feat[next-dace]: lowering to SchedTree by edopao · Pull Request #2690 · GridTools/gt4py

edopao · 2026-07-03T12:27:22Z

Summary

Adds a second GTIR→SDFG lowering path for the gt4py.next DaCe backend that goes through
DaCe's schedule tree (ScheduleTreeRoot) instead of building the SDFG directly, and
wires six new run_dace_stree_* backends into the test matrix:

run_dace_stree_fview_{cpu,gpu}[_noopt] — fieldview pipeline (as today)
run_dace_stree_iview_{cpu,gpu}[_noopt] — with the ITIR common-transforms pipeline
(apply_common_transforms), which is now supported only on the stree lowering

This follows the direction set by the cartesian schedule-tree decision
(ADR: DaCe backends – schedule tree):
express the program's execution schedule at tree level and let ScheduleTreeRoot.as_sdfg()
produce the state machine, so that macro-level optimizations (loop merging, re-ordering)
can later operate on the tree instead of pattern-matching SDFGs.

Design

Direct GTIR → schedule-tree lowering (no intermediate TreeIR), in the new
runners/dace/lowering_stree/ module (+5144 lines), alongside the existing lowering/:

gtir_to_stree.py — GTIRToScheduleTree visitor: statements, storage, let-lambdas,
dispatch. All program storage is pre-allocated on root.containers/root.symbols;
let-lambdas are inlined into the same root (dict-union shadowing, call stack = implicit
push/pop) — no nested SDFGs produced by the lowering itself. (as_sdfg() may still
introduce them downstream; see scan_carry_scalarization.py.)
gtir_to_stree_codegen.py — StreePythonCodegen: each stencil body becomes one Python
code string inside a TaskletNode. No library nodes — reductions are lowered
directly to tasklet code.
Scan is a ForScope inside a MapScope. For scalar leaves the carry lives in a
scalar transient that is pre-initialized with init before the loop and read/written
each iteration. List-typed leaves still read their carry from the result array at the
previous index and therefore keep a carry if k > start else init ternary to guard the
first iteration. Backward scans are handled as well.

The legacy lowering is untouched by this PR and remains the default.

Test coverage

Six new backend IDs in tests/next_tests/definitions.py, each mapped to a dedicated
DACE_STREE_SKIP_TEST_LIST in BACKEND_SKIP_TEST_MATRIX, so the full next test suite
runs against the stree backends in the test_next (dace, *) nox sessions.
Currently unsupported there (skipped): USES_LIFT, USES_SCAN_IN_STENCIL,
USES_SPARSE_FIELDS, USES_TUPLE_ITERATOR.
The six new backend entries carry the existing pytest.mark.uses_dace marker (GPU ones
also requires_gpu), so the established nox matrix selection applies to them unchanged.
Remaining test adjustments are expect/xfail bookkeeping.

Dependencies

This branch bundles the following transformation-layer fixes, which are still open as
standalone PRs: #2811, #2812, #2813, #2815, #2816, #2817, #2818. Each is a latent hazard
that the stree-lowered SDFGs — structured around regions and boundary states — trip and
that the legacy lowering never produced. They appear in this diff
(transformations/{simplify,state_fusion,redundant_array_removers,utils}.py and
test_distributed_buffer_relocator.py); once they merge individually, rebasing will drop
them from here.

The visitor-dispatch refactor this builds on, #2824, has already merged. Note that it
renamed gtir_dataflow.py → gtir_to_sdfg_lambda.py; gtir_to_sdfg.py kept its name.

Copilot

Pull request overview

This PR refactors the DaCe lowering pipeline for gt4py.next by renaming key lowering entry points and restructuring how builtin-function visitors are dispatched in both the fieldview and iterator lowering paths.

Changes:

Renames/reshapes lowering entry points (e.g., build_sdfg_from_gtir → lower_program_to_sdfg) and adjusts imports/types accordingly.
Refactors builtin dispatch in GTIRToSDFG (fieldview) and LambdaToDataflow (iterator) via centralized handler maps.
Updates scan/primitives/type helpers to use the split fieldview/iterator lowering modules.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_types.py	Updates local-view helpers to return iterator-based `IteratorExpr`/`MemletExpr` types.
src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_scan.py	Switches scan lowering to the new fieldview/iterator module split and updated dataflow types.
src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_primitives.py	Retypes primitives to the split builders and iterator dataflow edges; small typing-related adjustments.
src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_iterator.py	Refactors builtin dispatch in lambda-to-dataflow and retargets builder typing to fieldview interfaces.
src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_fieldview.py	Refactors builtin dispatch in program lowering; introduces `_visit_*` helpers; renames program lowering entry point.
src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_concat_where.py	Updates concat_where lowering to use the fieldview lowering module split.
src/gt4py/next/program_processors/runners/dace/lowering/init.py	Updates lowering package exports to point at the refactored lowering entry points.

Comments suppressed due to low confidence (3)

src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_iterator.py:58

gtir_to_sdfg_iterator.py imports gtir_to_sdfg_fieldview using the src.gt4py... package path, which will fail at runtime outside the repository checkout layout. Also, since this module uses from __future__ import annotations, this import is only needed for type checking and can be placed under TYPE_CHECKING to avoid creating an import cycle (fieldview -> primitives -> iterator -> fieldview).
src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_iterator.py:18
The new TYPE_CHECKING-guarded import for gtir_to_sdfg_fieldview (recommended below) requires TYPE_CHECKING to be imported from typing in this module.
src/gt4py/next/program_processors/runners/dace/lowering/init.py:23
lowering/__init__.py re-exports build_sdfg_from_gtir from gtir_to_sdfg_fieldview, but that function was renamed to lower_program_to_sdfg (and build_sdfg_from_gtir no longer exists). This currently breaks imports like gtx_dace_lowering.build_sdfg_from_gtir(...) used elsewhere in the codebase.

from gt4py.next.program_processors.runners.dace.lowering.gtir_to_sdfg_fieldview import (
    build_sdfg_from_gtir,
)
from gt4py.next.program_processors.runners.dace.lowering.gtir_to_sdfg_utils import (
    flatten_tuple_fields,
    get_map_variable,
)


__all__ = [
    "build_sdfg_from_gtir",
    "flatten_tuple_fields",
    "get_map_variable",
]

Copilot

Pull request overview

Copilot reviewed 9 out of 9 changed files in this pull request and generated 3 comments.

Comments suppressed due to low confidence (1)

src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_fieldview.py:1231

This comment line has a stray "to." at the end (likely meant to be "symbols."), which makes the sentence ungrammatical.

Copilot

Pull request overview

Copilot reviewed 9 out of 9 changed files in this pull request and generated 1 comment.

Comments suppressed due to low confidence (1)

src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_iterator.py:52

gtir_to_sdfg_fieldview is imported here only to satisfy type annotations, but this creates an avoidable import edge (and contributes to circular imports: fieldview -> primitives -> iterator -> fieldview). Since this module uses from __future__ import annotations, you can move the import behind typing.TYPE_CHECKING (or otherwise defer it) to keep runtime imports acyclic.

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

edopao · 2026-09-07T06:32:42Z

cscs-ci run default

edopao · 2026-09-11T15:46:24Z

cscs-ci run default

edopao · 2026-09-14T12:02:54Z

cscs-ci run default

edopao · 2026-09-14T13:30:12Z

cscs-ci run default

edopao · 2026-09-14T14:17:56Z

cscs-ci run default

edopao · 2026-09-15T07:43:31Z

cscs-ci run default

edopao · 2026-09-15T11:44:08Z

cscs-ci run default

edopao · 2026-09-15T13:14:41Z

cscs-ci run default

edopao · 2026-09-19T06:29:11Z

cscs-ci run default

Lower an `as_fieldop` whose body is a `reduce` using library nodes instead of a native Python accumulation loop, matching the SDFG lowering: - `dace.libraries.standard.Reduce` for dense connectivities, and - `ReduceWithSkipValues` for connectivities with skip values. The fused schedule-tree reduce `reduce(λ(acc, e0..ek) → op(acc, f(...)), init)(l0..lk)` is decomposed into a per-neighbor element list `map_list(λ → f(...))(l0..lk)` (materialized via the existing ListType path) reduced with WCR `op`. Nested reductions fall back to the native loop. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

…ee lowering" This reverts commit cb24e93.

edopao · 2026-09-19T22:33:50Z

cscs-ci run default

edopao · 2026-09-21T06:14:30Z

cscs-ci run default

…op bodies (#2818) ## Summary `DistributedBufferRelocator` can move the final write back of a temporary into a state nested inside a `LoopRegion`. The write back then executes in **every loop iteration** on data that the loop has only written partially, racing with the per-iteration partial writes and producing nondeterministic garbage in the unaffected parts of the output. This is what broke the `scan` tests on the `run_dace_stree_*` backends in #2690 (e.g. [CI run 32715667466](https://github.com/GridTools/gt4py/actions/runs/32715667466)): for a scan, the state that (partially) defines the scan temporary is the scan loop body, and the full-array write back after the loop was relocated into it. **Enabler:** `SDFG.states()` recurses into control-flow scopes (`return list(self.all_states())`), so `find_upstream_states()` now also discovers loop-body states as candidate definition locations — on top-level-only semantics this relocation could never have happened. ## Fix Never relocate into a state that is (transitively) nested inside a `LoopRegion`: `_find_candidates()` now skips such definition states via the new `_is_inside_loop()` helper. Relocation into conditional branches — the use case this pass was designed for — remains legal, as does any relocation into states that are not executed repeatedly. ## Test New unit test `test_distributed_buffer_definition_in_loop` builds the scan-shaped SDFG (per-iteration write of `t[k]` inside a loop body, full write back `t -> b` in the state after the loop) and asserts that the write back stays after the loop and that no write to the output is added inside the loop. The test fails without the fix (the pass relocates into the loop body) and passes with it. ## Validation - New reproducer test: fails without the fix, passes with it. - Full `transformation_tests/` unit suite: 326 passed, 3 xfailed (pre-existing). - On #2690's branch, an equivalent guard made all 32 CI-failing scan tests pass locally (incl. 5/5 stress runs of `test_scan.py::test_solve_triag` and `test_icon_like_scan.py` across all four `run_dace_stree_*` backends).

edopao changed the title ~~Dace refactor lowering~~ refactor[next-dace]: change dispatching of visitors for builtin functions Jul 3, 2026

edopao force-pushed the dace_refactor_lowering branch from 6f99a07 to 057605c Compare July 3, 2026 12:29

edopao requested a review from Copilot July 3, 2026 12:32

Copilot started reviewing on behalf of edopao July 3, 2026 12:32 View session

edopao force-pushed the dace_refactor_lowering branch from 057605c to 90d9ce6 Compare July 3, 2026 12:34

Copilot AI reviewed Jul 3, 2026

View reviewed changes

Comment thread src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_primitives.py Outdated

edopao force-pushed the dace_refactor_lowering branch 2 times, most recently from 4d8ca3b to e92eade Compare July 3, 2026 12:44

edopao requested a review from Copilot July 3, 2026 12:49

Copilot started reviewing on behalf of edopao July 3, 2026 12:52 View session

Copilot AI reviewed Jul 3, 2026

View reviewed changes

edopao requested a review from Copilot July 3, 2026 14:25

Copilot started reviewing on behalf of edopao July 3, 2026 14:29 View session

Copilot AI reviewed Jul 3, 2026

View reviewed changes

Comment thread src/gt4py/next/program_processors/runners/dace/lowering/gtir_to_sdfg_primitives.py Outdated

edopao and others added 16 commits August 18, 2026 11:35

add IR-pass to inline literals into lambda expressions

2d7545d

edit

f52e7a2

edit

47be34c

edit

0f60c7f

Enable tasklet fusion in dataflow optimization

652ece6

edit

281d2a6

edit

9c73ff5

fix lowering of scan

fabb7c6

edit

f0b9046

undo extra changes

76d3f81

edit

9ab6f2f

skip enabling the new pass

6b4f86c

undo extra change

ac3013b

dispatch visitor for builtin functions

c6699ea

rename modules for sdfg lowering

b690929

Edit code comment

222f4ee

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

use max value for missing neighbor value

e9664bd

edopao commented Sep 11, 2026

View reviewed changes

Comment thread _dacegraphs/invalid.sdfgz Outdated

edopao added 2 commits September 11, 2026 17:45

unstage file

f8158d3

Merge remote-tracking branch 'upstream/main' into dace_refactor_lowering

21eb270

edopao added 2 commits September 14, 2026 13:48

fix parentheses in ternary operator

ff77f79

Merge remote-tracking branch 'origin/main' into dace_refactor_lowering

5538f27

edopao added 2 commits September 14, 2026 15:14

support concat_where with a neighbor-list result

31a1476

fix pre-commit

dd69498

edopao added 2 commits September 14, 2026 16:09

fix parentheses in ternary operator

a1f82ea

update uv lock

cd8a87b

edopao added 2 commits September 15, 2026 09:20

add support for inlined concat_where

a563f06

Merge remote-tracking branch 'origin/main' into dace_refactor_lowering

b9c7608

edit

9a6bd11

revert changes

5445e8d

Merge branch 'main' into dace_refactor_lowering

14229cf

edopao and others added 2 commits September 20, 2026 00:17

Revert "feat[next-dace]: lower reduce to library nodes in schedule-tr…

44966e5

…ee lowering" This reverts commit cb24e93.

align common transforms to dace_mixed_ir

11386c3

Conversation

edopao commented Jul 3, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Design

Test coverage

Dependencies

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

edopao commented Sep 7, 2026

Uh oh!

Uh oh!

edopao commented Sep 11, 2026

Uh oh!

edopao commented Sep 14, 2026

Uh oh!

edopao commented Sep 14, 2026

Uh oh!

edopao commented Sep 14, 2026

Uh oh!

edopao commented Sep 15, 2026

Uh oh!

edopao commented Sep 15, 2026

Uh oh!

edopao commented Sep 15, 2026

Uh oh!

edopao commented Sep 19, 2026

Uh oh!

edopao commented Sep 19, 2026

Uh oh!

edopao commented Sep 21, 2026

Uh oh!

This branch has not been deployed

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

edopao commented Jul 3, 2026 •

edited

Loading