Jan Kotas [Sat, 31 Mar 2018 07:30:53 +0000 (00:30 -0700)]
Make AsSpan(this string) ForceInline to be on par with AsSpan(this T[]) (#17368)
Jan Kotas [Sat, 31 Mar 2018 07:29:33 +0000 (00:29 -0700)]
Delete dead code to support OSes prior to Windows 7 (#17367)
Brian Sullivan [Sat, 31 Mar 2018 06:50:07 +0000 (23:50 -0700)]
Merge pull request #17329 from briansull/struct-gscheck-fix
Reject struct promotion of parameters when -GS checks are enabled
David Mason [Sat, 31 Mar 2018 06:35:38 +0000 (23:35 -0700)]
fix for HasCriticalFinalizer in local gc (#17369)
Bruce Forstall [Sat, 31 Mar 2018 04:45:47 +0000 (21:45 -0700)]
Merge pull request #17319 from janvorli/disable-gc-stress-incompatible-test
Mark GCCollect test as GC stress excluded
Egor Chesakov [Sat, 31 Mar 2018 03:43:28 +0000 (20:43 -0700)]
Fix#16827 Stack probing for Linux (#17360)
* Add stack probing algorithm that moves sp while touching pages under _TARGET_UNIX_ only
dotnet-maestro-bot [Sat, 31 Mar 2018 03:42:27 +0000 (22:42 -0500)]
Update BuildTools, CoreClr, CoreFx to preview3-02631-01, preview3-26331-01, preview3-26331-02, respectively (#17346)
Konstantin Baladurin [Fri, 30 Mar 2018 07:18:04 +0000 (10:18 +0300)]
PEImageLayout: clear instruction cache after relocations
It fixes crashes on arm when using AOT images.
Eric StJohn [Sat, 31 Mar 2018 01:37:06 +0000 (18:37 -0700)]
Fix ilasm COR header (#17357)
Ilasm was emitting the wrong metadata version in the COR header.
This created a difference when examining the metadata diff between
desktop and core.
Brian Robbins [Sat, 31 Mar 2018 00:59:47 +0000 (17:59 -0700)]
Only allow the rundown thread to write events during rundown to avoid corruption of the trace file. (#17358)
Brian Sullivan [Thu, 29 Mar 2018 22:05:55 +0000 (15:05 -0700)]
Reject struct promotion of parameters when -GS checks are enabled
as we could introduce shadow copies of them.
Add new test case: GitHub_17329.cs
sergey ignatov [Sat, 31 Mar 2018 00:27:26 +0000 (03:27 +0300)]
[arm32] Fixed RBM_PROFILER_* (#17291)
* [arm32] Fixed RBM_PROFILER_*
* Changed trash registers to RBM_NONE
Jan Vorlicek [Fri, 30 Mar 2018 23:46:03 +0000 (01:46 +0200)]
Fix remote unwind (#17351)
The change that removed dependency on the external libunwind has broken
remote unwinding. I have not realized that the unw_create_addr_space
and other remote unwinding functions are just dummies.
This change adds the remote unwind related sources to the libunwind
compilation, enables remote unwinding support in the remote-unwind.cpp
and also does some magic in the CMakeLists.txt to fix issues caused
by the fact that the local and remote unwinding code was not expected
to be compiled into the same binary. There was one general and one
arm related function whose names needed to be different for local and
remote unwind code. And there was one function that was the same for
both local and remote case and so I had to ensure that it gets compiled
in just once.
Fortunately, all of these could still be achieved without changing the
libunwind sources.
Sergey Andreenko [Fri, 30 Mar 2018 23:14:25 +0000 (16:14 -0700)]
replace unreached with assert(false) (#17359)
Sergey Andreenko [Fri, 30 Mar 2018 00:11:02 +0000 (17:11 -0700)]
add an additional assert for the future
Sergey Andreenko [Thu, 29 Mar 2018 23:53:00 +0000 (16:53 -0700)]
fix the bug
Sergey Andreenko [Thu, 29 Mar 2018 23:33:52 +0000 (16:33 -0700)]
add repro
Sergey Andreenko [Thu, 29 Mar 2018 22:17:06 +0000 (15:17 -0700)]
fix ifdefs comments
Carol Eidt [Fri, 30 Mar 2018 18:36:04 +0000 (11:36 -0700)]
Merge pull request #17314 from fiigii/fixgctests
Fix GCStress failures from hardware intrinsic tests
dotnet-maestro-bot [Fri, 30 Mar 2018 15:40:27 +0000 (10:40 -0500)]
Update CoreClr, CoreFx to preview3-26330-06, preview3-26330-06, respectively (#17344)
Ahson Khan [Fri, 30 Mar 2018 15:27:25 +0000 (08:27 -0700)]
Change OwnedMemory to MemoryManager and add an IMemoryOwner. (#17340)
* Change OwnedMemory to MemoryManager and add an IMemoryOwner.
* Fix comments.
* Reset start and length if TryGetMemoryManager returns false.
* Reset start and length if TryGetMemoryManager returns false [actually].
* Re-order MemoryHandle ctor parameters and rename TOwner to TManager.
* Fix formatting, remove unused code, and fix impl of Pin()
dotnet-maestro-bot [Fri, 30 Mar 2018 07:53:45 +0000 (02:53 -0500)]
Update CoreClr to preview3-26330-01 (#17328)
Wes Haggard [Thu, 29 Mar 2018 23:33:04 +0000 (16:33 -0700)]
Merge pull request #17327 from weshaggard/FixIdentityPackageBuild
Fix identity package filtering
Fei Peng [Thu, 29 Mar 2018 22:39:04 +0000 (15:39 -0700)]
Fix GCStress failures
Bruce Forstall [Thu, 29 Mar 2018 22:29:07 +0000 (15:29 -0700)]
Merge pull request #17322 from BruceForstall/FixTestBuild
Add back hosting to split builds
Carol Eidt [Thu, 29 Mar 2018 21:57:03 +0000 (14:57 -0700)]
Merge pull request #17211 from mikedn/dyn-hwinfo
Remove HWIntrinsicFlag bitwise OR operator
Wes Haggard [Thu, 29 Mar 2018 17:48:52 +0000 (10:48 -0700)]
Fix identity package filtering
We only want to filter out identity package builds on
official builds that aren't windows x64, to avoid building
and publishing them more then once in an official build.
We also need to account for source-build which sets the
OfficialBuildId property to replicate the official builds,
so we also look at the DotNetBuildOffline property to
determine that we should still build the identity package.
Luqun Lou [Thu, 29 Mar 2018 20:59:18 +0000 (13:59 -0700)]
Fix for incorrectly handle invalid UTF8 characters issue (#17302)
There are some behavior difference between C/C++ UTF8 encoder/decoder and Encoding.UTF8 as mentioned by #16786.
Daniel Harvey [Thu, 29 Mar 2018 18:59:19 +0000 (11:59 -0700)]
Fix AssemblyLoadContext.Unloading and ProcessExit for Windows Docker containers (#17265)
On Windows, we need to shutdown the EE when receiving a CTRL_CLOSE_EVENT to we run ProcessExit handlers and other code that relies on ProcessExit working (like AssemblyLoadContext.Unloading). One way we receive this event is when there's a running process in a docker container that has the stop command run against it.
Divino César [Thu, 29 Mar 2018 18:01:36 +0000 (11:01 -0700)]
Merge pull request #17304 from dotnet/RemovingObsoleteCleanupBuilds
This BuildTask has been deprecated in favor of maintenance jobs.
dotnet-maestro-bot [Thu, 29 Mar 2018 17:32:32 +0000 (12:32 -0500)]
Update CoreFx to preview3-26329-05 (#17318)
Bruce Forstall [Thu, 29 Mar 2018 17:20:23 +0000 (10:20 -0700)]
Add back hosting to split builds
Ben Adams [Thu, 29 Mar 2018 17:06:42 +0000 (18:06 +0100)]
Dict cache default comparer for object types (#17285)
* Dict cache default comparer for object types
* Improves
* Fix EventRegistrationToken Equals
Maryam Ariyan [Thu, 29 Mar 2018 16:43:58 +0000 (09:43 -0700)]
StringBuilder.Equals: value to span (#17311)
Jan Vorlicek [Thu, 29 Mar 2018 14:25:56 +0000 (16:25 +0200)]
Mark GCCollect test as GC stress excluded
This test is incompatible with GC stress. The tests compares results of
GC.GetTotalMemory before and after GC.Collect and fails if the size
before the collection is smaller than after the collection. But GC
stress breaks this assumption.
Sergey Andreenko [Thu, 29 Mar 2018 05:44:32 +0000 (22:44 -0700)]
change arm maxIter to 20 (#17308)
dotnet-maestro-bot [Thu, 29 Mar 2018 03:55:21 +0000 (22:55 -0500)]
Update BuildTools, CoreClr, CoreFx to preview3-02629-02, preview3-26329-01, preview3-26329-01, respectively (#17309)
Steve MacLean [Thu, 29 Mar 2018 00:40:11 +0000 (20:40 -0400)]
[Arm64] Correct cross/arm64/tryrun.cmake (#17293)
dotnet-maestro-bot [Thu, 29 Mar 2018 00:05:58 +0000 (19:05 -0500)]
Update BuildTools, CoreClr to preview3-02628-01, preview3-26328-08, respectively (#17299)
Wes Haggard [Wed, 28 Mar 2018 23:55:49 +0000 (16:55 -0700)]
Merge pull request #17280 from weshaggard/SyncInitToolsWithSourceBuild
Sync init-tools scripts with what is in source-build
Ahson Khan [Wed, 28 Mar 2018 23:33:56 +0000 (16:33 -0700)]
Use ROSpan.IndexOf as the workhorse for string.IndexOf (#17284)
* Use ROSpan.IndexOf as the workhorse for string.IndexOf
* Make changes to Span.IndexOf to follow what string.IndexOf did.
* Address PR feedback.
* Use Unsafe.Read instead of Unsafe.ReadUnaligned.
* Remove special casing for count == 0
* Fix up debug assert to use vector count instead of intptr.size
* Use size of Vector<ushort> instead of Vector<byte>.Count
Ben Adams [Wed, 28 Mar 2018 23:11:38 +0000 (00:11 +0100)]
Fix Dictionary CopyTo regression (#17300)
Cesar Soares Lucas [Wed, 28 Mar 2018 21:44:57 +0000 (14:44 -0700)]
This BuildTask has been deprecated in favor of maintenance jobs.
Bruce Forstall [Wed, 28 Mar 2018 21:07:26 +0000 (14:07 -0700)]
Merge pull request #17303 from BruceForstall/EnableUbuntuArmGCStressJobsMaster
Enable Linux arm32 GCStress jobs
Noah Falk [Wed, 28 Mar 2018 20:56:24 +0000 (13:56 -0700)]
Removing 'EXPERIMENTAL' from tiered compilation env var (#17283)
Things have progressed far enough that its time to use a friendlier name. The feature still still has performance aspects that need to be investigated and improved, but I don't want to scare people off simply because it isn't as fast as it could be.
This also updates to use a newer CoreFX version for JitBench since that appeared to be broken, and updated some comments and usage of the tieredcompilation variable.
Bruce Forstall [Tue, 27 Mar 2018 19:22:38 +0000 (12:22 -0700)]
Enable Linux arm32 GCStress jobs
Non-PR jobs (cron and push) are disabled for now.
Russ Keldorph [Wed, 28 Mar 2018 19:34:59 +0000 (12:34 -0700)]
Unique-ify name of TestList.json to avoid collisions
Fixes #16840
Bruce Forstall [Wed, 28 Mar 2018 19:50:44 +0000 (12:50 -0700)]
Merge pull request #17090 from debayang/corefx_tests
Fix run-corefx-tests.py for Arm64 Linux cross build
Jan Vorlicek [Wed, 28 Mar 2018 18:16:57 +0000 (20:16 +0200)]
Fix some more unwinding in R2R ARM (#17292)
There was one more change needed to make the unwinding work properly.
Pushes in some prologs were missing the unwinder annotation.
The fix is to use PROLOG_PUSH for them.
To make things in this file consistent, I've also replaced pops in
epilogs with EPILOG_POP macro and vpush / vpop with PROLOG_VPUSH /
PROLOG_VPOP, although it is not functionally necessary.
With these changes, all the EH related issues are gone.
Maryam Ariyan [Wed, 28 Mar 2018 18:09:42 +0000 (11:09 -0700)]
Nit changes in API diff between .NET Core 2.0 and .NET Core 2.1 (#17288)
* 'value' to 'other': bool System.IEquatable<System.UIntPtr>.Equals(UIntPtr value);
* 'value' to 'other': bool System.IEquatable<System.IntPtr>.Equals(IntPtr value);
Wes Haggard [Tue, 27 Mar 2018 23:56:11 +0000 (16:56 -0700)]
Sync init-tools scripts with what is in source-build
dotnet-maestro-bot [Wed, 28 Mar 2018 18:02:57 +0000 (13:02 -0500)]
Update BuildTools, CoreClr, CoreFx to preview3-02627-03, preview3-26328-07, preview3-26328-06, respectively (#17248)
Carol Eidt [Wed, 28 Mar 2018 18:01:03 +0000 (11:01 -0700)]
Merge pull request #17126 from CarolEidt/Fix545500
ARM: Fix reg resolution for doubles
Egor Chesakov [Tue, 27 Mar 2018 21:12:56 +0000 (14:12 -0700)]
Add crossbitness support to ClrJit:
* Add FEATURE_CROSSBITNESS in crosscomponents.cmake
* Exclude mscordaccore mscordbi sos from CLR_CROSS_COMPONENTS_LIST when FEATURE_CROSSBITNESS is defined in crosscomponents.cmake
* Introduce target_size_t in src/jit/target.h
* Use size_t value in genMov32RelocatableImmediate in src/jit/codegen.h src/jit/codegencommon.cpp
* Fix definition/declaration inconsistency for emitter::emitIns_R_I in emitarm.cpp
* Zero HiVal when GetTree::SetOper GenTreeLngCon->GetTreeIntCon in src/jit/compiler.hpp
* Explicity specify roundUp(expr, TARGET_POINTER_SIZE)
* Use target_size_t* target in emitOutputDataSec in src/jit/emit.cpp
Jan Kotas [Wed, 28 Mar 2018 14:53:49 +0000 (07:53 -0700)]
Root methods used by debugger for IL linker (#17287)
Fei Peng [Wed, 28 Mar 2018 14:32:28 +0000 (07:32 -0700)]
Remove tests of unimplemented ISAs (#17286)
David Mason [Wed, 28 Mar 2018 02:21:17 +0000 (19:21 -0700)]
fix for logalways eventing level (#17277)
Debayan Ghosh [Wed, 21 Mar 2018 06:36:39 +0000 (12:06 +0530)]
Fix run-corefx-tests.py for Arm64
Enable cross build of corefx on x64, creates runtime for arm64
and builds corefx tests.
Jan Vorlicek [Wed, 28 Mar 2018 00:41:47 +0000 (02:41 +0200)]
Fix DelayLoad_MethodCall unwinding (#17275)
* Fix DelayLoad_MethodCall unwinding
Unwinding through DelayLoad_MethodCall was broken due to the overwriting
of R7 which is used as a frame pointer. That caused some managed
exceptions to cause abort with unhandled PAL_SEHException.
This change fixes the problem by using a different register.
* Fix one more spot with the same issue
Wes Haggard [Wed, 28 Mar 2018 00:18:33 +0000 (17:18 -0700)]
Merge pull request #17161 from dotnetrt/testbuild
Improve Windows test build performance by building projects in groups - 4x less memory 326% faster build.
Ahson Khan [Wed, 28 Mar 2018 00:05:33 +0000 (17:05 -0700)]
Adding Memory.Pin() to eventually replace Memory.Retain(bool) (#17269)
* Adding Memory.Pin() to eventually replace Memory.Retain(bool)
* Fix copy/paste error and return default for when object is null.
* Fix XML comments.
Jan Vorlicek [Tue, 27 Mar 2018 23:31:20 +0000 (01:31 +0200)]
Fix handling of FCalls in ExternalMethodFixupWorker (#17252)
When running ready to run code on ARM, the ExternalMethodFixupWorker
doesn't handle the entrypoints of FCalls correctly. It tries to handle
them as compact entrypoints, but those use a different machine code
instructions and it results in an assert in debug / checked build.
This change detects the runtime supplied calls before trying to check
for the compact entrypoint.
Stephen Toub [Tue, 27 Mar 2018 22:11:15 +0000 (15:11 -0700)]
Fix regression in StreamWriter.Write perf for small inputs (#17251)
* Streamline StreamWriter/Reader.CheckAsyncTaskInProgress and make it inlineable
* Re-inline StreamWriter.WriteSpan
Prior to my span changes, the code in WriteCore was manually inlined into each of its call sites. I'd refactored it out into a shared helper to reduce code duplication, but that introduced a regression for small inputs. I'm keeping the factoring but making it AggressiveInlining to avoid that regression while still keeping the C# code reuse.
* Use NoInlining on WriteSpan call sites
In case methods get devirtualized and JIT attempts to inline them and WriteSpan.
Carol Eidt [Tue, 27 Mar 2018 22:10:59 +0000 (15:10 -0700)]
Fix comment
Sung Yoon Whang [Tue, 27 Mar 2018 21:42:12 +0000 (14:42 -0700)]
Failfast windows event log (#16875)
* Add FailFast error log to Windows Event Log
* change const wchar * to lpcwstr
* Enable sending unhandled exception info to Windows Event Log
* Change log format to match console output and address PR comments
* Remove more comments
* Change the order DoReportForUnhandledException to do a safety check first before calling managed code
* Fix parameter name in header file
* Add Windows Event logging in DefaultCatchHandler and remove DoReportForUnhandledException
* Add back event reporting for ignored unhandled exception cases, fix broken UNIX builds
* Fix more broken unix builds
* Fix typo
* Address PR comments
Steve MacLean [Tue, 27 Mar 2018 21:32:46 +0000 (17:32 -0400)]
[Arm64] Fix forcerelocs (#17264)
Luqun Lou [Tue, 27 Mar 2018 21:20:09 +0000 (14:20 -0700)]
Enable reflection load ComImport assembly and Type.IsComObjectType (#16943)
* Enable reflection load ComImport assembly and Type.IsComObjectType
* Update Enable reflection load ComImport assembly
Carol Eidt [Tue, 27 Mar 2018 20:54:11 +0000 (13:54 -0700)]
Fix comment in test.
Michelle McDaniel [Tue, 27 Mar 2018 20:23:45 +0000 (13:23 -0700)]
Turn off PGO unprofiled warning on Linux (#17262)
This warning will break the build when new files are added to the build that do not have PGO data, meaning that release builds will be broken until a new PGO collection can be run. We should not break the build if there are not counts for new files.
Maoni Stephens [Tue, 27 Mar 2018 20:04:26 +0000 (13:04 -0700)]
restrict memory limit by taking VM into consideration (#17177)
Carol Eidt [Tue, 27 Mar 2018 19:35:19 +0000 (12:35 -0700)]
Merge pull request #17233 from fiigii/portvectortests
Port VectorReturn test to Vector128/256<T>
Bruce Forstall [Tue, 27 Mar 2018 18:48:06 +0000 (11:48 -0700)]
Merge pull request #17255 from BruceForstall/UpdateArmWhitelist
Update arm64 whitelist
Bruce Forstall [Tue, 27 Mar 2018 18:47:24 +0000 (11:47 -0700)]
Merge pull request #17258 from dotnet/dev/unix_test_workflow
Update arm Ubuntu cross build docker image
Jan Kotas [Tue, 27 Mar 2018 17:51:22 +0000 (10:51 -0700)]
Cleanup InternalsVisibleTo (#17185)
Remove InternalsVisibleTo for mscorlib and make sure that corefx still builds with it.
Delete global constants that were left-overs from .NET Framework build system.
Carol Eidt [Tue, 27 Mar 2018 16:28:57 +0000 (09:28 -0700)]
Merge pull request #17246 from debayang/clr_jitstress_arm64
[ARM64] Fix UnspillReg instruction generation
Bruce Forstall [Tue, 27 Mar 2018 16:06:14 +0000 (09:06 -0700)]
Update arm64 whitelist
Steve MacLean [Tue, 27 Mar 2018 14:20:19 +0000 (10:20 -0400)]
[Arm64] reserve for jump stubs (#17244)
Jan Kotas [Tue, 27 Mar 2018 14:06:36 +0000 (07:06 -0700)]
Move String.Comparison.cs to shared CoreLib partition (#17247)
Jan Vorlicek [Tue, 27 Mar 2018 13:27:01 +0000 (15:27 +0200)]
Fix build without libunwind installed (#17221)
* Fix build without libunwind installed
I have removed libunwind dependency in a recent commit, but
it turns out that the build was using incorrect include
paths and it was still depending on the libunwind installation
for the libunwind.h header.
This change fixes it.
* Fix clang warning
The libunwind headers contain two empty structs and clang
issues a warning since the size of such structs is
0 in C and 1 in C++.
This is benign for our purposes, since the affected
structs are not directly used by our code.
dotnet-maestro-bot [Tue, 27 Mar 2018 07:53:56 +0000 (02:53 -0500)]
Update BuildTools, CoreClr, CoreFx to preview3-02627-01, preview3-26327-02, preview3-26327-01, respectively (#17239)
Jan Kotas [Tue, 27 Mar 2018 07:31:55 +0000 (00:31 -0700)]
Delete MemoryExtensions.AsBytes (#17245)
Moved to MemoryMarshal
Jan Kotas [Tue, 27 Mar 2018 07:27:21 +0000 (00:27 -0700)]
Vectorized SequenceCompareTo for Span<char> (#17237)
- This change makes the compare for very short Span strings a bit slower and for longer Span strings many times faster.
- Switch several places where it was a clear benefit to use it.
`String.CompareOrdinal(string,string)` is notable exception that I have left intact for now. It is fine tuned for current string layout, and replacing with a call of vectorized SequenceCompareTo gives mixed results.
Bruce Forstall [Tue, 27 Mar 2018 07:09:33 +0000 (00:09 -0700)]
Merge pull request #17240 from BruceForstall/RemoveTizenPRTrigger
Remove Tizen armel automatic PR triggered jobs
Debayan Ghosh [Tue, 27 Mar 2018 06:00:19 +0000 (11:30 +0530)]
[ARM64] Fix UnspillReg instruction generation
Fixes the runtime issues (#16359,#15389) seen with COMPlus_JitStressRegs=0x2* on ARM64.
Maryam Ariyan [Tue, 27 Mar 2018 04:56:35 +0000 (21:56 -0700)]
New chunk should be equal/larger than the one it replaces (#17219)
* The new chunk should be equal or larger than the one it is replacing. Fixes #17205
* Adding .exe.stackdump to .gitignore
Jacek Blaszczynski [Thu, 22 Mar 2018 21:17:35 +0000 (22:17 +0100)]
Enable partially implemented Intel HW intrinsics ISAs - CoreCLR part of changes
Wes Haggard [Tue, 27 Mar 2018 00:28:31 +0000 (17:28 -0700)]
Merge pull request #17226 from weshaggard/MergeRestoreTargets
Merge various restore tasks into one Sync target
Bruce Forstall [Tue, 27 Mar 2018 00:27:31 +0000 (17:27 -0700)]
Remove Tizen armel automatic PR triggered jobs
These are failing almost every job with "qeumu: Unsupported
syscall: 389". Issue: https://github.com/dotnet/coreclr/issues/12972.
Fei Peng [Tue, 27 Mar 2018 00:09:22 +0000 (17:09 -0700)]
Port VectorReturn test to Vector128/256<T>
mikedn [Mon, 26 Mar 2018 23:41:46 +0000 (02:41 +0300)]
Remove a few bogus reinterpret_cast(nullptr_t) (#17229)
reinterpret_cast cannot be used to convert nullptr_t values, so does C++ std says in a note. VC++ ignores this restriction claiming that notes are not normative. Clang enforces the restriction. Oh well...
Bruce Forstall [Mon, 26 Mar 2018 23:36:58 +0000 (16:36 -0700)]
Merge pull request #17236 from BruceForstall/UpdateArmDockerImage
Update arm Ubuntu cross build docker image
Bruce Forstall [Mon, 26 Mar 2018 23:35:46 +0000 (16:35 -0700)]
Update arm Ubuntu cross build docker image
Now, use:
microsoft/dotnet-buildtools-prereqs:ubuntu-14.04-cross-e435274-
20180323032140
Carol Eidt [Mon, 26 Mar 2018 23:25:54 +0000 (16:25 -0700)]
Merge pull request #15301 from mikedn/cast-un
Fix inconsistent handling of zero extending casts
Wes Haggard [Mon, 26 Mar 2018 23:13:54 +0000 (16:13 -0700)]
Disable MSBuildEventLogging for optdata alias
We don't need the binclash logger for optdata so disable it
by pass a single space (note empty string means default).
The default value points at net46 which isn't what we want on
linux.
Andy Ayers [Mon, 26 Mar 2018 22:11:39 +0000 (15:11 -0700)]
Use movsdsse2 for HW intrinsic SSE2_ConvertToDouble (#17110)
Instead of movsd, which is for GPRs. Fixes one of the stress failures
seen in #17027.
Bruce Forstall [Mon, 26 Mar 2018 22:05:32 +0000 (15:05 -0700)]
Merge pull request #17018 from dotnet/dev/unix_test_workflow
Enable arm32 Ubuntu testing in CI
Adam Stark [Mon, 26 Mar 2018 20:46:13 +0000 (13:46 -0700)]
Only use FEATURE_PERFTRACING and FEATURE_EVENT_TRACE features when turned on (#17232)
Carol Eidt [Mon, 26 Mar 2018 20:33:28 +0000 (13:33 -0700)]
Fix condition
David Mason [Mon, 26 Mar 2018 20:25:32 +0000 (13:25 -0700)]
[local gc] Fix for BitScanForward64, it was ignoring the high byte check (#17142)
dotnet-maestro-bot [Mon, 26 Mar 2018 20:00:05 +0000 (15:00 -0500)]
Update CoreFx to preview3-26326-03 (#17222)