Update sta ptr to 08/24 - #11226
Draft
minjukim55 wants to merge 3 commits into
Draft
Conversation
Signed-off-by: Minju Kim <mkim@precisioninno.com>
OpenSTA 3a595df0 (parallaxsw/OpenSTA#470) added %typemap(freearg) for InstanceSeq*, CellSeq* and the other Seq pointer types in tcl/StaTclTypes.i, so the SWIG wrapper now frees the container it allocated in the (in) typemap. Upstream dropped its own matching deletes in search/Search.i, verilog/Verilog.i and sdc/FilterObjects.cc at the same time. Two wrappers here still deleted the same container, so the callee and the wrapper both freed it: remove_buffers_cmd(InstanceSeq*) src/rsz/src/Resizer.i write_verilog_cmd(..., CellSeq *remove_cells) src/dbSta/src/dbSta.i write_verilog_cmd is annotated "Copied from sta/verilog/Verilog.i"; upstream fixed the original and this copy went stale. The double free shows up as "Signal 11 received" with free in the stack trace, failing 13 tests: dbSta.write_verilog7, write_verilog9, write_verilog9_hier rsz.remove_buffers3, remove_buffers6, remove_buffers_hier1, repair_fanout3 grt.remove_buffers1, remove_buffers2, remove_buffers3 grt.remove_buffers1_cugr, remove_buffers2_cugr, remove_buffers3_cugr Dropping the deletes does not leak: the freearg typemap frees the container exactly once. find_fanin_fanouts and insert_buffer_before_loads_cmd also delete their argument, but that argument is PinSet*, which got no freearg typemap, so they are left alone. Signed-off-by: Minju Kim <mkim@precisioninno.com>
OpenSTA d2c508dd (parallaxsw/OpenSTA#489) rewrote the unsized-constant path in verilog/VerilogReader.cc. It used to be size_t size = std::stol(csize); which threw std::invalid_argument on 'd1 and on a negative width, so read_verilog/link_design aborted and OpenROAD reported an error. It is now a warning plus a 1-bit default: if (csize.empty()) reader->warn(2724, ..., "unsized constant {} not supported.", constant); else { auto [size_value, valid] = stringLong(csize); ... } Two consequences, both recorded here. STRUCTURAL_EXPECTED_FAIL: read_verilog now accepts the input, so these three round-trip and the framework reports them as unexpected passes. bx_constants_unsized_b0.v bx_constants_unsized_d1.v structural/wb_sta_reader_const_negative_width.v Dropped from both the flat and hier lists. The remaining entries in those lists still throw, from the stoi in the attribute parser, so they stay. CONFORMANCE_EXPECTED_FAIL: the same two constant cases still fail, but no longer as "or-error". OpenROAD now emits a netlist in which the constant has been dropped, and kepler-formal refuses to load it: Netlist loading failed: In SNLVRLConstructor construct: setTermsNets only supported when terms (size: 1 [A2]) and nets share same size (size: 0) so the observed mode is "tool-error". Recorded, with a comment noting that losing the connection silently instead of stopping is a behaviour change worth raising with upstream. Updating the entry does not endorse it. Signed-off-by: Minju Kim <mkim@precisioninno.com>
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Code Review
This pull request removes redundant memory deletions in dbSta.i and Resizer.i, updates the src/sta subproject commit, and adjusts the expected failure test configurations in hier_expected_fail.bzl to align with the updated OpenSTA behavior regarding unsized constants. There are no review comments, so I have no feedback to provide.
This was referenced Aug 25, 2026
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
1 participant
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.
Bump
src/stato the upstream OpenSTA merge inThe-OpenROAD-Project/OpenSTA#404, plus the two OpenROAD-side changes that merge
requires.
rsz,dbSta: stop deleting SWIG-owned Seq containers.parallaxsw/OpenSTA#470 added
%typemap(freearg)forInstanceSeq*,CellSeq*and the other Seq pointer types, so the wrapper now frees the container it
allocated in the
(in)typemap, and upstream dropped its own matching deletes.remove_buffers_cmdandwrite_verilog_cmdstill deleted it - a double free,Signal 11withfreein the stack trace, 13 tests acrossdbSta,rszandgrt.write_verilog_cmdis annotated "Copied from sta/verilog/Verilog.i";upstream fixed the original and this copy went stale. Dropping the deletes does
not leak.
find_fanin_fanoutsandinsert_buffer_before_loads_cmdtakePinSet*, which got nofreearg, so they are untouched.dbSta/test: refresh constant-width expected-fail entries.parallaxsw/OpenSTA#489 replaced the
std::stolthat threw on an unsizedconstant with
STA-2724/STA-2725warnings and a 1-bit default.bx_constants_unsized_b0.v,bx_constants_unsized_d1.vandstructural/wb_sta_reader_const_negative_width.vnow round-trip, so they leaveSTRUCTURAL_EXPECTED_FAIL. The twobx_constants_*conformance entries stillfail but as
tool-errorrather thanor-error: OpenROAD now emits a netlistwith the constant dropped, and kepler-formal will not load it -
Losing the connection silently instead of stopping is a behaviour change worth
raising upstream; recording the new mode does not endorse it.