-
Notifications
You must be signed in to change notification settings - Fork 12.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add regression test for #105501 #106264
Add regression test for #105501 #106264
Conversation
(rustbot has picked a reviewer for you, use r? to override) |
Could you move this test from |
The test was minified from the published `msf-ice:0.2.1` crate which failed in a crater run. A faulty compiler was triggering a `higher-ranked lifetime error`: > could not prove `[async block@...]: Send`
1c66744
to
42e7df9
Compare
Rebased and moved the test @rustbot ready |
@bors r+ |
☀️ Test successful - checks-actions |
Finished benchmarking commit (f6cc345): comparison URL. Overall result: ❌ regressions - ACTION NEEDEDNext Steps: If you can justify the regressions found in this perf run, please indicate this with @rustbot label: +perf-regression Instruction countThis is a highly reliable metric that was used to determine the overall result at the top of this comment.
Max RSS (memory usage)This benchmark run did not return any relevant results for this metric. CyclesResultsThis 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.
|
Of course, this is again unfortunate noise (which seems to have started appearing around #106210 (comment)). @rustbot label: +perf-regression-triaged |
This changes from the stdlib supported `ResumeTy`, and converting that to a `&mut Context<'_>` to directly use `&mut Context<'_>` in lowering. It pretty much reverts rust-lang#105977 and re-applies an updated version of rust-lang#105250. The PR is currently failing as it depends on `-Zdrop-tracking-mir` becoming the default, so that the compiler does not falsly believe the context is being held across await points,which would make async blocks `!Send` and `!UnwindSafe`. However it also still fails the testcase added in rust-lang#106264 for reasons I don’t understand.
Instead of using the stdlib supported `ResumeTy`, which is being converting to a `&mut Context<'_>` during the Generator MIR pass, this will use `&mut Context<'_>` directly in HIR lowering. It pretty much reverts rust-lang#105977 and re-applies an updated version of rust-lang#105250. This still fails the testcase added in rust-lang#106264 however, for reasons I don’t understand.
Instead of using the stdlib supported `ResumeTy`, which is being converting to a `&mut Context<'_>` during the Generator MIR pass, this will use `&mut Context<'_>` directly in HIR lowering. It pretty much reverts rust-lang#105977 and re-applies an updated version of rust-lang#105250. This still fails the testcase added in rust-lang#106264 however, for reasons I don’t understand.
Instead of using the stdlib supported `ResumeTy`, which is being converting to a `&mut Context<'_>` during the Generator MIR pass, this will use `&mut Context<'_>` directly in HIR lowering. It pretty much reverts rust-lang#105977 and re-applies an updated version of rust-lang#105250. This still fails the testcase added in rust-lang#106264 however, for reasons I don’t understand.
Instead of using the stdlib supported `ResumeTy`, which is being converting to a `&mut Context<'_>` during the Generator MIR pass, this will use `&mut Context<'_>` directly in HIR lowering. It pretty much reverts rust-lang#105977 and re-applies an updated version of rust-lang#105250. This still fails the testcase added in rust-lang#106264 however, for reasons I don’t understand.
Instead of using the stdlib supported `ResumeTy`, which is being converting to a `&mut Context<'_>` during the Generator MIR pass, this will use `&mut Context<'_>` directly in HIR lowering. It pretty much reverts rust-lang#105977 and re-applies an updated version of rust-lang#105250. This still fails the testcase added in rust-lang#106264 however, for reasons I don’t understand.
Instead of using the stdlib supported `ResumeTy`, which is being converting to a `&mut Context<'_>` during the Generator MIR pass, this will use `&mut Context<'_>` directly in HIR lowering. It pretty much reverts rust-lang#105977 and re-applies an updated version of rust-lang#105250. This still fails the testcase added in rust-lang#106264 however, for reasons I don’t understand.
Instead of using the stdlib supported `ResumeTy`, which is being converting to a `&mut Context<'_>` during the Generator MIR pass, this will use `&mut Context<'_>` directly in HIR lowering. It pretty much reverts rust-lang#105977 and re-applies an updated version of rust-lang#105250. This still fails the testcase added in rust-lang#106264 however, for reasons I don’t understand.
The test was minified from the published crate
msf-ice:0.2.1
which failed in a crater run.A faulty compiler was triggering a
higher-ranked lifetime error
:The testcase has some complexity, as it has a simplified subset of
futures::StreamExt
in it, but the error is only being triggered by a few layers of nesting. For example removing the noopthen
call would have been enough to make the error go away.