de)serialization.

Closes tensorflow/mlir#130

COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/130 from denis0x0D:sandbox/memory_barrier 2e3fff16bca44904dc1039592cb9a65d526faea8
PiperOrigin-RevId: 270457478
2019-09-21 10:18:34 -07:00
Mahesh Ravishankar
9a4f5d2ee3 Allow specification of decorators on SPIR-V StructType members.
Allow specification of decorators on SPIR-V StructType members. If the
struct has layout information, these decorations are to be specified
after the offset specification of the member. These decorations are
emitted as OpMemberDecorate instructions on the struct <id>. Update
(de)serialization to handle these decorations.

PiperOrigin-RevId: 270130136
2019-09-19 14:50:05 -07:00
Mahesh Ravishankar
9330c1b9a1 Add (de)serialization support for OpRuntimeArray.
Update the SPIR-V (de)serialization to handle RuntimeArrayType.

PiperOrigin-RevId: 269667196
2019-09-17 15:21:57 -07:00
Mahesh Ravishankar
2d86ad79f0 Autogenerate (de)serialization for Extended Instruction Sets
A generic mechanism for (de)serialization of extended instruction sets
is added with this CL. To facilitate this, a new class
"SPV_ExtendedInstSetOp" is added which is a base class for all
operations corresponding to extended instruction sets. The methods to
(de)serialization such ops as well as its dispatch is generated
automatically.

The behavior controlled by autogenSerialization and hasOpcode is also
slightly modified to enable this. They are now decoupled.
1) Setting hasOpcode=1 means the operation has a corresponding
   opcode in SPIR-V binary format, and its dispatch for
   (de)serialization is automatically generated.
2) Setting autogenSerialization=1 generates the function for
   (de)serialization automatically.
So now it is possible to have hasOpcode=0 and autogenSerialization=1
(for example SPV_ExtendedInstSetOp).

Since the dispatch functions is also auto-generated, the input file
needs to contain all operations. To this effect, SPIRVGLSLOps.td is
included into SPIRVOps.td. This makes the previously added
SPIRVGLSLOps.h and SPIRVGLSLOps.cpp unnecessary, and are deleted.

The SPIRVUtilsGen.cpp is also changed to make better use of
formatv,making the code more readable.

PiperOrigin-RevId: 269456263
2019-09-16 17:12:33 -07:00
Denis Khalikov
8a34d5d18c [spirv] Add support for function calls.
Add spv.FunctionCall operation and (de)serialization.

Closes tensorflow/mlir#137

COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/137 from denis0x0D:sandbox/function_call_op e2e6f07d21e7f23e8b44c7df8a8ab784f3356ce4
PiperOrigin-RevId: 269437167
2019-09-16 15:39:54 -07:00
Lei Zhang
a84bc68acc [spirv] Add support for spv.loop (de)serialization
This CL adds support for serializing and deserializing spv.loop ops.
This adds support for spv.Branch and spv.BranchConditional op
(de)serialization, too, because they are needed for spv.loop.

PiperOrigin-RevId: 268536962
2019-09-11 14:02:59 -07:00
Lei Zhang
3af6b53381 [spirv] Fix the entry block to start with OpLabel
Each basic block in SPIR-V must start with an OpLabel instruction.
We don't support control flow yet, so this CL just makes sure that
the entry block follows this rule and is valid.

PiperOrigin-RevId: 265718841
2019-08-27 10:51:26 -07:00
Denis Khalikov
8f2dfb51d4 [spirv] Add Block decoration for spv.struct.
Add Block decoration for top-level spv.struct.

Closes tensorflow/mlir#102

PiperOrigin-RevId: 265716241
2019-08-27 10:41:42 -07:00
Lei Zhang
51cbf97b53 [spirv] Add support for extension (de)serialization
Only a few important KHR extensions are registered to the
SPIR-V dialect for now.

PiperOrigin-RevId: 264939428
2019-08-22 16:01:35 -07:00
Lei Zhang
27ed82f99c [spirv] Add support for capability (de)serialization
This CL pulls in capabilities defined in the spec and adds
support for (de)serialize capabilities of a spv.module.

PiperOrigin-RevId: 264877413
2019-08-22 11:15:41 -07:00
Lei Zhang
f4934bcc3e Add spv.specConstant and spv._reference_of
Similar to global variables, specialization constants also live
in the module scope and can be referenced by instructions in
functions in native SPIR-V. A direct modelling would be to allow
functions in the SPIR-V dialect to implicit capture, but it means
we are losing the ability to write passes for Functions. While
in SPIR-V normally we want to process the module as a whole,
it's not common to see multiple functions get used so we'd like
to leave the door open for those cases. Therefore, similar to
global variables, we introduce spv.specConstant to model three
SPIR-V instructions: OpSpecConstantTrue, OpSpecConstantFalse,
and OpSpecConstant. They do not return SSA value results;
instead they have symbols and can only be referenced by the
symbols. To use it in a function, we need to have another op
spv._reference_of to turn the symbol into an SSA value. This
breaks the tie and makes functions still explicit capture.
Previously specialization constants were handled similarly as
normal constants. That is incorrect given that specialization
constant actually acts more like variable (without need to
load and store). E.g., they cannot be de-duplicated like normal
constants.

This CL also refines various documents and comments.

PiperOrigin-RevId: 264455172
2019-08-20 13:34:13 -07:00
Denis Khalikov
82cf6051ee [spirv] Support (de)serialization of spv.struct
Support (de)serialization of spv.struct with offset decorations.

Closes tensorflow/mlir#94

PiperOrigin-RevId: 264421427
2019-08-20 11:03:42 -07:00
Mahesh Ravishankar
d745101339 Add spirv::GlobalVariableOp that allows module level definition of variables
FuncOps in MLIR use explicit capture. So global variables defined in
module scope need to have a symbol name and this should be used to
refer to the variable within the function. This deviates from SPIR-V
spec, which assigns an SSA value to variables at all scopes that can
be used to refer to the variable, which requires SPIR-V functions to
allow implicit capture. To handle this add a new op,
spirv::GlobalVariableOp that can be used to define module scope
variables.
Since instructions need an SSA value, an new spirv::AddressOfOp is
added to convert a symbol reference to an SSA value for use with other
instructions.
This also means the spirv::EntryPointOp instruction needs to change to
allow initializers to be specified using symbol reference instead of
SSA value
The current spirv::VariableOp which returns an SSA value (as defined
by SPIR-V spec) can still be used to define function-scope variables.
PiperOrigin-RevId: 263951109
2019-08-17 10:20:13 -07:00
Denis Khalikov
cf358017e6 [spirv] Extend spv.array with Layoutinfo
Extend spv.array with Layoutinfo to support (de)serialization.

Closes tensorflow/mlir#80

PiperOrigin-RevId: 263795304
2019-08-16 10:18:14 -07:00
Mahesh Ravishankar
d71915420b Add BuiltIn EnumAttr to SPIR-V dialect
Generate the EnumAttr to represent BuiltIns in SPIR-V dialect. The
builtIn can be specified as a StringAttr with value being the
name of the builtin. Extend Decoration (de)serialization to handle
BuiltIns.
Also fix an error in the SPIR-V dialect generator script.

PiperOrigin-RevId: 263596624
2019-08-15 10:52:59 -07:00
River Riddle
a481032a33 Refactor ElementsAttr::getValue and DenseElementsAttr::getSplatValue.
All 'getValue' variants now require that the index is valid, queryable via 'isValidIndex'. 'getSplatValue' now requires that the attribute is a proper splat. This allows for querying these methods on DenseElementAttr with all possible value types; e.g. float, int, APInt, etc. This also allows for removing unnecessary conversions to Attribute that really want the underlying value.

PiperOrigin-RevId: 263437337
2019-08-14 15:03:53 -07:00
Lei Zhang
00a7b6706d [spirv] Add support for specialization constant
This CL extends the existing spv.constant op to also support
specialization constant by adding an extra unit attribute
on it.

PiperOrigin-RevId: 261194869
2019-08-01 14:13:37 -07:00
Mahesh Ravishankar
1de519a753 Add support for (de)serialization of SPIR-V Op Decorations
All non-argument attributes specified for an operation are treated as
decorations on the result value and (de)serialized using OpDecorate
instruction. An error is generated if an attribute is not an argument,
and the name doesn't correspond to a Decoration enum. Name of the
attributes that represent decoerations are to be the snake-case-ified
version of the Decoration name.
Add utility methods to convert to snake-case and camel-case.

PiperOrigin-RevId: 260792638
2019-07-30 14:15:03 -07:00
Lei Zhang
4a55bd5f28 [spirv] Add basic infrastructure for negative deserializer tests
We are relying on serializer to construct positive cases to drive
the test for deserializer. This leaves negative cases untested.

This CL adds a basic test fixture for covering the negative
corner cases to enforce a more robust deserializer.

Refactored common SPIR-V building methods out of serializer to
share it with the deserialization test.

PiperOrigin-RevId: 260742733
2019-07-30 11:55:33 -07:00
Lei Zhang
8b8caa888d (De)serialize composite spv.constant
This CL covers the case of composite spv.constant. We encode/decode
them into/from OpConstantComposite/OpConstantNull.

PiperOrigin-RevId: 259394700
2019-07-22 13:31:00 -07:00
Lei Zhang
83c97a6784 (De)serialize float scalar spv.constant
This CL adds support for float scalar spv.constant in (de)serialization.

PiperOrigin-RevId: 259311776
2019-07-22 06:02:32 -07:00
Lei Zhang
c1844220cd (De)serialize bool and integer scalar spv.constant
SPIR-V has multiple constant instructions covering different
constant types:

* `OpConstantTrue` and `OpConstantFalse` for boolean constants
* `OpConstant` for scalar constants
* `OpConstantComposite` for composite constants
* `OpConstantNull` for null constants
* ...

We model them all with a single spv.constant op for uniformity
and friendliness to transformations. This does mean that when
doing (de)serialization, we need to poke spv.constant's type
to determine which SPIR-V binary instruction to use.

This CL only covers the case of bool and integer spv.constant.
The rest will follow.

PiperOrigin-RevId: 259311698
2019-07-22 06:02:08 -07:00
Lei Zhang
9d52ceaf16 [spirv] NFC: adjust encode* function signatures in Serializer
* Let them return `LogicalResult` so we can chain them together
  with other functions returning `LogicalResult`.
* Added "Into" as the suffix to the function name and made the
  `binary` as the first parameter so that it reads more naturally.

PiperOrigin-RevId: 259311636
2019-07-22 06:01:19 -07:00
Lei Zhang
17c18840da [spirv] Remove one level of indirection: processOp to processOpImpl
We already have two levels of controls in SPIRVBase.td: hasOpcode and
autogenSerialization. The former controls whether to add an entry to
the dispatch table, while the latter controls whether to autogenerate
the op's (de)serialization method specialization. This is enough for
our cases. Remove the indirection from processOp to processOpImpl
to simplify the picture.

PiperOrigin-RevId: 259308711
2019-07-22 05:37:39 -07:00
Mahesh Ravishankar
2fb53e65ab Add (de)serialization of EntryPointOp and ExecutionModeOp
Since the serialization of EntryPointOp contains the name of the
function as well, the function serialization emits the function name
using OpName instruction, which is used during deserialization to get
the correct function name.

PiperOrigin-RevId: 259158784
2019-07-20 18:12:05 -07:00
Lei Zhang
1331c84fe3 Wrap op (de)serialization methods in anonymous namespace
It's a known bug that older GCC is not happy with method specialization in
the enclosing (global) namespace:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56480

This CL wraps the generated specialization methods in the anonymous namespace
to make sure the specialization is in the same namespace as the class.

PiperOrigin-RevId: 258983181
2019-07-19 11:41:27 -07:00
Lei Zhang
36a26e0033 [spirv] group methods better and improve comments
This CL groups (de)serialization methods logically and improves comments
at various places. It also sorted method implementations to follow the
order of their declarations. There is NFC.

PiperOrigin-RevId: 258843490
2019-07-19 11:40:12 -07:00
Mahesh Ravishankar
c6cfebf1af Automatically generate (de)serialization methods for SPIR-V ops
For ops in SPIR-V dialect that are a direct mirror of SPIR-V
operations, the serialization/deserialization methods can be
automatically generated from the Op specification. To enable this an
'autogenSerialization' field is added to SPV_Ops. When set to
non-zero, this will enable the automatic (de)serialization function
generation

Also adding tests that verify the spv.Load, spv.Store and spv.Variable
ops are serialized and deserialized correctly. To fully support these
tests also add serialization and deserialization of float types and
spv.ptr types

PiperOrigin-RevId: 258684764
2019-07-19 11:39:22 -07:00
Lei Zhang
d36dd94c75 NFC: Move SPIR-V dialect to Dialect/ subdirectory
PiperOrigin-RevId: 258345603
2019-07-16 13:45:09 -07:00