Home
last modified time | relevance | path

Searched refs:operations (Results 1 – 25 of 945) sorted by relevance

12345678910>>...38

/llvm-project-15.0.7/llvm/lib/Target/PowerPC/
H A DP10InstrResources.td1204 // 3 Cycles ALU operations, and 3 Cycles ALU operations, 0 input operands
1214 // 3 Cycles ALU operations, and 3 Cycles ALU operations, 1 input operands
1256 // 3 Cycles ALU operations, 3 Cycles ALU operations, 3 Cycles ALU operations, and 3 Cycles ALU oper…
1379 // 6 Cycles Load operations, 6 Cycles Load operations, 6 Cycles Load operations, and 6 Cycles Load …
1386 // 6 Cycles Load operations, 6 Cycles Load operations, 6 Cycles Load operations, and 6 Cycles Load …
1439 // 6 Cycles Load operations, 4 Cycles ALU2 operations, 3 Cycles Simple Fixed-point (SFX) operations…
1572 …ix Multiply Engine operations, 3 Cycles ALU operations, 10 Cycles SIMD Matrix Multiply Engine oper…
1917 …operations, 3 Cycles ALU operations, 3 Cycles Store operations, 3 Cycles ALU operations, 3 Cycles …
1925 // 3 Cycles Store operations, 3 Cycles ALU operations, 3 Cycles Store operations, and 3 Cycles Stor…
1933 // 3 Cycles Store operations, 3 Cycles Store operations, 3 Cycles Store operations, and 3 Cycles St…
[all …]
/llvm-project-15.0.7/mlir/test/python/ir/
H A Doperation.py59 operations = list(blocks[0].operations)
139 func = module.body.operations[0]
154 func = module.body.operations[0]
213 func = module.body.operations[0]
240 func = module.body.operations[0]
507 op = module.body.operations[0]
655 print(module.body.operations[2])
828 foo = m1.body.operations[0]
829 bar = m2.body.operations[0]
830 qux = m2.body.operations[1]
[all …]
H A Dinsertion_point.py25 entry_block = module.body.operations[0].regions[0].blocks[0]
46 entry_block = module.body.operations[0].regions[0].blocks[0]
47 ip = InsertionPoint(entry_block.operations[1])
67 entry_block = module.body.operations[0].regions[0].blocks[0]
96 entry_block = module.body.operations[0].regions[0].blocks[0]
116 entry_block = module.body.operations[0].regions[0].blocks[0]
137 entry_block = module.body.operations[0].regions[0].blocks[0]
158 entry_block = module.body.operations[0].regions[0].blocks[0]
H A Dsymbol_table.py63 qux = m2.body.operations[0]
69 foo2 = m2.body.operations[0]
82 symbol_table.insert(m2.body.operations[0])
101 foo, bar = list(m.operation.regions[0].blocks[0].operations)[0:2]
124 foo = m.operation.regions[0].blocks[0].operations[0]
H A Dvalue.py51 func = module.body.operations[0]
55 op = func.regions[0].blocks[0].operations[0]
72 [func] = module.body.operations
74 op, ret = block.operations
/llvm-project-15.0.7/mlir/include/mlir/Conversion/
H A DPasses.td20 "operations";
23 Convert operations from the affine dialect into operations from the SCF and
26 `affine.for` operations are converted to `scf.for` operations that are free
166 This pass converts bufferization operations into memref operations.
234 Convert ControlFlow operations into LLVM IR dialect operations.
237 IR dialect operations, the pass will fail. Any LLVM IR operations or types
272 Convert Func dialect operations into the LLVM IR dialect operations.
282 IR dialect operations, the pass will fail. Any LLVM IR operations or types
347 let summary = "Generate NVVM operations for gpu operations";
366 let summary = "Generate ROCDL operations for gpu operations";
[all …]
/llvm-project-15.0.7/flang/test/Fir/
H A Dbasic-program.fir20 // PASSES-NEXT: (S) 0 num-cse'd - Number of operations CSE'd
21 // PASSES-NEXT: (S) 0 num-dce'd - Number of operations DCE'd
31 // PASSES-NEXT: (S) 0 num-cse'd - Number of operations CSE'd
32 // PASSES-NEXT: (S) 0 num-dce'd - Number of operations DCE'd
40 // PASSES-NEXT: (S) 0 num-cse'd - Number of operations CSE'd
41 // PASSES-NEXT: (S) 0 num-dce'd - Number of operations DCE'd
50 // PASSES-NEXT: (S) 0 num-cse'd - Number of operations CSE'd
51 // PASSES-NEXT: (S) 0 num-dce'd - Number of operations DCE'd
58 // PASSES-NEXT: (S) 0 num-dce'd - Number of operations eliminated
/llvm-project-15.0.7/mlir/include/mlir/Dialect/Async/
H A DPasses.td15 let summary = "Convert scf.parallel operations to multiple async compute ops "
28 "The number of available workers to execute async operations. If `-1` "
45 "explicit async.runtime and async.coro operations";
58 let summary = "Automatic reference counting for Async runtime operations";
61 `async.execute` and `async.await` operations are lowered to the async
62 runtime API calls, and async coroutine operations.
65 the correct placing of the reference counting operations.
75 let summary = "Optimize automatic reference counting operations for the"
76 "Async runtime by removing redundant operations";
86 `async.execute` and `async.await` operations are lowered to the async
[all …]
/llvm-project-15.0.7/mlir/include/mlir/Dialect/Transform/IR/
H A DTransformDialect.td26 chunks of data in a way that can be decomposed into operations on
68 operations. This is conceptually a set of operations and no assumptions
77 by the operations they contain, e.g., `transform.sequence` executes
92 It is recommended to prefix injected operations with one or several
126 mapping between transform IR values and payload IR operations.
160 by subsequent operations.
204 with operations nested in the operations associated with the consumed
210 operations associated with it are not nested in payload IR operations
282 transform operations and perform client-specific bookkeeping.
356 /// operations in this dialect.
[all …]
H A DTransformInterfaces.td20 This interface is to be implemented by operations that identify
21 transformations to be performed on other operations. The former are referred
22 to as transform IR operations. The latter are referred to as payload IR
23 operations. Such transform IR operations provide a fine-grain control
26 operations in the payload IR. Transformations are applied starting from the
27 operations identified by handles, but may affect other operations as well.
29 operations to control transformations.
40 used for queries, e.g., to obtain the list of operations on which the
/llvm-project-15.0.7/mlir/include/mlir/IR/
H A DBlock.h42 operations.pop_back(); in clear()
128 OpListType &getOperations() { return operations; } in getOperations()
134 iterator begin() { return operations.begin(); } in begin()
135 iterator end() { return operations.end(); } in end()
136 reverse_iterator rbegin() { return operations.rbegin(); } in rbegin()
137 reverse_iterator rend() { return operations.rend(); } in rend()
139 bool empty() { return operations.empty(); } in empty()
143 Operation &back() { return operations.back(); } in back()
144 Operation &front() { return operations.front(); } in front()
337 return &Block::operations; in getSublistAccess()
[all …]
/llvm-project-15.0.7/mlir/test/Pass/
H A Dpipeline-stats.mlir7 // LIST-NEXT: (S) {{0|8}} num-ops - Number of operations counted
8 // LIST-NEXT: (S) {{0|8}} num-ops2 - Number of operations counted one more time
14 // PIPELINE-NEXT: (S) {{0|4}} num-ops - Number of operations counted
15 // PIPELINE-NEXT: (S) {{0|4}} num-ops2 - Number of operations counted one more time
17 // PIPELINE-NEXT: (S) {{0|4}} num-ops - Number of operations counted
18 // PIPELINE-NEXT: (S) {{0|4}} num-ops2 - Number of operations counted one more time
/llvm-project-15.0.7/mlir/include/mlir/Transforms/
H A DPasses.td20 let summary = "Canonicalize operations";
41 let summary = "Sink operations into conditional blocks";
43 This pass implements control-flow sink on operations that implement
60 Statistic<"numSunk", "num-sunk", "Number of operations sunk">,
75 Statistic<"numCSE", "num-cse'd", "Number of operations CSE'd">,
76 Statistic<"numDCE", "num-dce'd", "Number of operations DCE'd">
100 generate new locations for the operations.
146 let summary = "Print statistics of operations";
169 let summary = "Strip debug info from all operations";
186 or other non-Symbol operations.
[all …]
/llvm-project-15.0.7/mlir/include/mlir/Dialect/Tosa/Transforms/
H A DPasses.td19 let summary = "Fold layerwise operations on constant tensors";
21 Pass that enables folding of full-layer operations on constant tensors.
28 let summary = "Propagate shapes across TOSA operations";
30 Pass that uses operand types and propagates shapes to TOSA operations.
46 number of dimensions. Insert RESHAPE operations to prepend dimensions
57 let summary = "Applies Tosa operations optional decompositions";
59 Pass to apply the Tosa operations decompositions
/llvm-project-15.0.7/mlir/docs/
H A DTraits.md6 operations, and types that suit a specific level of abstraction. `Traits` are a
8 across many different attributes/operations/types/etc.. `Traits` may be used to
20 `AttributeTrait::TraitBase`. For operations, this is `OpTrait::TraitBase`. For
42 operations in the regions will be verified before the verification of this
207 this trait includes all operations in its region excluding operations that are
208 nested inside of other operations that themselves have this trait.
266 within regions that are attached to operations of `ParentOpType`.
295 operations can be modified so that the `MemRef` has an identity layout
320 This trait is used for operations that define a
328 to be [terminators](LangRef.md#terminator-operations).
[all …]
H A DSymbolsAndSymbolTables.md7 around this nesting structure; including the processing of operations within the
9 design is that it is able to process operations in parallel, utilizing multiple
14 operations defined above. Making this thread-safe requires expensive locking in
23 operations defined above regions that were defined as `IsolatedFromAbove` in a
24 safe way. It also allows for symbolically referencing operations define below
32 result value, and may be referred to by other operations to provide a symbolic
35 symbol name, which is [referred to](#referencing-a-symbol) by operations like
41 necessary verification and accessors; it also supports operations, such as
63 the [`Symbol`](#symbol) operations. It verifies that all `Symbol` operations
136 operations that materialize SSA values from a symbol reference. Each has
[all …]
H A DLangRef.md45 and the set of operations usable in MLIR can be arbitrarily extended.
190 consuming the MLIR syntax. [Operations](#operations),
245 ecosystem. They allow for defining new [operations](#operations), as well as
265 ### Target specific operations
272 operations). LLVM intrinsics in MLIR are represented via operations that start
308 [target-independent operations](Dialects/MemRef.md),
309 [affine operations](Dialects/Affine.md), and
310 [target-specific machine operations](#target-specific-operations).
363 A *Block* is a list of operations. In
515 of terminator operations is determined by the specific dialect operations
[all …]
H A DDialectConversion.md5 illegal operations to those supported by a provided conversion target, via a set
18 When applying a conversion to a set of operations, there are several different
23 - A partial conversion will legalize as many operations to the target as
24 possible, but will allow pre-existing operations that were not
27 operations.
45 operations.
90 /// Mark all operations within the LLVM dialect are legal.
99 /// Mark all operations within Affine dialect have dynamic legality
107 /// Treat unknown operations, i.e. those without a legalization action
114 /// All operations within the GPU dialect are illegal.
[all …]
/llvm-project-15.0.7/mlir/include/mlir/Interfaces/
H A DCallInterfaces.td10 // related to call-like and callable operations. Each of which are defined along
24 /// Interface for call-like operations.
28 another. These operations may be traditional direct calls `call @foo`, or
29 indirect calls to other operations `call_indirect %foo`. An operation that
61 /// Interface for callable operations.
66 above). These operations may be traditional functional operation
67 `func @foo(...)`, as well as function producing operations
68 `%foo = dialect.create_function(...)`. These operations may only contain a
/llvm-project-15.0.7/mlir/test/IR/
H A Dprint-ir-nesting.mlir6 // CHECK: Block with 0 arguments, 0 successors, and 2 operations
22 // CHECK: Block with 0 arguments, 0 successors, and 1 operations
30 // CHECK: Block with 0 arguments, 2 successors, and 2 operations
37 // CHECK: Block with 1 arguments, 0 successors, and 2 operations
45 // CHECK: Block with 1 arguments, 0 successors, and 2 operations
/llvm-project-15.0.7/mlir/include/mlir/Dialect/PDLInterp/IR/
H A DPDLInterpOps.td30 interpreter code generation. The dialect operations encapsulates
32 IR (Operation::getOperand), creating new operations (OpBuilder::create),
65 // Check operations evaluate a predicate on a positional value and then
75 // Switch operations evaluate a predicate on a positional value and then
151 `pdl_interp.are_equal` operations compare two positional values for
196 `pdl_interp.check_attribute` operations compare the value of a given
335 `pdl_interp.check_types` operations compare a range of types with a
488 `pdl_interp.create_types` operations generate a handle within the
639 callable SSA-region operations that contain other interpreter operations.
789 `pdl_interp.get_operands` operations try to get a specific operand
[all …]
/llvm-project-15.0.7/mlir/lib/Tools/PDLL/ODS/
H A DDialect.cpp28 std::unique_ptr<Operation> &operation = operations[name]; in insertOperation()
38 auto it = operations.find(name); in lookupOperation()
39 return it != operations.end() ? it->second.get() : nullptr; in lookupOperation()
/llvm-project-15.0.7/mlir/include/mlir/Dialect/PDL/IR/
H A DPDLOps.td1 //===- PDLOps.td - Pattern descriptor operations -----------*- tablegen -*-===//
9 // This file declares the Pattern Descriptor Language dialect operations.
36 `pdl.apply_native_constraint` operations apply a native C++ constraint, that
61 `pdl.apply_native_rewrite` operations apply a native C++ function, that has
172 node that originate from operations or block arguments not otherwise
210 `pdl.operands` operations capture external operand range edges into an
211 operation node that originate from operations or block arguments not
213 `pdl.results`). These operations define groups of input operands into a
251 operations correspond to input operations, or those that already existing
252 within the MLIR module. Inside of a `pdl.rewrite`, these operations
[all …]
/llvm-project-15.0.7/flang/include/flang/Optimizer/Transforms/
H A DPasses.td41 Convert fir operations which satisfy affine constraints to the affine
50 the fir condition's parent operations to construct the integer set.
64 let summary = "Converts `affine.{load,store}` back to fir operations";
68 the Fortran runtime. Therefore, conversion of memory operations back to
78 let summary = "Annotate constant operands to all FIR operations";
95 let summary = "Convert array value operations to memory operations.";
102 to determine if there are potential dependences between these operations.
103 If not, these array operations can be lowered to work directly on the memory
105 along with appropriate copy-in/copy-out operations. Here, a more refined
136 plain old test and branch operations. Removing the high-level control
[all …]
/llvm-project-15.0.7/mlir/test/python/dialects/
H A Dmemref.py31 func_body = module.body.operations[0].regions[0].blocks[0]
32 subview = func_body.operations[6]
66 f = module.body.operations[0]

12345678910>>...38