Constant PoisonErrors

[lovesegfault]

I haven't used Rust in a few weeks, so yesterday I updated my toolchain and started a new project. I use VSCode with the official RLS extension, both of which are at the latest version.

$ rustup update
info: syncing channel updates for 'stable-x86_64-unknown-linux-gnu'
info: syncing channel updates for 'nightly-x86_64-unknown-linux-gnu'
   stable-x86_64-unknown-linux-gnu unchanged - rustc 1.24.1 (d3ae9a9e0 2018-02-27)
  nightly-x86_64-unknown-linux-gnu unchanged - rustc 1.26.0-nightly (259e4a678 2018-03-04)
$ code --version
1.20.1
f88bbf9137d24d36d968ea6b2911786bfe103002
x64

In trying to write even the simplest code, RLS constantly crashes with the following error, grabbed from VSCode's Output window:

thread '<unnamed>' panicked at 'no entry found for key', libcore/option.rs:917:5
note: Run with `RUST_BACKTRACE=1` for a backtrace.
thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: "PoisonError { inner: .. }"', libcore/result.rs:945:5

It happens constantly, independent of which project I am editing, and causes RLS to crash. Also, whenever I have an open file and I'm in the midst of writing something, say I type fn foo( and go get more tea, it ramps my CPU usage to 100% while I don't make it syntactically correct.

Edit: Apparently it causes the whole Extension Host to crash as well.

[nrc]

Could you try running with RUST_BACKTRACE=1 to get a backtrace please? The relevant error is the 'no entry found for key' one, the poison errors are just an effect of the earlier one. Also, what project are you editing?

[kamyuentse]

Same here:

thread '<unnamed>' panicked at 'no entry found for key', libcore/option.rs:917:5
stack backtrace:
   0:        0x113327573 - std::sys::unix::backtrace::tracing::imp::unwind_backtrace::hc7e7373e4b839c9b
   1:        0x11331e682 - std::sys_common::backtrace::_print::h6a4974acad20e757
   2:        0x113323f7d - std::panicking::default_hook::{{closure}}::h78c62f8e395d91c7
   3:        0x113323c08 - std::panicking::default_hook::h907911f1269dc7bf
   4:        0x1118f34c0 - core::ops::function::Fn::call::had89f3dbf8bade74
   5:        0x113324440 - std::panicking::rust_panic_with_hook::h8ce3bca5760f9a14
   6:        0x11332423a - std::panicking::begin_panic_fmt::hcd1bdff30aebfae7
   7:        0x113324132 - rust_begin_unwind
   8:        0x11337f375 - core::panicking::panic_fmt::h3a5d7d1561eabb2a
   9:        0x11337f3e8 - core::option::expect_failed::h79bc204e6880241e
  10:        0x10f75d23f - rls::build::plan::Plan::prepare_work::h1ce43ba631af6e15
  11:        0x10f760046 - rls::build::Internals::run_build::h2cc49a6f84506a06
  12:        0x10f64a50b - std::sys_common::backtrace::__rust_begin_short_backtrace::heab24dcd036098a7
  13:        0x11333f2ae - __rust_maybe_catch_panic
  14:        0x10f6ca40f - <F as alloc::boxed::FnBox<A>>::call_box::h71e7bf5f02600718
  15:        0x113333f1b - std::sys::unix::thread::Thread::new::thread_start::hc558e44d3000c9fa
  16:     0x7fff6649d6c0 - _pthread_body
  17:     0x7fff6649d56c - _pthread_start
thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: "PoisonError { inner: .. }"', libcore/result.rs:945:5
stack backtrace:
   0:        0x113327573 - std::sys::unix::backtrace::tracing::imp::unwind_backtrace::hc7e7373e4b839c9b
   1:        0x11331e682 - std::sys_common::backtrace::_print::h6a4974acad20e757
   2:        0x113323f7d - std::panicking::default_hook::{{closure}}::h78c62f8e395d91c7
   3:        0x113323c08 - std::panicking::default_hook::h907911f1269dc7bf
   4:        0x1118f34c0 - core::ops::function::Fn::call::had89f3dbf8bade74
   5:        0x113324440 - std::panicking::rust_panic_with_hook::h8ce3bca5760f9a14
   6:        0x11332423a - std::panicking::begin_panic_fmt::hcd1bdff30aebfae7
   7:        0x113324132 - rust_begin_unwind
   8:        0x11337f375 - core::panicking::panic_fmt::h3a5d7d1561eabb2a
   9:        0x10f6c887f - core::result::unwrap_failed::h905f4c1cea1445ec
  10:        0x10f75ec72 - rls::build::BuildQueue::request_build::h393f6d93bee23cbf
  11:        0x10f740d1b - rls::actions::InitActionContext::build_current_project::h33d71442d9b4bb2e
  12:        0x10f78e54a - <rls::server::LsService<O>>::dispatch_message::hb04b01480cb399d5
  13:        0x10f79b816 - rls::main::h286de24f3dde7ae1
  14:        0x10f64b7b5 - std::rt::lang_start::{{closure}}::hbc03afc0e9e830d1
  15:        0x113324097 - std::panicking::try::do_call::ha563b1ad4e069264
  16:        0x11333f2ae - __rust_maybe_catch_panic
  17:        0x11332476d - std::rt::lang_start_internal::h0fed5c2d94673dee
  18:        0x10f7b920b - main

I don't know which RPC request cause these panic, but the two panic reason shown above occurred repeatedly.

Edit: I got some output before rls panic:

error: expected identifier, found keyword `in`
 --> bogofile:1:5
  |
1 | use in
  |     ^^

thread '<unnamed>' panicked at 'no entry found for key', libcore/option.rs:917:5
note: Run with `RUST_BACKTRACE=1` for a backtrace.
thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: "PoisonError { inner: .. }"', libcore/result.rs:945:5

[nrc]

The error in the backtrace here is due to a bad lookup in a HashMap. There are only two places that can happen, I think: line 328: self.compiler_jobs[x] and line 231: self.rev_dep_graph[&top].

[nrc]

I believe this is fixed now (with latest nightly RLS), please comment if not and I'll re-open