Skip to content

[FIX] fix bug when normalize iter with different lower bounds #17360

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

Merged
tqchen merged 1 commit into from
Sep 14, 2024
Merged

[FIX] fix bug when normalize iter with different lower bounds #17360

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
2 changes: 1 addition & 1 deletion src/arith/iter_affine_map.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -696,7 +696,7 @@ class IterMapRewriter : public ExprMutator {
// the delta of iter_min when it is updated when the lower bound predicate is present
PrimExpr iter_min_delta = make_const(iter_min.dtype(), 0);
if (predicate_induced_min.defined()) {
iter_min_delta = predicate_induced_min.value() - iter_min;
iter_min_delta = max(predicate_induced_min.value(), iter_min) - iter_min;
iter_min = max(predicate_induced_min.value(), iter_min);
}
if (predicate_induced_max.defined()) {
Expand Down
21 changes: 21 additions & 0 deletions tests/python/arith/test_arith_iter_affine_map.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -346,6 +346,27 @@ def test_predicate():
predicate=tvm.tir.all(2 <= j * 2 + k, 0 <= i * 4 + j),
)

# constraint with differnent lower bound
assert_iter_sum_pattern(
{
(i * 16 + j) // 23 * 8
+ (i * 16 + j) % 23
- 15: (
64,
0,
1,
(i * 16 + j) // 23 * 8 + ((i * 16 + j) % 23 + tvm.tir.IntImm("int32", -15)),
)
},
var_dom([(i, 12), (j, 16)]),
predicate=tvm.tir.And(
tvm.tir.And(
i * 16 + j < 184, tvm.tir.LE(tvm.tir.IntImm("int32", 8), (i * 16 + j) % 23)
),
tvm.tir.LE(tvm.tir.IntImm("int32", 15), (i * 16 + j) % 23),
),
)

# constraint on many disjoint fused iters, case 1
# i4 * 6 + i5 in [3, 9), extent=6 (= scale of i2)
# i2 * 30 + i3 * 15 in [30, 90), extent=60 (= scale of i1)
Expand Down