10056c821a
This commit makes reductions part of the terminator. Instead of
`scf.yield`, `scf.reduce` now terminates the body of `scf.parallel` ops.
`scf.reduce` may contain an arbitrary number of reductions, with one
region per reduction.
Example:
```mlir
%init = arith.constant 0.0 : f32
%r:2 = scf.parallel (%iv) = (%lb) to (%ub) step (%step) init (%init, %init)
-> f32, f32 {
%elem_to_reduce1 = load %buffer1[%iv] : memref<100xf32>
%elem_to_reduce2 = load %buffer2[%iv] : memref<100xf32>
scf.reduce(%elem_to_reduce1, %elem_to_reduce2 : f32, f32) {
^bb0(%lhs : f32, %rhs: f32):
%res = arith.addf %lhs, %rhs : f32
scf.reduce.return %res : f32
}, {
^bb0(%lhs : f32, %rhs: f32):
%res = arith.mulf %lhs, %rhs : f32
scf.reduce.return %res : f32
}
}
```
`scf.reduce` operations can no longer be interleaved with other ops in
the body of `scf.parallel`. This simplifies the op and makes it possible
to assign the `RecursiveMemoryEffects` trait to `scf.reduce`. (This was
not possible before because the op was not a terminator, causing the op
to be DCE'd.)
98 lines
4.0 KiB
MLIR
98 lines
4.0 KiB
MLIR
// RUN: mlir-opt %s -convert-linalg-to-parallel-loops -split-input-file | FileCheck %s
|
|
|
|
#map0 = affine_map<(d0, d1) -> (d0, d1)>
|
|
func.func @linalg_generic_sum(%lhs: memref<2x2xf32>,
|
|
%rhs: memref<2x2xf32>,
|
|
%sum: memref<2x2xf32>) {
|
|
linalg.generic {
|
|
indexing_maps = [#map0, #map0, #map0],
|
|
iterator_types = ["parallel", "parallel"]}
|
|
ins(%lhs, %rhs : memref<2x2xf32>, memref<2x2xf32>)
|
|
outs(%sum : memref<2x2xf32>) {
|
|
^bb0(%lhs_in: f32, %rhs_in: f32, %sum_out: f32):
|
|
%0 = arith.addf %lhs_in, %rhs_in : f32
|
|
linalg.yield %0 : f32
|
|
}
|
|
return
|
|
}
|
|
// CHECK-LABEL: @linalg_generic_sum
|
|
// CHECK: (%[[LHS:.*]]:{{.*}}, %[[RHS:.*]]:{{.*}}, %[[SUM:.*]]:{{.*}})
|
|
// CHECK-DAG: %[[C2:.*]] = arith.constant 2
|
|
// CHECK-DAG: %[[C0:.*]] = arith.constant 0
|
|
// CHECK-DAG: %[[C1:.*]] = arith.constant 1
|
|
// CHECK: scf.parallel (%[[I:.*]], %[[J:.*]]) = {{.*}}
|
|
// CHECK: %[[LHS_ELEM:.*]] = memref.load %[[LHS]][%[[I]], %[[J]]]
|
|
// CHECK: %[[RHS_ELEM:.*]] = memref.load %[[RHS]][%[[I]], %[[J]]]
|
|
// CHECK: %[[SUM:.*]] = arith.addf %[[LHS_ELEM]], %[[RHS_ELEM]] : f32
|
|
// CHECK: store %[[SUM]], %{{.*}}[%[[I]], %[[J]]]
|
|
// CHECK: scf.reduce
|
|
|
|
// -----
|
|
|
|
#accesses = [
|
|
affine_map<(d0, d1, d2, d3) -> (d0, d1, d2, d3)>,
|
|
affine_map<(d0, d1, d2, d3) -> (d0, d1, d3)>
|
|
]
|
|
#trait = {
|
|
iterator_types = ["parallel", "parallel", "reduction", "parallel"],
|
|
indexing_maps = #accesses
|
|
}
|
|
|
|
func.func @lower_outer_parallel(%A: memref<?x?x?x?xf32>, %B: memref<?x?x?xf32>) {
|
|
linalg.generic #trait
|
|
ins(%A : memref<?x?x?x?xf32>)
|
|
outs(%B : memref<?x?x?xf32>) {
|
|
^bb0(%a: f32, %b: f32):
|
|
linalg.yield %a: f32
|
|
}
|
|
return
|
|
}
|
|
// CHECK-LABEL: @lower_outer_parallel
|
|
// CHECK-DAG: %[[C0:.*]] = arith.constant 0
|
|
// CHECK-DAG: %[[C1:.*]] = arith.constant 1
|
|
// CHECK-DAG: %[[D0:.*]] = memref.dim %{{.*}}, %c0
|
|
// CHECK-DAG: %[[D1:.*]] = memref.dim %{{.*}}, %c1
|
|
// CHECK-DAG: %[[D2:.*]] = memref.dim %{{.*}}, %c2
|
|
// CHECK-DAG: %[[D3:.*]] = memref.dim %{{.*}}, %c3
|
|
// CHECK: scf.parallel (%[[IV0:.*]], %[[IV1:.*]]) = (%[[C0]], %[[C0]]) to (%[[D0]], %[[D1]]) step (%[[C1]], %[[C1]])
|
|
// CHECK: scf.for %[[IV2:.*]] = %[[C0]] to %[[D2]] step %[[C1]]
|
|
// CHECK: scf.parallel (%[[IV3:.*]]) = (%[[C0]]) to (%[[D3]]) step (%[[C1]])
|
|
// CHECK: memref.load %{{.*}}[%[[IV0]], %[[IV1]], %[[IV2]], %[[IV3]]]
|
|
// CHECK: store %{{.*}}, %{{.*}}[%[[IV0]], %[[IV1]], %[[IV3]]]
|
|
|
|
// -----
|
|
|
|
#accesses = [
|
|
affine_map<(d0, d1, d2, d3, d4, d5) -> (d0, d1, d2, d3, d4, d5)>,
|
|
affine_map<(d0, d1, d2, d3, d4, d5) -> (d0, d1, d4, d3)>
|
|
]
|
|
#trait = {
|
|
iterator_types = ["parallel", "parallel", "reduction", "parallel", "parallel", "reduction"],
|
|
indexing_maps = #accesses
|
|
}
|
|
|
|
func.func @lower_mixed_parallel(%A: memref<?x?x?x?x?x?xf32>, %B: memref<?x?x?x?xf32>) {
|
|
linalg.generic #trait
|
|
ins(%A : memref<?x?x?x?x?x?xf32>)
|
|
outs(%B : memref<?x?x?x?xf32>) {
|
|
^bb0(%a: f32, %b: f32):
|
|
linalg.yield %a: f32
|
|
}
|
|
return
|
|
}
|
|
// CHECK-LABEL: @lower_mixed_parallel
|
|
// CHECK-DAG: %[[C0:.*]] = arith.constant 0
|
|
// CHECK-DAG: %[[C1:.*]] = arith.constant 1
|
|
// CHECK-DAG: %[[D0:.*]] = memref.dim %{{.*}}, %c0
|
|
// CHECK-DAG: %[[D1:.*]] = memref.dim %{{.*}}, %c1
|
|
// CHECK-DAG: %[[D2:.*]] = memref.dim %{{.*}}, %c2
|
|
// CHECK-DAG: %[[D3:.*]] = memref.dim %{{.*}}, %c3
|
|
// CHECK-DAG: %[[D4:.*]] = memref.dim %{{.*}}, %c4
|
|
// CHECK-DAG: %[[D5:.*]] = memref.dim %{{.*}}, %c5
|
|
// CHECK: scf.parallel (%[[IV0:.*]], %[[IV1:.*]]) = (%[[C0]], %[[C0]]) to (%[[D0]], %[[D1]]) step (%[[C1]], %[[C1]])
|
|
// CHECK: scf.for %[[IV2:.*]] = %[[C0]] to %[[D2]] step %[[C1]]
|
|
// CHECK: scf.parallel (%[[IV3:.*]], %[[IV4:.*]]) = (%[[C0]], %[[C0]]) to (%[[D3]], %[[D4]]) step (%[[C1]], %[[C1]])
|
|
// CHECK: scf.for %[[IV5:.*]] = %[[C0]] to %[[D5]] step %[[C1]]
|
|
// CHECK: memref.load %{{.*}}[%[[IV0]], %[[IV1]], %[[IV2]], %[[IV3]], %[[IV4]], %[[IV5]]]
|
|
// CHECK: store %{{.*}}, %{{.*}}[%[[IV0]], %[[IV1]], %[[IV4]], %[[IV3]]]
|