Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

dialects: (memref_stream) make bounds in dialect index typed not IntAttr #2810

Merged
merged 2 commits into from
Jul 1, 2024
Merged

dialects: (memref_stream) make bounds in dialect index typed not IntAttr #2810

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
02c9118
testing: unpeel matmul in bottom-up tests
superlopuh Jul 1, 2024
185a889
dialects: (memref_stream) make bounds in dialect index typed not IntAttr
superlopuh Jul 1, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 12 additions & 12 deletions tests/filecheck/dialects/memref_stream/ops.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,7 +44,7 @@ memref_stream.streaming_region {


memref_stream.generic {
bounds = [#builtin.int<3>, #builtin.int<2>],
bounds = [3, 2],
indexing_maps = [
affine_map<(d0, d1) -> (d0)>,
affine_map<(d0, d1) -> (d1)>,
Expand All @@ -56,18 +56,18 @@ memref_stream.generic {
memref_stream.yield %arg3 : f32
}

// CHECK-NEXT: memref_stream.generic {bounds = [#builtin.int<3>, #builtin.int<2>], indexing_maps = [affine_map<(d0, d1) -> (d0)>, affine_map<(d0, d1) -> (d1)>, affine_map<(d0, d1) -> (d0, d1)>], iterator_types = ["parallel", "parallel"]} ins(%A, %B : memref<2xf32>, memref<3xf32>) outs(%C : memref<3x2xf64>) attrs = {"hello" = "world"} {
// CHECK-NEXT: memref_stream.generic {bounds = [3, 2], indexing_maps = [affine_map<(d0, d1) -> (d0)>, affine_map<(d0, d1) -> (d1)>, affine_map<(d0, d1) -> (d0, d1)>], iterator_types = ["parallel", "parallel"]} ins(%A, %B : memref<2xf32>, memref<3xf32>) outs(%C : memref<3x2xf64>) attrs = {"hello" = "world"} {
// CHECK-NEXT: ^1(%arg3 : f32, %arg4 : f32):
// CHECK-NEXT: memref_stream.yield %arg3 : f32
// CHECK-NEXT: }

// CHECK-GENERIC-NEXT: "memref_stream.generic"(%A, %B, %C) <{"bounds" = [#builtin.int<3>, #builtin.int<2>], "init_indices" = [], "indexing_maps" = [affine_map<(d0, d1) -> (d0)>, affine_map<(d0, d1) -> (d1)>, affine_map<(d0, d1) -> (d0, d1)>], "iterator_types" = [#memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<parallel>], "operandSegmentSizes" = array<i32: 2, 1, 0>}> ({
// CHECK-GENERIC-NEXT: "memref_stream.generic"(%A, %B, %C) <{"bounds" = [3 : index, 2 : index], "init_indices" = [], "indexing_maps" = [affine_map<(d0, d1) -> (d0)>, affine_map<(d0, d1) -> (d1)>, affine_map<(d0, d1) -> (d0, d1)>], "iterator_types" = [#memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<parallel>], "operandSegmentSizes" = array<i32: 2, 1, 0>}> ({
// CHECK-GENERIC-NEXT: ^1(%arg3 : f32, %arg4 : f32):
// CHECK-GENERIC-NEXT: "memref_stream.yield"(%arg3) : (f32) -> ()
// CHECK-GENERIC-NEXT: }) {"hello" = "world"} : (memref<2xf32>, memref<3xf32>, memref<3x2xf64>) -> ()

memref_stream.generic {
bounds = [#builtin.int<3>, #builtin.int<2>],
bounds = [3, 2],
indexing_maps = [
affine_map<(d0, d1) -> ()>,
affine_map<(d0, d1) -> (d0, d1)>
Expand All @@ -78,12 +78,12 @@ memref_stream.generic {
memref_stream.yield %d : f64
}

// CHECK-NEXT: memref_stream.generic {bounds = [#builtin.int<3>, #builtin.int<2>], indexing_maps = [affine_map<(d0, d1) -> ()>, affine_map<(d0, d1) -> (d0, d1)>], iterator_types = ["parallel", "parallel"]} ins(%D : f64) outs(%C : memref<3x2xf64>) {
// CHECK-NEXT: memref_stream.generic {bounds = [3, 2], indexing_maps = [affine_map<(d0, d1) -> ()>, affine_map<(d0, d1) -> (d0, d1)>], iterator_types = ["parallel", "parallel"]} ins(%D : f64) outs(%C : memref<3x2xf64>) {
// CHECK-NEXT: ^2(%d : f64, %c_1 : f64):
// CHECK-NEXT: memref_stream.yield %d : f64
// CHECK-NEXT: }

// CHECK-GENERIC-NEXT: "memref_stream.generic"(%D, %C) <{"bounds" = [#builtin.int<3>, #builtin.int<2>], "init_indices" = [], "indexing_maps" = [affine_map<(d0, d1) -> ()>, affine_map<(d0, d1) -> (d0, d1)>], "iterator_types" = [#memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<parallel>], "operandSegmentSizes" = array<i32: 1, 1, 0>}> ({
// CHECK-GENERIC-NEXT: "memref_stream.generic"(%D, %C) <{"bounds" = [3 : index, 2 : index], "init_indices" = [], "indexing_maps" = [affine_map<(d0, d1) -> ()>, affine_map<(d0, d1) -> (d0, d1)>], "iterator_types" = [#memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<parallel>], "operandSegmentSizes" = array<i32: 1, 1, 0>}> ({
// CHECK-GENERIC-NEXT: ^2(%d : f64, %c_1 : f64):
// CHECK-GENERIC-NEXT: "memref_stream.yield"(%d) : (f64) -> ()
// CHECK-GENERIC-NEXT: }) : (f64, memref<3x2xf64>) -> ()
Expand All @@ -93,7 +93,7 @@ memref_stream.generic {
// CHECK-GENERIC-NEXT: %E, %F, %G, %H = "test.op"() : () -> (memref<4x2xf64>, memref<2x3xf64>, memref<4x3xf64>, f64)

memref_stream.generic {
bounds = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>],
bounds = [4, 2, 3],
indexing_maps = [
affine_map<(d0, d1, d2) -> (d0, d2)>,
affine_map<(d0, d1, d2) -> (d2, d1)>,
Expand All @@ -107,14 +107,14 @@ memref_stream.generic {
linalg.yield %acc_new : f64
}

// CHECK-NEXT: memref_stream.generic {bounds = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>], indexing_maps = [affine_map<(d0, d1, d2) -> (d0, d2)>, affine_map<(d0, d1, d2) -> (d2, d1)>, affine_map<(d0, d1) -> (d0, d1)>], iterator_types = ["parallel", "parallel", "reduction"]} ins(%{{.*}}, %{{.*}} : memref<4x2xf64>, memref<2x3xf64>) outs(%{{.*}} : memref<4x3xf64>) inits(%H : f64) {
// CHECK-NEXT: memref_stream.generic {bounds = [4, 2, 3], indexing_maps = [affine_map<(d0, d1, d2) -> (d0, d2)>, affine_map<(d0, d1, d2) -> (d2, d1)>, affine_map<(d0, d1) -> (d0, d1)>], iterator_types = ["parallel", "parallel", "reduction"]} ins(%{{.*}}, %{{.*}} : memref<4x2xf64>, memref<2x3xf64>) outs(%{{.*}} : memref<4x3xf64>) inits(%H : f64) {
// CHECK-NEXT: ^{{.*}}(%{{.*}} : f64, %{{.*}} : f64, %{{.*}} : f64):
// CHECK-NEXT: %{{.*}} = arith.mulf %{{.*}}, %{{.*}} : f64
// CHECK-NEXT: %{{.*}} = arith.addf %{{.*}}, %{{.*}} : f64
// CHECK-NEXT: linalg.yield %{{.*}} : f64
// CHECK-NEXT: }

// CHECK-GENERIC-NEXT: "memref_stream.generic"(%{{.*}}, %{{.*}}, %{{.*}}, %H) <{"bounds" = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>], "init_indices" = [#builtin.int<0>], "indexing_maps" = [affine_map<(d0, d1, d2) -> (d0, d2)>, affine_map<(d0, d1, d2) -> (d2, d1)>, affine_map<(d0, d1) -> (d0, d1)>], "iterator_types" = [#memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<reduction>], "operandSegmentSizes" = array<i32: 2, 1, 1>}> ({
// CHECK-GENERIC-NEXT: "memref_stream.generic"(%{{.*}}, %{{.*}}, %{{.*}}, %H) <{"bounds" = [4 : index, 2 : index, 3 : index], "init_indices" = [#builtin.int<0>], "indexing_maps" = [affine_map<(d0, d1, d2) -> (d0, d2)>, affine_map<(d0, d1, d2) -> (d2, d1)>, affine_map<(d0, d1) -> (d0, d1)>], "iterator_types" = [#memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<reduction>], "operandSegmentSizes" = array<i32: 2, 1, 1>}> ({
// CHECK-GENERIC-NEXT: ^{{.*}}(%{{.*}} : f64, %{{.*}} : f64, %{{.*}} : f64):
// CHECK-GENERIC-NEXT: %{{.*}} = "arith.mulf"(%{{.*}}, %{{.*}}) <{"fastmath" = #arith.fastmath<none>}> : (f64, f64) -> f64
// CHECK-GENERIC-NEXT: %{{.*}} = "arith.addf"(%{{.*}}, %{{.*}}) <{"fastmath" = #arith.fastmath<none>}> : (f64, f64) -> f64
Expand All @@ -126,7 +126,7 @@ memref_stream.generic {
// CHECK-GENERIC-NEXT: %I = "test.op"() : () -> memref<4x3xf64>

memref_stream.generic {
bounds = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>],
bounds = [4, 2, 3],
indexing_maps = [
affine_map<(d0, d1, d2) -> (d0, d2)>,
affine_map<(d0, d1, d2) -> (d2, d1)>,
Expand All @@ -142,15 +142,15 @@ memref_stream.generic {
linalg.yield %acc_new_0, %acc_new_1 : f64, f64
}

// CHECK-NEXT: memref_stream.generic {bounds = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>], indexing_maps = [affine_map<(d0, d1, d2) -> (d0, d2)>, affine_map<(d0, d1, d2) -> (d2, d1)>, affine_map<(d0, d1) -> (d0, d1)>, affine_map<(d0, d1) -> (d0, d1)>], iterator_types = ["parallel", "parallel", "reduction"]} ins(%{{.*}}, %{{.*}} : memref<4x2xf64>, memref<2x3xf64>) outs(%{{.*}}, %{{.*}} : memref<4x3xf64>, memref<4x3xf64>) inits(%{{.*}} : f64, None) {
// CHECK-NEXT: memref_stream.generic {bounds = [4, 2, 3], indexing_maps = [affine_map<(d0, d1, d2) -> (d0, d2)>, affine_map<(d0, d1, d2) -> (d2, d1)>, affine_map<(d0, d1) -> (d0, d1)>, affine_map<(d0, d1) -> (d0, d1)>], iterator_types = ["parallel", "parallel", "reduction"]} ins(%{{.*}}, %{{.*}} : memref<4x2xf64>, memref<2x3xf64>) outs(%{{.*}}, %{{.*}} : memref<4x3xf64>, memref<4x3xf64>) inits(%{{.*}} : f64, None) {
// CHECK-NEXT: ^{{.*}}(%{{.*}} : f64, %{{.*}} : f64, %{{.*}} : f64, %{{.*}} : f64):
// CHECK-NEXT: %{{.*}} = arith.mulf %{{.*}}, %{{.*}} : f64
// CHECK-NEXT: %{{.*}} = arith.addf %{{.*}}, %{{.*}} : f64
// CHECK-NEXT: %{{.*}} = arith.addf %{{.*}}, %{{.*}} : f64
// CHECK-NEXT: linalg.yield %{{.*}}, %{{.*}} : f64, f64
// CHECK-NEXT: }

// CHECK-GENERIC-NEXT: "memref_stream.generic"(%{{.*}}, %{{.*}}, %{{.*}}, %{{.*}}, %{{.*}}) <{"bounds" = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>], "init_indices" = [#builtin.int<0>], "indexing_maps" = [affine_map<(d0, d1, d2) -> (d0, d2)>, affine_map<(d0, d1, d2) -> (d2, d1)>, affine_map<(d0, d1) -> (d0, d1)>, affine_map<(d0, d1) -> (d0, d1)>], "iterator_types" = [#memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<reduction>], "operandSegmentSizes" = array<i32: 2, 2, 1>}> ({
// CHECK-GENERIC-NEXT: "memref_stream.generic"(%{{.*}}, %{{.*}}, %{{.*}}, %{{.*}}, %{{.*}}) <{"bounds" = [4 : index, 2 : index, 3 : index], "init_indices" = [#builtin.int<0>], "indexing_maps" = [affine_map<(d0, d1, d2) -> (d0, d2)>, affine_map<(d0, d1, d2) -> (d2, d1)>, affine_map<(d0, d1) -> (d0, d1)>, affine_map<(d0, d1) -> (d0, d1)>], "iterator_types" = [#memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<parallel>, #memref_stream.iterator_type<reduction>], "operandSegmentSizes" = array<i32: 2, 2, 1>}> ({
// CHECK-GENERIC-NEXT: ^{{.*}}(%{{.*}} : f64, %{{.*}} : f64, %{{.*}} : f64, %{{.*}} : f64):
// CHECK-GENERIC-NEXT: %{{.*}} = "arith.mulf"(%{{.*}}, %{{.*}}) <{"fastmath" = #arith.fastmath<none>}> : (f64, f64) -> f64
// CHECK-GENERIC-NEXT: %{{.*}} = "arith.addf"(%{{.*}}, %{{.*}}) <{"fastmath" = #arith.fastmath<none>}> : (f64, f64) -> f64
Expand Down
10 changes: 5 additions & 5 deletions tests/filecheck/dialects/memref_stream/verify.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
%A, %B, %C = "test.op"() : () -> (memref<4x2xf64>, memref<2x3xf64>, memref<4x3xf64>)

memref_stream.generic {
bounds = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>],
bounds = [4, 2, 3],
indexing_maps = [
affine_map<(d0, d1, d2) -> (d0, d1)>,
affine_map<(d0, d1, d2) -> (d1, d2)>,
Expand All @@ -24,7 +24,7 @@ memref_stream.generic {
%A, %B, %C = "test.op"() : () -> (memref<4x2xf64>, memref<2x3xf64>, memref<4x3xf64>)

memref_stream.generic {
bounds = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>],
bounds = [4, 2, 3],
indexing_maps = [
affine_map<(d0, d1, d2) -> (d0, d1)>,
affine_map<(d0, d1, d2) -> (d1, d2)>
Expand All @@ -44,7 +44,7 @@ memref_stream.generic {
%A, %B, %C = "test.op"() : () -> (memref<4x2xf64>, memref<2x3xf64>, memref<4x3xf64>)

memref_stream.generic {
bounds = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>],
bounds = [4, 2, 3],
indexing_maps = [
affine_map<(d0, d1, d2, d3) -> (d0, d1)>,
affine_map<(d0, d1, d2) -> (d1, d2)>,
Expand All @@ -65,7 +65,7 @@ memref_stream.generic {
%A, %B, %C = "test.op"() : () -> (memref<4x2xf64>, memref<2x3xf64>, memref<4x3xf64>)

memref_stream.generic {
bounds = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>],
bounds = [4, 2, 3],
indexing_maps = [
affine_map<(d0, d1, d2) -> (d0, d1)>,
affine_map<(d0, d1, d2) -> (d1, d2)>,
Expand All @@ -86,7 +86,7 @@ memref_stream.generic {
%A, %B, %C, %D = "test.op"() : () -> (memref<4x2xf64>, memref<2x3xf64>, memref<4x3xf64>, memref<4x3xf64>)

memref_stream.generic {
bounds = [#builtin.int<4>, #builtin.int<2>, #builtin.int<3>],
bounds = [4, 2, 3],
indexing_maps = [
affine_map<(d0, d1, d2) -> (d0, d1)>,
affine_map<(d0, d1, d2) -> (d1, d2)>,
Expand Down
155 changes: 63 additions & 92 deletions tests/filecheck/projects/riscv-backend-paper/bottom_up.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -240,13 +240,7 @@ func.func public @conv_2d_nchw_fchw_d1_s1_3x3(
%Y : memref<8x8xf64>,
%G : memref<8x8xf64>
) {
%c0 = arith.constant 0 : index
%c1 = arith.constant 1 : index
%c2 = arith.constant 2 : index
%c4 = arith.constant 4 : index
%c8 = arith.constant 8 : index
%frep_count = arith.constant 6 : index

%zero_float = arith.constant 0.0 : f64
memref_stream.streaming_region {
patterns = [
#memref_stream.stride_pattern<ub = [8, 2, 8, 4], index_map = (m, n, k, j) -> (m, k)>,
Expand All @@ -255,22 +249,14 @@ func.func public @conv_2d_nchw_fchw_d1_s1_3x3(
]
} ins(%X, %Y : memref<8x8xf64>, memref<8x8xf64>) outs(%G : memref<8x8xf64>) {
^0(%x_stream : !stream.readable<f64>, %y_stream : !stream.readable<f64>, %g_stream : !stream.writable<f64>):
%c0 = arith.constant 0 : index
%c1 = arith.constant 1 : index
%c2 = arith.constant 2 : index
%c8 = arith.constant 8 : index
%frep_count = arith.constant 8 : index
scf.for %i0 = %c0 to %c8 step %c1 {
scf.for %i1 = %c0 to %c2 step %c1 {
%x00 = memref_stream.read from %x_stream : f64
%y00 = memref_stream.read from %y_stream : f64
%init0 = arith.mulf %x00, %y00 fastmath<fast> : f64
%x01 = memref_stream.read from %x_stream : f64
%y01 = memref_stream.read from %y_stream : f64
%init1 = arith.mulf %x01, %y01 fastmath<fast> : f64
%x02 = memref_stream.read from %x_stream : f64
%y02 = memref_stream.read from %y_stream : f64
%init2 = arith.mulf %x02, %y02 fastmath<fast> : f64
%x03 = memref_stream.read from %x_stream : f64
%y03 = memref_stream.read from %y_stream : f64
%init3 = arith.mulf %x03, %y03 fastmath<fast> : f64

%g00, %g01, %g02, %g03 = scf.for %inner_i = %c0 to %frep_count step %c1 iter_args(%acc0 = %init0, %acc1 = %init1, %acc2 = %init2, %acc3 = %init3) -> (f64, f64, f64, f64) {
%g00, %g01, %g02, %g03 = scf.for %inner_i = %c0 to %frep_count step %c1 iter_args(%acc0 = %zero_float, %acc1 = %zero_float, %acc2 = %zero_float, %acc3 = %zero_float) -> (f64, f64, f64, f64) {
%x10 = memref_stream.read from %x_stream : f64
%y10 = memref_stream.read from %y_stream : f64
%prod10 = arith.mulf %x10, %y10 fastmath<fast> : f64
Expand All @@ -290,26 +276,10 @@ func.func public @conv_2d_nchw_fchw_d1_s1_3x3(
scf.yield %res0, %res1, %res2, %res3 : f64, f64, f64, f64
}

%x20 = memref_stream.read from %x_stream : f64
%y20 = memref_stream.read from %y_stream : f64
%prod20 = arith.mulf %x20, %y20 fastmath<fast> : f64
%g10 = arith.addf %prod20, %g00 fastmath<fast> : f64
memref_stream.write %g10 to %g_stream : f64
%x21 = memref_stream.read from %x_stream : f64
%y21 = memref_stream.read from %y_stream : f64
%prod21 = arith.mulf %x21, %y21 fastmath<fast> : f64
%g11 = arith.addf %prod21, %g01 fastmath<fast> : f64
memref_stream.write %g11 to %g_stream : f64
%x22 = memref_stream.read from %x_stream : f64
%y22 = memref_stream.read from %y_stream : f64
%prod22 = arith.mulf %x22, %y22 fastmath<fast> : f64
%g12 = arith.addf %prod22, %g02 fastmath<fast> : f64
memref_stream.write %g12 to %g_stream : f64
%x23 = memref_stream.read from %x_stream : f64
%y23 = memref_stream.read from %y_stream : f64
%prod23 = arith.mulf %x23, %y23 fastmath<fast> : f64
%g13 = arith.addf %prod23, %g03 fastmath<fast> : f64
memref_stream.write %g13 to %g_stream : f64
memref_stream.write %g00 to %g_stream : f64
memref_stream.write %g01 to %g_stream : f64
memref_stream.write %g02 to %g_stream : f64
memref_stream.write %g03 to %g_stream : f64
}
}
}
Expand All @@ -321,66 +291,67 @@ func.func public @conv_2d_nchw_fchw_d1_s1_3x3(
// CHECK-NEXT: .globl matmul
// CHECK-NEXT: .p2align 2
// CHECK-NEXT: matmul:
// CHECK-NEXT: mv t3, a0
// CHECK-NEXT: mv t0, a1
// CHECK-NEXT: mv t1, a2
// CHECK-NEXT: li t4, 3
// CHECK-NEXT: scfgwi t4, 64
// CHECK-NEXT: li t4, 7
// CHECK-NEXT: scfgwi t4, 96
// CHECK-NEXT: li t4, 1
// CHECK-NEXT: scfgwi t4, 128
// CHECK-NEXT: li t4, 7
// CHECK-NEXT: scfgwi t4, 160
// CHECK-NEXT: mv t0, a0
// CHECK-NEXT: mv t1, a1
// CHECK-NEXT: mv t2, a2
// CHECK-NEXT: fcvt.d.w ft3, zero
// CHECK-NEXT: li t3, 3
// CHECK-NEXT: scfgwi t3, 64
// CHECK-NEXT: li t3, 7
// CHECK-NEXT: scfgwi t3, 96
// CHECK-NEXT: li t3, 1
// CHECK-NEXT: scfgwi t3, 128
// CHECK-NEXT: li t3, 7
// CHECK-NEXT: scfgwi t3, 160
// CHECK-NEXT: scfgwi zero, 192
// CHECK-NEXT: li t4, 8
// CHECK-NEXT: scfgwi t4, 224
// CHECK-NEXT: li t4, -56
// CHECK-NEXT: scfgwi t4, 256
// CHECK-NEXT: li t4, 8
// CHECK-NEXT: scfgwi t4, 288
// CHECK-NEXT: li t4, 3
// CHECK-NEXT: scfgwi t4, 65
// CHECK-NEXT: li t4, 7
// CHECK-NEXT: scfgwi t4, 97
// CHECK-NEXT: li t4, 1
// CHECK-NEXT: scfgwi t4, 129
// CHECK-NEXT: li t4, 7
// CHECK-NEXT: scfgwi t4, 161
// CHECK-NEXT: li t4, 8
// CHECK-NEXT: scfgwi t4, 193
// CHECK-NEXT: li t4, 40
// CHECK-NEXT: scfgwi t4, 225
// CHECK-NEXT: li t4, -440
// CHECK-NEXT: scfgwi t4, 257
// CHECK-NEXT: li t4, -504
// CHECK-NEXT: scfgwi t4, 289
// CHECK-NEXT: li t4, 63
// CHECK-NEXT: scfgwi t4, 66
// CHECK-NEXT: li t4, 8
// CHECK-NEXT: scfgwi t4, 194
// CHECK-NEXT: scfgwi t3, 864
// CHECK-NEXT: scfgwi t0, 865
// CHECK-NEXT: scfgwi t1, 898
// CHECK-NEXT: li t3, 8
// CHECK-NEXT: scfgwi t3, 224
// CHECK-NEXT: li t3, -56
// CHECK-NEXT: scfgwi t3, 256
// CHECK-NEXT: li t3, 8
// CHECK-NEXT: scfgwi t3, 288
// CHECK-NEXT: li t3, 3
// CHECK-NEXT: scfgwi t3, 65
// CHECK-NEXT: li t3, 7
// CHECK-NEXT: scfgwi t3, 97
// CHECK-NEXT: li t3, 1
// CHECK-NEXT: scfgwi t3, 129
// CHECK-NEXT: li t3, 7
// CHECK-NEXT: scfgwi t3, 161
// CHECK-NEXT: li t3, 8
// CHECK-NEXT: scfgwi t3, 193
// CHECK-NEXT: li t3, 40
// CHECK-NEXT: scfgwi t3, 225
// CHECK-NEXT: li t3, -440
// CHECK-NEXT: scfgwi t3, 257
// CHECK-NEXT: li t3, -504
// CHECK-NEXT: scfgwi t3, 289
// CHECK-NEXT: li t3, 63
// CHECK-NEXT: scfgwi t3, 66
// CHECK-NEXT: li t3, 8
// CHECK-NEXT: scfgwi t3, 194
// CHECK-NEXT: scfgwi t0, 864
// CHECK-NEXT: scfgwi t1, 865
// CHECK-NEXT: scfgwi t2, 898
// CHECK-NEXT: csrrsi zero, 1984, 1
// CHECK-NEXT: li t1, 16
// CHECK-NEXT: mv t0, zero
// CHECK-NEXT: # Constant folded riscv_cf.bge
// CHECK-NEXT: scf_body_1_for:
// CHECK-NEXT: fmul.d ft6, ft0, ft1
// CHECK-NEXT: fmul.d ft5, ft0, ft1
// CHECK-NEXT: fmul.d ft4, ft0, ft1
// CHECK-NEXT: fmul.d ft3, ft0, ft1
// CHECK-NEXT: li t3, 5
// CHECK-NEXT: scf_body_{{\d+}}_for:
// CHECK-NEXT: fmv.d ft7, ft3
// CHECK-NEXT: fmv.d ft6, ft3
// CHECK-NEXT: fmv.d ft5, ft3
// CHECK-NEXT: fmv.d ft4, ft3
// CHECK-NEXT: li t3, 7
// CHECK-NEXT: frep.o t3, 4, 0, 0
// CHECK-NEXT: fmadd.d ft7, ft0, ft1, ft7
// CHECK-NEXT: fmadd.d ft6, ft0, ft1, ft6
// CHECK-NEXT: fmadd.d ft5, ft0, ft1, ft5
// CHECK-NEXT: fmadd.d ft4, ft0, ft1, ft4
// CHECK-NEXT: fmadd.d ft3, ft0, ft1, ft3
// CHECK-NEXT: fmadd.d ft2, ft0, ft1, ft6
// CHECK-NEXT: fmadd.d ft2, ft0, ft1, ft5
// CHECK-NEXT: fmadd.d ft2, ft0, ft1, ft4
// CHECK-NEXT: fmadd.d ft2, ft0, ft1, ft3
// CHECK-NEXT: fmv.d ft2, ft7
// CHECK-NEXT: fmv.d ft2, ft6
// CHECK-NEXT: fmv.d ft2, ft5
// CHECK-NEXT: fmv.d ft2, ft4
// CHECK-NEXT: addi t0, t0, 1
// CHECK-NEXT: blt t0, t1, scf_body_{{\d+}}_for
// CHECK-NEXT: scf_body_end_{{\d+}}_for:
Expand Down
Loading
Loading