Stepping with "next" in a debugger skips past end-of-scope drops #128971

saethlin · 2024-08-11T14:05:04Z

I am trying to re-enable a lot of disabled debuginfo tests, and in doing that I learned that tests/debuginfo/drop-locations https://github.com/rust-lang/rust/blob/c9bd03cb724e13cca96ad320733046cbdb16fbbe/tests/debuginfo/drop-locations.rs was broken at some point in the past 6 years. No idea when, I will not be doing a bisection.

The bug is that in code like this:

{
    let s = String::from("s");
}

stepping through this with the gdb command next should step from the String::from call to the close brace. But gdb never stops on the close brace, even though there is indeed a call to drop_in_place for the String, and the debuginfo for the call correctly points at the close brace.

The text was updated successfully, but these errors were encountered:

…=compiler-errors Enable debuginfo tests that have been "temporarily disabled" for the past 6 years The PR history is a bit of a mess because I had to test this a lot with try-jobs, so I'll try to summarize the non-obvious changes here. A number of tests now have `min-lldb-version: 1800`. Those tests should have gotten an lldb version jump either in rust-lang#124781 or long ago. Note that all such tests with that lldb version requirement do not run in Apple CI. `tests/debuginfo/drop-locations.rs` is staying disabled for now because gdb doesn't know to stop on the drop calls produced by a `}`: rust-lang#128971 `tests/debuginfo/function-arg-initialization.rs` now has `-Zmir-enable-passes=-SingleUseConsts`; without that we initialize the const before the function prelude: rust-lang#128945 `tests/debuginfo/by-value-non-immediate-argument.rs` fails because we don't generate a function prelude for unused non-immediate arguments, even with all optimizations disabled, and this seems to confuse debuggers on aarch64: rust-lang#128973 `tests/debuginfo/pretty-std.rs` is staying disabled on windows-gnu because our test harness doesn't know how to load our pretty-printers on that target: rust-lang#128981 `tests/debuginfo/method-on-enum.rs` and `tests/debuginfo/option-like-enum.rs` encounter some kind of gdb bug on i686-pc-windows-gnu. I don't know enough about that situation to write a good issue. I plan on doing more work on this test suite. There's clearly a lot more basic cleanup work to do here.

…=<try> Enable debuginfo tests that have been "temporarily disabled" for the past 6 years The PR history is a bit of a mess because I had to test this a lot with try-jobs, so I'll try to summarize the non-obvious changes here. A number of tests now have `min-lldb-version: 1800`. Those tests should have gotten an lldb version jump either in rust-lang#124781 or long ago. Note that all such tests with that lldb version requirement do not run in Apple CI. `tests/debuginfo/drop-locations.rs` is staying disabled for now because gdb doesn't know to stop on the drop calls produced by a `}`: rust-lang#128971 `tests/debuginfo/function-arg-initialization.rs` now has `-Zmir-enable-passes=-SingleUseConsts`; without that we initialize the const before the function prelude: rust-lang#128945 `tests/debuginfo/by-value-non-immediate-argument.rs` fails because we don't generate a function prelude for unused non-immediate arguments, even with all optimizations disabled, and this seems to confuse debuggers on aarch64: rust-lang#128973 `tests/debuginfo/pretty-std.rs` is staying disabled on windows-gnu because our test harness doesn't know how to load our pretty-printers on that target: rust-lang#128981 `tests/debuginfo/method-on-enum.rs` and `tests/debuginfo/option-like-enum.rs` encounter some kind of gdb bug on i686-pc-windows-gnu. I don't know enough about that situation to write a good issue. I plan on doing more work on this test suite. There's clearly a lot more basic cleanup work to do here. try-job: i686-gnu try-job: aarch64-gnu try-job: x86_64-msvc try-job: i686-mingw

…=compiler-errors Enable debuginfo tests that have been "temporarily disabled" for the past 6 years The PR history is a bit of a mess because I had to test this a lot with try-jobs, so I'll try to summarize the non-obvious changes here. A number of tests now have `min-lldb-version: 1800`. Those tests should have gotten an lldb version jump either in rust-lang#124781 or long ago. Note that all such tests with that lldb version requirement do not run in Apple CI. `tests/debuginfo/drop-locations.rs` is staying disabled for now because gdb doesn't know to stop on the drop calls produced by a `}`: rust-lang#128971 `tests/debuginfo/function-arg-initialization.rs` now has `-Zmir-enable-passes=-SingleUseConsts`; without that we initialize the const before the function prelude: rust-lang#128945 `tests/debuginfo/by-value-non-immediate-argument.rs` fails because we don't generate a function prelude for unused non-immediate arguments, even with all optimizations disabled, and this seems to confuse debuggers on aarch64: rust-lang#128973 `tests/debuginfo/pretty-std.rs` is staying disabled on windows-gnu because our test harness doesn't know how to load our pretty-printers on that target: rust-lang#128981 `tests/debuginfo/method-on-enum.rs` and `tests/debuginfo/option-like-enum.rs` encounter some kind of gdb bug on i686-pc-windows-gnu. I don't know enough about that situation to write a good issue. I plan on doing more work on this test suite. There's clearly a lot more basic cleanup work to do here.

…=<try> Enable debuginfo tests that have been "temporarily disabled" for the past 6 years The PR history is a bit of a mess because I had to test this a lot with try-jobs, so I'll try to summarize the non-obvious changes here. A number of tests now have `min-lldb-version: 1800`. Those tests should have gotten an lldb version jump either in rust-lang#124781 or long ago. Note that all such tests with that lldb version requirement do not run in Apple CI. `tests/debuginfo/drop-locations.rs` is staying disabled for now because gdb doesn't know to stop on the drop calls produced by a `}`: rust-lang#128971 `tests/debuginfo/function-arg-initialization.rs` now has `-Zmir-enable-passes=-SingleUseConsts`; without that we initialize the const before the function prelude: rust-lang#128945 `tests/debuginfo/by-value-non-immediate-argument.rs` fails because we don't generate a function prelude for unused non-immediate arguments, even with all optimizations disabled, and this seems to confuse debuggers on aarch64: rust-lang#128973 `tests/debuginfo/pretty-std.rs` is staying disabled on windows-gnu because our test harness doesn't know how to load our pretty-printers on that target: rust-lang#128981 `tests/debuginfo/method-on-enum.rs` and `tests/debuginfo/option-like-enum.rs` encounter some kind of gdb bug on i686-pc-windows-gnu. I don't know enough about that situation to write a good issue. I plan on doing more work on this test suite. There's clearly a lot more basic cleanup work to do here. try-job: x86_64-gnu-llvm-17 try-job: x86_64-gnu-tools try-job: aarch64-gnu try-job: arm-android try-job: armhf-gnu try-job: i686-gnu try-job: i686-gnu-nopt try-job: mingw-check try-job: test-various try-job: x86_64-fuchsia try-job: x86_64-rust-for-linux try-job: x86_64-gnu try-job: x86_64-gnu-stable try-job: x86_64-gnu-aux try-job: x86_64-gnu-debug try-job: x86_64-gnu-llvm-18 try-job: x86_64-gnu-nopt try-job: x86_64-apple-1 try-job: aarch64-apple try-job: x86_64-msvc try-job: i686-msvc try-job: i686-mingw try-job: x86_64-mingw

…=<try> Enable debuginfo tests that have been "temporarily disabled" for the past 6 years The PR history is a bit of a mess because I had to test this a lot with try-jobs, so I'll try to summarize the non-obvious changes here. A number of tests now have `min-lldb-version: 1800`. Those tests should have gotten an lldb version jump either in rust-lang#124781 or long ago. Note that all such tests with that lldb version requirement do not run in Apple CI. `tests/debuginfo/drop-locations.rs` is staying disabled for now because gdb doesn't know to stop on the drop calls produced by a `}`: rust-lang#128971 `tests/debuginfo/function-arg-initialization.rs` now has `-Zmir-enable-passes=-SingleUseConsts`; without that we initialize the const before the function prelude: rust-lang#128945 `tests/debuginfo/by-value-non-immediate-argument.rs` fails because we don't generate a function prelude for unused non-immediate arguments, even with all optimizations disabled, and this seems to confuse debuggers on aarch64: rust-lang#128973 `tests/debuginfo/pretty-std.rs` is staying disabled on windows-gnu because our test harness doesn't know how to load our pretty-printers on that target: rust-lang#128981 `tests/debuginfo/method-on-enum.rs` and `tests/debuginfo/option-like-enum.rs` encounter some kind of gdb bug on i686-pc-windows-gnu. I don't know enough about that situation to write a good issue. I plan on doing more work on this test suite. There's clearly a lot more basic cleanup work to do here. try-job: aarch64-gnu try-job: arm-android try-job: armhf-gnu try-job: i686-gnu try-job: i686-gnu-nopt try-job: test-various try-job: x86_64-fuchsia try-job: x86_64-rust-for-linux try-job: x86_64-gnu try-job: x86_64-gnu-stable

…=<try> Enable debuginfo tests that have been "temporarily disabled" for the past 6 years The PR history is a bit of a mess because I had to test this a lot with try-jobs, so I'll try to summarize the non-obvious changes here. A number of tests now have `min-lldb-version: 1800`. Those tests should have gotten an lldb version jump either in rust-lang#124781 or long ago. Note that all such tests with that lldb version requirement do not run in Apple CI. `tests/debuginfo/drop-locations.rs` is staying disabled for now because gdb doesn't know to stop on the drop calls produced by a `}`: rust-lang#128971 `tests/debuginfo/function-arg-initialization.rs` now has `-Zmir-enable-passes=-SingleUseConsts`; without that we initialize the const before the function prelude: rust-lang#128945 `tests/debuginfo/by-value-non-immediate-argument.rs` fails because we don't generate a function prelude for unused non-immediate arguments, even with all optimizations disabled, and this seems to confuse debuggers on aarch64: rust-lang#128973 `tests/debuginfo/pretty-std.rs` is staying disabled on windows-gnu because our test harness doesn't know how to load our pretty-printers on that target: rust-lang#128981 `tests/debuginfo/method-on-enum.rs` and `tests/debuginfo/option-like-enum.rs` encounter some kind of gdb bug on i686-pc-windows-gnu. I don't know enough about that situation to write a good issue. I plan on doing more work on this test suite. There's clearly a lot more basic cleanup work to do here. try-job: aarch64-apple try-job: aarch64-gnu try-job: dist-i586-gnu-i586-i686-musl try-job: i686-gnu try-job: i686-gnu-nopt try-job: i686-mingw try-job: i686-msvc try-job: test-various try-job: x86_64-apple-1

…=<try> Enable debuginfo tests that have been "temporarily disabled" for the past 6 years The PR history is a bit of a mess because I had to test this a lot with try-jobs, so I'll try to summarize the non-obvious changes here. A number of tests now have `min-lldb-version: 1800`. Those tests should have gotten an lldb version jump either in rust-lang#124781 or long ago. Note that all such tests with that lldb version requirement do not run in Apple CI. `tests/debuginfo/drop-locations.rs` is staying disabled for now because gdb doesn't know to stop on the drop calls produced by a `}`: rust-lang#128971 `tests/debuginfo/function-arg-initialization.rs` now has `-Zmir-enable-passes=-SingleUseConsts`; without that we initialize the const before the function prelude: rust-lang#128945 `tests/debuginfo/by-value-non-immediate-argument.rs` fails because we don't generate a function prelude for unused non-immediate arguments, even with all optimizations disabled, and this seems to confuse debuggers on aarch64: rust-lang#128973 `tests/debuginfo/pretty-std.rs` is staying disabled on windows-gnu because our test harness doesn't know how to load our pretty-printers on that target: rust-lang#128981 `tests/debuginfo/method-on-enum.rs` and `tests/debuginfo/option-like-enum.rs` encounter some kind of gdb bug on i686-pc-windows-gnu. I don't know enough about that situation to write a good issue. I plan on doing more work on this test suite. There's clearly a lot more basic cleanup work to do here. try-job: x86_64-gnu try-job: x86_64-gnu-distcheck try-job: x86_64-gnu-llvm-17 try-job: x86_64-gnu-llvm-18 try-job: x86_64-gnu-nopt try-job: x86_64-gnu-stable try-job: x86_64-mingw try-job: x86_64-msvc

…=compiler-errors Enable debuginfo tests that have been "temporarily disabled" for the past 6 years The PR history is a bit of a mess because I had to test this a lot with try-jobs, so I'll try to summarize the non-obvious changes here. A number of tests now have `min-lldb-version: 1800`. Those tests should have gotten an lldb version jump either in rust-lang#124781 or long ago. Note that all such tests with that lldb version requirement do not run in Apple CI. `tests/debuginfo/drop-locations.rs` is staying disabled for now because gdb doesn't know to stop on the drop calls produced by a `}`: rust-lang#128971 `tests/debuginfo/function-arg-initialization.rs` now has `-Zmir-enable-passes=-SingleUseConsts`; without that we initialize the const before the function prelude: rust-lang#128945 `tests/debuginfo/by-value-non-immediate-argument.rs` fails because we don't generate a function prelude for unused non-immediate arguments, even with all optimizations disabled, and this seems to confuse debuggers on aarch64: rust-lang#128973 `tests/debuginfo/pretty-std.rs` is staying disabled on windows-gnu because our test harness doesn't know how to load our pretty-printers on that target: rust-lang#128981 `tests/debuginfo/method-on-enum.rs` and `tests/debuginfo/option-like-enum.rs` encounter some kind of gdb bug on i686-pc-windows-gnu. I don't know enough about that situation to write a good issue. I plan on doing more work on this test suite. There's clearly a lot more basic cleanup work to do here.

…-errors Enable debuginfo tests that have been "temporarily disabled" for the past 6 years The PR history is a bit of a mess because I had to test this a lot with try-jobs, so I'll try to summarize the non-obvious changes here. A number of tests now have `min-lldb-version: 1800`. Those tests should have gotten an lldb version jump either in rust-lang/rust#124781 or long ago. Note that all such tests with that lldb version requirement do not run in Apple CI. `tests/debuginfo/drop-locations.rs` is staying disabled for now because gdb doesn't know to stop on the drop calls produced by a `}`: rust-lang/rust#128971 `tests/debuginfo/function-arg-initialization.rs` now has `-Zmir-enable-passes=-SingleUseConsts`; without that we initialize the const before the function prelude: rust-lang/rust#128945 `tests/debuginfo/by-value-non-immediate-argument.rs` fails because we don't generate a function prelude for unused non-immediate arguments, even with all optimizations disabled, and this seems to confuse debuggers on aarch64: rust-lang/rust#128973 `tests/debuginfo/pretty-std.rs` is staying disabled on windows-gnu because our test harness doesn't know how to load our pretty-printers on that target: rust-lang/rust#128981 `tests/debuginfo/method-on-enum.rs` and `tests/debuginfo/option-like-enum.rs` encounter some kind of gdb bug on i686-pc-windows-gnu. I don't know enough about that situation to write a good issue. I plan on doing more work on this test suite. There's clearly a lot more basic cleanup work to do here.

Introduce and use specialized `//@ ignore-auxiliary` for test support files instead of using `//@ ignore-test` ### Summary Add a semantically meaningful directive for ignoring test *auxiliary* files. This is for auxiliary files that *participate* in actual tests but should not be built by `compiletest` (i.e. these files are involved through `mod xxx;` or `include!()` or `#[path = "xxx"]`, etc.). ### Motivation A specialized directive like `//@ ignore-auxiliary` makes it way easier to audit disabled tests via `//@ ignore-test`. - These support files cannot use the canonical `auxiliary/` dir because they participate in module resolution or are included, or their relative paths can be important for test intention otherwise. Follow-up to: - rust-lang#139705 - rust-lang#139783 - rust-lang#139740 See also discussions in: - [#t-compiler > Directive name for non-test aux files?](https://rust-lang.zulipchat.com/#narrow/channel/131828-t-compiler/topic/Directive.20name.20for.20non-test.20aux.20files.3F/with/512773817) - [#t-compiler > Handling disabled &rust-lang#96;//@ ignore-test&rust-lang#96; tests](https://rust-lang.zulipchat.com/#narrow/channel/131828-t-compiler/topic/Handling.20disabled.20.60.2F.2F.40.20ignore-test.60.20tests/with/512005974) - [#t-compiler/meetings > &rust-lang#91;steering&rust-lang#93; 2025-04-11 Dealing with disabled tests](https://rust-lang.zulipchat.com/#narrow/channel/238009-t-compiler.2Fmeetings/topic/.5Bsteering.5D.202025-04-11.20Dealing.20with.20disabled.20tests/with/511717981) ### Remarks on remaining unconditionally disabled tests under `tests/` After this PR, against commit 79a272c, only **14** remaining test files are disabled through `//@ ignore-test`: <details> <summary>Remaining `//@ ignore-test` files under `tests/`</summary> ``` tests/debuginfo/drop-locations.rs 4://@ ignore-test (broken, see rust-lang#128971) tests/rustdoc/macro-document-private-duplicate.rs 1://@ ignore-test (fails spuriously, see issue rust-lang#89228) tests/rustdoc/inline_cross/assoc-const-equality.rs 3://@ ignore-test (FIXME: rust-lang#125092) tests/ui/match/issue-27021.rs 7://@ ignore-test (rust-lang#54987) tests/ui/match/issue-26996.rs 7://@ ignore-test (rust-lang#54987) tests/ui/issues/issue-49298.rs 9://@ ignore-test (rust-lang#54987) tests/ui/issues/issue-59756.rs 2://@ ignore-test (rustfix needs multiple suggestions) tests/ui/precondition-checks/write.rs 5://@ ignore-test (unimplemented) tests/ui/precondition-checks/read.rs 5://@ ignore-test (unimplemented) tests/ui/precondition-checks/write_bytes.rs 5://@ ignore-test (unimplemented) tests/ui/explicit-tail-calls/drop-order.rs 2://@ ignore-test: tail calls are not implemented in rustc_codegen_ssa yet, so this causes 🧊 tests/ui/panics/panic-short-backtrace-windows-x86_64.rs 3://@ ignore-test (rust-lang#92000) tests/ui/json/json-bom-plus-crlf-multifile-aux.rs 3://@ ignore-test Not a test. Used by other tests tests/ui/traits/next-solver/object-soundness-requires-generalization.rs 2://@ ignore-test (see rust-lang#114196) ``` </details> Of these, most are either **unimplemented**, or **spurious**, or **known-broken**. The outstanding one is `tests/ui/json/json-bom-plus-crlf-multifile-aux.rs` which I did not want to touch in *this* PR -- that aux file has load-bearing BOM and carriage returns and byte offset matters. I think those test files that require special encoding / BOM probably are better off as `run-make` tests. See rust-lang#139968 for that aux file. ### Review advice - Best reviewed commit-by-commit. - The directive name diverged from the most voted `//@ auxiliary` because I think that's easy to confuse with `//@ aux-{crate,dir}`. r? compiler

Rollup merge of rust-lang#139967 - jieyouxu:auxiliary, r=wesleywiser Introduce and use specialized `//@ ignore-auxiliary` for test support files instead of using `//@ ignore-test` ### Summary Add a semantically meaningful directive for ignoring test *auxiliary* files. This is for auxiliary files that *participate* in actual tests but should not be built by `compiletest` (i.e. these files are involved through `mod xxx;` or `include!()` or `#[path = "xxx"]`, etc.). ### Motivation A specialized directive like `//@ ignore-auxiliary` makes it way easier to audit disabled tests via `//@ ignore-test`. - These support files cannot use the canonical `auxiliary/` dir because they participate in module resolution or are included, or their relative paths can be important for test intention otherwise. Follow-up to: - rust-lang#139705 - rust-lang#139783 - rust-lang#139740 See also discussions in: - [#t-compiler > Directive name for non-test aux files?](https://rust-lang.zulipchat.com/#narrow/channel/131828-t-compiler/topic/Directive.20name.20for.20non-test.20aux.20files.3F/with/512773817) - [#t-compiler > Handling disabled &rust-lang#96;//@ ignore-test&rust-lang#96; tests](https://rust-lang.zulipchat.com/#narrow/channel/131828-t-compiler/topic/Handling.20disabled.20.60.2F.2F.40.20ignore-test.60.20tests/with/512005974) - [#t-compiler/meetings > &rust-lang#91;steering&rust-lang#93; 2025-04-11 Dealing with disabled tests](https://rust-lang.zulipchat.com/#narrow/channel/238009-t-compiler.2Fmeetings/topic/.5Bsteering.5D.202025-04-11.20Dealing.20with.20disabled.20tests/with/511717981) ### Remarks on remaining unconditionally disabled tests under `tests/` After this PR, against commit 79a272c, only **14** remaining test files are disabled through `//@ ignore-test`: <details> <summary>Remaining `//@ ignore-test` files under `tests/`</summary> ``` tests/debuginfo/drop-locations.rs 4://@ ignore-test (broken, see rust-lang#128971) tests/rustdoc/macro-document-private-duplicate.rs 1://@ ignore-test (fails spuriously, see issue rust-lang#89228) tests/rustdoc/inline_cross/assoc-const-equality.rs 3://@ ignore-test (FIXME: rust-lang#125092) tests/ui/match/issue-27021.rs 7://@ ignore-test (rust-lang#54987) tests/ui/match/issue-26996.rs 7://@ ignore-test (rust-lang#54987) tests/ui/issues/issue-49298.rs 9://@ ignore-test (rust-lang#54987) tests/ui/issues/issue-59756.rs 2://@ ignore-test (rustfix needs multiple suggestions) tests/ui/precondition-checks/write.rs 5://@ ignore-test (unimplemented) tests/ui/precondition-checks/read.rs 5://@ ignore-test (unimplemented) tests/ui/precondition-checks/write_bytes.rs 5://@ ignore-test (unimplemented) tests/ui/explicit-tail-calls/drop-order.rs 2://@ ignore-test: tail calls are not implemented in rustc_codegen_ssa yet, so this causes 🧊 tests/ui/panics/panic-short-backtrace-windows-x86_64.rs 3://@ ignore-test (rust-lang#92000) tests/ui/json/json-bom-plus-crlf-multifile-aux.rs 3://@ ignore-test Not a test. Used by other tests tests/ui/traits/next-solver/object-soundness-requires-generalization.rs 2://@ ignore-test (see rust-lang#114196) ``` </details> Of these, most are either **unimplemented**, or **spurious**, or **known-broken**. The outstanding one is `tests/ui/json/json-bom-plus-crlf-multifile-aux.rs` which I did not want to touch in *this* PR -- that aux file has load-bearing BOM and carriage returns and byte offset matters. I think those test files that require special encoding / BOM probably are better off as `run-make` tests. See rust-lang#139968 for that aux file. ### Review advice - Best reviewed commit-by-commit. - The directive name diverged from the most voted `//@ auxiliary` because I think that's easy to confuse with `//@ aux-{crate,dir}`. r? compiler

saethlin added the C-bug Category: This is a bug. label Aug 11, 2024

rustbot added the needs-triage This issue may need triage. Remove it if it has been sufficiently triaged. label Aug 11, 2024

saethlin mentioned this issue Aug 11, 2024

Enable debuginfo tests that have been "temporarily disabled" for the past 6 years #128913

Merged

saethlin added the WG-debugging Working group: Bad Rust debugging experiences label Aug 11, 2024

jieyouxu mentioned this issue Dec 23, 2024

Tracking Issue for debuginfo testsuite test infra woes #134682

Open

github-actions bot mentioned this issue Apr 18, 2025

Rollup of 8 pull requests #139992

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Stepping with "next" in a debugger skips past end-of-scope drops #128971

Stepping with "next" in a debugger skips past end-of-scope drops #128971

saethlin commented Aug 11, 2024

Stepping with "next" in a debugger skips past end-of-scope drops #128971

Stepping with "next" in a debugger skips past end-of-scope drops #128971

Comments

saethlin commented Aug 11, 2024