Aot Analyzer shouldn't warn when passing a const string to RequiresDynamicCode applied to Types

[eerhardt]

If I have a class like:

    [RequiresDynamicCode(RequiresDynamicCodeMessage)]
    public class XmlDsigXsltTransform : Transform
    {
        internal const string RequiresDynamicCodeMessage = "XmlDsigXsltTransform uses XslCompiledTransform which requires dynamic code.";

I am getting a warning by the analyzer:

XmlDsigXsltTransform.cs(12,26): error IL3050: Using member 'System.Security.Cryptography.Xml.XmlDsigXsltTransform.RequiresDynamicCodeMessage' which has RequiresDynamicCodeAttribute' can break functionality when AOT compiling. XmlDsigXsltTransform uses XslCompiledTransform which equires dynamic code. [C:\git\runtime2\src\libraries\System.Security.Cryptography.Xml\src\System.Security.Cryptography.Xml.csproj::TargetFramework=net8.0]

Similarly, if I reference the const string from another RequiresDynamicCode attribute applied to a class:

    [RequiresDynamicCode(XmlDsigXsltTransform.RequiresDynamicCodeMessage)]
    public class SignedInfo : ICollection
    {

produces:

SignedInfo.cs(10,26): error IL3050: Using member  'System.Security.Cryptography.Xml.XmlDsigXsltTransform.RequiresDynamicCodeMessage' which has 'RequiresDynamicCodeAttribute' can break functionality when AOT compiling. XmlDsigXsltTransform uses XslCompiledTransform which requires dynamic code. [C:\git\runtime2\src\libraries\System.Security.Cryptography.Xml\src\System.Security.Cryptography.Xml.csproj::TargetFramework=net8.0]

But if I reference the const string from a RequiresDynamicCode attribute applied to a method, it doesn't warn:

        [RequiresDynamicCode(XmlDsigXsltTransform.RequiresDynamicCodeMessage)]
        internal void LoadXml(XmlElement value)
        {

I am declaring a RequiresDynamicCode attribute, the parameter I pass to the ctor of the attribute should also be considered as being in a "RequiresDynamicCode" scope - since by definition, I'm inside RequiresDynamicCode.

@vitek-karas @sbomer

Tagging subscribers to this area: @agocke, @sbomer, @vitek-karas
See info in area-owners.md if you want to be subscribed.

Issue Details

---

If I have a class like:

    [RequiresDynamicCode(RequiresDynamicCodeMessage)]
    public class XmlDsigXsltTransform : Transform
    {
        internal const string RequiresDynamicCodeMessage = "XmlDsigXsltTransform uses XslCompiledTransform which requires dynamic code.";

I am getting a warning by the analyzer:

XmlDsigXsltTransform.cs(12,26): error IL3050: Using member 'System.Security.Cryptography.Xml.XmlDsigXsltTransform.RequiresDynamicCodeMessage' which has RequiresDynamicCodeAttribute' can break functionality when AOT compiling. XmlDsigXsltTransform uses XslCompiledTransform which equires dynamic code. [C:\git\runtime2\src\libraries\System.Security.Cryptography.Xml\src\System.Security.Cryptography.Xml.csproj::TargetFramework=net8.0]

Similarly, if I reference the const string from another RequiresDynamicCode attribute applied to a class:

    [RequiresDynamicCode(XmlDsigXsltTransform.RequiresDynamicCodeMessage)]
    public class SignedInfo : ICollection
    {

produces:

SignedInfo.cs(10,26): error IL3050: Using member  'System.Security.Cryptography.Xml.XmlDsigXsltTransform.RequiresDynamicCodeMessage' which has 'RequiresDynamicCodeAttribute' can break functionality when AOT compiling. XmlDsigXsltTransform uses XslCompiledTransform which requires dynamic code. [C:\git\runtime2\src\libraries\System.Security.Cryptography.Xml\src\System.Security.Cryptography.Xml.csproj::TargetFramework=net8.0]

But if I reference the const string from a RequiresDynamicCode attribute applied to a method, it doesn't warn:

        [RequiresDynamicCode(XmlDsigXsltTransform.RequiresDynamicCodeMessage)]
        internal void LoadXml(XmlElement value)
        {

I am declaring a RequiresDynamicCode attribute, the parameter I pass to the ctor of the attribute should also be considered as being in a "RequiresDynamicCode" scope - since by definition, I'm inside RequiresDynamicCode.

@vitek-karas @sbomer

Author:	eerhardt
Assignees:	-
Labels:	area-Tools-ILLink
Milestone:	-

[agocke]

A simple solution here might be that Requires on types does not apply to consts in the type as though they were statics.

[agocke]

This area is somewhat problematic as Requires on consts actually doesn't work in the trimmer/AOT because consts don't exist in IL -- they're always inlined.

But they do exist for the source generator, and I've previously mentioned that it could be useful to place RUC on a particular enum value to indicate that that setting is not trim compatible.

Curious on what you think @sbomer @vitek-karas

[vitek-karas]

A simple solution here might be that Requires on types does not apply to consts in the type as though they were statics.

RUC on type applies to statics and .ctors - that's the rule. That's why the warning kicks in. The question here is if there's any situation where the .cctor would be triggered by an access to a const field.
If not - then we can safely ignore const fields in this case.

[agocke]

RUC on type applies to statics and .ctors - that's the rule. That's why the warning kicks in.

Right, and the warning behavior rests on the definition of consts, from my perspective. If we see const values as "functionally static" then the current behavior is consistent.

If we decide consts are a separate category, we can apply a different rule to them.

[vitek-karas]

I've previously mentioned that it could be useful to place RUC on a particular enum value to indicate that that setting is not trim compatible.

There's been several discussions on this - the problem is that from IL this is really hard to tell because the values are inlined by the compiler and they look like integers (not even typed as the enum)

[agocke]

Sure -- leaving aside the IL analysis, is it useful to consider only having support in the analyzer?

[vitek-karas]

Hmm - maybe in this case it would be beneficial.
I image also being able to "Obsolete" enum values would be nice as well.

[agocke]

Obsolete already works: https://sharplab.io/#v2:CYLg1APgAgTAjAWAFBQMwAJboMLoN7LpGYZQAs6AsgBQCU+hxTAbgIYBO6AHugLzoBRAHTYA3IyIBfZNKTIApgDsArgFtByAkiYBBADQT0AIQPbiAbQDyAIwDOAewA28gC7yAuoewygA