ICE in librustc_resolve/macros.rs

[durka]

New ICE in nightly, found by Travis. This is a regression (works on stable and beta). The first nightly that fails is rustc 1.13.0-nightly (1265cbf4e 2016-09-15) (that is, nightly-2016-09-16).

Backtrace:

[root@li1424-173 macrolisp]# cargo +nightly-2016-09-16 test
   Compiling macrolisp v0.2.0 (file:///root/rust/macrolisp)
error: internal compiler error: unexpected panic

note: the compiler unexpectedly panicked. this is a bug.

note: we would appreciate a bug report: https://github.com/rust-lang/rust/blob/master/CONTRIBUTING.md#bug-reports

note: run with `RUST_BACKTRACE=1` for a backtrace

thread 'rustc' panicked at 'assertion failed: `(left == right)` (left: `28`, right: `27`)', ../src/librustc_resolve/macros.rs:138
stack backtrace:
   1:     0x7f087aa56773 - std::sys::backtrace::tracing::imp::write::h4b09e6e8c01db097
   2:     0x7f087aa6ff3d - std::panicking::default_hook::{{closure}}::h1d3243f546573ff4
   3:     0x7f087aa6508e - std::panicking::default_hook::h96c288d728df3ebf
   4:     0x7f087aa65778 - std::panicking::rust_panic_with_hook::hb1322e5f2588b4db
   5:     0x7f087aa65612 - std::panicking::begin_panic::hfbeda5aad583dc32
   6:     0x7f087aa65550 - std::panicking::begin_panic_fmt::h4fe9fb9d5109c4bf
   7:     0x7f087909e08a - rustc_resolve::macros::ExpansionVisitor::visit_invoc::hd3288ffa1e19e7a3
   8:     0x7f0879094b22 - syntax::visit::walk_expr::hac8fda8588c2ef13
   9:     0x7f087909c3a2 - rustc_resolve::macros::<impl syntax::ext::base::Resolver for rustc_resolve::Resolver<'a>>::visit_expansion::hbc70678eeb0f70e7
  10:     0x7f08737d4c15 - syntax::ext::expand::MacroExpander::collect_invocations::h394c2a558f484088
  11:     0x7f08737d1fd0 - syntax::ext::expand::MacroExpander::expand::h959b262d25638add
  12:     0x7f08737d0619 - syntax::ext::expand::MacroExpander::expand_crate::h08dcfa5db99b42c9
  13:     0x7f08737dad84 - syntax::ext::expand::expand_crate::hc240d93cfea63653
  14:     0x7f087ae6b8ed - rustc_driver::driver::phase_2_configure_and_expand::{{closure}}::h2d39fc9f02dd8655
  15:     0x7f087ae1fb18 - rustc_driver::driver::phase_2_configure_and_expand::hb41c4aa3b639bda4
  16:     0x7f087ae192d8 - rustc_driver::driver::compile_input::hc0edbed7edb3eb18
  17:     0x7f087ae466b6 - rustc_driver::run_compiler::h22d678d32fb7c300
  18:     0x7f087ad80aa3 - std::panicking::try::do_call::h4d040997e2efdaf3
  19:     0x7f087aa77e36 - __rust_maybe_catch_panic
  20:     0x7f087ad9fb59 - <F as alloc::boxed::FnBox<A>>::call_box::hba0b436c79e56b23
  21:     0x7f087aa63ac0 - std::sys::thread::Thread::new::thread_start::h022e3887023c6290
  22:     0x7f0872b59453 - start_thread
  23:     0x7f087a6bc7de - __GI___clone
  24:                0x0 - <unknown>

error: Could not compile `macrolisp`.

To learn more, run the command again with --verbose.

[est31]

I can reproduce it with the vorbis crate.

[est31]

Err sorry, its my crate lewton which shows the bug.

[TimNN]

This could be fixed by the next nightly (not sure without looking at the code).

[est31]

Minified example that reproduces the issue:

fn main() {
    let mut a = 0;
    macro_rules! b {
        (-$x:expr) => a
    }
    b![-1] = b![-0];
}

[TimNN]

That probably won't be fixed by the next nightly, cc @jseyfried.

Edit: Or maybe it will, the previous PR was more expansive than I thought.

[durka]

It's still failing on Travis up through today, but I'll report back tomorrow.

[est31]

Even more minified, and made compile on rust stable/beta:

fn main() {
    let mut a = 0;
    macro_rules! b { () => {a} }
    b![] = b![];
}

[jseyfried]

This should be fixed by #36555.

[est31]

Awesome!

[est31]

I can confirm that this bug is fixed in rust nightly version 55bf6a4f8 2016-09-18.

[durka]

Close or needstest?

On Sep 20, 2016 11:41, "Eduard-Mihai Burtescu" [email protected]
wrote:

Closed #36575 #36575.

—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
#36575 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AAC3n6w7oFt2L3fNBjVgHhU45G2qdGf4ks5qr6pPgaJpZM4KAXC1
.

[TimNN]

#36555 already added a test, as far as I can tell.