Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Tests mpmc_stream and mpmc fails on s390x: assertion left == right failed #99

Open
jonassmedegaard opened this issue Jul 29, 2024 · 1 comment

Comments

@jonassmedegaard
Copy link

jonassmedegaard commented Jul 29, 2024

Building crate async-channel v2.3.1 fails tests mpmc_stream and mpmc on s390x, like this:

Running `CARGO=/usr/bin/cargo CARGO_MANIFEST_DIR=/usr/share/cargo/registry/async-channel-2.3.1 CARGO_PKG_AUTHORS='Stjepan Glavina <[email protected]>' CARGO_PKG_DESCRIPTION='Async multi-producer multi-consumer channel' CARGO_PKG_HOMEPAGE='' CARGO_PKG_LICENSE='Apache-2.0 OR MIT' CARGO_PKG_LICENSE_FILE='' CARGO_PKG_NAME=async-channel CARGO_PKG_README=README.md CARGO_PKG_REPOSITORY='https://github.com/smol-rs/async-channel' CARGO_PKG_RUST_VERSION=1.60 CARGO_PKG_VERSION=2.3.1 CARGO_PKG_VERSION_MAJOR=2 CARGO_PKG_VERSION_MINOR=3 CARGO_PKG_VERSION_PATCH=1 CARGO_PKG_VERSION_PRE='' LD_LIBRARY_PATH='/tmp/tmp.5edrX768DR/target/s390x-unknown-linux-gnu/release/deps:/tmp/tmp.5edrX768DR/target/s390x-unknown-linux-gnu/release:/usr/lib/rustlib/s390x-unknown-linux-gnu/lib' /tmp/tmp.5edrX768DR/target/s390x-unknown-linux-gnu/release/deps/bounded-9d69132e446b210a`
104s 
104s running 22 tests
104s test capacity ... ok
104s test len_empty_full ... ok
104s test receiver_count ... ok
104s test recv_after_close ... ok
104s test len ... ok
104s test send_after_close ... ok
104s test sender_count ... ok
104s test smoke ... ok
104s test smoke_blocking ... ok
104s test mpmc_stream ... FAILED
105s test mpmc ... FAILED
105s test spsc ... ok
105s test weak ... ok
105s test forget_blocked_sender ... ok
105s test forget_blocked_receiver ... ok
105s test close_wakes_sender ... ok
105s test close_wakes_receiver ... ok
106s test try_recv ... ok
107s test try_send ... ok
107s test force_send ... ok
108s test recv ... ok
108s test send ... ok
108s 
108s failures:
108s 
108s ---- mpmc_stream stdout ----
108s thread 'mpmc_stream' panicked at tests/bounded.rs:501:9:
108s assertion `left == right` failed
108s   left: 3
108s  right: 4
108s stack backtrace:
108s    0: rust_begin_unwind
108s    1: core::panicking::panic_fmt
108s    2: core::panicking::assert_failed_inner
108s    3: core::panicking::assert_failed
108s    4: core::ops::function::FnOnce::call_once
108s note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.
108s 
108s ---- mpmc stdout ----
108s thread 'mpmc' panicked at tests/bounded.rs:468:9:
108s assertion `left == right` failed
108s   left: 5
108s  right: 4
108s stack backtrace:
108s    0: rust_begin_unwind
108s    1: core::panicking::panic_fmt
108s    2: core::panicking::assert_failed_inner
108s    3: core::panicking::assert_failed
108s    4: core::ops::function::FnOnce::call_once
108s note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.
108s 
108s 
108s failures:
108s     mpmc
108s     mpmc_stream
108s 
108s test result: FAILED. 20 passed; 2 failed; 0 ignored; 0 measured; 0 filtered out; finished in 3.50s

Full build log

Curiously, a previous build from virtually identical source (only change was specific to armel) succeeded.
The build environment had changed slightly:
systemd 256.2 -> 256-4
libc6 2.38 -> 2.39
linux-libc-dev 6.9.9 -> 6.9.10

Perhaps this is related to #78 - except here the failure has persisted in 5 contiguous rebuilds.

@jonassmedegaard
Copy link
Author

Another curious detail: The tests succeed with --no-default-features.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Development

No branches or pull requests

1 participant