…fault

The tst Fixture unconditionally created a dbSta instance, so every test
that used it linked OpenSTA + dbSta even when it only manipulated an odb
database (e.g. the odb helper-based tests). OpenSTA/dbSta debug symbols
dominate those binaries (~160 MB each).

Split Fixture into a base tst::DbFixture (tst/db_fixture.h, odb-only: db
manipulation, LEF tech loading, file lookup) and a derived tst::Fixture
(tst/fixture.h) that adds the dbSta/OpenSTA instance, readLiberty, and
the LEF library loaders that notify STA. The base lives in a new tst_base
library (//src/tst:db_fixture in Bazel) that does not link STA.

odb DB-only tests and SimpleDbFixture now derive from DbFixture and link
tst_base; STA consumers (PDK fixtures, IntegratedFixture, rsz/dbSta/est
tests) are unchanged. Mirrored in Bazel.

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

…d-fields

rcx: rm ununsed fields in dbCreateNetUtil

Signed-off-by: Naveen Venkat <archgen.guest@nyayanidhi.in>

LEF power and ground pins may legally use INPUT direction. Exclude supply MTerms when identifying a buffer signal input and output so those cells remain usable for repair operations.

Signed-off-by: Naveen Venkat <archgen.guest@nyayanidhi.in>

…actor

NesterovBase::getDensityGradient re-fetched three loop-invariant values on
every inner bin iteration: getBinCntX() (two non-inlined .cpp calls),
bg_.getBinsConst(), and gCell->getDensityScale(). Hoist them to locals before
the loop and index a raw const Bin* base pointer.

Arithmetic-neutral: identical loop nesting, operands, and accumulation order,
so placement output is bit-identical. Verified the three values are not mutated
inside the loop (binCntX_ member, bins_ not realloced, densityScale_ on the
const gCell param).

medium03 global_placement (DEF read excluded), 3 runs each on a shared
36-core host, -j8 Release:
  before median 199910 ms, after median 185648 ms (~7.1%, within run noise).
QoR bit-identical: iter 486, overflow 0.0994, HPWL 1.162648e+06 unchanged.
gpl ctest suite 65/65 green (log-compare enforces identical output).




Signed-off-by: Saurav Singh <saurav.singh@fermions.co>

Signed-off-by: Augusto Berndt <augusto.berndt@precisioninno.com>

Debug info dominates the size of these binaries (template-heavy C++;
TestAccessPoint was ~97% DWARF). Enable -gz=zlib for GNU/Clang in the
Debug and RelWithDebInfo configs, at both compile and link, to compress
the .debug_* sections.

This is transparent to gdb/lldb/readelf/addr2line and a no-op for builds
without -g (e.g. Release). TestAccessPoint drops 330 MB -> 89 MB (-73%).

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

The default bazel build is -c opt with no -g, so it carries no debug
info. The asan and profile configs add -g (and keep symbols via
--strip=never), so their binaries carry DWARF that benefits from
compression, mirroring the CMake -gz change.

Add -gz=zlib (compile + link) to the asan and profile configs. It is
transparent to gdb/perf/addr2line.

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

The tst_base fixture library links GTest::gtest, but that imported
target only exists when find_package(GTest) runs, which is guarded by
ENABLE_TESTS. add_subdirectory(tst) was unconditional, so a -no-tests
build failed at generate time with 'Target tst_base links to
GTest::gtest but the target was not found'.

Guard add_subdirectory(tst) with ENABLE_TESTS. This is safe because
every consumer of tst/tst_base/tst_integrated_fixture lives in a
test/cpp directory that is itself gated by ENABLE_TESTS.

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

…minimal-deps

Trim odb test binary bloat: minimal deps, fixture split, debug compression

GlobalRouter::removeNet only erased the net from dirty_nets_ in the
FastRoute (non-CUGR) branch. When use_cugr_ is true, a net destroyed
during incremental routing stayed in dirty_nets_, so updateDirtyRoutes
later passed the dangling dbNet to CUGR::updateNet, which dereferenced
it via Design::updateNet (db_net->getSigType()).

In release builds (NDEBUG) this use-after-free is silent, so
grt.incremental_deleted_net.tcl passed. With assertions enabled
(coverage / RELEASEWITHASSERTS builds) the stale sig type tripped
dbSigType::isSupply()'s assert(false) and aborted.

Move the dirty_nets_.erase to run for both routers.

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

…dirty-net-destroy

grt: erase destroyed net from dirty set in CUGR path

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

SteinerTreeBuilder stored a dbDatabase* that was never used. Remove the
constructor parameter and the db_ member, and update the callers
(OpenRoad::init and the gpl mbff unit test) to match.

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

The dependabot bump to v3.1.0 (30efff6) regressed contributor
detection. v3 OR-combines isFirstIssue and isFirstPullRequest on every
event, and the PR check dropped author filtering. As a result, any
contributor who has opened PRs but no plain issue in the repo is greeted
as a first-time PR author (upstream actions/first-interaction#369).

v1.3.0 branches on event type and matches PRs by author, so it does not
misfire. Pin back to v1.3.0 and restore its kebab-case input names.

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

…unused-db-param

stt: drop unused dbDatabase* from SteinerTreeBuilder

…bCallBack-network_

rsz: rm unused OdbCallBack::network_

…rst-interaction-v1

ci: revert first-interaction to v1.3.0 (fix false first-PR greetings)

gpl: hoist loop-invariant accessors in getDensityGradient (bit-identical)

…ings

Enable the external_include_paths toolchain feature so include paths into
external/ (bzlmod/BCR deps) are passed as -isystem instead of -I. Clang
suppresses diagnostics from system headers, so this silences warnings from
ALL third-party headers pulled into our TUs via template instantiation
(boost, Eigen, googletest, ...) at the source -- no per-warning, per-dep
-Wno list to maintain. The existing '.*external/.*@-w' per_file_copt only
covers compiling external *source* files, not their headers included by ours.

The feature is global (target and exec config) and also demotes a from-source
dep's own -iquote entries to -isystem. Two deps need adjusting so their own
quoted includes keep precedence over glibc/system headers:

- git (built as an exec tool for the lint tests): builtin/get-tar-commit-id.c
  includes git's own tar.h (struct ustar_header, TYPEFLAG_GLOBAL_HEADER),
  which must shadow glibc's <tar.h>. Patch GIT_COPTS to re-add the repo root
  as -iquote. Same class of fix as the existing bison/sed/gawk/m4 overrides.

- qt-bazel: external_include_paths moves Qt's include dirs into
  compilation_context.external_includes, which the moc rule did not gather.
  Bump to 541ebf6d, which teaches moc to also collect external_includes.

A full 'bazel build //...' passes; no other from-source dep (abc, yosys,
or-tools, ...) needed changes.

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

…IpChecker-sta_

dbsta: rm unused IpChecker::sta_

…ernal-include-paths

bazel: pass external headers as -isystem to silence third-party warnings

This test is listed in src/dpl/test/CMakeLists.txt but was missing from
the Bazel BUILD file, so it only ran under CMake. Add it to
COMPULSORY_TESTS to restore CMake/Bazel parity.

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

…-multi-height-power-align

dpl: register multi_height_power_align regression test in Bazel

…d Abseil

Two fixes for installer breakage first reported after the Yosys v0.67 bump:

1. Yosys v0.67 dropped its Makefile in favor of CMake, so _install_yosys
   failed with "make: *** No targets specified and no makefile found".
   Detect a missing Makefile and build with CMake instead, mirroring the
   flags ORFS uses (Release + install prefix). The make path is kept for
   older versions.

2. Prebuilt or-tools links its own bundled (shared) Abseil. _install_abseil
   preferred an Abseil copy in the install prefix over the or-tools bundle,
   so machines with a leftover /usr/local Abseil from earlier installer runs
   built OpenROAD against two Abseil copies and crashed at startup with:

     ERROR: Inconsistency between flag object and registration for flag
     'flagfile', likely due to duplicate flags or an ODR violation.

   Prefer the or-tools bundled Abseil when it exists, purge stale Abseil
   artifacts from the prefix before installing a new copy, and warn when a
   stale prefix copy shadows the or-tools bundle or when the pinned
   ABSL_VERSION diverges from the bundled one.

Signed-off-by: Joao Luis Sombrio <sombrio@sombrasoft.dev>

Resolve upf_aes .ok/.defok golden conflicts by regenerating against the
merged binary (buffering-boost rsz + master DPL/GPL negotiation changes).

Signed-off-by: Minju Kim <mkim@precisioninno.com>

…sue1737-hold-buffer-count-fix

rsz: fix hold-buffer count reset on repair_timing -hold rollback

Review feedback: operator at end of line is less fragile than
backslash continuation with a multi-line conditional.

Signed-off-by: Joao Luis Sombrio <sombrio@sombrasoft.dev>

…to rsz-buffering-boost

rsz sources auto-merged. Regenerate aes_sky130hd / jpeg_sky130hd metric +
limit goldens against the merged binary (buffering-boost + #10835): e.g. aes
RSZ::hold_buffer_count settles at 330 (neither the pre-fix 19 nor master 343).

Signed-off-by: Minju Kim <mkim@precisioninno.com>

rsz: Avoid SIGILL in net tree stitching if subtree can't build

…z-buffering-boost

rsz: Merge rsz-buffering-boost for Rebuffer QoR gain

Signed-off-by: Simon Dorrer <simon.dorrer@jku.at>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

Address review: the double negative ("disable" in the command name plus -disable/-enable flags) was confusing. Rename to `set_routing_auto_taper` and require exactly one of -enable/-disable (no default), so it reads as a direct on/off switch. Behavior and the odb flag (`disable_auto_taper`) are unchanged.

Signed-off-by: Simon Dorrer <simon.dorrer@jku.at>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

…essed`

Address review from @mole99: the 'NDR net whose auto-taper is suppressed' predicate
was duplicated inline in three places (FlexPA_unique once, FlexPA_acc_point
twice). Extract it into `frNet::isAutoTaperSuppressed(bool auto_taper_enabled)`
and reuse it. Behavior-preserving.

Signed-off-by: Simon Dorrer <simon.dorrer@jku.at>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

Signed-off-by: Simon Dorrer <simon.dorrer@jku.at>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

…gument checks

Signed-off-by: Simon Dorrer <simon.dorrer@jku.at>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

…ble` flag

Signed-off-by: Simon Dorrer <simon.dorrer@jku.at>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

Signed-off-by: Simon Dorrer <simon.dorrer@jku.at>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

set_routing_auto_taper was defined in OpenRoad.tcl but documented in
src/odb/README.md, so odb_man_tcl_check saw README count (22) != help
count (21).  Move the sta::define_cmd_args/proc into odb.tcl and renumber
its errors ORD 1016-1018 -> ODB 1023-1025 to match the odb namespace.

Also refresh the stale goldens: bump odb_readme_msgs_check.ok to 24 for
the added README section, and regenerate ndr_no_auto_taper.defok after
the QoR shift from rebasing onto master.

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

Address review: isAutoTaperDisabled and isAutoTaperSuppressed were too
similar and negative. The frNet per-net flag is now positive
(auto_taper_enabled_, default true) with setAutoTaperEnabled(), and the
effective query is autoTaperEnabled(bool global_enabled), matching the
existing drNet::autoTaperEnabled (which now delegates to it). Call sites
state the NDR condition explicitly. Behavior unchanged.

Signed-off-by: Simon Dorrer <simon.dorrer@jku.at>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

gpl: update large tests

Signed-off-by: Matt Liberty <mliberty@precisioninno.com>

drt: allow NDR auto-taper to be disabled per net

The FastRouteRenderer.h is reachable via
src/fastroute/include/FastRouteRenderer.h, but no include
path was given, so it worked by accident.
Make that explicit.

Signed-off-by: Henner Zeller <h.zeller@acm.org>

…dency-installer-yosys-absl

etc: fix DependencyInstaller for CMake-only Yosys and or-tools bundled Abseil

Relevant release notes:
https://github.com/hzeller/bant/releases/tag/v0.3.1

Useful new feature: select a column of the tabular outputs
right with bant without extra use of awk.

Signed-off-by: Henner Zeller <h.zeller@acm.org>

FastRouteRenderer.h is declared by //src/grt:ui, use correct prefix.

Update bant.

Signed-off-by: dsengupta0628 <dsengupta@precisioninno.com>

Signed-off-by: dsengupta0628 <dsengupta@precisioninno.com>

…_sta_july10

bump sta ptr to sta master