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

saethlin · 2024-08-10T02:35:35Z

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 #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 }: #128971

tests/debuginfo/function-arg-initialization.rs now has -Zmir-enable-passes=-SingleUseConsts; without that we initialize the const before the function prelude: #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: #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: #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.

rustbot · 2024-08-10T02:35:42Z

r? @Mark-Simulacrum

rustbot has assigned @Mark-Simulacrum.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

saethlin · 2024-08-10T12:28:50Z

@bors try

bors · 2024-08-10T12:29:59Z

⌛ Trying commit 1bc1383 with merge fd2f290...

…=<try> Enable debuginfo tests that were "temporarily disabled" for the past 6 years `tests/debuginfo/drop-locations.rs` No idea why we are setting `-O -Cno-prepopulate-passes`. Nowadays that turns on MIR opts, no idea what it did in 2018. Turning off optimizations makes just loc2 missing. That seems like a bug in debuginfo generation. Turning off all MIR opts and even adding a `drop(s)` on that line doesn't help. Cursed. Keeping this test ignored. `tests/debuginfo/option-like-enum.rs` Debuggers are better at Rust! Also one inexplicable change. No idea why we used to expect that. Deleted. `tests/debuginfo/function-arg-initialization.rs` SingleUseConsts produces incoherent gdb behavior. Seems like a bug. `tests/debuginfo/macro-stepping.rs` Also broken by SingleUseConsts. I hate tests like this, nobody uses a debugger on code that's just assigning consts to unused variables. Test should be rewritten with realistic code. `tests/debuginfo/basic-types-metadata.rs` Debuggers are better at Rust! Diverging functions don't have -> ! in their whatis output anymore? Not sure if that is a bug. try-job: test-various try-job: aarch64-apple try-job: x86_64-msvc try-job: x86_64-mingw try-job: i686-msvc try-job: i686-mingw try-job: armhf-gnu try-job: dist-various-1 try-job: aarch64-gnu

bors · 2024-08-10T13:03:21Z

💔 Test failed - checks-actions

saethlin · 2024-08-10T13:26:30Z

@bors try

…=<try> Enable debuginfo tests that were "temporarily disabled" for the past 6 years `tests/debuginfo/drop-locations.rs` No idea why we are setting `-O -Cno-prepopulate-passes`. Nowadays that turns on MIR opts, no idea what it did in 2018. Turning off optimizations makes just loc2 missing. That seems like a bug in debuginfo generation. Turning off all MIR opts and even adding a `drop(s)` on that line doesn't help. Cursed. Keeping this test ignored. `tests/debuginfo/option-like-enum.rs` Debuggers are better at Rust! Also one inexplicable change. No idea why we used to expect that. Deleted. `tests/debuginfo/function-arg-initialization.rs` SingleUseConsts produces incoherent gdb behavior. Seems like a bug. `tests/debuginfo/macro-stepping.rs` Also broken by SingleUseConsts. I hate tests like this, nobody uses a debugger on code that's just assigning consts to unused variables. Test should be rewritten with realistic code. `tests/debuginfo/basic-types-metadata.rs` Debuggers are better at Rust! Diverging functions don't have -> ! in their whatis output anymore? Not sure if that is a bug. try-job: aarch64-apple

bors · 2024-08-10T13:27:41Z

⌛ Trying commit b37d94f with merge ecaa186...

bors · 2024-08-10T14:00:16Z

💔 Test failed - checks-actions

saethlin · 2024-08-10T18:35:39Z

@bors try

…=<try> Enable debuginfo tests that were "temporarily disabled" for the past 6 years `tests/debuginfo/drop-locations.rs` No idea why we are setting `-O -Cno-prepopulate-passes`. Nowadays that turns on MIR opts, no idea what it did in 2018. Turning off optimizations makes just loc2 missing. That seems like a bug in debuginfo generation. Turning off all MIR opts and even adding a `drop(s)` on that line doesn't help. Cursed. Keeping this test ignored. `tests/debuginfo/option-like-enum.rs` Debuggers are better at Rust! Also one inexplicable change. No idea why we used to expect that. Deleted. `tests/debuginfo/function-arg-initialization.rs` SingleUseConsts produces incoherent gdb behavior. Seems like a bug. `tests/debuginfo/macro-stepping.rs` Also broken by SingleUseConsts. I hate tests like this, nobody uses a debugger on code that's just assigning consts to unused variables. Test should be rewritten with realistic code. `tests/debuginfo/basic-types-metadata.rs` Debuggers are better at Rust! Diverging functions don't have -> ! in their whatis output anymore? Not sure if that is a bug. --- More failures on apple! `tests/debuginfo/by-value-non-immediate-argument.rs` `tests/debuginfo/function-arg-initialization.rs` `tests/debuginfo/function-prologue-stepping-regular.rs` I think all of these indicate that lldb on aarch64-apple does not understand indirect by-value non-immediate arguments. Is that known? `tests/debuginfo/method-on-enum.rs` `tests/debuginfo/option-like-enum.rs` `tests/debuginfo/struct-in-enum.rs` These look like lldb on aarch64-apple can't print enums. Except that we have other tests where it looks like it can? It looks like all the other enum-related tests were using `min-lldb-version: 1800`. I don't know what effect that has. try-job: aarch64-apple

bors · 2024-08-10T18:36:49Z

⌛ Trying commit 5241f87 with merge 996a340...

saethlin · 2024-08-10T19:35:10Z

@bors try

bors · 2024-08-10T19:36:20Z

⌛ Trying commit c3cf903 with merge 29bb0d6...

…=<try> Enable debuginfo tests that were "temporarily disabled" for the past 6 years `tests/debuginfo/drop-locations.rs` No idea why we are setting `-O -Cno-prepopulate-passes`. Nowadays that turns on MIR opts, no idea what it did in 2018. Turning off optimizations makes just loc2 missing. That seems like a bug in debuginfo generation. Turning off all MIR opts and even adding a `drop(s)` on that line doesn't help. Cursed. Keeping this test ignored. `tests/debuginfo/option-like-enum.rs` Debuggers are better at Rust! Also one inexplicable change. No idea why we used to expect that. Deleted. `tests/debuginfo/function-arg-initialization.rs` SingleUseConsts produces incoherent gdb behavior. Seems like a bug. `tests/debuginfo/macro-stepping.rs` Also broken by SingleUseConsts. I hate tests like this, nobody uses a debugger on code that's just assigning consts to unused variables. Test should be rewritten with realistic code. `tests/debuginfo/basic-types-metadata.rs` Debuggers are better at Rust! Diverging functions don't have -> ! in their whatis output anymore? Not sure if that is a bug. --- More failures on apple! `tests/debuginfo/by-value-non-immediate-argument.rs` `tests/debuginfo/function-arg-initialization.rs` `tests/debuginfo/function-prologue-stepping-regular.rs` I think all of these indicate that lldb on aarch64-apple does not understand indirect by-value non-immediate arguments. Is that known? `tests/debuginfo/method-on-enum.rs` `tests/debuginfo/option-like-enum.rs` `tests/debuginfo/struct-in-enum.rs` These look like lldb on aarch64-apple can't print enums. Except that we have other tests where it looks like it can? It looks like all the other enum-related tests were using `min-lldb-version: 1800`. I don't know what effect that has. try-job: aarch64-apple

bors · 2024-08-10T20:11:06Z

💔 Test failed - checks-actions

saethlin · 2024-08-10T21:17:14Z

@bors try

…=<try> Enable debuginfo tests that were "temporarily disabled" for the past 6 years `tests/debuginfo/drop-locations.rs` No idea why we are setting `-O -Cno-prepopulate-passes`. Nowadays that turns on MIR opts, no idea what it did in 2018. Turning off optimizations makes just loc2 missing. That seems like a bug in debuginfo generation. Turning off all MIR opts and even adding a `drop(s)` on that line doesn't help. Cursed. Keeping this test ignored. `tests/debuginfo/option-like-enum.rs` Debuggers are better at Rust! Also one inexplicable change. No idea why we used to expect that. Deleted. `tests/debuginfo/function-arg-initialization.rs` SingleUseConsts produces incoherent gdb behavior. Seems like a bug. `tests/debuginfo/macro-stepping.rs` Also broken by SingleUseConsts. I hate tests like this, nobody uses a debugger on code that's just assigning consts to unused variables. Test should be rewritten with realistic code. `tests/debuginfo/basic-types-metadata.rs` Debuggers are better at Rust! Diverging functions don't have -> ! in their whatis output anymore? Not sure if that is a bug. --- More failures on apple! `tests/debuginfo/by-value-non-immediate-argument.rs` `tests/debuginfo/function-arg-initialization.rs` `tests/debuginfo/function-prologue-stepping-regular.rs` I think all of these indicate that lldb on aarch64-apple does not understand indirect by-value non-immediate arguments. Is that known? `tests/debuginfo/method-on-enum.rs` `tests/debuginfo/option-like-enum.rs` `tests/debuginfo/struct-in-enum.rs` lldb on aarch64-apple can't print enums until at least llvm 18. We use an old lldb on apple. try-job: aarch64-apple

saethlin · 2024-08-15T21:10:21Z

We have an incredible amount of rollup=never PRs in queue and I'm quite tired of this bouncing out.

I've added every job to this PR as a try-job except for those that are -dist, -2, -ext, or are run in PR CI.

@bors try

bors · 2024-08-15T21:11:34Z

⌛ Trying commit 23273e0 with merge 4a0fe6f...

…=<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

rust-log-analyzer · 2024-08-15T21:11:57Z

A job failed! Check out the build log: (web) (plain)

Click to see the possible cause of the failure (guessed by this bot)

i686-msvc
i686-mingw
x86_64-mingw
##[endgroup]
INFO:root:Job type: TryRunType(custom_jobs=['x86_64-gnu-llvm-17', 'x86_64-gnu-tools', 'aarch64-gnu', 'arm-android', 'armhf-gnu', 'i686-gnu', 'i686-gnu-nopt', 'mingw-check', 'test-various', 'x86_64-fuchsia', 'x86_64-rust-for-linux', 'x86_64-gnu', 'x86_64-gnu-stable', 'x86_64-gnu-aux', 'x86_64-gnu-debug', 'x86_64-gnu-llvm-18', 'x86_64-gnu-nopt', 'x86_64-apple-1', 'aarch64-apple', 'x86_64-msvc', 'i686-msvc', 'i686-mingw', 'x86_64-mingw'])
  File "/home/runner/work/rust/rust/src/ci/github-actions/calculate-job-matrix.py", line 195, in <module>
  File "/home/runner/work/rust/rust/src/ci/github-actions/calculate-job-matrix.py", line 195, in <module>
    jobs = calculate_jobs(run_type, data)
  File "/home/runner/work/rust/rust/src/ci/github-actions/calculate-job-matrix.py", line 124, in calculate_jobs
    raise Exception(
Exception: It is only possible to schedule up to 10 custom jobs, received 23 jobs
##[error]Process completed with exit code 1.

saethlin · 2024-08-15T21:15:01Z

Well that's deeply unfortunate. I'll have to do this in batches.

saethlin · 2024-08-15T21:16:11Z

@bors try

bors · 2024-08-15T21:17:21Z

⌛ Trying commit 23273e0 with merge 7237a68...

…=<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

saethlin · 2024-08-15T21:28:05Z

@bors try-

saethlin · 2024-08-15T21:29:14Z

@bors try

bors · 2024-08-15T21:30:24Z

⌛ Trying commit 23273e0 with merge dc0d722...

…=<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

bors · 2024-08-15T23:42:48Z

☀️ Try build successful - checks-actions
Build commit: dc0d722 (dc0d722c0d1b73f4031ac499936a3810c18b032b)

saethlin · 2024-08-15T23:50:55Z

@bors try

saethlin · 2024-08-16T00:05:49Z

@bors retry

bors · 2024-08-16T00:07:01Z

⌛ Trying commit 23273e0 with merge 1e46c35...

…=<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

bors · 2024-08-16T02:13:58Z

☀️ Try build successful - checks-actions
Build commit: 1e46c35 (1e46c3571acb5ef1e0c7b799aaceca252dd00778)

saethlin · 2024-08-16T02:19:10Z

@bors r=compiler-errors

bors · 2024-08-16T02:19:13Z

📌 Commit 23273e0 has been approved by compiler-errors

It is now in the queue for this repository.

bors · 2024-08-16T06:41:20Z

⌛ Testing commit 23273e0 with merge d626fbd...

bors · 2024-08-16T09:13:32Z

☀️ Test successful - checks-actions
Approved by: compiler-errors
Pushing d626fbd to master...

rust-timer · 2024-08-16T10:29:56Z

Finished benchmarking commit (d626fbd): comparison URL.

Overall result: no relevant changes - no action needed

@rustbot label: -perf-regression

Instruction count

This benchmark run did not return any relevant results for this metric.

Max RSS (memory usage)

Results (primary 4.0%)

This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.

	mean	range	count
Regressions ❌ (primary)	4.0%	[4.0%, 4.0%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	4.0%	[4.0%, 4.0%]	1

Cycles

This benchmark run did not return any relevant results for this metric.

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 751.18s -> 749.902s (-0.17%)
Artifact size: 339.11 MiB -> 339.17 MiB (0.02%)

rustbot assigned Mark-Simulacrum Aug 10, 2024

rustbot added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. label Aug 10, 2024

This comment has been minimized.

Sign in to view

jieyouxu added A-testsuite Area: The testsuite used to check the correctness of rustc A-debuginfo Area: Debugging information in compiled programs (DWARF, PDB, etc.) labels Aug 10, 2024

This comment has been minimized.

Sign in to view

bors added S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. and removed S-waiting-on-review Status: Awaiting review from the assignee but also interested parties. labels Aug 10, 2024

This comment has been minimized.

Sign in to view

saethlin mentioned this pull request Aug 16, 2024

Is anyone running our gdb-based Android debuginfo tests? #129139

Open

bors added S-waiting-on-bors Status: Waiting on bors to run and complete tests. Bors will change the label on completion. and removed S-waiting-on-author Status: This is awaiting some action (such as code changes or more information) from the author. labels Aug 16, 2024

bors added the merged-by-bors This PR was explicitly merged by bors. label Aug 16, 2024

bors merged commit d626fbd into rust-lang:master Aug 16, 2024
7 checks passed

rustbot added this to the 1.82.0 milestone Aug 16, 2024

saethlin deleted the unignore-debuginfo-tests branch August 25, 2024 17:16

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

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

Conversation

saethlin commented Aug 10, 2024 • edited Loading

rustbot commented Aug 10, 2024

This comment has been minimized.

saethlin commented Aug 10, 2024

bors commented Aug 10, 2024

This comment has been minimized.

bors commented Aug 10, 2024

saethlin commented Aug 10, 2024

bors commented Aug 10, 2024

This comment has been minimized.

bors commented Aug 10, 2024

saethlin commented Aug 10, 2024

bors commented Aug 10, 2024

This comment has been minimized.

This comment has been minimized.

saethlin commented Aug 10, 2024

bors commented Aug 10, 2024

This comment has been minimized.

bors commented Aug 10, 2024

saethlin commented Aug 10, 2024

saethlin commented Aug 15, 2024

bors commented Aug 15, 2024

rust-log-analyzer commented Aug 15, 2024

saethlin commented Aug 15, 2024

saethlin commented Aug 15, 2024

bors commented Aug 15, 2024

saethlin commented Aug 15, 2024

saethlin commented Aug 15, 2024

bors commented Aug 15, 2024

bors commented Aug 15, 2024

saethlin commented Aug 15, 2024

saethlin commented Aug 16, 2024

bors commented Aug 16, 2024

bors commented Aug 16, 2024

saethlin commented Aug 16, 2024

bors commented Aug 16, 2024

bors commented Aug 16, 2024

bors commented Aug 16, 2024

rust-timer commented Aug 16, 2024

Overall result: no relevant changes - no action needed

Instruction count

Max RSS (memory usage)

Cycles

Binary size

saethlin commented Aug 10, 2024 •

edited

Loading