run-pass tests are failing on travis with "fatal runtime error: failed to initiate panic, error 3"

[TimNN]

For example https://travis-ci.org/rust-lang/rust/builds/147001444

Log excerpt:

test [run-pass] run-pass/autoderef-method-priority.rs ... ok
test [run-pass] run-pass/autoderef-method-twice.rs ... ok
test [run-pass] run-pass/autoderef-method-twice-but-not-thrice.rs ... ok
test [run-pass] run-pass/autoderef-method.rs ... ok
test [run-pass] run-pass/autoderef-privacy.rs ... ok
test [run-pass] run-pass/backtrace-debuginfo-aux.rs ... ignored
test [run-pass] run-pass/autoref-intermediate-types-issue-3585.rs ... ok
fatal runtime error: failed to initiate panic, error 3
Aborted (core dumped)
make: *** [tmp/check-stage2-T-x86_64-unknown-linux-gnu-H-x86_64-unknown-linux-gnu-rpass.ok] Error 134
/build/mk/tests.mk:769: recipe for target 'tmp/check-stage2-T-x86_64-unknown-linux-gnu-H-x86_64-unknown-linux-gnu-rpass.ok' failed

From the timing I suspect this is related to #34832, cc @alexcrichton @vadimcn

Edit: The failing test is probably either run-pass/backtrace.rs or run-pass/backtrace-debuginfo.rs.

Edit2: very likely caused by #34832, since it's travis build is the first to exhibit this error.

[vadimcn]

Any ideas what's different about Travis? This test passed on my local builds (as well as on buildbots, obviously).

(BTW, am I the only one who's always getting confused about which test has failed? I don't see anything in the compiletest's output to indicate the culprit).

[TimNN]

No idea what's different about travis, everything worked fine locally for me as well. We may want to just do a PR with some additional logging to see which one (if any) of the return _URC_FATAL_PHASE1_ERROR triggers the error.

Yeah, I can also only guess, I'm believe compiletest only prints something after running the test...

[alexcrichton]

@vadimcn I wonder if it's different ubuntu versions on Travis here? It may be some glibc version thing going on. If you execute tests in the dockerfile it should in theory reproduce.

[cgswords]

I'm also experiencing this on #35018 and it's working fine locally.

[cgswords]

This has also caused the following PRs to fail:

• Fix HashMap's values_mut example to use println! #35001
• rustdoc: Remove item types from some title pages #35003
• Fix build of compiler-rt on FreeBSD #35004
• Dir entry doc #35009
• Improve Open doc #35010
• Fix formatting generation for rustdoc code examples #35012
• move coerce-match{,-calls} into run-pass-valgrind #35013
• Properly enforce the "patterns aren't allowed in foreign functions" rule #35015
• Rewrite/expansion of slice::split doc examples. #35019

[vadimcn]

I wonder if it's different ubuntu versions on Travis here? It may be some glibc version thing going on.

If I read our Dockerfile correctly, it uses LLVM 3.7 release, not whatever dev version is currently in rust-lang/llvm? I think that's far more likely to cause codegen differences than a different glibc.

[alexcrichton]

@vadimcn ah yeah that makes sense too yeah

[jhod0]

This is also causing #34970 to fail

[jhod0]

Looks like the PRs were merged - should this be closed?

[alexcrichton]

Indeed!