bump ui test crate #3008

oli-obk · 2023-08-03T10:25:04Z

The recommended way to run tests locally is ./miri bless -- -- --quiet, which will show

progress bars
the currently running tests (allowing you to see which ones are still running towards the end of the test suite)
the output of the currently running tests (if they are slow). This means slow running tests can output lines to stderr and the last line will be shown after the test name and updated every few hundred milliseconds.

As a side effect this PR also fixes #2998 and only builds dependencies if any tests actually need them (this means that with the next ui_test update we'll be able to merge all our test suites).

Also fixes #3052.

oli-obk · 2023-08-03T11:17:30Z

huh how did I manage to break mir-opt-level 4 tests?

RalfJung · 2023-08-03T11:27:38Z

./miri bless -- -- --quiet

That's a lot of dashes... why?^^ (Also it's ./miri test --bless now.)

(this means that with the next ui_test update we'll be able to merge all our test suites).

One reason we did the split was so that the pass tests can run even when cargo-miri is borked, so we might want to keep the split.

oli-obk · 2023-08-03T11:52:22Z

One reason we did the split was so that the pass tests can run even when cargo-miri is borked, so we might want to keep the split.

We can keep the directory split, but we can run all tests in one big runner (without accidentally allowing dep tests in the non-dep folder). Maybe I should have said "merge the test runners" not "merge the test suites"

oli-obk · 2023-08-03T11:52:52Z

You can then still filter by directory, and the dependencies won't get built at all if the dep folder is not part of the filter

oli-obk · 2023-08-03T11:54:29Z

That's a lot of dashes... why?

idk yet 😆 something with the test runner

oli-obk · 2023-08-03T11:54:54Z

huh how did I manage to break mir-opt-level 4 tests?

uh... those are broken for me locally on master, too

RalfJung · 2023-08-03T12:11:42Z

One reason we did the split was so that the pass tests can run even when cargo-miri is borked, so we might want to keep the split.

We can keep the directory split, but we can run all tests in one big runner (without accidentally allowing dep tests in the non-dep folder). Maybe I should have said "merge the test runners" not "merge the test suites"

Ah, fancy. I guess I'll see how that works. :)

RalfJung · 2023-08-03T12:13:26Z

idk yet laughing something with the test runner

One set of -- I get, it's distinguishing cargo test arguments from arguments passed to the runner.
But the 2nd -- should not be needed.

Maybe we should use an env var scheme so that we can have ./miri test --quiet --bless? Anyway I'll need to see locally what it looks like with and without quiet anyway, I don't get yet why --quiet should be recommended (and if it is recommended, why it shouldn't be the default).

oli-obk · 2023-08-03T12:16:44Z

why it shouldn't be the default).

because it means CI will not show anything (same reason we don't use --quiet in CI right now, you'd just get a bunch of dots)

RalfJung · 2023-08-03T12:26:31Z

Well CI can use a non-default config. But honestly I prefer the version without the dots. ;) So I'm fine with non-quiet being the default.

I gave it quick shot locally. Some first feedback:

Now building the test runner takes longer than building Miri. 😂 might be worth investigating whether all these dependencies carry their weight. E.g. do we truly need clap? It pulls in syn and that's really slow.
The live output, seeing which tests are still running, is really neat. :D
Its non-quiet version however can be improved: the last test that still runs is one of the earliest ones, so what happens is that output just stops and nothing moves any more -- because the "this is still running" indicator is scrolled out already, outside my viewport. The tests that are still running should always be at the bottom.
I got some errors in fail tests. And when I scroll back up, I just see the same stuff printed over and over again. Like, when I search for tests/fail/both_borrows/buggy_as_mut_slice.rs in my terminal output it has a dozen hits. The same does not happen for pass tests so it does look like something went wrong. I mean obviously something went wrong because I am getting errors:

Backtrace omitted. Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.
The application panicked (crashed).
Message:  could not execute "/home/r/src/rust/miri/target/debug/miri" "--error-format=json" "-Dwarnings" "-Dunused" "-Zui-testing" "--target" "x86_64-unknown-linux-gnu" "--extern" "getrandom=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libgetrandom-ea1755338c1459a8.rlib" "--extern" "getrandom=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libgetrandom-ea1755338c1459a8.rmeta" "--extern" "getrandom_1=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libgetrandom-964643f2b0add773.rlib" "--extern" "getrandom_1=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libgetrandom-964643f2b0add773.rmeta" "--extern" "libc=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/liblibc-f255c48072fae688.rlib" "--extern" "libc=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/liblibc-f255c48072fae688.rmeta" "--extern" "num_cpus=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libnum_cpus-5080ae5f42796c75.rlib" "--extern" "num_cpus=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libnum_cpus-5080ae5f42796c75.rmeta" "--extern" "rand=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/librand-fa9573179d9c7292.rlib" "--extern" "rand=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/librand-fa9573179d9c7292.rmeta" "--extern" "page_size=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libpage_size-c5bca0befe654b9c.rlib" "--extern" "page_size=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libpage_size-c5bca0befe654b9c.rmeta" "--extern" "tokio=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libtokio-c962404ad4b80287.rlib" "--extern" "tokio=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libtokio-c962404ad4b80287.rmeta" "--extern" "miri_test_deps=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/miri-test-deps" "-L" "/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps" "-L" "/home/r/src/rust/miri/target/ui/miri/debug/deps" "-L" "/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug" "--out-dir" "/home/r/src/rust/miri/target/ui" "tests/fail/validity/invalid_char.rs" "--edition" "2021": Too many open files (os error 24)
Location: /home/r/.cargo/registry/src/index.crates.io-6f17d22bba15001f/ui_test-0.13.0/src/lib.rs:620

Backtrace omitted. Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.
The application panicked (crashed).
Message:  could not execute "/home/r/src/rust/miri/target/debug/miri" "--error-format=json" "-Dwarnings" "-Dunused" "-Zui-testing" "--target" "x86_64-unknown-linux-gnu" "--extern" "getrandom=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libgetrandom-ea1755338c1459a8.rlib" "--extern" "getrandom=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libgetrandom-ea1755338c1459a8.rmeta" "--extern" "getrandom_1=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libgetrandom-964643f2b0add773.rlib" "--extern" "getrandom_1=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libgetrandom-964643f2b0add773.rmeta" "--extern" "libc=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/liblibc-f255c48072fae688.rlib" "--extern" "libc=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/liblibc-f255c48072fae688.rmeta" "--extern" "num_cpus=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libnum_cpus-5080ae5f42796c75.rlib" "--extern" "num_cpus=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libnum_cpus-5080ae5f42796c75.rmeta" "--extern" "rand=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/librand-fa9573179d9c7292.rlib" "--extern" "rand=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/librand-fa9573179d9c7292.rmeta" "--extern" "page_size=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libpage_size-c5bca0befe654b9c.rlib" "--extern" "page_size=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libpage_size-c5bca0befe654b9c.rmeta" "--extern" "tokio=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libtokio-c962404ad4b80287.rlib" "--extern" "tokio=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps/libtokio-c962404ad4b80287.rmeta" "--extern" "miri_test_deps=/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/miri-test-deps" "-L" "/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug/deps" "-L" "/home/r/src/rust/miri/target/ui/miri/debug/deps" "-L" "/home/r/src/rust/miri/target/ui/miri/x86_64-unknown-linux-gnu/debug" "--out-dir" "/home/r/src/rust/miri/target/ui" "tests/fail/unaligned_pointers/unaligned_ptr_zst.rs" "-Zmiri-disable-validation" "-Cdebug-assertions=no" "--edition" "2021": Too many open files (os error 24)
Location: /home/r/.cargo/registry/src/index.crates.io-6f17d22bba15001f/ui_test-0.13.0/src/lib.rs:620

I don't know where one test ends and the next one begins. "Too many open files" sounds like maybe it went a bit overboard with the concurrency or so?

The beginning of the output now looks like:

     Running unittests src/lib.rs (target/debug/deps/miri-73bfc5312ddef86b)

running 21 tests
test borrow_tracker::tree_borrows::perms::propagation_optimization_checks::access_transitions_progress_increasing ... ok
test borrow_tracker::tree_borrows::unimap::tests::consistency_small ... ok
test borrow_tracker::tree_borrows::perms::propagation_optimization_checks::all_transitions_idempotent ... ok
test borrow_tracker::tree_borrows::unimap::tests::consistency_large ... ok
test borrow_tracker::tree_borrows::unimap::tests::extend_to_length ... ok
test concurrency::range_object_map::tests::boundaries ... ok
test concurrency::range_object_map::tests::empty_map ... ok
test concurrency::range_object_map::tests::perfectly_overlapping ... ok
test borrow_tracker::tree_borrows::perms::propagation_optimization_checks::foreign_read_is_noop_after_write ... ok
test concurrency::range_object_map::tests::straddling ... ok
test eval::tests::windows_argv0_no_escape ... ok
test borrow_tracker::tree_borrows::tree::commutation_tests::all_read_accesses_commute ... ok
test concurrency::vector_clock::tests::test_equal ... ok
test range_map::tests::out_of_range_iter - should panic ... ok
test concurrency::vector_clock::tests::test_partial_order ... ok
test concurrency::range_object_map::tests::no_overlapping_inserts - should panic ... ok
test eval::tests::windows_argv0_panic_on_quote - should panic ... ok
test intptrcast::tests::test_align_addr ... ok
test range_map::tests::out_of_range_iter_mut - should panic ... ok
test range_map::tests::gaps ... ok
test range_map::tests::basic_insert ... ok

test result: ok. 21 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 0.00s

     Running tests/compiletest.rs (target/debug/deps/compiletest-cdf3132f0a720797)
## Running ui tests in tests/pass against miri for x86_64-unknown-linux-gnu
   Compiler: /home/r/src/rust/miri/target/debug/miri "--error-format=json" "-Dwarnings" "-Dunused" "-Zui-testing" "--target" "x86_64-unknown-linux-gnu" "--out-dir" OUT_DIR
  tests/pass/arrays.rs ... ok
  Building dependencies ... ok
  tests/pass/align.rs ... ok
  tests/pass/associated-const.rs (stack) ... ok
  tests/pass/align_repeat_into_well_aligned_array.rs ... ok
  tests/pass/assume_bug.rs (stack) ... ok
  tests/pass/align_repeat_into_packed_field.rs ... ok
  tests/pass/async-fn.rs ... ok
  tests/pass/align_strange_enum_discriminant_offset.rs ... ok
  tests/pass/available-parallelism-miri-num-cpus.rs ... ok
  tests/pass/atomic-compare-exchange-weak-never-fail.rs ... ok
[...]
  tests/pass/loop-break-value.rs ... ok
  tests/pass/loops.rs ... ok
test result: ok. 306 tests passed, 5 ignored, 0 filtered out

libtest prints the test <name> ... ok at the beginning of the line, ui-test adds some spaces, that looks inconsistent. Also the Compiler: lines is indented more than the others. And finally, the test result has an empty line above it with libtest. And if we are trying to make it super compatible anyway, N tests passed should become N passed, and there should be ; between the numbers instead of ,. (Skipping 0 failed sounds good though.)

oli-obk · 2023-08-03T12:32:43Z

"Too many open files" sounds like maybe it went a bit overboard with the concurrency or so?

uh. I have 96 cores and I don't get this issue... weird. Not sure how to investigate this :/

RalfJung · 2023-08-03T12:35:12Z

Also filtering doesn't seem to work any more: ./miri test fail runs all the tests. (That might be why mir-opt=4 fails, it relies on filtering.)

On my next attempt I am getting some even wilder errors:

  tests/fail/data_race/dealloc_write_race2.rs ... FAILED
  tests/fail/data_race/dealloc_write_race_stack.rs ... FAILEDThe application panicked (crashed).
Message:  `tests/fail/function_calls/arg_inplace_mutate.rs` not found
Location: /home/r/.cargo/registry/src/index.crates.io-6f17d22bba15001f/ui_test-0.13.0/src/status_emitter.rs:103

Backtrace omitted. Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.

That file definitely exists.

oli-obk · 2023-08-03T12:37:44Z

That file definitely exists.

that's an internal check, not about the file. Uh... what XD I want your desktop's thread switching capabilities, they seem to be great at fuzzing

oli-obk · 2023-08-03T12:38:50Z

Also filtering doesn't seem to work any more: ./miri test fail runs all the tests. (That might be why mir-opt=4 fails, it relies on filtering.)

oh yea, that's the -- issue lol. I'll drop clap in favor of manual parsing again. But note that we'll still have syn because we need serde for parsing the json output of tests

RalfJung · 2023-08-03T12:48:46Z

You might be running into clap-rs/clap#5055.

saethlin · 2023-08-03T12:54:54Z

uh. I have 96 cores and I don't get this issue... weird. Not sure how to investigate this :/

I don't know what OS yall are on, but MacOS has a default open files limit that is shockingly low.

oli-obk · 2023-08-03T13:00:04Z

I don't know what OS yall are on, but MacOS has a default open files limit that is shockingly low.

I really don't want to start figuring out the open file limit. Maybe a few quiet retries (bounded to a hard limit) with some sleeps will make it work out?

oli-obk · 2023-08-03T13:01:41Z

libtest prints the test <name> ... ok at the beginning of the line, ui-test adds some spaces, that looks inconsistent.

I replicated compiletest-rs behaviour. I'll switch to mirroring libtest

saethlin · 2023-08-03T13:35:10Z

I really don't want to start figuring out the open file limit.

I can look into this later, or we could invite someone else to look into it.

RalfJung · 2023-08-03T15:13:01Z

I'm on Linux. I've never seen "too many open files" before so retry to me files like it would paper over whatever is going on...

Also, Linux error code are awfully unspecific. Can I get it to tell me which operation raised the error? It might also appear when spawning too many threads too quickly or whatever.

oli-obk · 2023-08-03T15:15:28Z

it's probably during child process spawning and setting up pipes to it (from your panic line and message).

saethlin · 2023-08-03T16:19:08Z

Can I get it to tell me which operation raised the error?

I would/will figure this out by using strace. Probably something like strace -fk -otrace -eopen,openat ./miri test then the trace will be in a file called trace, and I'd look for the syscall that returned the error code in question. -f attaches to children, -k prints a backtrace. There's a flag to print only failing syscalls, but eh.

oli-obk · 2023-08-03T16:19:22Z

Everything but the spurious failures and

Now building the test runner takes longer than building Miri. 😂 might be worth investigating whether all these dependencies carry their weight. E.g. do we truly need clap? It pulls in syn and that's really slow.

has been fixed now. I got rid of clap, but ui_test still needs around 8-10s to build on my machine (vs 30 for miri).

RalfJung · 2023-08-03T18:02:29Z

Nice. :)

I am still seeing errors, though new ones:

The application panicked (crashed).
Message:  called `Result::unwrap()` on an `Err` value: "SendError(..)"
Location: /home/r/.cargo/registry/src/index.crates.io-6f17d22bba15001f/ui_test-0.14.0/src/status_emitter.rs:262

Backtrace omitted. Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.

Also there are some glitches while tests are running. "at " or so? Text that very briefly appears in a few places. After Ctrl-C only tiny parts of it are left

tests/pass/backtrace/backtrace-global-alloc.rs ⠒              at 
tests/pass/backtrace/backtrace-std.rs ⠒              at 
tests/pass/concurrency/sync.rs (stack) ⠐ 
tests/pass/panic/nested_panic_caught.rs ⠤    9:           0x3f56e1 - 
tests/pass/shims/time-with-isolation2.rs ⠄ 
tests/pass/concurrency/sync.rs (tree) ⠙

oli-obk · 2023-08-03T18:12:16Z

Also there are some glitches while tests are running. "at " or so? Text that very briefly appears in a few places. After Ctrl-C only tiny parts of it are left

there shouldn't be. These lines are from filtered backtraces from when miri interprets a panic. Any test that takes longer than 1s starts putting its last line from stderr after the test spinner

Ctrl+C may cause random stuff to be left over, yes. A full run should clean it up tho

oli-obk · 2023-08-07T10:42:11Z

Why are there so many threads anyway? Are you not using a thread pool?

Uh, this thread issue is in miri not in ui_test. ui_test has exactly n_cores + 3 threads, where the +3 is just some threads that are mostly asleep and don't do much.

oli-obk · 2023-08-07T10:42:45Z

You said this happens even without a ulimit for threads?

No, I have never seen this before. Only with a ulimit for processes

RalfJung · 2023-08-07T11:30:52Z

Uh, this thread issue is in miri not in ui_test. ui_test has exactly n_cores + 3 threads, where the +3 is just some threads that are mostly asleep and don't do much.

Yeah I've seen that. The reports I linked are this happening in rustc.

No, I have never seen this before. Only with a ulimit for processes

Threads are basically processes that share the address space on Linux, so I wouldn't be surprised if the process limit also includes threads.

oli-obk · 2023-08-07T11:45:25Z

Threads are basically processes that share the address space on Linux, so I wouldn't be surprised if the process limit also includes threads.

oh that makes sense. I guess it'll be fine then if ppl reduce the number of threads or increase their ulimit -u. Orthogonally we could start figuring out how to harden everything against such spurious failures. This will need some work in the ecosystem combined with fancy testing under very low ulimits.

RalfJung · 2023-08-07T15:25:25Z

If the default limits are fine then I wouldn't worry about it.

saethlin · 2023-08-07T16:08:53Z

Yes, I would only worry about defaults on common dev systems. So like Ubuntu or whatever common Linux with available_parallelism() of 64 or 128 should work. Same with MacOS, which often has lower resource limits but similarly lower core counts.

Manually lowering the resource limits can break all kinds of software, including systemd. So let's handle users being creative if anyone reports a problem. Same deal with a 448-vCPU VM.

oli-obk · 2023-08-08T07:53:50Z

Oh great I need a windows host to reproduce the i686-pc-windows-msvc failure.

oli-obk · 2023-08-08T14:02:23Z

Found a windows host. CI passes, locally tests pass with ulimit -Su 1000 at 96 threads

RalfJung

That looks very nice here. :) Please squash all the "bump ui-test" commits though.

RalfJung · 2023-08-11T21:08:09Z

tests/compiletest.rs

+        std::env::var("MIRI_TEST_THREADS").map_or_else(
+            |_| std::thread::available_parallelism().unwrap(),
+            |threads| NonZeroUsize::new(threads.parse().unwrap()).unwrap(),
+        ),


Please write this as a match, I find map_or_else much harder to read than a direct match

bors · 2023-08-15T12:59:08Z

☔ The latest upstream changes (presumably #2972) made this pull request unmergeable. Please resolve the merge conflicts.

…at wasm tests don't build dependencies

oli-obk · 2023-09-22T16:09:53Z

I squashed all the chained ui test bumps, but left the ones that had other changes in between. It also helps having individual commits to review the changes required by the bumps instead of skipping over too many versions at once

RalfJung

I tried this locally for a bit and it looks great, thanks a ton for all the work you put into this. :) Just got some nits here.

tests/compiletest.rs

RalfJung · 2023-09-25T09:08:10Z

tests/pass/backtrace/backtrace-api-v0.stdout

@@ -1,5 +1,5 @@
 $DIR/backtrace-api-v0.rs:24:14 (func_d)
 $DIR/backtrace-api-v0.rs:20:5 (func_c)
-$DIR/backtrace-api-v0.rs:9:5 (func_b)
+$DIR/backtrace-api-v0.rs:9:5 (func_b::<u8>)


Fascinating, why does this change?

I have absolutely no clue

(I think I changed some default path filters and they may affect this but I don't know why)

It seems like somehow this filter no longer applies?

miri/tests/pass/backtrace/backtrace-api-v0.rs

Line 1 in 7a6a812

//@normalize-stderr-test: "::<.*>" -> ""

Ah wait, that's a stderr filter. Was there a bug previously where those applied to stdout as well?

EDIT: Yes, there was. So this is a bugfix in ui_test. Nice. :) Does the ui-test test suite have a test to ensure normalization is for stdout/stderr only?

RalfJung · 2023-09-25T11:29:06Z

@bors r+

bors · 2023-09-25T11:29:08Z

📌 Commit 4a3b941 has been approved by RalfJung

It is now in the queue for this repository.

bors · 2023-09-25T11:30:14Z

⌛ Testing commit 4a3b941 with merge c004fc1...

bors · 2023-09-25T12:06:51Z

☀️ Test successful - checks-actions
Approved by: RalfJung
Pushing c004fc1 to master...

oli-obk force-pushed the ui_test_progress_bars branch from 63b608c to 3ee9f49 Compare August 3, 2023 12:09

oli-obk mentioned this pull request Aug 3, 2023

Remove clap dependency oli-obk/ui_test#127

Merged

oli-obk mentioned this pull request Aug 3, 2023

Make output more like libtest oli-obk/ui_test#128

Merged

This was referenced Aug 3, 2023

Some indicatif cleanups oli-obk/ui_test#129

Merged

Some performance related changes oli-obk/ui_test#130

Merged

RalfJung reviewed Aug 11, 2023

View reviewed changes

RalfJung mentioned this pull request Aug 15, 2023

windows: provide more accurate result for available_parallelism if cores > 64 rust-lang/rust#114856

Closed

RalfJung added the S-waiting-on-author Status: Waiting for the PR author to address review comments label Aug 21, 2023

RalfJung mentioned this pull request Aug 21, 2023

auto-detect no_std where possible #3000

Merged

oli-obk force-pushed the ui_test_progress_bars branch 2 times, most recently from 684c291 to 80dd9a2 Compare September 22, 2023 15:46

oli-obk added 5 commits September 22, 2023 16:08

Bump ui test crate

7514c69

Move fail tests that need dependencies into their own folder, so th…

f151e35

…at wasm tests don't build dependencies

Bump ui_test crate

263477b

Allow limiting the number of threads running in parallel

1bf0eb0

Bump ui_test crate

78b4c6c

oli-obk force-pushed the ui_test_progress_bars branch from 80dd9a2 to 78b4c6c Compare September 22, 2023 16:08

RalfJung reviewed Sep 25, 2023

View reviewed changes

oli-obk added 2 commits September 25, 2023 10:31

Document new env var

25d6df8

Explain single element vector

4a3b941

bors merged commit c004fc1 into rust-lang:master Sep 25, 2023

bump ui test crate #3008

bump ui test crate #3008

Conversation

oli-obk commented Aug 3, 2023 • edited by RalfJung Loading

oli-obk commented Aug 3, 2023

RalfJung commented Aug 3, 2023

oli-obk commented Aug 3, 2023

oli-obk commented Aug 3, 2023

oli-obk commented Aug 3, 2023

oli-obk commented Aug 3, 2023

RalfJung commented Aug 3, 2023

RalfJung commented Aug 3, 2023

oli-obk commented Aug 3, 2023

RalfJung commented Aug 3, 2023

oli-obk commented Aug 3, 2023

RalfJung commented Aug 3, 2023

oli-obk commented Aug 3, 2023

oli-obk commented Aug 3, 2023

RalfJung commented Aug 3, 2023

saethlin commented Aug 3, 2023

oli-obk commented Aug 3, 2023

oli-obk commented Aug 3, 2023

saethlin commented Aug 3, 2023

RalfJung commented Aug 3, 2023

oli-obk commented Aug 3, 2023

saethlin commented Aug 3, 2023

oli-obk commented Aug 3, 2023

RalfJung commented Aug 3, 2023

oli-obk commented Aug 3, 2023 • edited Loading

oli-obk commented Aug 7, 2023

oli-obk commented Aug 7, 2023

RalfJung commented Aug 7, 2023

oli-obk commented Aug 7, 2023

RalfJung commented Aug 7, 2023

saethlin commented Aug 7, 2023

oli-obk commented Aug 8, 2023

oli-obk commented Aug 8, 2023

RalfJung left a comment

Choose a reason for hiding this comment

RalfJung Aug 11, 2023

Choose a reason for hiding this comment

bors commented Aug 15, 2023

oli-obk commented Sep 22, 2023

RalfJung left a comment

Choose a reason for hiding this comment

RalfJung Sep 25, 2023

Choose a reason for hiding this comment

oli-obk Sep 25, 2023

Choose a reason for hiding this comment

oli-obk Sep 25, 2023

Choose a reason for hiding this comment

RalfJung Sep 25, 2023 • edited Loading

Choose a reason for hiding this comment

RalfJung Sep 25, 2023 • edited Loading

Choose a reason for hiding this comment

RalfJung commented Sep 25, 2023

bors commented Sep 25, 2023

bors commented Sep 25, 2023

bors commented Sep 25, 2023

oli-obk commented Aug 3, 2023 •

edited by RalfJung

Loading

oli-obk commented Aug 3, 2023 •

edited

Loading

RalfJung Sep 25, 2023 •

edited

Loading

RalfJung Sep 25, 2023 •

edited

Loading