Repository navigation
[Only for CI] Extensions - #2475
Draft
ThrudPrimrose wants to merge 9312 commits into
Draft
ThrudPrimrose wants to merge 9312 commits into
ThrudPrimrose wants to merge 9312 commits into
Conversation
ThrudPrimrose
added
no-ci
ThrudPrimrose
force-pushed
the
extended
branch
from
August 12, 2026 11:16
269d9cf to
f9c1368
Compare
ThrudPrimrose
added a commit
that referenced
this pull request
Aug 20, 2026
…skip_gpu_outputs; restore OMP thread default; fix extras - codegen.py: restore scalar_emission conditional form; pass skip_gpu_outputs=True so GPU map outputs are not scalarized (the unconditional preserve_abi run in PR #2475 introduced broad CPU regressions). - conftest.py: do not force OMP_NUM_THREADS=1; keep setdefault so callers control it. - pyproject.toml: re-add fastgraph/polyhedral extras and testing z3-solver/matplotlib that were dropped when moving from setup.py.
ThrudPrimrose
added a commit
that referenced
this pull request
Sep 3, 2026
Restores a mergeable PR #2475 so the extended workflows fire again. Main carried only 036a963, the upstream form of the __-prefixed operator round-trip fix that extended had already landed locally as 6e25a02. Every conflict was a diff-alignment artifact around code extended adds on top; no behaviour changed. Resolutions: - dace/symbolic.py, _SerializedSymbolicParser._functions: kept extended. Main's side of the hunk was empty; the block holds extended-only entries (ITE, id, diag, jn) that main never had. - dace/symbolic.py, _print_ceiling: kept extended's body, took main's comment. The is_integer early return is byte-identical on both sides, so only the comment differed; upstream wording keeps the hunk from conflicting again. Extended's MPR lowering, the integral_index_expression int64 cast, and _print_Rational/_fp_literal/_print_Max/_print_Min/ _print_minmax are all preserved. - tests/symbolic/test_int_div_rewrites.py, imports: kept extended, a strict superset of main's line (adds symbol, dtypes, dace). - tests/symbolic/test_int_div_rewrites.py, tests: kept extended. Main's side was empty; the block is two extended-only ceiling tests. Main's test_ceiling_of_an_integer_prints_as_its_argument was already present and auto-merged. - tests/symbolic_serialization_test.py: kept extended. Main's side was empty; all four of main's new tests were already present, byte-identical. Tests: 305 passed, 1 skipped (absent CloudSC fixture), 2 xfailed across tests/symbolic/ and the serialization suite. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
ThrudPrimrose
added a commit
that referenced
this pull request
Oct 15, 2026
- ipow of integer operands is integer arithmetic: a shape extent, and a ceiling/floor that needs no rounding (TSVC offload "ipow(LEN_2D, 2) is not integral", stockham_fft's ceil() loop bounds). - Printing reads what it may not see in an expression from its surroundings, and its memos key on it: a symbol's dtype comes from the SDFG being generated (the scope's declared types), and symstr / sym2cpp key on those types and on the printing facts. A modulo printed as C's % where the facts proved it was served from the memo where they did not. - Canonicalization holds back a DOALL sibling of a sibling-loop chain with a refused sibling until WavefrontSkew (lu/ludcmp keep their (i, j) wavefront now that LoopToMap proves the j >= i loop). - Frontend: an annotation's free symbol never replaces the program's declared global (backported to #2673); a promoted symbol already declared is not re-declared from the visitor scope; an extent respelled as its source is undeclared once. - Tests stop asserting the pre-registry contract (dtype as symbol identity, simplify's own cache, sympy assumptions, declared iterators) and read signs from printing facts. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ThrudPrimrose
added
no-ci
…sing loop is live, a one-trip loop keeps its live-out check
…signed grid compiles
… bump or unlike its entry value, and a start read by name past a store refuse; subscript_index_symbols takes C++ text
… is a name; compare names
… in place, a buffer would lose what the write skips" This reverts commit 8f96f3f.
…ts leaves its map" This reverts commit 93d8fe7.
…s in place, a buffer would lose what the write skips" This reverts commit c0e86e9.
…ny more, a later fusion deleted its descriptor
…wn, the edge between them stays
…holds, no copy of a container onto itself
…costs CI 40 minutes
… their own, the edge between them stays" This reverts commit 838def2.
…ng its writers do not" This reverts commit e9796fc.
…rnel's thread block The collective walked up one state only, missed the kernel above a nested loop body, and took the default block size, which disagrees with the launch geometry and leaves rows unwritten.
…ets its own state A part of a state that holds a map runs in every thread with no barrier inside, so a tasklet or a copy behind the map raced it. Between states the lowering puts a grid barrier and guards what writes global memory.
…ve in device memory A kernel's copy of a host array is a thread local, so the value init publishes never reached the BFS kernel. The test declares the unsigned types it passes.
…mPy one hides from every analysis
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.