* Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that.
* Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2
* Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided).
* Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement).
Differential Revision: https://reviews.llvm.org/D105070
In order to allow large matmul operations using the MMA ops we need to chain
operations this is not possible unless "DOp" and "COp" type have matching
layout so remove the "DOp" layout and force accumulator and result type to
match.
Added a test for the case where the MMA value is accumulated.
Differential Revision: https://reviews.llvm.org/D103023
The current design uses a unique entry for each argument/result attribute, with the name of the entry being something like "arg0". This provides for a somewhat sparse design, but ends up being much more expensive (from a runtime perspective) in-practice. The design requires building a string every time we lookup the dictionary for a specific arg/result, and also requires N attribute lookups when collecting all of the arg/result attribute dictionaries.
This revision restructures the design to instead have an ArrayAttr that contains all of the attribute dictionaries for arguments and another for results. This design reduces the number of attribute name lookups to 1, and allows for O(1) lookup for individual element dictionaries. The major downside is that we can end up with larger memory usage, as the ArrayAttr contains an entry for each element even if that element has no attributes. If the memory usage becomes too problematic, we can experiment with a more sparse structure that still provides a lot of the wins in this revision.
This dropped the compilation time of a somewhat large TensorFlow model from ~650 seconds to ~400 seconds.
Differential Revision: https://reviews.llvm.org/D102035
Just a pure method renaming.
It is a preparation step for replacing "memory space as raw integer"
with more generic "memory space as attribute", which will be done in
separate commit.
The `MemRefType::getMemorySpace` method will return `Attribute` and
become the main API, while `getMemorySpaceAsInt` will be declared as
deprecated and will be replaced in all in-tree dialects (also in separate
commits).
Reviewed By: mehdi_amini, rriddle
Differential Revision: https://reviews.llvm.org/D97476
This is part of a larger refactoring the better congregates the builtin structures under the BuiltinDialect. This also removes the problematic "standard" naming that clashes with the "standard" dialect, which is not defined within IR/. A temporary forward is placed in StandardTypes.h to allow time for downstream users to replaced references.
Differential Revision: https://reviews.llvm.org/D92435
Given that OpState already implicit converts to Operator*, this seems reasonable.
The alternative would be to add more functions to OpState which forward to Operation.
Reviewed By: rriddle, ftynse
Differential Revision: https://reviews.llvm.org/D92266
These includes have been deprecated in favor of BuiltinDialect.h, which contains the definitions of ModuleOp and FuncOp.
Differential Revision: https://reviews.llvm.org/D91572
This is a roll-forward of rGec7780ebdab4, now that the remaining
gpu.launch_func have been converted to custom form in rGb22f111023ba.
Reviewed By: antiagainst
Differential Revision: https://reviews.llvm.org/D90420
This combines two separate ops (D88972: `gpu.create_token`, D89043: `gpu.host_wait`) into one.
I do after all like the idea of combining the two ops, because it matches exactly the pattern we are
going to have in the other gpu ops that will implement the AsyncOpInterface (launch_func, copies, alloc):
If the op is async, we return a !gpu.async.token. Otherwise, we synchronize with the host and don't return a token.
The use cases for `gpu.wait async` and `gpu.wait` are further apart than those of e.g. `gpu.h2d async` and `gpu.h2d`,
but I like the consistent meaning of the `async` keyword in GPU ops.
Reviewed By: herhut
Differential Revision: https://reviews.llvm.org/D89160
- Use TypeRange instead of ArrayRef<Type> where possible.
- Change some of the custom builders to also use TypeRange
Differential Revision: https://reviews.llvm.org/D87944
Now backends spell out which namespace they want to be in, instead of relying on
clients #including them inside already-opened namespaces. This also means that
cppNamespaces should be fully qualified, and there's no implicit "::mlir::"
prepended to them anymore.
Reviewed By: mehdi_amini
Differential Revision: https://reviews.llvm.org/D86811
This patch moves the registration to a method in the MLIRContext: getOrCreateDialect<ConcreteDialect>()
This method requires dialect to provide a static getDialectNamespace()
and store a TypeID on the Dialect itself, which allows to lazyily
create a dialect when not yet loaded in the context.
As a side effect, it means that duplicated registration of the same
dialect is not an issue anymore.
To limit the boilerplate, TableGen dialect generation is modified to
emit the constructor entirely and invoke separately a "init()" method
that the user implements.
Differential Revision: https://reviews.llvm.org/D85495
- Arguments of the first block of a region are considered region arguments.
- Add API on Region class to deal with these arguments directly instead of
using the front() block.
- Changed several instances of existing code that can use this API
- Fixes https://bugs.llvm.org/show_bug.cgi?id=46535
Differential Revision: https://reviews.llvm.org/D83599
Use ::Adaptor alias instead uniformly. Makes the naming more consistent as
adaptor can refer to attributes now too.
Differential Revision: https://reviews.llvm.org/D81789
This is a wrapper around vector of NamedAttributes that keeps track of whether sorted and does some minimal effort to remain sorted (doing more, e.g., appending attributes in sorted order, could be done in follow up). It contains whether sorted and if a DictionaryAttr is queried, it caches the returned DictionaryAttr along with whether sorted.
Change MutableDictionaryAttr to always return a non-null Attribute even when empty (reserve null cases for errors). To this end change the getter to take a context as input so that the empty DictionaryAttr could be queried. Also create one instance of the empty dictionary attribute that could be reused without needing to lock context etc.
Update infer type op interface to use DictionaryAttr and use NamedAttrList to avoid incurring multiple conversion costs.
Fix bug in sorting helper function.
Differential Revision: https://reviews.llvm.org/D79463
As we start defining more complex Ops, we increasingly see the need for
Ops-with-regions to be able to construct Ops within their regions in
their ::build methods. However, these methods only have access to
Builder, and not OpBuilder. Creating a local instance of OpBuilder
inside ::build and using it fails to trigger the operation creation
hooks in derived builders (e.g., ConversionPatternRewriter). In this
case, we risk breaking the logic of the derived builder. At the same
time, OpBuilder::create, which is by far the largest user of ::build
already passes "this" as the first argument, so an OpBuilder instance is
already available.
Update all ::build methods in all Ops in MLIR and Flang to take
"OpBuilder &" instead of "Builder *". Note the change from pointer and
to reference to comply with the common style in MLIR, this also ensures
all other users must change their ::build methods.
Differential Revision: https://reviews.llvm.org/D78713
Summary:
Use a nested symbol to identify the kernel to be invoked by a `LaunchFuncOp` in the GPU dialect.
This replaces the two attributes that were used to identify the kernel module and the kernel within seperately.
Differential Revision: https://reviews.llvm.org/D78551
These have proved incredibly useful for interleaving values between a range w.r.t to streams. After this revision, the mlir/Support/STLExtras.h is empty. A followup revision will remove it from the tree.
Differential Revision: https://reviews.llvm.org/D78067
Summary: This generates the class declarations for dialects using the existing 'Dialect' tablegen classes.
Differential Revision: https://reviews.llvm.org/D76185
Summary:
This patch add some builtin operation for the gpu.all_reduce ops.
- for Integer only: `and`, `or`, `xor`
- for Float and Integer: `min`, `max`
This is useful for higher level dialect like OpenACC or OpenMP that can lower to the GPU dialect.
Differential Revision: https://reviews.llvm.org/D75766
Summary:
This patch add some builtin operation for the gpu.all_reduce ops.
- for Integer only: `and`, `or`, `xor`
- for Float and Integer: `min`, `max`
This is useful for higher level dialect like OpenACC or OpenMP that can lower to the GPU dialect.
Differential Revision: https://reviews.llvm.org/D75766
Summary:
NFC - Moved StandardOps/Ops.h to a StandardOps/IR dir to better match surrounding
directories. This is to match other dialects, and prepare for moving StandardOps
related transforms in out for Transforms and into StandardOps/Transforms.
Differential Revision: https://reviews.llvm.org/D74940
Thus far IntegerType has been signless: a value of IntegerType does
not have a sign intrinsically and it's up to the specific operation
to decide how to interpret those bits. For example, std.addi does
two's complement arithmetic, and std.divis/std.diviu treats the first
bit as a sign.
This design choice was made some time ago when we did't have lots
of dialects and dialects were more rigid. Today we have much more
extensible infrastructure and different dialect may want different
modelling over integer signedness. So while we can say we want
signless integers in the standard dialect, we cannot dictate for
others. Requiring each dialect to model the signedness semantics
with another set of custom types is duplicating the functionality
everywhere, considering the fundamental role integer types play.
This CL extends the IntegerType with a signedness semantics bit.
This gives each dialect an option to opt in signedness semantics
if that's what they want and helps code sharing. The parser is
modified to recognize `si[1-9][0-9]*` and `ui[1-9][0-9]*` as
signed and unsigned integer types, respectively, leaving the
original `i[1-9][0-9]*` to continue to mean no indication over
signedness semantics. All existing dialects are not affected (yet)
as this is a feature to opt in.
More discussions can be found at:
https://groups.google.com/a/tensorflow.org/d/msg/mlir/XmkV8HOPWpo/7O4X0Nb_AQAJ
Differential Revision: https://reviews.llvm.org/D72533
In the previous state, we were relying on forcing the linker to include
all libraries in the final binary and the global initializer to self-register
every piece of the system. This change help moving away from this model, and
allow users to compose pieces more freely. The current change is only "fixing"
the dialect registration and avoiding relying on "whole link" for the passes.
The translation is still relying on the global registry, and some refactoring
is needed to make this all more convenient.
Differential Revision: https://reviews.llvm.org/D74461
The existing (default) calling convention for memrefs in standard-to-LLVM
conversion was motivated by interfacing with LLVM IR produced from C sources.
In particular, it passes a pointer to the memref descriptor structure when
calling the function. Therefore, the descriptor is allocated on stack before
the call. This convention leads to several problems. PR44644 indicates a
problem with stack exhaustion when calling functions with memref-typed
arguments in a loop. Allocating outside of the loop may lead to concurrent
access problems in case the loop is parallel. When targeting GPUs, the contents
of the stack-allocated memory for the descriptor (passed by pointer) needs to
be explicitly copied to the device. Using an aggregate type makes it impossible
to attach pointer-specific argument attributes pertaining to alignment and
aliasing in the LLVM dialect.
Change the default calling convention for memrefs in standard-to-LLVM
conversion to transform a memref into a list of arguments, each of primitive
type, that are comprised in the memref descriptor. This avoids stack allocation
for ranked memrefs (and thus stack exhaustion and potential concurrent access
problems) and simplifies the device function invocation on GPUs.
Provide an option in the standard-to-LLVM conversion to generate auxiliary
wrapper function with the same interface as the previous calling convention,
compatible with LLVM IR porduced from C sources. These auxiliary functions
pack the individual values into a descriptor structure or unpack it. They also
handle descriptor stack allocation if necessary, serving as an allocation
scope: the memory reserved by `alloca` will be freed on exiting the auxiliary
function.
The effect of this change on MLIR-generated only LLVM IR is minimal. When
interfacing MLIR-generated LLVM IR with C-generated LLVM IR, the integration
only needs to require auxiliary functions and change the function name to call
the wrapper function instead of the original function.
This also opens the door to forwarding aliasing and alignment information from
memrefs to LLVM IR pointers in the standrd-to-LLVM conversion.
Summary:
In the original design, gpu.launch required explicit capture of uses
and passing them as operands to the gpu.launch operation. This was
motivated by infrastructure restrictions rather than design. This
change lifts the requirement and removes the concept of kernel
arguments from gpu.launch. Instead, the kernel outlining
transformation now does the explicit capturing.
This is a breaking change for users of gpu.launch.
Differential Revision: https://reviews.llvm.org/D73769
Summary:
The 'gpu.terminator' operation is used as the terminator for the
regions of gpu.launch. This is to disambugaute them from the
return operation on 'gpu.func' functions.
This is a breaking change and users of the gpu dialect will need
to adapt their code when producting 'gpu.launch' operations.
Reviewers: nicolasvasilache
Subscribers: mehdi_amini, rriddle, jpienaar, burmako, shauheen, antiagainst, csigg, arpith-jacob, mgester, lucyrfox, liufengdb, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D73620
Stella Laurenzo