- 27 11月, 2020 4 次提交
-
-
由 Michal Strehovský 提交于
We never test managed tools in Debug configuration because we never build the runtime partition in the Debug configuration - we only build Checked. But managed tools don't know what Checked means because it's not an MSBuild concept. So we never run the tools with asserts turned on.
-
由 Jan Kotas 提交于
* Convert several FCalls to QCalls, plus misc cleanup Co-authored-by: NStephen Toub <stoub@microsoft.com>
-
由 Jan Kotas 提交于
Avoid code duplication and contribute to #45237 and #45032
-
由 yowl 提交于
-
- 26 11月, 2020 23 次提交
-
-
https://github.com/dotnet/xharness由 dotnet-maestro[bot] 提交于
[master] Update dependencies from dotnet/xharness
-
由 Peter Sollich 提交于
* Changes to allow us to use the mark list for foreground GCs. The key point is to clear the background GC mark bits for objects that the foreground GC found to be dead. The existing code walked all the dead objects individually and cleared their mark bits, but as it turns out, it is significantly cheaper to turn off the mark bits in bulk. * Address code review feedback - factor out common while-loop.
-
由 Peter Sollich 提交于
This change implements some primitive work stealing by executing GCScan::GcWeakPtrScanBySingleThread on the first thread that finishes GCScan::GcWeakPtrScan and sort_mark_list.
-
由 Adeel Mujahid 提交于
* Share unix getexepath() definition via src/native * Address CR feedback * Remove duplicate property * Inline variable with single usage * Fix include path in tests * Remove unixcoreruncommon static lib dependency * Merge coreruncommon.{cpp,h} into corerun.cpp * Include local headers before the first use * Remove Unix case from (Win32 only) corerun * Use realpath in getauxval case * Delete obsolete comment
-
由 Marek Safar 提交于
-
由 Stephen Toub 提交于
* Remove closure/delegate allocation from ConsoleLoggerProvider.CreateLogger * Update src/libraries/Microsoft.Extensions.Logging.Console/src/ConsoleLoggerProvider.cs Co-authored-by: NJan Kotas <jkotas@microsoft.com>
-
由 Khalid Abuhakmeh 提交于
Add AggressiveInline To BitConverter.ToXXX methods that take ReadOnlySpan<byte> Fixes #2106 Co-authored-by: NJan Kotas <jkotas@microsoft.com>
-
由 Ryan Lucia 提交于
* Ifdef out removed icalls on netcore * Fix format string for cross builds desc here is based on the pointer size of the target, whereas void* is based on the host * Re-enable -WError on CI for OSX and Wasm This got lost in the CMake transition * Improve ep_rt_thread_id_t handling. * Make types_str local to mono_fdhandle_insert * Fix OSX build error. * Align os thread id to uint64_t. * Update src/mono/mono/sgen/sgen-descriptor.c Co-authored-by: NAleksey Kliger (λgeek) <akliger@gmail.com> * Update src/mono/mono/sgen/sgen-descriptor.c Co-authored-by: NAleksey Kliger (λgeek) <akliger@gmail.com> * Update src/mono/mono/sgen/sgen-descriptor.c Co-authored-by: NAleksey Kliger (λgeek) <akliger@gmail.com> * Revert "Make types_str local to mono_fdhandle_insert" This reverts commit 04f56efcb8e9d33dca12f831f9c05591c649678c. * Put types_str under appropriate ifdef * Disable Werror on wasm Co-authored-by: NlateralusX <lateralusx.github@gmail.com> Co-authored-by: NAleksey Kliger (λgeek) <akliger@gmail.com> Co-authored-by: NAlexander Köplinger <alex.koeplinger@outlook.com>
-
由 Levi Broderick 提交于
- Use modern C# calli features to invoke allocator and ctor - Share arg validation code between CreateInstance and GetUninitializedObject - Improve exception message when CreateInstance fails - Lay foundation for future work in Activator Co-authored-by: NJan Kotas <jkotas@microsoft.com>
-
由 Justin Kotalik 提交于
-
由 Tom Deseyn 提交于
NetworkInterface.Linux: take into account physical link status for OperationalStatus and GetIsNetworkAvailable (#44867)
-
由 Alexander Köplinger 提交于
They're no longer needed after https://github.com/dotnet/runtime/pull/44976
-
https://github.com/dotnet/xharness由 dotnet-maestro[bot] 提交于
Microsoft.DotNet.XHarness.CLI , Microsoft.DotNet.XHarness.TestRunners.Xunit From Version 1.0.0-prerelease.20574.2 -> To Version 1.0.0-prerelease.20575.2 Co-authored-by: Ndotnet-maestro[bot] <dotnet-maestro[bot]@users.noreply.github.com>
-
由 Stephen Toub 提交于
* Reduce allocation from OptionsCache's concurrent dictionary This type is primarily used for getting and rarely mutated after startup; we don't need to pay for lots of lock objects to optimize for mutation. * Avoid closure/delegate allocations in `OptionsManager<T>.Value` * Update src/libraries/Microsoft.Extensions.Options/src/OptionsCache.cs Co-authored-by: NDavid Fowler <davidfowl@gmail.com> Co-authored-by: NDavid Fowler <davidfowl@gmail.com>
-
由 Stephen Toub 提交于
-
由 Marie Píchová 提交于
-
由 Buyaa 提交于
* Handle browser warnings * Apply feedback, revert updates handled with different PRs * Add misssing diagnotic id * Address multitargeted warnings * Apply feedback * Small type/comment updates * Apply more feedback * Use project settings instead adding Directory.Build.props * Annotate APIs injecting unsupported type through DI * Fix window support related warnings found with generic type parameter * Fix another browser warning found with generic type parameter bug fix * All public APIs of ConsoleLoggerExtensions are unsupported, so marking entire type as unsupported on browser * Try handle mono warnings * Revert mono related changes, it was mistake * Try handle browser warninga in mono * Apply feedback and fix new warnings caused from corelib changes * Review update * Apply feedback * Move platform specific section from HttpTelemetry * Revering unwanted changes * Remove redundant attributes * Exclude cross platform build with browser target * small updates * Annotate entire type DiagnosticCounter unsupported * Apply feedback, improve suppression comments * Remove Unsupported browser from TypeDescriptor.CreateInstance, TypeDescriptionProvider.CreateInstance and related updates
-
由 Andy Ayers 提交于
* Initial version of class profiling for PGO Add support to the jit and runtime so that PGO can determine the distribution of classes at virtual and indirect call sites. Use this information when jitting to enable guarded devirtualization, if there is a suitably likely class to guess for. Enable by setting: ``` COMPlus_TieredCompilation=1 COMPlus_TieredPGO=1 COMPlus_JitClassProfiling=1 COMPlus_JitEnableGuardedDevirtualization=1 ``` impact can be enhanced by also setting ``` COMPlus_TC_QuickJitForLoops=1 ``` to allow more methods to pass through Tier0.
-
由 Marie Píchová 提交于
-
由 Jeremy Koritzinsky 提交于
-
由 Alexander Köplinger 提交于
This unifies the build with the other platforms. Hooks up the versioning targets so the mono libraries get the correct version.
-
由 Marek Safar 提交于
-
由 Tammy Qiu 提交于
* change filtering system timezones from zone.tab as a task parameter Co-authored-by: NAnkit Jain <radical@gmail.com>
-
- 25 11月, 2020 13 次提交
-
-
由 dotnet-maestro[bot] 提交于
[master] Update dependencies from dotnet/icu dotnet/llvm-project dotnet/arcade dotnet/runtime-assets dotnet/xharness (#44459) [master] Update dependencies from dotnet/icu dotnet/llvm-project dotnet/arcade dotnet/runtime-assets dotnet/xharness - Merge branch 'master' into darc-master-f04f89fe-a712-45ce-96bf-a8d278fcda72 - Revert changes to dotnet sdk in global.json See https://github.com/dotnet/runtime/pull/45108#pullrequestreview-536374298
-
由 Stephen Toub 提交于
* Fix spelling of OverideEventProvider * Fix a few more "overrides" in comments
-
由 Anton Firszov 提交于
Implements the final version of the API proposal in #44025 except the [SupportedOSPlatform("windows10.0.2004")] bits
-
由 Stephen Toub 提交于
Based purely on code inspection, since I couldn't repro the failures happening in the lab, I believe what's happening is we're not outputting the right events if the connect ends up completing so fast that it's treated as a synchronous completion. The fix is to move the relevant tracing to be done when the work completes, regardless of the completion mode. I was able to simulate at least one set of failures by delaying the calling thread before it reaches a particular point, and this fixes that issue, so even if it's not fixing all known problems (hopefully it is), it's at least fixing some.
-
由 Levi Broderick 提交于
-
由 Tarek Mahmoud Sayed 提交于
-
由 Marek Safar 提交于
-
由 Stephen Toub 提交于
-
由 Sergey Andreenko 提交于
Create `GT_PUTARG_TYPE` when signature type does not match node type. Check in morph that this information has survived inlining and other phases between.
-
由 Alex Perovich 提交于
* Fix work item exit code for helix tests The helix work items are only supposed to return non-zero if they fail to report tests for some reason. * Fix if
-
由 Stephen Toub 提交于
And avoiding leaving the last invalid one for finalization.
-
由 Andy Ayers 提交于
Change the core data type for basic block weights from unsigned to float, to simplify overall calculations and allow for a wider dynamic range. Many changes are straightforward, but a few are worth noting: * LSRA needs a true max weight, so had to introduce infinity * I removed some of the overflow checking as floats naturally saturate. * The simple geometric loop weight scaling (*8 per loop nest level) leads to some very large counts in some tests (15 level loop nests). We may want to rethink this and scale less aggressively in deep nests. * Morph's use of the weighted ref counts for RCS_EARLY is nonstandard and the values are not actually weights, so I just added a cast back to unsigned. * Several places in the jit seem to try and compare or combine unweighted and weighted counts; I don't think this makes sense. But have left as is. * Lower, LIR, and Decompose were passing around weights but never using them. * I had to introduce a special new weight for the inline projection we do for the prejit root. These changes lead to small numbers of diffs, mostly places where small rounding changes have altered heuristics; notably: * cse weights * LSRA's initial take on whether a parameter should be enregistered Overall diff impact is a wash. There are almost no diffs without PGO/IBC data. Diffs are slightly more prominent in the Roslyn assemblies prejitted with some IBC. I've tried to keep the format of weights the same in dumps (in most places) and see minimal diffs in dumps too.
-
由 Tarek Mahmoud Sayed 提交于
-