[NativeAOT-LLVM] Merge oct 24 #2745
Draft
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
* Update dotnet10 feeds * Update hardcoded old TFM values * More hardcode updates and TFM checks * Updates and delete duplicate file in Bcl.Numerics * Linker stays on net9.0 * Fix .NET SDK max supported version * Update apicompat sfx apicompat baseline file * Update Microsoft.NET.WebAssembly.Threading APICompat settings * Fix DiagnosticSource build * Update SDK to RC1 for needed GetAlternateLookup API to be able to build System.Text.Json on net9.0. * Fix APICompat runtime regex pattern * Suppress Bcl.Memory compatibility error for dropped framework * Fix BinaryFormatter tests regex pattern for version subsitution * Update BinaryFormatterTests.cs * Pin SdkVersionForWorkloadTesting to the version in global.json * Treat net9 and net10 as current for now * Bump MicrosoftDotNetILCompilerVersion * Update BinaryFormatterTests.cs * Update System.Resources.Extensions * Update AutoImports as well * Fix ImporterTests.cs * Block failing SignalRClientTests * Update XmlSerialzation Schema Import tests. * Update Versions.props * Update CompatibilitySuppressions.xml for System.Numerics.Tensors * Don't set EnableDefaultCompileItems, it doesn't work since we set EnableDefaultItems=false higher up * Update regex in legacy BinaryFormatterTests.cs too * Hardcode BinaryFormatter test assembly version to 9.0 This allows us to not change the majority of the tests * Fix typo * Fix issue in Microsoft.NET.Workload.Mono.Toolchain.Current.Manifest.pkgproj * Replace ForceNet8Current in a few more places * Fix wasi workload * Fix TreatAsCurrent for 10.0 * Add net9 workload manifests * Update PackageId in icrosoft.NET.Runtime.WebAssembly.Templates.csproj * Update ApiCompatBaseline.NetCoreAppLatestStable.xml Changes from dotnet/runtime#107758 * Remove duplicate dotnet10 feeds from nuget.config * Update Versions.props * Fake SDK band version for now * Update Versions.props * Update the SDK band versions for real * Update the dotnet/sdk dependencies * Update emsdk to net10 workloads * Install a 9.0 runtime for workload tests too * Update the compat tasks * Merge in changes from net10-manifests * Tmp: Trying out @am11's branch. * Tmp2: Trying out @am11's branch. * Tmp3: Replace the last usage of main in clone action. * Update eng/pipelines/coreclr/templates/build-perf-bdn-app.yml Co-authored-by: Adeel Mujahid <[email protected]> * Update eng/pipelines/coreclr/templates/perf-job.yml Co-authored-by: Adeel Mujahid <[email protected]> * Update eng/pipelines/runtime-wasm-perf.yml Co-authored-by: Adeel Mujahid <[email protected]> --------- Co-authored-by: Larry Ewing <[email protected]> Co-authored-by: Jackson Schuster <[email protected]> Co-authored-by: Alexander Köplinger <[email protected]> Co-authored-by: Ilona Tomkowicz <[email protected]> Co-authored-by: Steve Molloy <[email protected]> Co-authored-by: Adeel Mujahid <[email protected]>
* Update dependencies from https://github.com/dotnet/arcade build 20240923.2 Microsoft.SourceBuild.Intermediate.arcade , Microsoft.DotNet.Arcade.Sdk , Microsoft.DotNet.Build.Tasks.Archives , Microsoft.DotNet.Build.Tasks.Feed , Microsoft.DotNet.Build.Tasks.Installers , Microsoft.DotNet.Build.Tasks.Packaging , Microsoft.DotNet.Build.Tasks.TargetFramework , Microsoft.DotNet.Build.Tasks.Templating , Microsoft.DotNet.Build.Tasks.Workloads , Microsoft.DotNet.CodeAnalysis , Microsoft.DotNet.GenAPI , Microsoft.DotNet.GenFacades , Microsoft.DotNet.Helix.Sdk , Microsoft.DotNet.PackageTesting , Microsoft.DotNet.RemoteExecutor , Microsoft.DotNet.SharedFramework.Sdk , Microsoft.DotNet.VersionTools.Tasks , Microsoft.DotNet.XliffTasks , Microsoft.DotNet.XUnitAssert , Microsoft.DotNet.XUnitConsoleRunner , Microsoft.DotNet.XUnitExtensions From Version 10.0.0-beta.24467.1 -> To Version 10.0.0-beta.24473.2 * Update dependencies from https://github.com/dotnet/arcade build 20240924.1 Microsoft.SourceBuild.Intermediate.arcade , Microsoft.DotNet.Arcade.Sdk , Microsoft.DotNet.Build.Tasks.Archives , Microsoft.DotNet.Build.Tasks.Feed , Microsoft.DotNet.Build.Tasks.Installers , Microsoft.DotNet.Build.Tasks.Packaging , Microsoft.DotNet.Build.Tasks.TargetFramework , Microsoft.DotNet.Build.Tasks.Templating , Microsoft.DotNet.Build.Tasks.Workloads , Microsoft.DotNet.CodeAnalysis , Microsoft.DotNet.GenAPI , Microsoft.DotNet.GenFacades , Microsoft.DotNet.Helix.Sdk , Microsoft.DotNet.PackageTesting , Microsoft.DotNet.RemoteExecutor , Microsoft.DotNet.SharedFramework.Sdk , Microsoft.DotNet.VersionTools.Tasks , Microsoft.DotNet.XliffTasks , Microsoft.DotNet.XUnitAssert , Microsoft.DotNet.XUnitConsoleRunner , Microsoft.DotNet.XUnitExtensions From Version 10.0.0-beta.24473.2 -> To Version 10.0.0-beta.24474.1 --------- Co-authored-by: dotnet-maestro[bot] <dotnet-maestro[bot]@users.noreply.github.com>
… dotnet/icu, dotnet/source-build-externals (#108183) * Update dependencies from https://github.com/dotnet/icu build 20240923.1 Microsoft.NETCore.Runtime.ICU.Transport From Version 10.0.0-alpha.1.24466.3 -> To Version 10.0.0-alpha.1.24473.1 * Update dependencies from https://github.com/dotnet/hotreload-utils build 20240923.1 Microsoft.DotNet.HotReload.Utils.Generator.BuildTool From Version 10.0.0-alpha.0.24466.1 -> To Version 10.0.0-alpha.0.24473.1 * Update dependencies from https://github.com/dotnet/cecil build 20240923.1 Microsoft.SourceBuild.Intermediate.cecil , Microsoft.DotNet.Cecil From Version 0.11.5-alpha.24467.1 -> To Version 0.11.5-alpha.24473.1 * Update dependencies from https://github.com/dotnet/source-build-externals build 20240923.2 Microsoft.SourceBuild.Intermediate.source-build-externals From Version 10.0.0-alpha.1.24466.1 -> To Version 10.0.0-alpha.1.24473.2 --------- Co-authored-by: dotnet-maestro[bot] <dotnet-maestro[bot]@users.noreply.github.com>
…923.3 (#108191) Microsoft.SourceBuild.Intermediate.roslyn , Microsoft.CodeAnalysis , Microsoft.CodeAnalysis.CSharp , Microsoft.Net.Compilers.Toolset From Version 4.12.0-3.24470.11 -> To Version 4.12.0-3.24473.3 Co-authored-by: dotnet-maestro[bot] <dotnet-maestro[bot]@users.noreply.github.com>
Fixes #107953.
* ARM64: Fix lsra for AdvSimd_LoadAndInsertScalar * Fix comment * Expand LoadAndInsertScalar testing * Only delayfree when register types match * fix formatting * Fix comment * Check that multiregister nodes use fp registers * Revert rmw assert
* Bring back Microsoft.BCL.Memory netstandard2.1 facade * Reorder TargetFrameworks Co-authored-by: Larry Ewing <[email protected]> --------- Co-authored-by: Larry Ewing <[email protected]>
* Fix handling of appending keywords to boolean schemas. * Add a few comments documenting boolean schemas.
* working * comments from PR * can always reshape to self * fixed tests * comments from PR * fixing tests
Match what we do for other android RIDs. The _Unwind_Backtrace support is only used when llvmonly mode is enabled, which is not something that we do for Android
* Added tests for server response DirectoryControls * Addressing platform inconsistencies * There are looser rules in OpenLDAP when an octet string exceeds the length of its containing sequence. * Windows 8.1 and below treats trailing 0x80 bytes differently to every other platform. * Added a test which covers an octet string containing invalid Unicode bytes. * Handle differing .NET Framework behaviour Also handle changes to the "a" format string: Winldap in Windows 10 is stricter when a SEQUENCE's contents overflow its length * Update SortResponseControlTests.cs
…925.5 (#108277) Microsoft.SourceBuild.Intermediate.arcade , Microsoft.DotNet.Arcade.Sdk , Microsoft.DotNet.Build.Tasks.Archives , Microsoft.DotNet.Build.Tasks.Feed , Microsoft.DotNet.Build.Tasks.Installers , Microsoft.DotNet.Build.Tasks.Packaging , Microsoft.DotNet.Build.Tasks.TargetFramework , Microsoft.DotNet.Build.Tasks.Templating , Microsoft.DotNet.Build.Tasks.Workloads , Microsoft.DotNet.CodeAnalysis , Microsoft.DotNet.GenAPI , Microsoft.DotNet.GenFacades , Microsoft.DotNet.Helix.Sdk , Microsoft.DotNet.PackageTesting , Microsoft.DotNet.RemoteExecutor , Microsoft.DotNet.SharedFramework.Sdk , Microsoft.DotNet.VersionTools.Tasks , Microsoft.DotNet.XliffTasks , Microsoft.DotNet.XUnitAssert , Microsoft.DotNet.XUnitConsoleRunner , Microsoft.DotNet.XUnitExtensions From Version 10.0.0-beta.24474.1 -> To Version 10.0.0-beta.24475.5 Co-authored-by: dotnet-maestro[bot] <dotnet-maestro[bot]@users.noreply.github.com>
…926.1 (#108280) Microsoft.SourceBuild.Intermediate.roslyn , Microsoft.CodeAnalysis , Microsoft.CodeAnalysis.CSharp , Microsoft.Net.Compilers.Toolset From Version 4.12.0-3.24473.3 -> To Version 4.12.0-3.24476.1 Co-authored-by: dotnet-maestro[bot] <dotnet-maestro[bot]@users.noreply.github.com>
…25.1 (#108279) Microsoft.SourceBuild.Intermediate.cecil , Microsoft.DotNet.Cecil From Version 0.11.5-alpha.24473.1 -> To Version 0.11.5-alpha.24475.1 Co-authored-by: dotnet-maestro[bot] <dotnet-maestro[bot]@users.noreply.github.com>
…24.1 (#108278) Microsoft.SourceBuild.Intermediate.emsdk , Microsoft.NET.Workload.Emscripten.Current.Manifest-10.0.100.Transport From Version 10.0.0-alpha.1.24468.9 -> To Version 10.0.0-alpha.1.24474.1 Co-authored-by: dotnet-maestro[bot] <dotnet-maestro[bot]@users.noreply.github.com>
…#108289) Work around dotnet/msbuild#9152 Possibly fixes dotnet/runtime#107936
We used to stackalloc a lot of data because the Target callback had to be an UnmanagedCallersOnly function pointer. But now it's just a delegate. So we can heap allocate the mock memory space and get rid of a bunch of incidental fixed statements We used to spell out the JSON contract descriptor in every test case. We dont' need to do that - most tests are not testing parsing. Hide all that stuff in MockMemorySpace.Builder * make MockMemorySpace.ReadContext a class Now that Target's reader callback is just a normal managed delegate, we don't have a lot of reason to make the testing ReadContext a ref struct. So simplify our resource usage and just store the descriptors and json as heap allocated arrays * remove uses of fixed in tests The ReadContext for the tests keeps a copy of the json payload, we don't need to use unmanaged pointers * make the MockMemorySpace.Builder a normal class we don't hold pointers to span data anymore * make MockMemorySpace.ReadContext private * begin hiding ContractDescriptorFill most of our tests that use a valid contract descriptor don't need to repeat the boilerplate to fill it in * start removing json and pointer data boilerplate * Don't fill the pointer data when there isn't any We previously did the wrong thing with TargetTestHelpers.MakeGlobalJson: it always filled in direct values. So technically filling in the pointer data in all the other tests is unnecessary. Add a second SetGlobals overload for the builder that takes a 4-tuple (either ulong? Value or uint IndirectIndex) and a second collection of actual values for the indices. * remove unused MockMemorySpace.Builder functions everything is using high level construction now * remove unneeded stackalloc * describe test addresses
reviously we used EmbeddedResource to collect the contract descriptor baselines into cdac-build-tool and we also embedded the C source template. But we only ever run cdac-build-tool during the runtime build - the files are in the same source tree. Just pass their paths when invoking the tool, don't embed them. Fixes #107958 * don't embed baselines into cdac-build-tool * don't embed the template source file in the tool
* Remove HelperMethodFrame from Delegate construction
Root the System.Runtime EventSource The System.Runtime EventSource (RuntimeEventSource), was unintentionally being garbage collected because it wasn't rooted. This caused runtime EventCounters to no longer be available after GC occurred. This was a regression from a recent change (dotnet/runtime#106014). That change accidentally converted the static field that was intended to the root the object into a property getter that returned a new instance each time it was called. This fix converts the property back to being initialized only once. This will fix #107919 once it is backported. Co-authored-by: Tarek Mahmoud Sayed <[email protected]>
…ser` templates (#108271) * Clean up icu tests (automatic mode). * Block the failing test, mark problems with issue links.
…8187) Enable and disable tests for hybrid globalization on Apple
Newer versions of Clang warn when you are doing enum comparisons of different enum types.
Remove a stray end-of-line space
The TLS helper expansion would produce `ADD(long, int)` trees like ``` ▌ STORE_LCL_VAR long V01 rat0 └──▌ ADD long ├──▌ ADD long │ ├──▌ LCL_VAR long V02 rat1 │ └──▌ CNS_INT int 72 $41 └──▌ CNS_INT long 192 ``` which is not legal IR. Fix that by inserting a cast. Also do some more aggressive folding to get rid of these casts in the normal cases, and to fold some constant arithmeticf in other cases.
…(#108873) It should be _extremely_ rare to need to disable a test during SuperPMI collection. However, one case was found, so introduce the ability to do so. Use `<SuperPmiCollectIncompatible>true</SuperPmiCollectIncompatible>` in a test project file. (which also requires `<RequiresProcessIsolation>true</RequiresProcessIsolation>`) Fixes #108215
This change is some preparatory refactoring for the randomized allocation sampling feature. We need to add more state onto allocation context but we don't want to do a breaking change of the GC interface. The new state only needs to be visible to the EE but we want it physically near the existing alloc context state for good cache locality. To accomplish this we created a new ee_alloc_context struct which contains an instance of gc_alloc_context within it. The new ee_alloc_context.combined_limit field should be used by fast allocation helpers to determine when to go down the slow path. Most of the time combined_limit has the same value as alloc_limit, but periodically we need to emit an allocation sampling event on an object that is somewhere in the middle of an AC. Using combined_limit rather than alloc_limit as the slow path trigger allows us to keep all the sampling event logic in the slow path. PR feedback Co-authored-by: Jan Kotas <[email protected]>
* Move IsDebuggerPresent to minipal, convert to QCall --------- Co-authored-by: Aaron Robinson <[email protected]>
…14.2 (#108884) Microsoft.SourceBuild.Intermediate.emsdk , Microsoft.NET.Workload.Emscripten.Current.Manifest-10.0.100.Transport From Version 10.0.0-alpha.1.24508.1 -> To Version 10.0.0-alpha.1.24514.2 Dependency coherency updates runtime.linux-arm64.Microsoft.NETCore.Runtime.JIT.Tools,runtime.linux-x64.Microsoft.NETCore.Runtime.JIT.Tools,runtime.linux-musl-arm64.Microsoft.NETCore.Runtime.JIT.Tools,runtime.linux-musl-x64.Microsoft.NETCore.Runtime.JIT.Tools,runtime.win-arm64.Microsoft.NETCore.Runtime.JIT.Tools,runtime.win-x64.Microsoft.NETCore.Runtime.JIT.Tools,runtime.osx-arm64.Microsoft.NETCore.Runtime.JIT.Tools,runtime.osx-x64.Microsoft.NETCore.Runtime.JIT.Tools,runtime.linux-arm64.Microsoft.NETCore.Runtime.Mono.LLVM.Sdk,runtime.linux-arm64.Microsoft.NETCore.Runtime.Mono.LLVM.Tools,runtime.linux-musl-arm64.Microsoft.NETCore.Runtime.Mono.LLVM.Sdk,runtime.linux-musl-arm64.Microsoft.NETCore.Runtime.Mono.LLVM.Tools,runtime.linux-x64.Microsoft.NETCore.Runtime.Mono.LLVM.Sdk,runtime.linux-x64.Microsoft.NETCore.Runtime.Mono.LLVM.Tools,runtime.linux-musl-x64.Microsoft.NETCore.Runtime.Mono.LLVM.Sdk,runtime.linux-musl-x64.Microsoft.NETCore.Runtime.Mono.LLVM.Tools,runtime.win-x64.Microsoft.NETCore.Runtime.Mono.LLVM.Sdk,runtime.win-x64.Microsoft.NETCore.Runtime.Mono.LLVM.Tools,runtime.osx-arm64.Microsoft.NETCore.Runtime.Mono.LLVM.Sdk,runtime.osx-arm64.Microsoft.NETCore.Runtime.Mono.LLVM.Tools,runtime.osx-x64.Microsoft.NETCore.Runtime.Mono.LLVM.Sdk,runtime.osx-x64.Microsoft.NETCore.Runtime.Mono.LLVM.Tools From Version 19.1.0-alpha.1.24501.1 -> To Version 19.1.0-alpha.1.24510.2 (parent: Microsoft.NET.Workload.Emscripten.Current.Manifest-10.0.100.Transport Co-authored-by: dotnet-maestro[bot] <dotnet-maestro[bot]@users.noreply.github.com>
…dbi interface (#108853)
This change is some preparatory refactoring for the randomized allocation sampling feature. We need to add more state onto allocation context but we don't want to do a breaking change of the GC interface. The new state only needs to be visible to the EE but we want it physically near the existing alloc context state for good cache locality. To accomplish this we created a new ee_alloc_context struct which contains an instance of gc_alloc_context within it. The new ee_alloc_context::combined_limit should be used by fast allocation helpers to determine when to go down the slow path. Most of the time combined_limit has the same value as alloc_limit, but periodically we need to emit an allocation sampling event on an object that is somewhere in the middle of an AC. Using combined_limit rather than alloc_limit as the slow path trigger allows us to keep all the sampling event logic in the slow path. * PR feedback Co-authored-by: Jan Kotas <[email protected]>
* Do not run tests with v8 scenario on CI. Leave code that supports running v8 scenario. * Remove console app, consolidate template base class, order inheritance, clean the code. * Fix build error: Keep all types of bundle options in `Common` dir. * Rename the param after merge with main. * Fix cast errors in providers. * Blazor and Browser bundle names got unified some time ago. * Fix running of browser published apps. * Fix the test transformed from console to browser. * Browser sdk does not allow appending RID. * Fix "item was already added" in build.
…04608) Context: dotnet/runtime#104356 (comment) Co-authored-by: Pavel Savara <[email protected]>
* Consolidate LocateNativeCompiler target * Set install_name on all Apple platforms --------- Co-authored-by: Jan Kotas <[email protected]>
* Generate vectorized left shift for Vector128<Byte> * Use lower ISA instead * Expand for larger size vector widths * jit format * Move transformation to gtNewSimdBinOp * Remove conditional, use assert similar to GT_RSZ case * Fix missing semicolon * Use fgMakeMultiUse instead of gtCloneExpr, re-use previous shift count calculation * Consolidate GT_LSH path into same path as GT_RSZ * Use TYP_INT instead of TYP_SHORT for GT_LSH * Also only use CORINFO_TYPE_INT --------- Co-authored-by: Alex Covington (Advanced Micro Devices <[email protected]> Co-authored-by: Tanner Gooding <[email protected]>
- Replace them with using the normal monitor helpers - Change CORINFO_HELP_GETSYNCFROMCLASSHANDLE to return an `object` instead of the underlying lock object, implementation of the JIT helper can be shared with other `TypeHandle` to `RuntimeType` scenarios - This will likely improve the performance of synchronized methods, but that is beside the point, since this feature is so rarely used
* Add -ANAME=xxx option to ilasm * Update IL SDK to add SynthesizeIlasmAssemblyName property * Update one test subfolder to use new property
# Conflicts: # .github/policies/resourceManagement.yml # eng/liveBuilds.targets # eng/pipelines/runtimelab.yml # src/coreclr/jit/compiler.h # src/coreclr/tools/aot/ILCompiler.Compiler/IL/ILImporter.Scanner.cs # src/tests/Common/Directory.Build.targets # src/tests/Directory.Build.targets
update docs 9->10 resolve conflicts Co-authored-by: egorbot <[email protected]> # Conflicts: # src/coreclr/jit/assertionprop.cpp
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR merges forward to October 16.
Update .net from 9 to 10
Disable IDynamicInterfaceCastable tests as they now use a thunk pool (and RhGetCommonStubAddress which we haven't implemented) dotnet/runtime#108235
Cherry pick the PR dotnet/runtime#109224 to avoid having to do another merge.
Remove if around call to
SetExpandedEarlyinimportCallas we assume virtual vtable calls are expanded into gtControlExprThe second merge is merging from NativeAOT-LLVM and essentially just removes a comment (after fixing a duplicate line removed in the last commit)
Last commit is reverting some formatting that VSCode decided it wanted.