Investigate panic while dropping OsIpcReceiverSet #32912
Comments
|
Back traces of all threads: |
|
if we delay dropping of ipc stuff the panic does not happen: diff --git a/components/net/tests/resource_thread.rs b/components/net/tests/resource_thread.rs
index 9a32143d71..c52884b243 100644
--- a/components/net/tests/resource_thread.rs
+++ b/components/net/tests/resource_thread.rs
@@ -35,6 +35,7 @@ fn test_exit() {
);
resource_thread.send(CoreResourceMsg::Exit(sender)).unwrap();
receiver.recv().unwrap();
+ std::thread::sleep(std::time::Duration::from_millis(1000));
}
#[test]but this doesn't help: #[test]
fn test_exit() {
+ {
let (tx, _rx) = ipc::channel().unwrap();
let (mtx, _mrx) = ipc::channel().unwrap();
let (sender, receiver) = ipc::channel().unwrap();
@@ -35,6 +36,8 @@ fn test_exit() {
);
resource_thread.send(CoreResourceMsg::Exit(sender)).unwrap();
receiver.recv().unwrap();
+ }
+ std::thread::sleep(std::time::Duration::from_millis(1000));
} |
|
What's the code that's panicking? |
mrobinson
changed the title
github-merge-queue bot
pushed a commit
that referenced
this issue
Aug 7, 2024
* Update for nix Signed-off-by: Hayashi Mikihiro <[email protected]> * Update to Rust 1.80.0 Signed-off-by: Hayashi Mikihiro <[email protected]> * Rename to BindingMode from BindingAnnotation rust-lang/rust#124047 Signed-off-by: Hayashi Mikihiro <[email protected]> * Remove TypeVariableOriginKind rust-lang/rust#123016 Signed-off-by: Hayashi Mikihiro <[email protected]> * Remove TypeVariableOrigin rust-lang/rust#124955 Signed-off-by: Hayashi Mikihiro <[email protected]> * Remove LintDiagnostic::msg rust-lang/rust#125410 Signed-off-by: Hayashi Mikihiro <[email protected]> * common.rs fmt mistake indents Signed-off-by: Hayashi Mikihiro <[email protected]> * trace_in_no_trace.rs remove mistake space Signed-off-by: Hayashi Mikihiro <[email protected]> * trace_in_no_trace.rs remove mistake head space Signed-off-by: Hayashi Mikihiro <[email protected]> * Workaround for #32912 It's dirty but it worked on 2000 runs Signed-off-by: sagudev <[email protected]> --------- Signed-off-by: Hayashi Mikihiro <[email protected]> Signed-off-by: sagudev <[email protected]> Co-authored-by: Samson <[email protected]>
Gae24
pushed a commit
to Gae24/servo
that referenced
this issue
Aug 7, 2024
* Update for nix Signed-off-by: Hayashi Mikihiro <[email protected]> * Update to Rust 1.80.0 Signed-off-by: Hayashi Mikihiro <[email protected]> * Rename to BindingMode from BindingAnnotation rust-lang/rust#124047 Signed-off-by: Hayashi Mikihiro <[email protected]> * Remove TypeVariableOriginKind rust-lang/rust#123016 Signed-off-by: Hayashi Mikihiro <[email protected]> * Remove TypeVariableOrigin rust-lang/rust#124955 Signed-off-by: Hayashi Mikihiro <[email protected]> * Remove LintDiagnostic::msg rust-lang/rust#125410 Signed-off-by: Hayashi Mikihiro <[email protected]> * common.rs fmt mistake indents Signed-off-by: Hayashi Mikihiro <[email protected]> * trace_in_no_trace.rs remove mistake space Signed-off-by: Hayashi Mikihiro <[email protected]> * trace_in_no_trace.rs remove mistake head space Signed-off-by: Hayashi Mikihiro <[email protected]> * Workaround for servo#32912 It's dirty but it worked on 2000 runs Signed-off-by: sagudev <[email protected]> --------- Signed-off-by: Hayashi Mikihiro <[email protected]> Signed-off-by: sagudev <[email protected]> Co-authored-by: Samson <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I think this happens because we panic in drop of
OsIpcReceiverSet, (it's actually panic while panic), but this is in separate thread. While before this was visible rarely (we have retry 3 on unit-tests), but now rust-lang/rust#119224 causes main test thread to wait for panic_hook to not be used (so it can be freed), so this panic is visible way more often (as it actually finishes in time).Originally posted by @sagudev in #32896 (comment)
The text was updated successfully, but these errors were encountered: