Summary: This op mirrors the llvm.intr counterpart and allows lowering + type conversions in a progressive fashion.
Differential Revision: https://reviews.llvm.org/D75775
Summary:
This revision adds intrinsics for transpose, columnwise.load and columnwise.store
achieving full coverage of the llvm.matrix intrinsics.
Differential Revision: https://reviews.llvm.org/D75852
This revision adds the first intrinsic for llvm.matrix.multiply.
This uses the more general `LLVM_OneResultOp` for now since the goal is
to use the
specific Matrix builders that @fhahn has created recently.
When piped through:
```
opt -O3 -enable-matrix | llc -O3 -march=x86-64 -mcpu=skylake-avx512
```
this has been verified to generate ymm instructions.
Additional function attribute support will be needed to generate proper
zmm instructions but at least things run end to end.
Benchmarking will be provided separately with the experimental
metaprogramming
[ModelBuilder](https://github.com/google/iree/tree/master/experimental/ModelBuilder)
tool when ready.
Summary:
This patch adds support for translation of the OpenMP barrier construct to LLVM
IR. The OpenMP IRBuilder is used for this translation. In this patch the code
for translation is added to the existing LLVM dialect translation to LLVM IR.
The patch includes code changes and a testcase.
Reviewers: jdoerfert, nicolasvasilache, ftynse, rriddle, mehdi_amini
Reviewed By: ftynse, rriddle, mehdi_amini
Differential Revision: https://reviews.llvm.org/D72962
Some attribute kinds are not supported as "value" attributes of
`llvm.mlir.constant` when translating to LLVM IR. We were correctly reporting
an error, but continuing the translation using an "undef" value instead,
leading to a surprising mix of error messages and output IR. Abort the
translation after the error is reported.
Differential Revision: https://reviews.llvm.org/D75450
Summary:
the .row.col variant turns out to be the popular one, contrary to what I
thought as .row.row. Since .row.col is so prevailing (as I inspect
cuDNN's behavior), I'm going to remove the .row.row support here, which
makes the patch a little bit easier.
Reviewers: ftynse
Subscribers: jholewinski, bixia, sanjoy.google, mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, liufengdb, Joonsoo, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D74655
Summary:
This revision exposes the portable `llvm.fma` intrinsic in LLVMOps and uses it
in lieu of `llvm.fmuladd` when lowering the `vector.outerproduct` op to LLVM.
This guarantees proper `fma` instructions will be emitted if the target ISA
supports it.
`llvm.fmuladd` does not have this guarantee in its semantics, despite evidence
that the proper x86 instructions are emitted.
For more details, see https://llvm.org/docs/LangRef.html#llvm-fmuladd-intrinsic.
Reviewers: ftynse, aartbik, dcaballe, fhahn
Reviewed By: aartbik
Subscribers: mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, arpith-jacob, mgester, lucyrfox, liufengdb, Joonsoo, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D74219
Summary:
This revision adds basic support for emitting line table information when exporting to LLVMIR. We don't yet have a story for supporting all of the LLVM debug metadata, so this revision stubs some features(like subprograms) to enable emitting line tables.
Differential Revision: https://reviews.llvm.org/D73934
Summary:
The intrinsic operation added multiple type annotations to the llvm intrinsic operations, but only one is needed.
The related tests in llvmir-intrinsics.mlir checked the wrong number and are adjusted as well.
Reviewers: nicolasvasilache, ftynse
Reviewed By: ftynse
Subscribers: merge_guards_bot, ftynse, mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, arpith-jacob, mgester, lucyrfox, liufengdb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D73470
Summary:
Implement the handling of llvm::ConstantDataSequential and
llvm::ConstantAggregate for (nested) array and vector types when imporitng LLVM
IR to MLIR. In all cases, the result is a DenseElementsAttr that can be used in
either a `llvm.mlir.global` or a `llvm.mlir.constant`. Nested aggregates are
unpacked recursively until an element or a constant data is found. Nested
arrays with innermost scalar type are represented as DenseElementsAttr of
tensor type. Nested arrays with innermost vector type are represented as
DenseElementsAttr with (multidimensional) vector type.
Constant aggregates of struct type are not yet supported as the LLVM dialect
does not have a well-defined way of modeling struct-type constants.
Differential Revision: https://reviews.llvm.org/D72834
Summary:
Add a `llvm.cmpxchg` op as a counterpart to LLVM IR's `cmpxchg` instruction.
Note that the `weak`, `volatile`, and `syncscope` attributes are not yet supported.
This will be useful for upcoming parallel versions of affine.for and generally
for reduction-like semantics (especially for reductions that can't make use
of `atomicrmw`, e.g. `fmax`).
Reviewers: ftynse, nicolasvasilache
Reviewed By: ftynse
Subscribers: merge_guards_bot, jfb, mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, arpith-jacob, mgester, lucyrfox, liufengdb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D72995
Summary:
This op is the counterpart to LLVM's atomicrmw instruction. Note that
volatile and syncscope attributes are not yet supported.
This will be useful for upcoming parallel versions of `affine.for` and generally
for reduction-like semantics.
Differential Revision: https://reviews.llvm.org/D72741
Summary:
MLIR unlike LLVM IR supports multidimensional vector types. Such types are
lowered to nested LLVM IR arrays wrapping an LLVM IR vector for the innermost
dimension of the MLIR vector. MLIR supports constants of such types using
ElementsAttr for values. Introduce support for converting ElementsAttr into
LLVM IR Constant Aggregates recursively. This enables translation of
multidimensional vector constants from MLIR to LLVM IR.
Differential Revision: https://reviews.llvm.org/D72846
The current implementation of the LLVM-to-MLIR translation could not handle
functions used as constant values in instructions. The handling is added
trivially as `llvm.mlir.constant` can define constants of function type using
SymbolRef attributes, which works even for functions that have not been
declared yet.
Summary:
When converting splat constants for nested sequential LLVM IR types wrapped in
MLIR, the constant conversion was erroneously assuming it was always possible
to recursively construct a constant of a sequential type given only one value.
Instead, wait until all sequential types are unpacked recursively before
constructing a scalar constant and wrapping it into the surrounding sequential
type.
Subscribers: mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, nicolasvasilache, arpith-jacob, mgester, lucyrfox, aartbik, liufengdb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D72688
Summary:
`mlir-translate -import-llvm test.ll` was going into segmentation fault if `test.ll` had `float` or `double` constants.
For example,
```
%3 = fadd double 3.030000e+01, %0
```
Now, it is handled in `Importer::getConstantAsAttr` (similar behaviour as normal integers)
Added tests for FP arithmetic
Reviewers: ftynse, mehdi_amini
Reviewed By: ftynse, mehdi_amini
Subscribers: shauheen, mehdi_amini, rriddle, jpienaar, burmako, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D71912
Rename the 'shlis' operation in the standard dialect to 'shift_left'. Add tests
for this operation (these have been missing so far) and add a lowering to the
'shl' operation in the LLVM dialect.
Add also 'shift_right_signed' (lowered to LLVM's 'ashr') and 'shift_right_unsigned'
(lowered to 'lshr').
The original plan was to name these operations 'shift.left', 'shift.right.signed'
and 'shift.right.unsigned'. This works if the operations are prefixed with 'std.'
in MLIR assembly. Unfortunately during import the short form is ambigous with
operations from a hypothetical 'shift' dialect. The best solution seems to omit
dots in standard operations for now.
Closestensorflow/mlir#226
PiperOrigin-RevId: 286803388
Added test cases for the newly added LLVM operations and lowering features.
Closestensorflow/mlir#300
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/300 from dfki-jugr:std_to_llvm da6168bbc1a369ae2e99ad3881fdddd82f075dd4
PiperOrigin-RevId: 286231169
Introduce affine.prefetch: op to prefetch using a multi-dimensional
subscript on a memref; similar to affine.load but has no effect on
semantics, but only on performance.
Provide lowering through std.prefetch, llvm.prefetch and map to llvm's
prefetch instrinsic. All attributes reflected through the lowering -
locality hint, rw, and instr/data cache.
affine.prefetch %0[%i, %j + 5], false, 3, true : memref<400x400xi32>
Signed-off-by: Uday Bondhugula <uday@polymagelabs.com>
Closestensorflow/mlir#225
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/225 from bondhugula:prefetch 4c3b4e93bc64d9a5719504e6d6e1657818a2ead0
PiperOrigin-RevId: 286212997
Both work for the current use case, but the latter allows implementing
prefix sums and is a little easier to understand for partial warps.
PiperOrigin-RevId: 285145287
LLVM IR supports linkage on global objects such as global variables and
functions. Introduce the Linkage attribute into the LLVM dialect, backed by an
integer storage. Use this attribute on LLVM::GlobalOp and make it mandatory.
Implement parsing/printing of the attribute and conversion to LLVM IR.
See tensorflow/mlir#277.
PiperOrigin-RevId: 283309328
MLIR translation tools can emit diagnostics and we want to be able to check if
it is indeed the case in tests. Reuse the source manager error handlers
provided for mlir-opt to support the verification in mlir-translate. This
requires us to change the signature of the functions that are registered to
translate sources to MLIR: it now takes a source manager instead of a memory
buffer.
PiperOrigin-RevId: 279132972
This allows GlobalOp to either take a value attribute (for simple constants) or a region that can
contain IR instructions (that must be constant-foldable) to create a ConstantExpr initializer.
Example:
// A complex initializer is constructed with an initializer region.
llvm.mlir.global constant @int_gep() : !llvm<"i32*"> {
%0 = llvm.mlir.addressof @g2 : !llvm<"i32*">
%1 = llvm.mlir.constant(2 : i32) : !llvm.i32
%2 = llvm.getelementptr %0[%1] : (!llvm<"i32*">, !llvm.i32) -> !llvm<"i32*">
llvm.return %2 : !llvm<"i32*">
}
PiperOrigin-RevId: 278717836
This adds an importer from LLVM IR or bitcode to the LLVM dialect. The importer is registered with mlir-translate.
Known issues exposed by this patch but not yet fixed:
* Globals' initializers are attributes, which makes it impossible to represent a ConstantExpr. This will be fixed in a followup.
* icmp returns i32 rather than i1.
* select and a couple of other instructions aren't implemented.
* llvm.cond_br takes its successors in a weird order.
The testing here is known to be non-exhaustive.
I'd appreciate feedback on where this functionality should live. It looks like the translator *from MLIR to LLVM* lives in Target/, but the SPIR-V deserializer lives in Dialect/ which is why I've put this here too.
PiperOrigin-RevId: 278711683
nvvm.shfl.sync.bfly optionally returns a predicate whether source lane was active. Support for this was added to clang in https://reviews.llvm.org/D68892.
Add an optional 'pred' unit attribute to the instruction to return this predicate. Specify this attribute in the partial warp reduction so we don't need to manually compute the predicate.
PiperOrigin-RevId: 275616564
Similarly to `llvm.mlir.undef`, this auxiliary operation creates an SSA value
that corresponds to `null` in LLVM IR. This operation is necessary to model
sizeof(<...>) behavior when allocating memory.
PiperOrigin-RevId: 274158760
This function-like operation allows one to define functions that have wrapped
LLVM IR function type, in particular variadic functions. The operation was
added in parallel to the existing lowering flow, this commit only switches the
flow to use it.
Using a custom function type makes the LLVM IR dialect type system more
consistent and avoids complex conversion rules for functions that previously
had to use the built-in function type instead of a wrapped LLVM IR dialect type
and perform conversions during the analysis.
PiperOrigin-RevId: 273910855
This is matching what the runtime library is expecting.
Closestensorflow/mlir#171
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/171 from deven-amd:deven-rocdl-device-func-i64 80762629a8c34e844ebdc542b34dd783990db9db
PiperOrigin-RevId: 273640767
This commit introduces the ROCDL Dialect (i.e. the ROCDL ops + the code to lower those ROCDL ops to LLWM intrinsics/functions). Think of ROCDL Dialect as analogous to the NVVM Dialect, but for AMD GPUs. This patch contains just the essentials needed to get a simple example up and running. We expect to make further additions to the ROCDL Dialect.
This is the first of 3 commits, the follow-up will be:
* add a pass that lowers GPU Dialect to ROCDL Dialect
* add a "mlir-rocm-runner" utility
Closestensorflow/mlir#146
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/146 from deven-amd:deven-rocdl-dialect e78e8005c75a78912631116c78dc844fcc4b0de9
PiperOrigin-RevId: 271511259
This adds sign- and zero-extension and truncation of integer types to the
standard dialects. This allows to perform integer type conversions without
having to go to the LLVM dialect and introduce custom type casts (between
standard and LLVM integer types).
Closestensorflow/mlir#134
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/134 from ombre5733:sext-zext-trunc-in-std c7657bc84c0ca66b304e53ec03797e09152e4d31
PiperOrigin-RevId: 270479722
Make GlobalOp's value attribute an OptionalAttr. Change code that uses the value to handle 'nullopt'. Translate an unitialized value attribute to llvm::UndefValue.
PiperOrigin-RevId: 270423646
Some of the operations in the LLVM dialect are required to model the LLVM IR in
MLIR, for example "constant" operations are needed to declare a constant value
since MLIR, unlike LLVM, does not support immediate values as operands. To
avoid confusion with actual LLVM operations, we prefix such axuiliary
operations with "mlir.".
PiperOrigin-RevId: 266942838
This will allow iterating the values of a non-opaque ElementsAttr, with all of the types currently supported by DenseElementsAttr. This should help reduce the amount of specialization on DenseElementsAttr.
PiperOrigin-RevId: 264968151
This will allow iterating the values of a non-opaque ElementsAttr, with all of the types currently supported by DenseElementsAttr. This should help reduce the amount of specialization on DenseElementsAttr.
PiperOrigin-RevId: 264637293
LLVM intrinsics have an open name space and their names can potentially overlap
with names of LLVM instructions (LLVM intrinsics are functions, not
instructions). In MLIR, LLVM intrinsics are modeled as operations, so it needs
to make sure their names cannot clash with the instructions. Use the "intr."
prefix for intrinsics in the LLVM dialect.
PiperOrigin-RevId: 264372173
This operation is important to achieve decent performance in computational
kernels. In LLVM, it is implemented as an intrinsic (through function
declaration and function call). Thanks to MLIR's extendable set of operations,
it does not have to differentiate between built-ins and intrinsics, so fmuladd
is introduced as a general type-polymorphic operation. Custom printing and
parsing will be added later.
PiperOrigin-RevId: 263106305
This instruction is a local counterpart of llvm.global that takes a symbol
reference to a global and produces an SSA value containing the pointer to it.
Used in combination, these two operations allow one to use globals with other
operations expecting SSA values. At a cost of IR indirection, we make sure the
functions don't implicitly capture the surrounding SSA values and remain
suitable for parallel processing.
PiperOrigin-RevId: 262908622
Unlike regular constant values, strings must be placed in some memory and
referred to through a pointer to that memory. Until now, they were not
supported in function-local constant declarations with `llvm.constant`.
Introduce support for global strings using `llvm.global`, which would translate
them into global arrays in LLVM IR and thus make sure they have some memory
allocated for storage.
PiperOrigin-RevId: 262569316
This CL is step 1/n towards building a simple, programmable and portable vector abstraction in MLIR that can go all the way down to generating assembly vector code via LLVM's opt and llc tools.
This CL adds the 3 instructions `llvm.extractelement`, `llvm.insertelement` and `llvm.shufflevector` as documented in the LLVM LangRef "Vector Instructions" section.
The "Experimental Vector Reduction Intrinsics" are left out for now and can be added in the future on a per-need basis.
Appropriate roundtrip and LLVM Target tests are added.
PiperOrigin-RevId: 262542095
This adds support for fcmp to the LLVM dialect and adds any necessary lowerings, as well as support for EDSCs.
Closestensorflow/mlir#69
PiperOrigin-RevId: 262475255
llvm ir printer was changed at LLVM r367755.
Prints value numbers for unnamed functions argument.
Closestensorflow/mlir#67
COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/67 from denis0x0D:sandbox/fix_mlir_translate ae46844e66f34a02e0cf86782ddadc5bce58b30d
PiperOrigin-RevId: 261640048
The current syntax separates the name and value with ':', but ':' is already overloaded by several other things(e.g. trailing types). This makes the syntax difficult to parse in some situtations:
Old:
"foo: 10 : i32"
New:
"foo = 10 : i32"
PiperOrigin-RevId: 255097928
This is the standard syntax for types on operations, and is also already used by IntegerAttr and FloatAttr.
Example:
dense<5> : tensor<i32>
dense<[3]> : tensor<1xi32>
PiperOrigin-RevId: 255069157
PTX backend in LLVM expects additional module-level metadata
`!nvvm.annotations` that lists functions that can be used as GPU kernels.
Generate this metadata based on the `gpu.kernel` attribute attached to
functions. This attribute is added automatically by the kernel outlining pass
in the GPU dialect lowering flow.
PiperOrigin-RevId: 254957345
* There is no longer a need to explicitly remap function attrs.
- This removes a potentially expensive call from the destructor of Function.
- This will enable some interprocedural transformations to now run intraprocedurally.
- This wasn't scalable and forces dialect defined attributes to override
a virtual function.
* Replacing a function is now a trivial operation.
* This is a necessary first step to representing functions as operations.
--
PiperOrigin-RevId: 249510802
This is only teaching the LLVM converter to propagate the attribute onto
the function type. MLIR will not recognize this arguments, so it would only
be useful when calling for example `printf` with the same arguments across
a module. Since varargs is part of the ABI lowering, this is not NFC.
--
PiperOrigin-RevId: 242382427
making the IR dumps much nicer.
This is part 2/3 of the path to making dialect types more nice. Part 3/3 will
slightly generalize the set of characters allowed in pretty types and make it
more principled.
--
PiperOrigin-RevId: 242249955
Historically, the LLVM IR dialect has been using the generic form of MLIR
operation syntax. It is verbose and often redundant. Introduce the custom
printing and parsing for all existing operations in the LLVM IR dialect.
Update the relevant documentation and tests.
--
PiperOrigin-RevId: 241617393
When the LLVM IR dialect was implemented, TableGen operation definition scheme
did not support operations with variadic results. Therefore, the `call`
instruction was split into `call` and `call0` for the single- and zero-result
calls (LLVM does not support multi-result operations). Unify `call` and
`call0` using the recently added TableGen support for operations with Variadic
results. Explicitly verify that the new operation has 0 or 1 results. As a
side effect, this change enables clean-ups in the conversion to the LLVM IR
dialect that no longer needs to rely on wrapped LLVM IR void types when
constructing zero-result calls.
PiperOrigin-RevId: 236119197
Since the goal of the LLVM IR dialect is to reflect LLVM IR in MLIR, the
dialect and the conversion procedure must account for the differences betweeen
block arguments and LLVM IR PHI nodes. In particular, LLVM IR disallows PHI
nodes with different values coming from the same source. Therefore, the LLVM IR
dialect now disallows `cond_br` operations that have identical successors
accepting arguments, which would lead to invalid PHI nodes. The conversion
process resolves the potential PHI source ambiguity by injecting dummy blocks
if the same block is used more than once as a successor in an instruction.
These dummy blocks branch unconditionally to the original successors, pass them
the original operands (available in the dummy block because it is dominated by
the original block) and are used instead of them in the original terminator
operation.
PiperOrigin-RevId: 235682798
Add support for lowering DivF and RemF to LLVM::FDiv and LLMV::FRem
respectively. The lowering is a trivial one-to-one transformation.
The corresponding operations already existed in the LLVM IR dialect and can be
lowered to the LLVM IR proper. Add the necessary tests for scalar and vector
forms.
PiperOrigin-RevId: 234984608
Add support for converting MLIR `call_indirect` instructions to the LLVM IR
dialect. In LLVM IR, the same instruction is used for direct and indirect
calls. In the dialect, we have `llvm.call` and `llvm.call0` to work around the
absence of the void type in MLIR. For direct calls, the callee is stored as
instruction attribute. Use the same pair of instructions for indirect calls by
omitting the callee attribute. In the MLIR to LLVM IR translator, check the
presence of attribute to decide whether to construct a direct or an indirect
call using different LLVM IR Builder functions.
Add support for converting constants of function type to the LLVM IR dialect
and for translating them to the LLVM IR proper. The `llvm.constant` operation
works similarly to other types: its attribute has MLIR function type but the
value it produces has LLVM IR function type wrapped in the dialect type. While
lowering, look up the pointer to the converted function in the corresponding
mapping.
PiperOrigin-RevId: 234132351
Original implementation of the translation from MLIR to LLVM IR operated on the
Standard+BuiltIn dialect, with a later addition of the SuperVector dialect.
This required the translation to be aware of a potetially large number of other
dialects as the infrastructure extended. With the recent introduction of the
LLVM IR dialect into MLIR, the translation can be switched to only translate
the LLVM IR dialect, and the translation of the operations becomes largely
mechanical.
The reimplementation of the translator follows the lines of the original
translator in function and basic block conversion. In particular, block
arguments are converted to LLVM IR PHI nodes, which are connected to their
sources after all blocks of a function had been converted. Thanks to LLVM IR
types being wrapped in the MLIR LLVM dialect type, type conversion is
simplified to only convert function types, all other types are simply
unwrapped. Individual instructions are constructed using the LLVM IRBuilder,
which has a great potential for being table-generated from the LLVM IR dialect
operation definitions.
The input of the test/Target/llvmir.mlir is updated to use the MLIR LLVM IR
dialect. While it is now redundant with the dialect conversion test, the point
of the exercise is to guarantee exactly the same LLVM IR is emitted. (Only the
name of the allocation function is changed from `__mlir_alloc` to `alloc` in
the CHECK lines.) It will be simplified in a follow-up commit.
PiperOrigin-RevId: 233842306
These operations trivially map to LLVM IR counterparts for operands of scalar
and (one-dimensional) vector type. Multi-dimensional vector and tensor type
operands would fail type conversion before the operation conversion takes
place. Add tests for scalar and vector cases. Also add a test for vector
`select` instruction for consistency with other tests.
PiperOrigin-RevId: 228077564
The entire compiler now looks at structural properties of the function (e.g.
does it have one block, does it contain an if/for stmt, etc) so the only thing
holding up this difference is round tripping through the parser/printer syntax.
Removing this shrinks the compile by ~140LOC.
This is step 31/n towards merging instructions and statements. The last step
is updating the docs, which I will do as a separate patch in order to split it
from this mostly mechanical patch.
PiperOrigin-RevId: 227540453
This commit adds support for the "select" operation that lowers directly into
its LLVM IR counterpart. A simple test is included.
PiperOrigin-RevId: 227527893
printing the entry block in a CFG function's argument line. Since I'm touching
all of the testcases anyway, change the argument list from printing as
"%arg : type" to "%arg: type" which is more consistent with bb arguments.
In addition to being more consistent, this is a much nicer look for cfg functions.
PiperOrigin-RevId: 227240069
The binary subtraction operations were not supported by the lowering because
they were not essential for the testing flow. Add support for these
operations.
PiperOrigin-RevId: 226941463
Introduce initial support for 1D vector operations. LLVM does not support
higher-dimensional vectors so the caller must make sure they don't appear in
the input MLIR. Handle the presence of higher-dimensional vectors by failing
gracefully.
Introduce the type conversion for 1D vector types and hook it up with the rest
of the type convresion system. Support "splat" constants for vector types. As
a side effect, this refactors constant operation emission by separating out
scalar integer constants into a separate case and by extracting out the helper
function for scalar float construction. Existing binary operations apply to
vectors transparently.
PiperOrigin-RevId: 225172349
Unlike MLIR, LLVM IR does not support functions that return multiple values.
Simulate this by packing values into the LLVM structure type in the same order
as they appear in the MLIR return. If the function returns only a single
value, return it directly without packing.
PiperOrigin-RevId: 223964886
Add support for translating 'dim' opreation on MemRefs to LLVM IR. For a
static size, this operation merely defines an LLVM IR constant value that may
not appear in the output IR if not used (and had not been removed before by
DCE). For a dynamic size, this operation is translated into an access to the
MemRef descriptor that contains the dynamic size.
PiperOrigin-RevId: 223160774
Introduce initial support for MemRef types, including type conversion,
allocation and deallocation, read and write element-wise access, passing
MemRefs to and returning from functions. Affine map compositions and
non-default memory spaces are NOT YET supported.
Lowered code needs to handle potentially dynamic sizes of the MemRef. To do
so, it replaces a MemRef-typed value with a special MemRef descriptor that
carries the data and the dynamic sizes together. A MemRef type is converted to
LLVM's first-class structure type with the first element being the pointer to
the data buffer with data layed out linearly, followed by as many integer-typed
elements as MemRef has dynamic sizes. The type of these elements is that of
MLIR index lowered to LLVM. For example, `memref<?x42x?xf32>` is converted to
`{ f32*, i64, i64 }` provided `index` is lowered to `i64`. While it is
possible to convert MemRefs with fully static sizes to simple pointers to their
elemental types, we opted for consistency and convert them to the
single-element structure. This makes the conversion code simpler and the
calling convention of the generated LLVM IR functions consistent.
Loads from and stores to a MemRef element are lowered to a sequence of LLVM
instructions that, first, computes the linearized index of the element in the
data buffer using the access indices and combining the static sizes with the
dynamic sizes stored in the descriptor, and then loads from or stores to the
buffer element indexed by the linearized subscript. While some of the index
computations may be redundant (i.e., consecutive load and store to the same
location in the same scope could reuse the linearized index), we emit them for
every operation. A subsequent optimization pass may eliminate them if
necessary.
MemRef allocation and deallocation is performed using external functions
`__mlir_alloc(index) -> i8*` and `__mlir_free(i8*)` that must be implemented by
the caller. These functions behave similarly to `malloc` and `free`, but can
be extended to support different memory spaces in future. Allocation and
deallocation instructions take care of casting the pointers. Prior to calling
the allocation function, the emitted code creates an SSA Value for the
descriptor and uses it to store the dynamic sizes of the MemRef passed to the
allocation operation. It further emits instructions that compute the dynamic
amount of memory to allocate in bytes. Finally, the allocation stores the
result of calling the `__mlir_alloc` in the MemRef descriptor. Deallocation
extracts the pointer to the allocated memory from the descriptor and calls
`__mlir_free` on it. The descriptor itself is not modified and, being
stack-allocated, ceases to exist when it goes out of scope.
MLIR functions that access MemRef values as arguments or return them are
converted to LLVM IR functions that accept MemRef descriptors as LLVM IR
structure types by value. This significantly simplifies the calling convention
at the LLVM IR level and avoids handling descriptors in the dynamic memory,
however is not always comaptible with LLVM IR functions emitted from C code
with similar signatures. A separate LLVM pass may be introduced in the future
to provide C-compatible calling conventions for LLVM IR functions generated
from MLIR.
PiperOrigin-RevId: 223134883
Initial restricted implementaiton of the MLIR to LLVM IR translation.
Introduce a new flow into the mlir-translate tool taking an MLIR module
containing CFG functions only and producing and LLVM IR module. The MLIR
features supported by the translator are as follows:
- primitive and function types;
- integer constants;
- cfg and ext functions with 0 or 1 return values;
- calls to these functions;
- basic block conversion translation of arguments to phi nodes;
- conversion between arguments of the first basic block and function arguments;
- (conditional) branches;
- integer addition and comparison operations.
Are NOT supported:
- vector and tensor types and operations on them;
- memrefs and operations on them;
- allocations;
- functions returning multiple values;
- LLVM Module triple and data layout (index type is hardcoded to i64).
Create a new MLIR library and place it under lib/Target/LLVMIR. The "Target"
library group is similar to the one present in LLVM and is intended to contain
all future public MLIR translation targets.
The general flow of MLIR to LLVM IR convresion will include several lowering
and simplification passes on the MLIR itself in order to make the translation
as simple as possible. In particular, ML functions should be transformed to
CFG functions by the recently introduced pass, operations on structured types
will be converted to sequences of operations on primitive types, complex
operations such as affine_apply will be converted into sequence of primitive
operations, primitive operations themselves may eventually be converted to an
LLVM dialect that uses LLVM-like operations.
Introduce the first translation test so that further changes make sure the
basic translation functionality is not broken.
PiperOrigin-RevId: 222400112
Shraiysh Vaishay