platform/upstream/coreclr.git
6 years agoStop using size_t for passing immediate values in CodeGen CodeGenInterface (#19544)
Egor Chesakov [Tue, 21 Aug 2018 03:19:08 +0000 (20:19 -0700)]
Stop using size_t for passing immediate values in CodeGen CodeGenInterface (#19544)

6 years agoR2RDump - Add image describing R2R format (#19528)
Amy [Tue, 21 Aug 2018 02:43:17 +0000 (19:43 -0700)]
R2RDump - Add image describing R2R format (#19528)

* Add image describing R2R format

* Remove help message stating -v option dumps raw bytes

6 years agoFix tiered compilation option for case-sensitive systems (#19567)
Andy Ayers [Tue, 21 Aug 2018 01:38:30 +0000 (18:38 -0700)]
Fix tiered compilation option for case-sensitive systems (#19567)

6 years agoMerge pull request #19334 from CarolEidt/Fix19243
Carol Eidt [Mon, 20 Aug 2018 23:57:40 +0000 (16:57 -0700)]
Merge pull request #19334 from CarolEidt/Fix19243

Fix evaluation order for block copy

6 years agoAdd IBC support (#19046)
Michelle McDaniel [Mon, 20 Aug 2018 22:21:58 +0000 (15:21 -0700)]
Add IBC support (#19046)

This change adds support to consume IBC data packages that are created by the dotnet/optimization repository. With these changes and dotnet/buildtools#2103 to enable IBC optimizations for corefx, we will see a 32% decrease in the size of NetCoreApp (and an overall decrease of 16%), a 30% reduction in ref set, a 5% improvement in time to first request and steady state performance for MusicStore.

Size

  Crossgen Partial Partial vs Crossgen (lower is better)
Shared (MB) 185.6 137.3 0.74
Microsoft.NETCore.App (MB) 115.6 67.4 0.58
RefSet

MusicStore

  Crossgen Partial Partial vs Crossgen (lower is better)
Total 19.389 13.472 0.69
AllReady

Crossgen Partial Partial vs Crossgen (lower is better)
Total 17.58 12.214
Performance

MusicStore

  Crossgen Partial Partial vs Crossgen (lower is better)
Server Start (ms) 870 870.6 1.00
First request (ms) 3532.6 3386.6 0.95
Steady State (ms) 2.926 2.79 0.95
AllReady

Crossgen Partial Partial vs Crossgen (lower is better)
Server Start (ms) 2102 1942.4
First Request (ms) 4263.2 4126
Steady State (ms) 5.69 5.68
TechEmpower Plaintext

Crossgen Partial Partial vs Crossgen (lower is better)
Requests per Second 1928649.8 1893183.8
First Request (ms) 76.14 80.11
Startup (ms) 391 372.8
Working Set (MB) 382.2 373.4

6 years agoJIT: remove incremental ref count updates (#19345)
Andy Ayers [Mon, 20 Aug 2018 22:03:29 +0000 (15:03 -0700)]
JIT: remove incremental ref count updates (#19345)

Remove almost all of the code in the jit that tries to maintain local ref
counts incrementally. Also remove `lvaSortAgain` and related machinery.

Explicitly sort locals before post-lower-liveness when optimizing to get the
best set of tracked locals.

Explicitly recount after post-lower liveness to get accurate counts after
dead stores. This can lead to tracked unreferenced arguments; tolerate this
during codegen.

6 years agoUpdate Microsoft.NETCore.CoreDisTools to version 1.0.1-prerelease-00005 (#19520)
Zach Montoya [Mon, 20 Aug 2018 21:08:41 +0000 (14:08 -0700)]
Update Microsoft.NETCore.CoreDisTools to version 1.0.1-prerelease-00005 (#19520)

* Update Microsoft.NETCore.CoreDisTools to version 1.0.1-prerelease-00005. Temporarily add a direct reference to the win-x64 and win-x86 runtime packages

* Change Microsoft.NETCore.CoreDisTools package references to the identity package.

Improve formatting of R2RDump.csproj properties.

* Add an ArgumentException to explain the currently faulty behavior where disassembling an R2R image whose architecture is different than the coredistools.dll architecture.

* Add Issue #19564 to the R2RDump.cs Disassembler comment

6 years agoFix dbgshim's > 1000 module bug (issue #19538) (#19553)
Mike McLaughlin [Mon, 20 Aug 2018 20:33:05 +0000 (13:33 -0700)]
Fix dbgshim's > 1000 module bug (issue #19538) (#19553)

Fix dbgshim's > 1000 module bug (issue #19538)

Cap cbNeeded on second EnumProcessModules call. Change the allocations
to HMODULE to make sure they are aligned properly.

6 years agoFix issues with calling convention in tests. (#19548)
Aaron Robinson [Mon, 20 Aug 2018 19:00:54 +0000 (12:00 -0700)]
Fix issues with calling convention in tests. (#19548)

6 years agoAdd comment
Carol Eidt [Mon, 20 Aug 2018 17:44:41 +0000 (10:44 -0700)]
Add comment

6 years agoNormalize whitespace/indentation
Clinton Ingram [Mon, 20 Aug 2018 11:09:45 +0000 (04:09 -0700)]
Normalize whitespace/indentation

6 years agoUpdate t4 templates to generate simplified type names
Clinton Ingram [Mon, 20 Aug 2018 11:06:30 +0000 (04:06 -0700)]
Update t4 templates to generate simplified type names

6 years agoMark some readonly fields readonly (#19559)
Ben Adams [Mon, 20 Aug 2018 16:30:37 +0000 (17:30 +0100)]
Mark some readonly fields readonly (#19559)

6 years agoOptimize number formatting (#19551)
Pent Ploompuu [Mon, 20 Aug 2018 16:29:55 +0000 (19:29 +0300)]
Optimize number formatting (#19551)

6 years agoMark more structs as readonly (#19557)
Ben Adams [Mon, 20 Aug 2018 16:25:47 +0000 (17:25 +0100)]
Mark more structs as readonly (#19557)

6 years agoMake DateTimeOffset a readonly struct (#19552)
Ben Adams [Mon, 20 Aug 2018 16:22:20 +0000 (17:22 +0100)]
Make DateTimeOffset a readonly struct (#19552)

6 years ago Free the alternate signal stack if this PAL's thread allocated it. (#19539)
Mike McLaughlin [Sun, 19 Aug 2018 23:04:02 +0000 (16:04 -0700)]
 Free the alternate signal stack if this PAL's thread allocated it. (#19539)

Free the alternate signal stack if this PAL's thread allocated it.

Move Ensure/FreeSignalAlternateStack to CPalThread class.

Frees the alternate stack a little sooner during thread termination.

6 years agoFix test case comment to reflect the actual test (#19531)
Andy Ayers [Sun, 19 Aug 2018 02:23:41 +0000 (19:23 -0700)]
Fix test case comment to reflect the actual test (#19531)

6 years agoEnable Microsoft.VisualBasic.dll in crossgen_comparison scenario
Egor Chesakov [Fri, 17 Aug 2018 18:40:05 +0000 (11:40 -0700)]
Enable Microsoft.VisualBasic.dll in crossgen_comparison scenario

6 years agoReturn "default" target-specific NaN value during constant folding of arithmetic...
Egor Chesakov [Thu, 9 Aug 2018 00:57:26 +0000 (17:57 -0700)]
Return "default" target-specific NaN value during constant folding of arithmetic expressions on ARM32 and ARM64

6 years agoFix wrong sprinf_s format string in gtGetArgMsg getGetLateArgMsg on ARM32 (#19536)
Egor Chesakov [Sun, 19 Aug 2018 01:03:06 +0000 (18:03 -0700)]
Fix wrong sprinf_s format string in gtGetArgMsg getGetLateArgMsg on ARM32 (#19536)

6 years agoExclude some assemblies from CrossGen Comparison job (#19534)
Egor Chesakov [Sat, 18 Aug 2018 21:38:14 +0000 (14:38 -0700)]
Exclude some assemblies from CrossGen Comparison job (#19534)

* System.Runtime.Intrinsics.Experimental.dll
* Xunit assemblies

6 years agoInitial support for R2R image diff (#19533)
Tomáš Rylek [Fri, 17 Aug 2018 23:41:22 +0000 (01:41 +0200)]
Initial support for R2R image diff (#19533)

* Initial support for R2R image diff

In this first cut I'm adding just three size diff statistics -
PE section size diff, R2R section size diff and method size diffs.
I assume we'll add more statistics as needed in the course of various
size investigations.

Thanks

Tomas

6 years agoRemove tests previously ported to CoreFX repo (#19527)
Aaron Robinson [Fri, 17 Aug 2018 18:09:31 +0000 (11:09 -0700)]
Remove tests previously ported to CoreFX repo (#19527)

* Remove tests previously ported to CoreFX repo

* Remove references to ported Marshal API tests

6 years agoMerge pull request #19530 from BruceForstall/DisableArmUbuntuCorefxLogging
Bruce Forstall [Fri, 17 Aug 2018 17:47:38 +0000 (10:47 -0700)]
Merge pull request #19530 from BruceForstall/DisableArmUbuntuCorefxLogging

Disable Ubuntu arm32 corefx xunit test result parsing

6 years agoMerge pull request #19378 from sergign60/unwind_fix
Bruce Forstall [Fri, 17 Aug 2018 17:32:16 +0000 (10:32 -0700)]
Merge pull request #19378 from sergign60/unwind_fix

[CoreRT armel]Fix for unwinding support for ARM #5874

6 years agoDisable Ubuntu arm32 corefx xunit test result parsing
Bruce Forstall [Fri, 17 Aug 2018 17:24:48 +0000 (10:24 -0700)]
Disable Ubuntu arm32 corefx xunit test result parsing

We have a problem with the xunit plug-in, where it is consistently
failing on Ubuntu arm32 test result uploading with this error:

```
The plugin hasn't been performed correctly
```

We haven't been able to identify the reason. So, do not add xunit parsing of the test data in this scenario.
This is tracked by: https://github.com/dotnet/coreclr/issues/19447.

6 years agoMerge pull request #19512 from dotnet-maestro-bot/master-UpdateDependencies
William Godbe [Fri, 17 Aug 2018 17:14:02 +0000 (10:14 -0700)]
Merge pull request #19512 from dotnet-maestro-bot/master-UpdateDependencies

Update BuildTools, CoreClr, CoreFx, CoreSetup, PgoData to preview1-03116-01, preview1-26817-05, preview1-26817-04, preview1-26817-01, master-20180817-0151, respectively (master)

6 years agoEnable Tiered Compilation by default (#19525)
Koundinya Veluri [Fri, 17 Aug 2018 15:31:44 +0000 (08:31 -0700)]
Enable Tiered Compilation by default (#19525)

Enable Tiered Compilation by default

1) Changes the default state of the tiered compilation feature check to be ON BY DEFAULT
2) Removed comments about the source about this being a work in progress. Although it will surely continue to evolve and improve, remaining issues would be better tracked in our issue tracking system with the same default presumption as other runtime features - assume it works unless noted otherwise.
3) Adjusts a number of tests and automated scripts that made assumptions that the default setting of this feature is off.
4) Stop accepting the deprecated env var COMPLUS_EXPERIMENTAL_TieredCompilation. I'm not aware it has any remaining usage but if so we're going to find out.
5) Adjust config names for JitBench

6 years agoUpdate BuildTools, CoreClr, CoreFx, CoreSetup, PgoData to preview1-03116-01, preview1...
dotnet-maestro-bot [Fri, 17 Aug 2018 13:47:35 +0000 (06:47 -0700)]
Update BuildTools, CoreClr, CoreFx, CoreSetup, PgoData to preview1-03116-01, preview1-26817-05, preview1-26817-04, preview1-26817-01, master-20180817-0151, respectively

6 years agoMerge pull request #19514 from dotnet/dev/unix_test_workflow
Egor Chesakov [Fri, 17 Aug 2018 04:56:41 +0000 (21:56 -0700)]
Merge pull request #19514 from dotnet/dev/unix_test_workflow

 Introduce CrossGen Comparison scenario

6 years agoAdd CrossGen Comparison scenario to netci.groovy
Egor Chesakov [Thu, 9 Aug 2018 23:36:17 +0000 (16:36 -0700)]
Add CrossGen Comparison scenario to netci.groovy

6 years agoAdd tests/scripts/crossgen_comparison.py
Egor Chesakov [Thu, 9 Aug 2018 23:37:39 +0000 (16:37 -0700)]
Add tests/scripts/crossgen_comparison.py

6 years agoRemove coreclr.*txt upload (#19524)
Jarret Shook [Fri, 17 Aug 2018 01:16:59 +0000 (18:16 -0700)]
Remove coreclr.*txt upload (#19524)

6 years agoR2RDump - Documentation (#19497)
Amy [Fri, 17 Aug 2018 00:05:23 +0000 (17:05 -0700)]
R2RDump - Documentation (#19497)

* Add comments

* Add README

* Use href links, disable disasm

6 years agoMoved registry.cs to shared (#19471)
Anirudh Agnihotry [Thu, 16 Aug 2018 20:21:55 +0000 (13:21 -0700)]
Moved registry.cs to shared (#19471)

* moved registry to shared

* using corefx version

* set value added

6 years agoDetermine memory load based on cgroup usage. (#19518)
Tom Deseyn [Thu, 16 Aug 2018 18:47:20 +0000 (20:47 +0200)]
Determine memory load based on cgroup usage. (#19518)

cgroup usage is used to trigger oom kills. It includes rss and file cache
of the cgroup.

The implementation was only using the process rss to determine memory load.
This is less than the cgroup usage and leads to oom kills due to GC not
being triggered soon enough.

6 years agoAdd an empty console application to measure startup of the runtime. (#19490)
José Rivero [Thu, 16 Aug 2018 18:25:31 +0000 (11:25 -0700)]
Add an empty console application to measure startup of the runtime. (#19490)

6 years agoDisable a regex test that is failing with minopts (#19488)
Koundinya Veluri [Thu, 16 Aug 2018 18:23:47 +0000 (11:23 -0700)]
Disable a regex test that is failing with minopts (#19488)

See dotnet/coreclr#18912. This test would always fail once tiered compilation is enabled by default in CoreFX test runs in the CoreCLR repo.

6 years agoBuild-test.sh handles native test assets (#19430)
Aaron Robinson [Thu, 16 Aug 2018 18:21:51 +0000 (11:21 -0700)]
Build-test.sh handles native test assets (#19430)

* Respect Windows script argument to skip package building.

* Bring build.sh logic closer to build.cmd with respect to passing CMAKE flags for building tests

* Make cmake gen script find override file without using script arguments

* build-test.sh can now build native test projects
Remove compileoptions.cmake and push contents into configurecompiler.cmake
 which contains compiler configuration from repo root CMakeLists.txt

6 years agoSome JitBench fixes (#19479)
Koundinya Veluri [Thu, 16 Aug 2018 18:20:55 +0000 (11:20 -0700)]
Some JitBench fixes (#19479)

Some JitBench fixes

- Fixed `dotnet run` from the `unofficial_dotnet` folder after `Directory.Build.targets` was added to `coreclr/tests/src`. This file is auto-imported and was adding dependencies and properties for targets that were not necessary. Added files to the root JitBench folder to override.
- Updated JitBench to be able to run against netcoreapp3.0. I haven't changed the default yet, could consider doing so once there are no pending perf issues being looked into.
  - Added a `RetargetProjects()` function to the setup steps to replace `TargetFramework` and `RuntimeFrameworkVersion` values in the relevant `.csproj` files to run against the expected runtime. This reduces some dependency on other repos containing the actual perf projects such that we don't have to update every repo to support environment variables to set those values.
- Disabled tiering for the `MinOpts` config

6 years agoJIT: bail out in optExtractArrIndex for constant array length (#19493)
Andy Ayers [Thu, 16 Aug 2018 16:22:05 +0000 (09:22 -0700)]
JIT: bail out in optExtractArrIndex for constant array length (#19493)

Generalize the bail out pattern to handle cases from spans where we may
see a known array length in a bounds check.

Fixes #19454

6 years agoOptimize NumberBufferToDecimal (#19072)
Pent Ploompuu [Thu, 16 Aug 2018 15:40:59 +0000 (18:40 +0300)]
Optimize NumberBufferToDecimal (#19072)

6 years agoMerge pull request #19396 from CarolEidt/Fix19288
Carol Eidt [Thu, 16 Aug 2018 15:11:00 +0000 (08:11 -0700)]
Merge pull request #19396 from CarolEidt/Fix19288

Fix non-lclVar odd-byte struct passing

6 years ago[CoreRT armel]Fix for unwinding support for ARM is not fully implemented #5874
Sergey Ignatov [Thu, 9 Aug 2018 08:52:10 +0000 (11:52 +0300)]
[CoreRT armel]Fix for unwinding support for ARM is not fully implemented #5874

6 years agoMake file handle checks accurate on Windows (#19508)
Jeremy Kuhne [Thu, 16 Aug 2018 04:16:43 +0000 (21:16 -0700)]
Make file handle checks accurate on Windows (#19508)

* Make handle checks accurate on Windows

We can actually check the async state of a handle on Windows, so check directly when we have the API available.

Also allow all filetypes through when explicitly using extended syntax `\\?\` .

* Fix nits

* Don't throw in the handle constructor overloads

* Whoops

* Removing the close handle entirely. No adverse effects from closing a bad handle- or leaving it open for that matter.

6 years agoRevert "Temporarily flag a SIMD test as optimization-sensitive (#19163)" (#19505)
Koundinya Veluri [Thu, 16 Aug 2018 02:43:15 +0000 (19:43 -0700)]
Revert "Temporarily flag a SIMD test as optimization-sensitive (#19163)" (#19505)

This reverts commit 855ddf52c7b05994cf6728b9169fa6cdc694a537. Issue https://github.com/dotnet/coreclr/issues/19124 was fixed by PR https://github.com/dotnet/coreclr/pull/19234, so re-enabling the test in minopts and with tiering.

6 years agoUpdate JapaneseLunisolarCalendar.cs (#19504)
elyoh [Thu, 16 Aug 2018 01:03:50 +0000 (02:03 +0100)]
Update JapaneseLunisolarCalendar.cs (#19504)

Corrects an issue with the conversion table for Gregorian to Japanese lunisolar dates.
See: #19450.

Issue: data discrepancy in year 1962.
Table entry for 1962 has incorrect days per month for months 6 and 7.

Notes
The DaysPerMonth flag uses the binary literal for ease of readability.

6 years agoJIT: properly handle special-case varargs params in minopts (#19489)
Andy Ayers [Thu, 16 Aug 2018 00:05:28 +0000 (17:05 -0700)]
JIT: properly handle special-case varargs params in minopts (#19489)

We were previously marking all lcl vars as implicitly referenced in minopts,
but there are some special varargs params that can't ever be marked as
referenced, either implicitly or explicitly.

Special case these when computing or checking ref counts in minopts.

Closes #19349.

6 years agoRevert "Change CoreLib native image to be R2R by default on all platforms (#19359...
Jan Kotas [Wed, 15 Aug 2018 22:55:24 +0000 (15:55 -0700)]
Revert "Change CoreLib native image to be R2R by default on all platforms (#19359)" (#19501)

This reverts commit d485659348af3528b688ac5b7e4c58dacba6de58.

6 years agoAdd Public API Marshal.IsTypeVisibleFromCom and Marshal.GetEndComSlot (#19507)
Luqun Lou [Wed, 15 Aug 2018 22:52:28 +0000 (15:52 -0700)]
Add Public API Marshal.IsTypeVisibleFromCom and Marshal.GetEndComSlot (#19507)

6 years agoR2RDump - GcInfo for x86
Amy Yu [Wed, 8 Aug 2018 23:52:53 +0000 (16:52 -0700)]
R2RDump - GcInfo for x86

x86 GcInfo headers

x86 GcSlotTable

x86 GcTransitions

Update expected xml for R2RDumpTests

Add license headers

Allow multiple GcTransitions with same codeOffset

Add index property in GcSlot, make GcSlot a class instead of struct, add missing spaces, update tests

Remove placeholder functions for parsing partially interruptible pointer tables

Implement partially interruptible GcInfo

Example GcInfo output:
    CodeLength: 22 bytes
        InfoHdr:
            PrologSize: 7
            EpilogSize: 4
            EpilogCount: 1
            EpilogAtEnd: yes
            Callee-saved regs  = EBP
            EbpFrame: yes
            Fully Interruptible: yes
            DoubleAlign: no
            Arguments Size: 0 DWORDs
            Stack Frame Size: 1 DWORDs
            UntrackedCnt: 1
            VarPtrTableSize: 0
            GenericsContext: 0
            GenericsContextIsMethodDesc: 0
            ReturnKind: RT_Scalar
            RevPInvokeOffset: 0
            Epilogs: 18

            GcSlots:
                -------------------------
                [EBP-4]
                Flags: GC_SLOT_UNTRACKED
                LowBits:
                -------------------------

    28fc: 55                    push    ebp
    28fd: 8b ec                 mov     ebp, esp
    28ff: 50                    push    eax
    2900: 89 4d fc              mov     dword ptr [ebp - 4], ecx
    2903: 8b 4d fc              mov     ecx, dword ptr [ebp - 4]
    2906: ff 15 10 10 01 10     call    dword ptr [268505104]
                                reg ECX becoming live
    290c: 90                    nop
                                reg ECX becoming dead
    290d: 90                    nop
    290e: 8b e5                 mov     esp, ebp
    2910: 5d                    pop     ebp
    2911: c3                    ret

6 years agoUpdate BuildTools, CoreClr, CoreFx, CoreSetup, PgoData to preview1-03115-01, preview1...
dotnet-maestro-bot [Wed, 15 Aug 2018 19:30:35 +0000 (12:30 -0700)]
Update BuildTools, CoreClr, CoreFx, CoreSetup, PgoData to preview1-03115-01, preview1-26815-04, preview1-26815-04, preview1-26815-01, master-20180815-0120, respectively (#19494)

6 years agoMerge pull request #19487 from acmyu/disable
Bruce Forstall [Wed, 15 Aug 2018 03:56:16 +0000 (20:56 -0700)]
Merge pull request #19487 from acmyu/disable

ExcludeList for R2RDumpTests in wrong ItemGroup

6 years agoAdd TypeConverter fallback to DefaultValueAttribute (#19354)
Marco Rossignoli [Wed, 15 Aug 2018 01:56:43 +0000 (03:56 +0200)]
Add TypeConverter fallback to DefaultValueAttribute (#19354)

6 years agoUpdate debugging and createdump docs. (#19495)
Mike McLaughlin [Wed, 15 Aug 2018 01:22:23 +0000 (18:22 -0700)]
Update debugging and createdump docs. (#19495)

6 years agoExcludeList for R2RDumpTests in wrong ItemGroup
Amy Yu [Tue, 14 Aug 2018 20:19:46 +0000 (13:19 -0700)]
ExcludeList for R2RDumpTests in wrong ItemGroup

6 years agoUpdate BuildTools, CoreClr, CoreFx, CoreSetup, PgoData to preview1-03110-01, preview1...
dotnet-maestro-bot [Tue, 14 Aug 2018 15:08:16 +0000 (08:08 -0700)]
Update BuildTools, CoreClr, CoreFx, CoreSetup, PgoData to preview1-03110-01, preview1-26814-06, preview1-26814-01, preview1-26814-01, master-20180814-0056, respectively (#19377)

6 years agoDisable R2RDumpTests (#19475)
Amy [Tue, 14 Aug 2018 14:02:49 +0000 (07:02 -0700)]
Disable R2RDumpTests (#19475)

6 years agoDelete dead code (#19452)
Jan Kotas [Tue, 14 Aug 2018 13:34:34 +0000 (06:34 -0700)]
Delete dead code (#19452)

Unused strong name signing paths and related code

6 years agoTranslate [rip +- offset] to absolute RVA's in R2RDump (#19219)
Tomáš Rylek [Tue, 14 Aug 2018 10:37:19 +0000 (12:37 +0200)]
Translate [rip +- offset] to absolute RVA's in R2RDump (#19219)

* Translate [rip +- offset] to absoluate RVA's in R2RDump

The existing logic for displaying rip-relative addressed on X64
make it very hard to calculate the final addresses. I have added
a horrendous hack using textual analysis of the disassembled
instruction to translate this notation to absolute RVA's.

As part of this effort I have also encapsulated the CorDisTools
helper in a new class Disassembler that also contains customizable
provisions for handling special assembly cases on the individual
architectures.

Thanks

Tomas

* Temporarily block out disassembly to make tests pass

In my initial commit I removed the line blocking out disassembly
however this ends up failing several lab tests so I'm putting the
line back.

Thanks

Tomas

6 years agoCode review feedback for the alternate stack changes (PR #19309). (#19476)
Mike McLaughlin [Tue, 14 Aug 2018 07:46:13 +0000 (00:46 -0700)]
Code review feedback for the alternate stack changes (PR #19309). (#19476)

Fixed SOS plugin for core dumps.

6 years agoUpdate URLs and exclusion list for staging (#19423)
Andon Andonov [Tue, 14 Aug 2018 03:43:10 +0000 (20:43 -0700)]
Update URLs and exclusion list for staging (#19423)

* Update URLs and exclusion list for staging

* Update Failing tests

* Fix IO test namespace

* Add 2 failing Ubuntu tests

* Exclude test under Ubuntu

* Add regressed test

* Roll-back CI URLs

* Add issue number

6 years agoMerge pull request #19463 from BruceForstall/FixArmHeapVerify
Bruce Forstall [Mon, 13 Aug 2018 23:53:22 +0000 (16:53 -0700)]
Merge pull request #19463 from BruceForstall/FixArmHeapVerify

Disable 3 tests for Windows/arm32 HeapVerify runs

6 years agodisable tests\src\JIT\Methodical\fp\exgen\10w5d_cs_do (#19465)
Sergey Andreenko [Mon, 13 Aug 2018 23:16:50 +0000 (16:16 -0700)]
disable tests\src\JIT\Methodical\fp\exgen\10w5d_cs_do (#19465)

in stress modes. The issue is #18988

6 years agoMerge pull request #19469 from BruceForstall/DisableVectorDot
Bruce Forstall [Mon, 13 Aug 2018 22:21:23 +0000 (15:21 -0700)]
Merge pull request #19469 from BruceForstall/DisableVectorDot

Disable VectorDot_ro.cmd_9267 for Windows arm64 R2R

6 years agoDisable VectorDot_ro.cmd_9267 for Windows arm64 R2R
Bruce Forstall [Mon, 13 Aug 2018 22:19:49 +0000 (15:19 -0700)]
Disable VectorDot_ro.cmd_9267 for Windows arm64 R2R

Tracked by #18991

6 years agoMerge pull request #19466 from BruceForstall/Disable163200
Bruce Forstall [Mon, 13 Aug 2018 22:04:49 +0000 (15:04 -0700)]
Merge pull request #19466 from BruceForstall/Disable163200

Disable test b163200 under GCStress

6 years agoDisable test b163200 under GCStress
Bruce Forstall [Mon, 13 Aug 2018 22:03:23 +0000 (15:03 -0700)]
Disable test b163200 under GCStress

Tracked by #19464

6 years agoDisable 3 tests for Windows/arm32 HeapVerify runs
Bruce Forstall [Mon, 13 Aug 2018 21:06:29 +0000 (14:06 -0700)]
Disable 3 tests for Windows/arm32 HeapVerify runs

They all time out.

Fixes #19462

6 years agoMerge pull request #19459 from BruceForstall/Disable437044
Bruce Forstall [Mon, 13 Aug 2018 19:42:25 +0000 (12:42 -0700)]
Merge pull request #19459 from BruceForstall/Disable437044

Exclude 437044.cmd_10304 from GCStress runs

6 years agoExclude 437044.cmd_10304 from GCStress runs
Bruce Forstall [Mon, 13 Aug 2018 19:40:23 +0000 (12:40 -0700)]
Exclude 437044.cmd_10304 from GCStress runs

This test times out on ARM Windows R2R GCStress15 runs.

6 years agoAdd a skipmanaged option to build.sh (#19111)
Omair Majid [Mon, 13 Aug 2018 19:38:37 +0000 (15:38 -0400)]
Add a skipmanaged option to build.sh (#19111)

This option is the opposite of -msbuildonunsupportedplatform, and an
managed version of of -skipnative.

This is useful for bootstrapping on new Linux/x64 distributions, where
dotnet may not work (yet) but coreclr will try and use it anyway,
failing the bootstrap scripts.

See also https://github.com/dotnet/source-build/issues/663 for
additional background information.

6 years agoMerge pull request #19458 from BruceForstall/DisableTests
Bruce Forstall [Mon, 13 Aug 2018 19:12:23 +0000 (12:12 -0700)]
Merge pull request #19458 from BruceForstall/DisableTests

Disable System.Linq.Expressions.Tests against #19457

6 years agoDisable System.Linq.Expressions.Tests against #19457
Bruce Forstall [Mon, 13 Aug 2018 19:10:50 +0000 (12:10 -0700)]
Disable System.Linq.Expressions.Tests against #19457

6 years agoRemove unused local from List<T>.RemoveRange (dotnet/corefx#31727)
Robin Sue [Mon, 13 Aug 2018 02:49:39 +0000 (04:49 +0200)]
Remove unused local from List<T>.RemoveRange (dotnet/corefx#31727)

`int i` appears to be unused so i removed it.

Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
6 years agonon shared changes and feedback addressed
Anipik [Sun, 12 Aug 2018 22:42:34 +0000 (15:42 -0700)]
non shared changes and feedback addressed

Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
6 years agoFix a couple of apartment state issues (#19384)
Koundinya Veluri [Mon, 13 Aug 2018 01:35:48 +0000 (18:35 -0700)]
Fix a couple of apartment state issues (#19384)

* Fix a couple of apartment state issues

Fix for https://github.com/dotnet/coreclr/issues/17822
- The apartment state now defaults to MTA for the main thread along with a CoInitialize
- Calling `Thread.SetApartmentState` with STA now fails as expected (different behavior from previous netcore, same behavior as netfx)

Fix for https://github.com/dotnet/coreclr/issues/17787
- `WaitHandle.WaitAll` for multiple handles is not supported on an STA thread due to issues described in https://github.com/dotnet/coreclr/issues/17787#issuecomment-385117537
- It now throws `NotSupportedException` as expected (different behavior from previous netcore, same behavior as netfx)

Fix for https://github.com/dotnet/coreclr/issues/19225

* Temporarily exclude invalid CoreFX test

6 years agoCreate Microsoft.IO.Redist (dotnet/corefx#31679)
Jeremy Kuhne [Sat, 11 Aug 2018 00:06:39 +0000 (17:06 -0700)]
Create Microsoft.IO.Redist (dotnet/corefx#31679)

* Create Microsoft.Internal.IO

Includes all of Path, Directory, *Info, etc. from System.IO with targets of NetFX 4.6.1 and NetCore 2.1.

* Tweaks / minor fixes

* Remove netstandard build.

* Add package

Update to build for 4.7.2 only

* Rename folder to prepare for project rename

* Change the name to Microsoft.IO.Redist

* Namespace tweaks

* Use open key, tweak references

* More tweaks on feedback

Also grab latest build tools for 4.7.x RIDs

Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
6 years agoSilence WinRT type assert in crossgen (#19442)
Simon Nattress [Mon, 13 Aug 2018 00:09:55 +0000 (17:09 -0700)]
Silence WinRT type assert in crossgen (#19442)

Disable type loader assert that fires when resolving a type from a WinMD. Instead, compilation should fail gracefully, or at runtime a type load exception will be thrown.

6 years ago[local gc] disable feature redhawk (#17769)
David Mason [Sat, 11 Aug 2018 17:11:49 +0000 (10:11 -0700)]
[local gc] disable feature redhawk  (#17769)

6 years agoMake RuntimeType sealed and deleted support for introspection only loads (#19427)
Jan Kotas [Sat, 11 Aug 2018 14:43:16 +0000 (07:43 -0700)]
Make RuntimeType sealed and deleted support for introspection only loads (#19427)

- Sealed Runtime makes `is RuntimeType` and similar checks faster. These checks are fairly common in reflection.
- Delete support for introspection only loads from the runtime. We do not plan to use in .NET Core. The support for introspection loads inherited from RuntimeType and thus it is incompatible with sealed RuntimeType.

6 years agomoved to shared (#19419)
Anirudh Agnihotry [Sat, 11 Aug 2018 07:31:01 +0000 (00:31 -0700)]
moved to shared (#19419)

6 years agoMoved SpinLock.cs to shared (#19391)
Anirudh Agnihotry [Sat, 11 Aug 2018 06:19:28 +0000 (23:19 -0700)]
Moved SpinLock.cs to shared (#19391)

* m removed from names, spaces after \\ and braces added

* Moved to shared

6 years agoDon't early terminate on null for 64bit NR HashCode (#19331)
Ben Adams [Sat, 11 Aug 2018 05:44:19 +0000 (06:44 +0100)]
Don't early terminate on null for 64bit NR HashCode (#19331)

* Don't early terminate on null for 64bit NR HashCode

* Improved GetNonRandomizedHashCode

* Update message on GetNonRandomizedHashCode

* Consume null terminator rather than special case odd lengths

6 years agoMerge pull request #19434 from BruceForstall/DisableR2rDumpTest2
Bruce Forstall [Sat, 11 Aug 2018 03:00:44 +0000 (20:00 -0700)]
Merge pull request #19434 from BruceForstall/DisableR2rDumpTest2

Disable R2RDumpTest for JIT stress modes

6 years agoDisable R2RDumpTest for JIT stress modes
Bruce Forstall [Sat, 11 Aug 2018 00:41:17 +0000 (17:41 -0700)]
Disable R2RDumpTest for JIT stress modes

6 years agoMerge pull request #19433 from BruceForstall/UpdateTriggerPhrases
Bruce Forstall [Sat, 11 Aug 2018 00:32:02 +0000 (17:32 -0700)]
Merge pull request #19433 from BruceForstall/UpdateTriggerPhrases

Update existing trigger phrases documentation

6 years agoUpdate existing trigger phrases documentation
Bruce Forstall [Sat, 11 Aug 2018 00:29:59 +0000 (17:29 -0700)]
Update existing trigger phrases documentation

6 years agoCORDbgSetInstruction: clear thumb bit of the address before flushing icache (#19409)
Konstantin Baladurin [Fri, 10 Aug 2018 23:10:53 +0000 (02:10 +0300)]
CORDbgSetInstruction: clear thumb bit of the address before flushing icache (#19409)

We should clear thumb bit of the address before flushing icache otherwise
we may not flush the cache for the first byte of the instruction.

6 years agoMoved to shared (#19399)
Anirudh Agnihotry [Fri, 10 Aug 2018 23:09:48 +0000 (16:09 -0700)]
Moved to shared (#19399)

6 years agoupdate order of locks so jitstress is unblocked (#19401)
David Mason [Fri, 10 Aug 2018 22:40:39 +0000 (15:40 -0700)]
update order of locks so jitstress is unblocked (#19401)

6 years agoFix non-lclVar odd-byte struct passing
Carol Eidt [Thu, 9 Aug 2018 23:31:11 +0000 (16:31 -0700)]
Fix non-lclVar odd-byte struct passing

When a non-power-of-2 struct is passed, we can only use an IND if the arg is a local.

Fix #19288

6 years agoMerge pull request #19344 from adityamandaleeka/aa64_vm_callconv_changes
Aditya Mandaleeka [Fri, 10 Aug 2018 20:38:09 +0000 (13:38 -0700)]
Merge pull request #19344 from adityamandaleeka/aa64_vm_callconv_changes

CallDescr/TransitionBlock changes for ARM64 calling convention

6 years agoAdd SoA raytracer as a CQ test for Intel hardware intrinsic
Fei Peng [Mon, 6 Aug 2018 22:10:50 +0000 (15:10 -0700)]
Add SoA raytracer as a CQ test for Intel hardware intrinsic

6 years agoMerge pull request #19414 from BruceForstall/ReduceCorefxInnerloopTesting
Bruce Forstall [Fri, 10 Aug 2018 18:43:05 +0000 (11:43 -0700)]
Merge pull request #19414 from BruceForstall/ReduceCorefxInnerloopTesting

Reduce the number of innerloop CoreFX jobs

6 years agoReduce the number of innerloop CoreFX jobs
Bruce Forstall [Fri, 10 Aug 2018 18:30:04 +0000 (11:30 -0700)]
Reduce the number of innerloop CoreFX jobs

1. Stop doing any OSX CoreFX jobs; they're too unreliable.
2. Stop running Debug jobs.

6 years agoMerge pull request #19387 from dotnetrt/parallel-test-build
Bruce Forstall [Fri, 10 Aug 2018 18:10:21 +0000 (11:10 -0700)]
Merge pull request #19387 from dotnetrt/parallel-test-build

Reenable parallel msbuild execution - 2x faster test build

6 years agoAlign the Contents of EventPipeBuffers (#19375)
Brian Robbins [Fri, 10 Aug 2018 17:22:40 +0000 (10:22 -0700)]
Align the Contents of EventPipeBuffers (#19375)

6 years agoUpdating the ilproj-sdk to also copy native dependencies from Microsoft.NETCore.JIT
Tanner Gooding [Thu, 9 Aug 2018 19:33:21 +0000 (12:33 -0700)]
Updating the ilproj-sdk to also copy native dependencies from Microsoft.NETCore.JIT