maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

[mlir][Linalg] Wrong tile size for convolutions fixed 

Sizes of tiles (subviews) are bigger by 1 than they should. Let's consider
1D convolution without batches or channels. Furthermore let m iterate over
the output and n over the kernel then input is accessed with m + n. In tiling
subview sizes for convolutions are computed by applying requested tile size
together with kernel size to the above mentioned expression thus let's say
for tile size of 2 the subview size is 2 + size(n), which is bigger by one
than it should since we move kernel only once. The problem behind it is that
range is not turned into closed interval before the composition. This commit
fixes the problem by turning ranges first into closed intervals by substracting
1 and after the composition back to half open by adding 1.

Differential Revision: https://reviews.llvm.org/D86638
This commit is contained in:
Jakub Lichman 2020-09-02 14:02:40 +00:00
parent 3746906193
commit 8d35080ebb
3 changed files with 10 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
mlir/lib/Dialect/Linalg/Transforms/Tiling.cpp
View File

@ -243,7 +243,9 @@ static SmallVector<Value, 4> makeTiledViews(OpBuilder &b, Location loc,
for (unsigned idx = 0, idxIvs = 0, e = tileSizes.size(); idx < e; ++idx) {
bool isTiled = !isZero(tileSizes[idx]);
lbs.push_back(isTiled ? ivs[idxIvs++] : (Value)std_constant_index(0));
subViewSizes.push_back(isTiled ? tileSizes[idx] : viewSizes[idx]);
// Before composing, we need to make range a closed interval.
Value size = isTiled ? tileSizes[idx] : viewSizes[idx];
subViewSizes.push_back(size - std_constant_index(1));
}
auto *op = linalgOp.getOperation();
@ -282,7 +284,9 @@ static SmallVector<Value, 4> makeTiledViews(OpBuilder &b, Location loc,
auto m = map.getSubMap({r});
auto offset = applyMapToValues(b, loc, m, lbs).front();
offsets.push_back(offset);
auto size = applyMapToValues(b, loc, m, subViewSizes).front();
auto closedIntSize = applyMapToValues(b, loc, m, subViewSizes).front();
// Resulting size needs to be made half open interval again.
auto size = closedIntSize + std_constant_index(1);
// The size of the subview should be trimmed to avoid out-of-bounds
// accesses, unless we statically know the subview size divides the view

2
mlir/test/Dialect/Linalg/tile_conv.mlir
View File

@ -1,7 +1,7 @@
// RUN: mlir-opt %s -linalg-tile="linalg-tile-sizes=2,3,0,0,4" | FileCheck %s -check-prefix=TILE-23004
// TILE-23004-DAG: #[[$D0x30pS0x10:.*]] = affine_map<(d0) -> (d0 * 30)>
// TILE-23004-DAG: #[[$S0x10p90D0x30pS1:.*]] = affine_map<(d0)[s0, s1] -> (s0 * 10 + 90, d0 * -30 + s1)>
// TILE-23004-DAG: #[[$S0x10p90D0x30pS1:.*]] = affine_map<(d0)[s0, s1] -> (s0 * 10 + 51, d0 * -30 + s1)>
// TILE-23004-DAG: #[[$strided4D:.*]] = affine_map<(d0, d1, d2, d3)[s0, s1, s2, s3] -> (d0 * s1 + s0 + d1 * s2 + d2 * s3 + d3)>
// TILE-23004-DAG: #[[$bound_map_4:.*]] = affine_map<(d0)[s0] -> (4, -d0 + s0)>

6
mlir/test/Dialect/Linalg/tile_simple_conv.mlir
View File

@ -1,8 +1,8 @@
// RUN: mlir-opt %s -linalg-tile="linalg-tile-sizes=2,3,4" | FileCheck %s
// CHECK-DAG: #[[MAP0:.*]] = affine_map<(d0)[s0] -> (2, -d0 + s0)>
// CHECK-DAG: #[[MAP1:.*]] = affine_map<(d0)[s0, s1] -> (s0 + 3, -d0 + s1)>
// CHECK-DAG: #[[MAP2:.*]] = affine_map<(d0)[s0, s1] -> (s0 + 4, -d0 + s1)>
// CHECK-DAG: #[[MAP1:.*]] = affine_map<(d0)[s0, s1] -> (s0 + 2, -d0 + s1)>
// CHECK-DAG: #[[MAP2:.*]] = affine_map<(d0)[s0, s1] -> (s0 + 3, -d0 + s1)>
// CHECK-DAG: #[[MAP4:.*]] = affine_map<(d0)[s0] -> (3, -d0 + s0)>
// CHECK-DAG: #[[MAP5:.*]] = affine_map<(d0)[s0] -> (4, -d0 + s0)>
@ -46,4 +46,4 @@ func @conv(%arg0 : memref<?x?x?x?xf32>, %arg1 : memref<?x?x?x?xf32>, %arg2 : mem
// CHECK: %[[T19:.*]] = dim %[[ARG2]], %[[C3]]
// CHECK: %[[SV2:.*]] = subview %[[ARG2]][%[[ARG3]], %[[ARG4]], %[[ARG5]], 0]
// CHECK-SAME: [%[[T14]], %[[T16]], %[[T18]], %[[T19]]]
// CHECK: linalg.conv(%[[ARG0]], %[[SV1]], %[[SV2]])
// CHECK: linalg.conv(%[[ARG0]], %[[SV1]], %[[SV2]])