maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/mlir/test/Transforms/loop-fusion-slice-computati...

146 lines
6.8 KiB
MLIR
Raw Normal View History

Rename -verify mlir-opt flag to -verify-expected-diagnostics This name has caused some confusion because it suggests that it's running op verification (and that this verification isn't getting run by default). PiperOrigin-RevId: 254035268
2019-06-20 02:21:41 +08:00
// RUN: mlir-opt %s -test-loop-fusion -test-loop-fusion-slice-computation -split-input-file -verify-diagnostics | FileCheck %s
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
// -----
// CHECK-LABEL: func @slice_depth1_loop_nest() {
func @slice_depth1_loop_nest() {
%0 = alloc() : memref<100xf32>
%cst = constant 7.000000e+00 : f32
affine.for %i0 = 0 to 16 {
// expected-remark@-1 {{slice ( src loop: 1, dst loop: 0, depth: 1 : insert point: (1, 1) loop bounds: [(d0) -> (d0), (d0) -> (d0 + 1)] )}}
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
affine.store %cst, %0[%i0] : memref<100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
affine.for %i1 = 0 to 5 {
// expected-remark@-1 {{slice ( src loop: 0, dst loop: 1, depth: 1 : insert point: (1, 0) loop bounds: [(d0) -> (d0), (d0) -> (d0 + 1)] )}}
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
%1 = affine.load %0[%i1] : memref<100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
return
}
// -----
// Loop %i0 writes to locations [2, 17] and loop %i0 reads from locations [3, 6]
// Slice loop bounds should be adjusted such that the load/store are for the
// same location.
// CHECK-LABEL: func @slice_depth1_loop_nest_with_offsets() {
func @slice_depth1_loop_nest_with_offsets() {
%0 = alloc() : memref<100xf32>
%cst = constant 7.000000e+00 : f32
affine.for %i0 = 0 to 16 {
// expected-remark@-1 {{slice ( src loop: 1, dst loop: 0, depth: 1 : insert point: (1, 2) loop bounds: [(d0) -> (d0 + 3), (d0) -> (d0 + 4)] )}}
%a0 = affine.apply (d0) -> (d0 + 2)(%i0)
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
affine.store %cst, %0[%a0] : memref<100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
affine.for %i1 = 4 to 8 {
// expected-remark@-1 {{slice ( src loop: 0, dst loop: 1, depth: 1 : insert point: (1, 0) loop bounds: [(d0) -> (d0 - 3), (d0) -> (d0 - 2)] )}}
%a1 = affine.apply (d0) -> (d0 - 1)(%i1)
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
%1 = affine.load %0[%a1] : memref<100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
return
}
// -----
// Slices at loop depth 1 should only slice the loop bounds of the first loop.
Minor spelling tweaks Closes tensorflow/mlir#304 PiperOrigin-RevId: 284568358
2019-12-10 01:23:15 +08:00
// Slices at loop depth 2 should slice loop bounds of both loops.
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
// CHECK-LABEL: func @slice_depth2_loop_nest() {
func @slice_depth2_loop_nest() {
%0 = alloc() : memref<100x100xf32>
%cst = constant 7.000000e+00 : f32
affine.for %i0 = 0 to 16 {
// expected-remark@-1 {{slice ( src loop: 1, dst loop: 0, depth: 1 : insert point: (1, 1) loop bounds: [(d0) -> (d0), (d0) -> (d0 + 1)] [(d0) -> (0), (d0) -> (8)] )}}
// expected-remark@-2 {{slice ( src loop: 1, dst loop: 0, depth: 2 : insert point: (2, 1) loop bounds: [(d0, d1) -> (d0), (d0, d1) -> (d0 + 1)] [(d0, d1) -> (d1), (d0, d1) -> (d1 + 1)] )}}
affine.for %i1 = 0 to 16 {
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
affine.store %cst, %0[%i0, %i1] : memref<100x100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
}
affine.for %i2 = 0 to 10 {
// expected-remark@-1 {{slice ( src loop: 0, dst loop: 1, depth: 1 : insert point: (1, 0) loop bounds: [(d0) -> (d0), (d0) -> (d0 + 1)] [(d0) -> (0), (d0) -> (8)] )}}
// expected-remark@-2 {{slice ( src loop: 0, dst loop: 1, depth: 2 : insert point: (2, 0) loop bounds: [(d0, d1) -> (d0), (d0, d1) -> (d0 + 1)] [(d0, d1) -> (d1), (d0, d1) -> (d1 + 1)] )}}
affine.for %i3 = 0 to 8 {
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
%1 = affine.load %0[%i2, %i3] : memref<100x100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
}
return
}
// -----
// The load at depth 1 in loop nest %i2 prevents slicing loop nest %i0 at depths
// greater than 1. However, loop nest %i2 can be sliced into loop nest %i0 at
// depths 1 and 2 because the dependent store in loop nest %i0 is at depth 2.
// CHECK-LABEL: func @slice_depth2_loop_nest_two_loads() {
func @slice_depth2_loop_nest_two_loads() {
%0 = alloc() : memref<100x100xf32>
%c0 = constant 0 : index
%cst = constant 7.000000e+00 : f32
affine.for %i0 = 0 to 16 {
Extend map canonicalization to propagate constant operands - extend canonicalizeMapAndOperands to propagate constant operands into the map's expressions (and thus drop those operands). - canonicalizeMapAndOperands previously only dropped duplicate and unused operands; however, operands that were constants were retained. This change makes IR maps/expressions generated by various utilities/passes even simpler; also makes some of the test checks more accurate and simpler -- for eg., 0' instead of symbol(%{{.*}}). Signed-off-by: Uday Bondhugula <uday@polymagelabs.com> Closes tensorflow/mlir#107 COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/107 from bondhugula:canonicalize-maps c889a51486d14fbf7db489f224f881e7e1ff7d72 PiperOrigin-RevId: 266085289
2019-08-29 16:13:01 +08:00
// expected-remark@-1 {{slice ( src loop: 1, dst loop: 0, depth: 1 : insert point: (1, 1) loop bounds: [(d0) -> (d0), (d0) -> (d0 + 1)] [(d0) -> (0), (d0) -> (8)] )}}
// expected-remark@-2 {{slice ( src loop: 1, dst loop: 0, depth: 2 : insert point: (2, 1) loop bounds: [(d0, d1) -> (d0), (d0, d1) -> (d0 + 1)] [(d0, d1) -> (0), (d0, d1) -> (8)] )}}
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
affine.for %i1 = 0 to 16 {
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
affine.store %cst, %0[%i0, %i1] : memref<100x100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
}
affine.for %i2 = 0 to 10 {
Extend map canonicalization to propagate constant operands - extend canonicalizeMapAndOperands to propagate constant operands into the map's expressions (and thus drop those operands). - canonicalizeMapAndOperands previously only dropped duplicate and unused operands; however, operands that were constants were retained. This change makes IR maps/expressions generated by various utilities/passes even simpler; also makes some of the test checks more accurate and simpler -- for eg., 0' instead of symbol(%{{.*}}). Signed-off-by: Uday Bondhugula <uday@polymagelabs.com> Closes tensorflow/mlir#107 COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/107 from bondhugula:canonicalize-maps c889a51486d14fbf7db489f224f881e7e1ff7d72 PiperOrigin-RevId: 266085289
2019-08-29 16:13:01 +08:00
// expected-remark@-1 {{slice ( src loop: 0, dst loop: 1, depth: 1 : insert point: (1, 0) loop bounds: [(d0) -> (d0), (d0) -> (d0 + 1)] [(d0) -> (0), (d0) -> (8)] )}}
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
affine.for %i3 = 0 to 8 {
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
%1 = affine.load %0[%i2, %i3] : memref<100x100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
%2 = affine.load %0[%i2, %c0] : memref<100x100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
return
}
// -----
// The store at depth 1 in loop nest %i0 prevents slicing loop nest %i2 at
// depths greater than 1 into loop nest %i0. However, loop nest %i0 can be
// sliced into loop nest %i2 at depths 1 and 2 because the dependent load in
// loop nest %i2 is at depth 2.
// CHECK-LABEL: func @slice_depth2_loop_nest_two_stores() {
func @slice_depth2_loop_nest_two_stores() {
%0 = alloc() : memref<100x100xf32>
%c0 = constant 0 : index
%cst = constant 7.000000e+00 : f32
affine.for %i0 = 0 to 16 {
Extend map canonicalization to propagate constant operands - extend canonicalizeMapAndOperands to propagate constant operands into the map's expressions (and thus drop those operands). - canonicalizeMapAndOperands previously only dropped duplicate and unused operands; however, operands that were constants were retained. This change makes IR maps/expressions generated by various utilities/passes even simpler; also makes some of the test checks more accurate and simpler -- for eg., 0' instead of symbol(%{{.*}}). Signed-off-by: Uday Bondhugula <uday@polymagelabs.com> Closes tensorflow/mlir#107 COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/107 from bondhugula:canonicalize-maps c889a51486d14fbf7db489f224f881e7e1ff7d72 PiperOrigin-RevId: 266085289
2019-08-29 16:13:01 +08:00
// expected-remark@-1 {{slice ( src loop: 1, dst loop: 0, depth: 1 : insert point: (1, 2) loop bounds: [(d0) -> (d0), (d0) -> (d0 + 1)] [(d0) -> (0), (d0) -> (8)] )}}
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
affine.for %i1 = 0 to 16 {
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
affine.store %cst, %0[%i0, %i1] : memref<100x100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
affine.store %cst, %0[%i0, %c0] : memref<100x100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
affine.for %i2 = 0 to 10 {
Extend map canonicalization to propagate constant operands - extend canonicalizeMapAndOperands to propagate constant operands into the map's expressions (and thus drop those operands). - canonicalizeMapAndOperands previously only dropped duplicate and unused operands; however, operands that were constants were retained. This change makes IR maps/expressions generated by various utilities/passes even simpler; also makes some of the test checks more accurate and simpler -- for eg., 0' instead of symbol(%{{.*}}). Signed-off-by: Uday Bondhugula <uday@polymagelabs.com> Closes tensorflow/mlir#107 COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/107 from bondhugula:canonicalize-maps c889a51486d14fbf7db489f224f881e7e1ff7d72 PiperOrigin-RevId: 266085289
2019-08-29 16:13:01 +08:00
// expected-remark@-1 {{slice ( src loop: 0, dst loop: 1, depth: 1 : insert point: (1, 0) loop bounds: [(d0) -> (d0), (d0) -> (d0 + 1)] [(d0) -> (0), (d0) -> (16)] )}}
// expected-remark@-2 {{slice ( src loop: 0, dst loop: 1, depth: 2 : insert point: (2, 0) loop bounds: [(d0, d1) -> (d0), (d0, d1) -> (d0 + 1)] [(d0, d1) -> (0), (d0, d1) -> (16)] )}}
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
affine.for %i3 = 0 to 8 {
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
%1 = affine.load %0[%i2, %i3] : memref<100x100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
}
return
}
// -----
// Test loop nest which has a smaller outer trip count than its inner loop.
// CHECK-LABEL: func @slice_loop_nest_with_smaller_outer_trip_count() {
func @slice_loop_nest_with_smaller_outer_trip_count() {
%0 = alloc() : memref<100x100xf32>
%c0 = constant 0 : index
%cst = constant 7.000000e+00 : f32
affine.for %i0 = 0 to 16 {
// expected-remark@-1 {{slice ( src loop: 1, dst loop: 0, depth: 1 : insert point: (1, 1) loop bounds: [(d0) -> (d0), (d0) -> (d0 + 1)] [(d0) -> (0), (d0) -> (10)] )}}
// expected-remark@-2 {{slice ( src loop: 1, dst loop: 0, depth: 2 : insert point: (2, 1) loop bounds: [(d0, d1) -> (d0), (d0, d1) -> (d0 + 1)] [(d0, d1) -> (d1), (d0, d1) -> (d1 + 1)] )}}
affine.for %i1 = 0 to 16 {
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
affine.store %cst, %0[%i0, %i1] : memref<100x100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
}
affine.for %i2 = 0 to 8 {
// expected-remark@-1 {{slice ( src loop: 0, dst loop: 1, depth: 1 : insert point: (1, 0) loop bounds: [(d0) -> (d0), (d0) -> (d0 + 1)] [(d0) -> (0), (d0) -> (10)] )}}
// expected-remark@-2 {{slice ( src loop: 0, dst loop: 1, depth: 2 : insert point: (2, 0) loop bounds: [(d0, d1) -> (d0), (d0, d1) -> (d0 + 1)] [(d0, d1) -> (d1), (d0, d1) -> (d1 + 1)] )}}
affine.for %i3 = 0 to 10 {
Globally change load/store/dma_start/dma_wait operations over to affine.load/store/dma_start/dma_wait. In most places, this is just a name change (with the exception of affine.dma_start swapping the operand positions of its tag memref and num_elements operands). Significant code changes occur here: *) Vectorization: LoopAnalysis.cpp, Vectorize.cpp *) Affine Transforms: Transforms/Utils/Utils.cpp PiperOrigin-RevId: 256395088
2019-07-04 01:35:03 +08:00
%1 = affine.load %0[%i2, %i3] : memref<100x100xf32>
LoopFusion: adds support for computing forward computation slices, which will enable fusion of consumer loop nests into their producers in subsequent CLs. PiperOrigin-RevId: 253601994
2019-06-18 00:59:35 +08:00
}
}
return
Extend map canonicalization to propagate constant operands - extend canonicalizeMapAndOperands to propagate constant operands into the map's expressions (and thus drop those operands). - canonicalizeMapAndOperands previously only dropped duplicate and unused operands; however, operands that were constants were retained. This change makes IR maps/expressions generated by various utilities/passes even simpler; also makes some of the test checks more accurate and simpler -- for eg., 0' instead of symbol(%{{.*}}). Signed-off-by: Uday Bondhugula <uday@polymagelabs.com> Closes tensorflow/mlir#107 COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/mlir/pull/107 from bondhugula:canonicalize-maps c889a51486d14fbf7db489f224f881e7e1ff7d72 PiperOrigin-RevId: 266085289
2019-08-29 16:13:01 +08:00
}