https://sourceware.org/bugzilla/show_bug.cgi?id=16446
Michael Tautschnig <mt at debian dot org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |mt at debian dot org
--- Comment #3 from Michael Tautschnig <mt at debian dot org> ---
It seems that the problem is tied to the use of start-group/end-group.
I have recently run into this when trying to build CBMC via its Makefile-based
build system. In a system where ld defaults to gold I see:
x86_64-pc-linux-gnu-g++ -o goto-synthesizer -Wl,--start-group
../ansi-c/ansi-c.a ../analyses/analyses.a ../assembler/assembler.a
../big-int/big-int.a ../cpp/cpp.a ../goto-checker/goto-checker.a
../goto-programs/goto-programs.a ../goto-instrument/contracts/contracts.o
../goto-instrument/contracts/havoc_assigns_clause_targets.o
../goto-instrument/contracts/inlining_decorator.o
../goto-instrument/contracts/instrument_spec_assigns.o
../goto-instrument/contracts/memory_predicates.o
../goto-instrument/contracts/utils.o ../goto-instrument/havoc_utils.o
../goto-instrument/loop_utils.o ../goto-instrument/nondet_static.o
../goto-instrument/unwind.o ../goto-instrument/unwindset.o
../goto-symex/goto-symex.a ../json/json.a ../langapi/langapi.a
../linking/linking.a ../pointer-analysis/pointer-analysis.a
../solvers/solvers.a ../util/util.a ../xmllang/xmllang.a cegis_evaluator.o
cegis_verifier.o dump_loop_contracts.o enumerative_loop_contracts_synthesizer.o
expr_enumerator.o goto_synthesizer_languages.o goto_synthesizer_main.o
goto_synthesizer_parse_options.o synthesizer_utils.o -Wl,--end-group
/x86_64-pc-linux-gnu/bin/ld: internal error in find_runnable_or_wait, at
/src/binutils-2.41/gold/workqueue.cc:263
collect2: error: ld returned 1 exit status
Appending `-fuse-ld=bfd` makes the above work just fine. It, however, also
works find with gold when removing start-group/end-group (and adjusting the
order of files as needed).
As this is perfectly reproducible for me (I also tried different versions of
binutils, the above is for binutils-2.41) I am happy to provide whatever other
information that may help to get this resolved after such a number of years.
--
You are receiving this mail because:
You are on the CC list for the bug.
