Skip to content

Move runtime async method validation into initial binding #78310

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

Merged

Conversation

333fred
Copy link
Member

@333fred 333fred commented Apr 24, 2025

We now do method construction and validation for runtime async helpers up front in initial binding, rather than doing it in RuntimeAsyncRewriter. I've also renamed the APIs as per dotnet/runtime#114310 (comment) (though I haven't added ConfigureAwait support yet, that will be the next PR). We now validate:

  • The helpers come from System.Runtime.CompilerServices.AsyncHelpers, defined in corelib. This means that I now need a fairly extensive corelib mock to be able to compile. When we have a testing runtime that defines these helpers, we can remove the giant mock and use the real one.
  • We properly error when expected helpers aren't present.
  • We properly check to make sure that constraints are satisfied when doing generic substitution in one of the runtime helpers.
  • Runtime async is not turned on if the async method does not return Task, Task<T>, ValueTask, or ValueTask<T>.

Relates to test plan #75960

@333fred 333fred requested a review from a team as a code owner April 24, 2025 23:39
@ghost ghost added Area-Compilers untriaged Issues and PRs which have not yet been triaged by a lead labels Apr 24, 2025
@333fred 333fred requested review from RikkiGibson and jcouv April 24, 2025 23:39
// Keep in sync with VB's AssemblySymbol.RuntimeSupportsAsyncMethods
internal bool RuntimeSupportsAsyncMethods
=> RuntimeSupportsFeature(SpecialMember.System_Runtime_CompilerServices_RuntimeFeature__Async)
|| _overrideRuntimeSupportsAsyncMethods;
=> GetSpecialType(InternalSpecialType.System_Runtime_CompilerServices_AsyncHelpers) is { TypeKind: TypeKind.Class, IsStatic: true };
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Should we remove the runtime feature entry from the design doc?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What was the resolution?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It was removed.

&& !ReferenceEquals(methodReturn, GetSpecialType(InternalSpecialType.System_Threading_Tasks_ValueTask))
&& !ReferenceEquals(methodReturn, GetSpecialType(InternalSpecialType.System_Threading_Tasks_ValueTask_T)))
{
return false;
Copy link
Member

@jcouv jcouv Apr 25, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is there a corresponding update to the design doc? Or should we have a follow-up comment to make the void-returning method scenario work at some point? #Pending

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Need to update the design doc, the runtime-side is very clear that only Task/ValueTask methods can be runtime async.

@jcouv jcouv self-assigned this Apr 25, 2025
Copy link
Member

@jcouv jcouv left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Done with review pass (iteration 10)

@333fred
Copy link
Member Author

333fred commented Apr 29, 2025

@jcouv addressed feedback. @RikkiGibson for review please

Copy link
Member

@jcouv jcouv left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM Thanks (iteration 12). Consider updating the design doc (removing runtime feature flag) in same PR

@RikkiGibson RikkiGibson self-assigned this Apr 30, 2025
@RikkiGibson
Copy link
Member

It looks like the following tests need to be updated

Microsoft.CodeAnalysis.CSharp.UnitTests.MissingSpecialMember.AllWellKnownTypesBeforeCSharp7()
    in /_/src/Compilers/CSharp/Test/Symbol/Symbols/MissingSpecialMember.cs:line 926
Microsoft.CodeAnalysis.VisualBasic.UnitTests.Symbols.CorLibrary.CorTypes.PresentCorLib()
    in /_/src/Compilers/VisualBasic/Test/Symbol/SymbolsTests/CorLibrary/CorTypes.vb:line 77

@333fred
Copy link
Member Author

333fred commented Apr 30, 2025

LGTM Thanks (iteration 12). Consider updating the design doc (removing runtime feature flag) in same PR

The doc is maintained in main.

break;

default:
Debug.Fail($"Unexpected RuntimeAsyncAwaitMethod: {RuntimeAsyncAwaitMethod.Name}");
Copy link
Member

@RikkiGibson RikkiGibson Apr 30, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Checking my understanding: it looks like we expect this condition to never occur. i.e. compiler will never produce a BoundAwaitableInfo whose RuntimeAsyncAwaitMethod.Name is not one of the above cases.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yep, that's correct.

Copy link
Member

@RikkiGibson RikkiGibson left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM with some minor comments/suggestions

return false;
}

Debug.Assert((runtimeAwaitHelper is { Arity: 1 } && maybeResultType is { }) || runtimeAwaitHelper.TypeParameters.Length == 0);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit: maybe slightly more uniform/compact check would be:

Suggested change
Debug.Assert((runtimeAwaitHelper is { Arity: 1 } && maybeResultType is { }) || runtimeAwaitHelper.TypeParameters.Length == 0);
Debug.Assert(runtimeAwaitHelper.Arity == (maybeResultType is null ? 0 : 1));

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

TypeWithAnnotations also has its own HasType which distinguishes default values from "meaningful" values, which could be used instead of System.Nullable, but, there's no deep need to do that.

@@ -286,7 +302,90 @@ private bool GetAwaitableExpressionInfo(
TypeSymbol awaiterType = getAwaiter.Type!;
return GetIsCompletedProperty(awaiterType, node, expression.Type!, diagnostics, out isCompleted)
&& AwaiterImplementsINotifyCompletion(awaiterType, node, diagnostics)
&& GetGetResultMethod(getAwaiter, node, expression.Type!, diagnostics, out getResult, out getAwaiterGetResultCall);
&& GetGetResultMethod(getAwaiter, node, expression.Type!, diagnostics, out getResult, out getAwaiterGetResultCall)
&& (!isRuntimeAsyncEnabled || getRuntimeAwaitAwaiter(awaiterType, out runtimeAsyncAwaitMethod, expression.Syntax, diagnostics));
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Checking my understanding on the sequence of checks here.

  • if runtime async, and got runtime await helper, use that
  • otherwise, need an "ordinary" GetAwaiterMethod to proceed--if we didn't get one just return false.
  • then check for IsCompleted, GetResult, ..
  • finally if runtime async enabled, then get a RuntimeAwaitAwaiter?

Is the idea that in the 'RuntimeAwaitAwaiter' case, we still need to use the type's GetAwaiter method etc? That seems to make sense.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is the idea that in the 'RuntimeAwaitAwaiter' case, we still need to use the type's GetAwaiter method etc?

Precisely. This the lowering mechanism in https://github.com/dotnet/roslyn/blob/main/docs/compilers/CSharp/Runtime%20Async%20Design.md#await-any-other-type, as opposed to the rest of the document which talks about await Task

}

var methodReturn = method.ReturnType.OriginalDefinition;
if (!ReferenceEquals(methodReturn, GetSpecialType(InternalSpecialType.System_Threading_Tasks_Task))
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It would be nice if there a way to say something like methodReturn.InternalSpecialType is InternalSpecialType.Task or Task_T or ValueTask or ValueTask_T. No need to try and dig up a way to do that though.

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actually, I think we can, since this is special types, not well-known types.

@333fred
Copy link
Member Author

333fred commented Apr 30, 2025

@jcouv @RikkiGibson for another look, a few small changes in the last commit.

@333fred
Copy link
Member Author

333fred commented May 2, 2025

@jcouv, since the last 2 commits were only minor changes, I'm going to go ahead and merge this now; if you have any comments on these last 2 bits, I can address them in the next PR.

@333fred 333fred merged commit e272201 into dotnet:features/runtime-async May 2, 2025
24 checks passed
@333fred 333fred deleted the initial-binding-validation branch May 2, 2025 17:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Area-Compilers Feature - Runtime Async untriaged Issues and PRs which have not yet been triaged by a lead
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants