platform/upstream/coreclr.git
5 years agoUpdate CoreClr to preview5-27615-71 (#23991)
dotnet-maestro-bot [Mon, 15 Apr 2019 16:39:47 +0000 (09:39 -0700)]
Update CoreClr to preview5-27615-71 (#23991)

5 years agoUpdate dependencies from https://github.com/dotnet/corefx build 20190414.16 (#23989)
dotnet-maestro[bot] [Mon, 15 Apr 2019 16:07:41 +0000 (09:07 -0700)]
Update dependencies from https://github.com/dotnet/corefx build 20190414.16 (#23989)

- Microsoft.NETCore.Platforms - 3.0.0-preview5.19214.16
- Microsoft.Private.CoreFx.NETCoreApp - 4.6.0-preview5.19214.16

5 years agoimprove entropy (#23591)
Marco Rossignoli [Mon, 15 Apr 2019 03:01:31 +0000 (05:01 +0200)]
improve entropy (#23591)

5 years agoUpdate CoreClr to preview5-27614-72 (#23985)
dotnet-maestro-bot [Mon, 15 Apr 2019 02:51:06 +0000 (19:51 -0700)]
Update CoreClr to preview5-27614-72 (#23985)

5 years agoFix typo in environment variable name - profiler loading doc (#23983)
Gergely Kalapos [Mon, 15 Apr 2019 02:44:53 +0000 (04:44 +0200)]
Fix typo in environment variable name - profiler loading doc (#23983)

5 years agoRemove unused offset calculation (#23793)
Raz Friman [Mon, 15 Apr 2019 01:29:10 +0000 (11:29 +1000)]
Remove unused offset calculation (#23793)

Utf8Parser has a redundant calculation of date time offset.

Fix #21834

5 years agoRemove old method names from Rune (#23938)
Levi Broderick [Mon, 15 Apr 2019 01:26:05 +0000 (18:26 -0700)]
Remove old method names from Rune (#23938)

* Remove old method names from Rune, reenable suppressed tests

* Revert re-enablement of failing ArraySegment tests

5 years agoUpdate CoreClr to preview5-27614-71 (#23978)
dotnet-maestro-bot [Sun, 14 Apr 2019 20:49:39 +0000 (13:49 -0700)]
Update CoreClr to preview5-27614-71 (#23978)

5 years agoFix broken new lines (#23979)
Gergely Kalapos [Sun, 14 Apr 2019 17:16:24 +0000 (19:16 +0200)]
Fix broken new lines (#23979)

These files don't render correctly on github due to broken new lines - fixing it. Only new line characters replaced, nothing else touched.

5 years agoMove GCMemoryInfo to shared partition
Jan Kotas [Sun, 14 Apr 2019 04:54:08 +0000 (21:54 -0700)]
Move GCMemoryInfo to shared partition

5 years agoFix AssemblyName regression
Jan Kotas [Sat, 13 Apr 2019 21:10:26 +0000 (14:10 -0700)]
Fix AssemblyName regression

Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
5 years agoMove GCMemoryInfo to shared partition
Jan Kotas [Sat, 13 Apr 2019 16:36:03 +0000 (09:36 -0700)]
Move GCMemoryInfo to shared partition

Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
5 years agoUpdate BuildTools, CoreClr to preview4-03913-01, preview5-27612-73, respectively...
dotnet-maestro-bot [Sun, 14 Apr 2019 03:18:15 +0000 (20:18 -0700)]
Update BuildTools, CoreClr to preview4-03913-01, preview5-27612-73, respectively (master) (#23916)

* Update BuildTools, CoreClr to preview4-03913-01, preview5-27612-73, respectively

* Fix nullable build errors

5 years agoDelete unused AssemblyName._hashForControl and related code (#23972)
Jan Kotas [Sun, 14 Apr 2019 03:15:48 +0000 (20:15 -0700)]
Delete unused AssemblyName._hashForControl and related code (#23972)

5 years agoEnterContextualReflection handle null (#23953)
Steve MacLean [Sat, 13 Apr 2019 23:03:56 +0000 (19:03 -0400)]
EnterContextualReflection handle null (#23953)

* EnterContextualReflection handle null

* Add ContextualReflection MockAssembly test

* Fix ContextualReflection typo

5 years agoUse xUnit TestOutputHelper to get proper test output in Azure DevOps Tests Explorer...
Egor Chesakov [Sat, 13 Apr 2019 19:09:53 +0000 (12:09 -0700)]
Use xUnit TestOutputHelper to get proper test output in Azure DevOps Tests Explorer (#23957)

5 years agoDelete outdated comment (#23967)
Jan Kotas [Sat, 13 Apr 2019 19:07:09 +0000 (12:07 -0700)]
Delete outdated comment (#23967)

5 years agoRemove unnecessary ThrowIfCreated call (#23839)
Next Turn [Sat, 13 Apr 2019 19:06:35 +0000 (03:06 +0800)]
Remove unnecessary ThrowIfCreated call (#23839)

5 years ago[master] Update dependencies from dotnet/arcade (#23926)
dotnet-maestro[bot] [Sat, 13 Apr 2019 19:04:11 +0000 (12:04 -0700)]
[master] Update dependencies from dotnet/arcade (#23926)

* Update dependencies from https://github.com/dotnet/arcade build 20190412.2

- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19212.2
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19212.2

5 years agoFix ARM's genPutArgStk codegen (#23867)
mikedn [Sat, 13 Apr 2019 17:39:20 +0000 (20:39 +0300)]
Fix ARM's genPutArgStk codegen (#23867)

* Fix ARM's genPutArgStk codegen

When the OBJ node wraps a LCL_VAR node the code uses the type information (struct size, GC layout) from LclVarDsc. This is not always correct because the OBJ may actually have a different struct type due to type reinterpretation (e.g. Unsafe.As<X, Y>).

* Fix genPutArgStk comment

5 years agoMerge pull request #23948 from BruceForstall/Fix23920
Bruce Forstall [Sat, 13 Apr 2019 17:34:26 +0000 (10:34 -0700)]
Merge pull request #23948 from BruceForstall/Fix23920

Update maximum allowed arm prolog size

5 years agoMove DateTime Windows implementation to shared CoreLib (#23888)
Filip Navara [Sat, 13 Apr 2019 15:17:04 +0000 (17:17 +0200)]
Move DateTime Windows implementation to shared CoreLib (#23888)

* Move DateTime Windows implementation to shared CoreLib

* Make all the BOOL/struct marshalling explicit

* internal -> private

* Use unsafe code on all layers, remove unnecessary fixed blocks

5 years agoDelete outdated comment
Jan Kotas [Sat, 13 Apr 2019 15:13:04 +0000 (08:13 -0700)]
Delete outdated comment

5 years agoUpdate dependencies from https://github.com/dotnet/corefx build 20190412.14 (#23964)
dotnet-maestro[bot] [Sat, 13 Apr 2019 15:10:29 +0000 (08:10 -0700)]
Update dependencies from https://github.com/dotnet/corefx build 20190412.14 (#23964)

- Microsoft.NETCore.Platforms - 3.0.0-preview5.19212.14
- Microsoft.Private.CoreFx.NETCoreApp - 4.6.0-preview5.19212.14

5 years agoDelete COMPlus_FinalizeOnShutdown (#23595)
Jan Kotas [Sat, 13 Apr 2019 15:09:31 +0000 (08:09 -0700)]
Delete COMPlus_FinalizeOnShutdown (#23595)

This compat quirk is increasingly more broken since the framework is generally not compatible with, and we have not heard anybody actually using it.

Changed Environment.HasShutdownStarted to unconditionally return false. It does not make sense for it to ever return true with shutdown finalization disabled.

5 years agoExclude ResolveEventTests.cmd and CallbackTests.cmd on Windows_NT x86 (#23942)
Egor Chesakov [Sat, 13 Apr 2019 01:09:41 +0000 (18:09 -0700)]
Exclude ResolveEventTests.cmd and CallbackTests.cmd on Windows_NT x86 (#23942)

5 years agoMerge pull request #23935 from BruceForstall/FixArm64LargeOutgoingWithLocalloc
Bruce Forstall [Sat, 13 Apr 2019 01:04:41 +0000 (18:04 -0700)]
Merge pull request #23935 from BruceForstall/FixArm64LargeOutgoingWithLocalloc

Fix arm64 localloc with large outgoing argument space

5 years agoFix EventPipe IPC streaming (#23768)
Sung Yoon Whang [Sat, 13 Apr 2019 00:11:49 +0000 (17:11 -0700)]
Fix EventPipe IPC streaming (#23768)

* Make events stream out faster when we are doing IPC

* Block size should be less during IPC session

* Modify some comments

* Address PR feedback

* more pr feedback

* Fix high volume scenario

* Fix flush

* cleanup

* Only flush when we have buffers written

5 years agoFix x86 stack probing (#23881)
Bruce Forstall [Fri, 12 Apr 2019 23:54:55 +0000 (16:54 -0700)]
Fix x86 stack probing (#23881)

* Fix x86 stack probing

On x86, structs are passed by value on the stack. We copy structs
to the stack in various ways, but one way is to first subtract the
size of the struct and then use a "rep movsb" instruction. If the
struct we are passing is sufficiently large, this can cause us to
miss the stack guard page.

So, introduce stack probes for these struct copies.

It turns out the stack pointer after prolog probing can be sitting
near the very end of the guard page (one `STACK_ALIGN` slot before
the end, which allows a "call" instruction which pushes its
return address to touch the guard page with the return address push).
We don't want to probe with every argument push, though. So change
the prolog probing to insert an "extra" touch at the final SP location
if the previous touch was "too far" away, leaving at least some
buffer zone for un-probed SP adjustments. I chose this to be the
size of the largest SIMD register, which also can get copied to the
argument stack with a "SUB;MOV" sequence.

Added several test case variations showing different large stack
probe situations.

Fixes #23796

* Increase the argument size probe buffer

* Formatting

5 years ago[master] Update dependencies from dotnet/corefx (#23893)
dotnet-maestro[bot] [Fri, 12 Apr 2019 23:40:50 +0000 (16:40 -0700)]
[master] Update dependencies from dotnet/corefx (#23893)

* Update dependencies from https://github.com/dotnet/corefx build 20190411.2

- Microsoft.NETCore.Platforms - 3.0.0-preview5.19211.2
- Microsoft.Private.CoreFx.NETCoreApp - 4.6.0-preview5.19211.2

* Update dependencies from https://github.com/dotnet/corefx build 20190411.22

- Microsoft.NETCore.Platforms - 3.0.0-preview5.19211.22
- Microsoft.Private.CoreFx.NETCoreApp - 4.6.0-preview5.19211.22

* Removing renamed HWIntrinsic APIs from S.P.Corelib

* Updating the hwintrinsiclistxarch to match the new intrinsic API names

* Updating the HWIntrinsic generated tests to use the new API names

5 years agoAdd R2R prefix to TestRunNamePrefix when running ReadyToRun jobs (#23945)
Egor Chesakov [Fri, 12 Apr 2019 23:33:54 +0000 (16:33 -0700)]
Add R2R prefix to TestRunNamePrefix when running ReadyToRun jobs (#23945)

5 years agoFix OSX debugging (#23924)
Jan Vorlicek [Fri, 12 Apr 2019 21:55:07 +0000 (23:55 +0200)]
Fix OSX debugging (#23924)

A recent change has incorrectly added dependency on libcoreclrpal.a to
libutilcodenohost.a. This in turn, due to the transitive propagation of
dependencies, caused the libmscordbi to be linked with libcoreclrpal.a.
So libmscordbi contained its own PAL after that change, which is wrong.
libmscordbi should depend on PAL APIs through libmscordaccore that
exports the symbols it needs for it in order to keep just one PAL in the
process.

The fix is to remove the libcoreclrpal.a dependency and export few
new symbols from libmscordaccore instead.

5 years agoFix DIM Test to run on mono (#23870)
Thays Grazia [Fri, 12 Apr 2019 21:49:26 +0000 (18:49 -0300)]
Fix DIM Test to run on mono (#23870)

Changing the test to run on runtimes that supports lookups with runtime determined boxing and on runtimes that doesn't support.

5 years agoDisplay the arm instruction size
Bruce Forstall [Fri, 12 Apr 2019 20:19:14 +0000 (13:19 -0700)]
Display the arm instruction size

When dumping an arm instruction, and when we don't have the actual
code bytes (such as during code generation, as opposed to emission),
display the (current) instruction size, either 2 bytes (2B),
4 bytes (4B), or 6 bytes (6B).

5 years agoUpdate maximum allowed arm prolog size
Bruce Forstall [Fri, 12 Apr 2019 20:15:42 +0000 (13:15 -0700)]
Update maximum allowed arm prolog size

Change #23715 changed the localloc stack probe loop to load a
negative constant instead of zero to start the probing loop.
This increased the size of the probing loop, and hence the
size of the maximum prolog, by 4 bytes. Bump the assert on
the maximum size to match.

Note that as the comment says, the maximum size there is not
actually a maximum (it was originally), it is just to alert
us when the maximum generated prolog size has gone up.

Fixes #23920

5 years agoDon't Free UpperVector (#23889)
Carol Eidt [Fri, 12 Apr 2019 18:31:55 +0000 (11:31 -0700)]
Don't Free UpperVector (#23889)

* Don't Free UpperVector

UpperVector regs are freed at their time of use, and shouldn't be freed when the last RefPosition is encountered.
Also, we need to specify all 3 operands for vinsertf128

Fix #23861
Fix #23904
Fix #23804

5 years agoFix arm64 localloc with large outgoing argument space
Bruce Forstall [Fri, 12 Apr 2019 17:50:56 +0000 (10:50 -0700)]
Fix arm64 localloc with large outgoing argument space

With a large outgoing argument space, we were attempting to
generate un-encodable ADD and SUB instructions. Call an existing
helper instead that uses the reserved register when needed to
load large constants into a register.

5 years agoConsolidate duplicate assignment (#23923)
bdebaere [Fri, 12 Apr 2019 16:29:44 +0000 (18:29 +0200)]
Consolidate duplicate assignment (#23923)

* Consolidate duplicate assignment.

5 years agoMerge pull request #23900 from sergiy-k/fixcrossarchr2r1
Sergiy Kuryata [Fri, 12 Apr 2019 16:00:04 +0000 (09:00 -0700)]
Merge pull request #23900 from sergiy-k/fixcrossarchr2r1

Enable precompiling CoreRoot overlay with crossgen for Linux arm/arm64

5 years agoAdd several links to deep-dive blog posts
Stephen Toub [Fri, 12 Apr 2019 13:39:40 +0000 (09:39 -0400)]
Add several links to deep-dive blog posts

5 years agoJIT: update byref null check logic to handle field chains (#23850)
Andy Ayers [Fri, 12 Apr 2019 07:44:39 +0000 (00:44 -0700)]
JIT: update byref null check logic to handle field chains (#23850)

The jit was not properly accumulating offsets when figuring out if a byref
should be null checked.

Use a non-null MorphAddressContext as indication that GT_FIELD and GT_IND
nodes are actually part of an ongoing address computation.

During field morphing propagate the current address context to the child node,
instead of starting a new one.

Fixes #23791.

5 years agoEnable precompiling CoreRoot overlay with crossgen for Linux arm/arm64
Sergiy Kuryata [Thu, 11 Apr 2019 16:54:49 +0000 (09:54 -0700)]
Enable precompiling CoreRoot overlay with crossgen for Linux arm/arm64

5 years agoEnable default Ready2Run testing in pull requests. Workaround an issue with custom...
Egor Chesakov [Fri, 12 Apr 2019 05:50:48 +0000 (22:50 -0700)]
Enable default Ready2Run testing in pull requests. Workaround an issue with custom xunit.console.dll runner. (#23848)

5 years agoRemove duplicates. (#23922)
bdebaere [Fri, 12 Apr 2019 05:32:56 +0000 (07:32 +0200)]
Remove duplicates. (#23922)

5 years agoAdd optimized UTF-8 validation and transcoding apis, hook them up to UTF8Encoding...
Levi Broderick [Fri, 12 Apr 2019 03:50:16 +0000 (20:50 -0700)]
Add optimized UTF-8 validation and transcoding apis, hook them up to UTF8Encoding (#21948)

* Add optimized UTF-8 validation and transcoding logic
- Hook it up through the existing Utf8 public static APIs
- Move some shared methods out of ASCIIUtility
- Hook it up through the Utf8String ctor

* Hook up new UTF-8 logic through UTF8Encoding
- Add vectorized UTF-16 validation and transcoded byte counts
- Move Utf16Utility into Unicode namespace alongside Utf8Utility
- Fix some bugs in DecoderNLS's draining logic

* Improve perf of "is ASCII?" inner loop in UTF-8 validation.

* Remove SSE41.X64 optimization from AsciiUtility
RyuJIT now handles this optimally

* Clarify that vector read is unaligned

* Simplify vectorized logic; remove unnecessary adjustment

* PR feedback: GetElement(0) -> Sse2.StoreLow

* PR feedback
- Simplify CountNumberOfLeadingAsciiBytesFrom24BitInteger
- Extract some consts out to top of file w/ comments

* PR feedback: Enable SSE2 in Utf16Utility code

* Expand masks in Utf8Utility, fix const in fallback path

* Temporarily disable failing CoreFX tests

* Fix incorrect Debug.Assert statements

* Add comments tracking JIT workarounds.

* Rename DWORD -> UInt32 throughout API surface

* Re-flow Utf8Utility.Helpers

* PR feedback: Fix typos

* PR feedback: CountNumberOfLeadingAsciiBytesFrom24BitInteger

* PR feedback: Remove redundant endianess checks

* PR feedback: Validate nint definitions

* PR feedback: Clarify charIsNonAscii vector usage

* PR feedback: document tempUtf8CodeUnitCountAdjustment usage

* Fix compilation failure in Utf16Utility

* PR feedback: Clarify 3-byte sequence processing

* Add missing check to 3-byte processing logic

* Clarify comment in 3-byte processing

5 years agoFix Assembly.LoadFromPartialName(string) (#23912)
Steve MacLean [Fri, 12 Apr 2019 03:48:59 +0000 (23:48 -0400)]
Fix Assembly.LoadFromPartialName(string)  (#23912)

Assembly.LoadFromPartialName(string) should not throw FileNotFoundException, but
should rather return null.

ArgumentException should use nameof(partialName)

Fixes #19817

5 years agoUpdate CoreClr to preview5-27611-71 (#23884)
dotnet-maestro-bot [Fri, 12 Apr 2019 00:47:59 +0000 (17:47 -0700)]
Update CoreClr to preview5-27611-71 (#23884)

5 years agoFix Version's comparer operators to allow null (#23898)
Stephen Toub [Fri, 12 Apr 2019 00:47:35 +0000 (20:47 -0400)]
Fix Version's comparer operators to allow null (#23898)

Previously for `<`/`<=` it would throw if `v1` argument was null, but for `>`/`>=` would throw if the `v1` argument was null, albeit using `v1` in the exception parameter name.  It's also not expected for comparisons like this to throw, in general.

5 years agoFix 23740 (#23906)
Steve MacLean [Fri, 12 Apr 2019 00:28:25 +0000 (20:28 -0400)]
Fix 23740 (#23906)

Adjust AssemblyLoadContextBaseObject packing for
windows arm.

5 years agoJIT: fix assert when there are mixed types in Enum.HasFlags optimization (#23902)
Andy Ayers [Thu, 11 Apr 2019 23:45:30 +0000 (16:45 -0700)]
JIT: fix assert when there are mixed types in Enum.HasFlags optimization (#23902)

In some cases the pre-boxed nodes may differ in type. Bail if they don't have
the same stack type, then compute the result using the stack type.

Extended the hasflags test with a case that shows this issue.

Closes #23847

5 years agoClarify comment in 3-byte processing
Levi Broderick [Thu, 11 Apr 2019 22:36:47 +0000 (15:36 -0700)]
Clarify comment in 3-byte processing

5 years agoAdd missing check to 3-byte processing logic
Levi Broderick [Thu, 11 Apr 2019 22:30:11 +0000 (15:30 -0700)]
Add missing check to 3-byte processing logic

5 years agoPR feedback: Clarify 3-byte sequence processing
Levi Broderick [Thu, 11 Apr 2019 20:39:55 +0000 (13:39 -0700)]
PR feedback: Clarify 3-byte sequence processing

5 years agoDisable nullability-related warnings
Jan Kotas [Wed, 10 Apr 2019 14:03:22 +0000 (07:03 -0700)]
Disable nullability-related warnings

Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
5 years agoFix compilation failure in Utf16Utility
Levi Broderick [Thu, 11 Apr 2019 20:19:21 +0000 (13:19 -0700)]
Fix compilation failure in Utf16Utility

5 years agoPR feedback: document tempUtf8CodeUnitCountAdjustment usage
Levi Broderick [Thu, 11 Apr 2019 18:48:44 +0000 (11:48 -0700)]
PR feedback: document tempUtf8CodeUnitCountAdjustment usage

5 years agoPR feedback: Clarify charIsNonAscii vector usage
Levi Broderick [Thu, 11 Apr 2019 18:37:00 +0000 (11:37 -0700)]
PR feedback: Clarify charIsNonAscii vector usage

5 years agoPR feedback: Validate nint definitions
Levi Broderick [Thu, 11 Apr 2019 18:23:53 +0000 (11:23 -0700)]
PR feedback: Validate nint definitions

5 years agoPR feedback: Remove redundant endianess checks
Levi Broderick [Thu, 11 Apr 2019 18:14:16 +0000 (11:14 -0700)]
PR feedback: Remove redundant endianess checks

5 years agoPR feedback: CountNumberOfLeadingAsciiBytesFrom24BitInteger
Levi Broderick [Thu, 11 Apr 2019 18:11:49 +0000 (11:11 -0700)]
PR feedback: CountNumberOfLeadingAsciiBytesFrom24BitInteger

5 years agoPR feedback: Fix typos
Levi Broderick [Thu, 11 Apr 2019 17:56:34 +0000 (10:56 -0700)]
PR feedback: Fix typos

5 years agoUpdate dependencies from https://github.com/dotnet/core-setup build 20190410.11 ...
dotnet-maestro[bot] [Thu, 11 Apr 2019 17:56:27 +0000 (13:56 -0400)]
Update dependencies from https://github.com/dotnet/core-setup build 20190410.11 (#23894)

- Microsoft.NETCore.App - 3.0.0-preview5-27610-11

5 years ago[master] Update dependencies from dotnet/arcade (#23834)
dotnet-maestro[bot] [Thu, 11 Apr 2019 16:59:29 +0000 (12:59 -0400)]
[master] Update dependencies from dotnet/arcade (#23834)

* Update dependencies from https://github.com/dotnet/arcade build 20190409.1

- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19209.1
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19209.1

* Update dependencies from https://github.com/dotnet/arcade build 20190409.2

- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19209.2
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19209.2

* Update dependencies from https://github.com/dotnet/arcade build 20190410.7

- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19210.7
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19210.7

5 years agoRenaming 'failure' in counter tests (#23891)
Sung Yoon Whang [Thu, 11 Apr 2019 11:52:48 +0000 (04:52 -0700)]
Renaming 'failure' in counter tests (#23891)

5 years agoIf a delegate type is a WinRT delegate or a WinRT-projected delegate, allow default...
Jeremy Koritzinsky [Thu, 11 Apr 2019 06:31:54 +0000 (23:31 -0700)]
If a delegate type is a WinRT delegate or a WinRT-projected delegate, allow default marshalling. (#23877)

5 years agoMerge remote-tracking branch 'origin/master' into utf8_validation_apis_3
Levi Broderick [Thu, 11 Apr 2019 05:56:33 +0000 (22:56 -0700)]
Merge remote-tracking branch 'origin/master' into utf8_validation_apis_3

5 years agoRemove overrelease in test (#23887)
Aaron Robinson [Thu, 11 Apr 2019 05:54:46 +0000 (22:54 -0700)]
Remove overrelease in test (#23887)

5 years agoContextual reflection (#23740)
Steve MacLean [Thu, 11 Apr 2019 04:20:39 +0000 (00:20 -0400)]
Contextual reflection (#23740)

* Add ContextualReflection APIs

Add ContextualReflection APIs approved in dotnet/corefx#36236
Fix issue #22213

* SetParentAssembly even when IsCollectible()
* ContextualReflection tests

* PR Feedback
* Add more usage tests

Add using statement tests
Add bad usage tests including Assert.Throws<>

* Only initialize on set
* Add XML API comments
* Unify VerifyIsolation
* Fix unused expectedAssembly
* Remove ContextualReflectionScope throw
* Clean up TestResolveMissingAssembly et. al
* Remove unused QCall::AppDomainHandle
* Remove AppDomainBaseObject
* Pass AssemblyLoadContext as managed object to native
* Fix AssemblyLoadContextBaseObject packing

* AssemblyLoadContext backing stores

Use explicit backing stores for events and properties

* Remove StaticAsyncLocalCurrentContextualReflectionContext
* Remove PermissionSetObject

5 years agoMerge pull request #23824 from janvorli/remove_unix_cpu_groups_emulation
Jan Vorlicek [Thu, 11 Apr 2019 00:26:32 +0000 (02:26 +0200)]
Merge pull request #23824 from janvorli/remove_unix_cpu_groups_emulation

Remove Unix CPU groups emulation

5 years agoRe-flow Utf8Utility.Helpers
Levi Broderick [Wed, 10 Apr 2019 23:52:48 +0000 (16:52 -0700)]
Re-flow Utf8Utility.Helpers

5 years agoRename DWORD -> UInt32 throughout API surface
Levi Broderick [Wed, 10 Apr 2019 23:51:30 +0000 (16:51 -0700)]
Rename DWORD -> UInt32 throughout API surface

5 years agoAdd comments tracking JIT workarounds.
Levi Broderick [Wed, 10 Apr 2019 23:47:11 +0000 (16:47 -0700)]
Add comments tracking JIT workarounds.

5 years agoMove AssemblyName to shared (#23737)
Egor Bogatov [Wed, 10 Apr 2019 23:36:14 +0000 (02:36 +0300)]
Move AssemblyName to shared (#23737)

5 years agoMove TimerQueue.TickCount Windows implementation to shared CoreLib (#23838)
Filip Navara [Wed, 10 Apr 2019 23:04:38 +0000 (01:04 +0200)]
Move TimerQueue.TickCount Windows implementation to shared CoreLib (#23838)

5 years agoMake GC.GetGCMemoryInfo public (#23779)
Ludovic Henry [Wed, 10 Apr 2019 22:57:00 +0000 (15:57 -0700)]
Make GC.GetGCMemoryInfo public (#23779)

* Make GC.GetGCMemoryInfo public

This is to be used to allow users to optimize memory consumption based
on what's available to the GC and on the system.

This is based on https://github.com/dotnet/corefx/issues/34631

* Address reviews

- Pass GCMemoryInfo struct to native in place of many arguments
- Make GCMemoryInfo.HeapSize and GCMemoryInfo.Fragmentation long in place of IntPtr

* Address reviews

- Mark GCMemoryInfo readonly

* Rearrange fields and match managed/native names

* Fix compilation

* Use getter shorthand

* Address API Design Review

https://github.com/dotnet/corefx/issues/34631#issuecomment-481358549

* Fix comments

* Fix comments

* Do everything in managed

* Address review

 - Remove unecessary [StructLayout.Sequential]
 - Remove "_" prefix for parameters variables

5 years agoMerge pull request #23876 from briansull/jit-dump
Brian Sullivan [Wed, 10 Apr 2019 22:19:11 +0000 (15:19 -0700)]
Merge pull request #23876 from briansull/jit-dump

Improved JitDump

5 years agoMerge pull request #23858 from sandreenko/fixOptPrepareTreeToReplacement
Sergey Andreenko [Wed, 10 Apr 2019 22:15:12 +0000 (15:15 -0700)]
Merge pull request #23858 from sandreenko/fixOptPrepareTreeToReplacement

Clean optVNConstantPropOnTree.

5 years agoCheck for a null import descriptor when fixing up vtables. (#23825)
Jeremy Koritzinsky [Wed, 10 Apr 2019 19:54:27 +0000 (12:54 -0700)]
Check for a null import descriptor when fixing up vtables. (#23825)

A fragile ngened image doesn't have an import descriptor directory in its PE image. Without this check, we AV when loading a fragile ngened image.

5 years agojit-format
Brian Sullivan [Wed, 10 Apr 2019 19:52:14 +0000 (12:52 -0700)]
jit-format

5 years agoImproved JitDump
Brian Sullivan [Wed, 10 Apr 2019 19:37:21 +0000 (12:37 -0700)]
Improved JitDump
- Prints the bbNatLoopNum when dumping the BasicBlocks
- Refactor the GenTree node dump to use gtDispCommonEndLine:
//     Utility function that prints the following node information
//       1: The associated zero field sequence (if any)
//       2. The register assigned to this node (if any)
//       2. The value number assigned (if any)
//       3. A newline character

5 years agoUpdate the comment.
Sergey Andreenko [Wed, 10 Apr 2019 19:29:46 +0000 (12:29 -0700)]
Update the comment.

5 years agoUpdate Range.GetOffsetAndLength (#23855)
Tarek Mahmoud Sayed [Wed, 10 Apr 2019 19:05:27 +0000 (20:05 +0100)]
Update Range.GetOffsetAndLength (#23855)

* Update Range.GetOffsetAndLength

Make it return offset and length tuple instead of the old Type Range.OffsetAndLength

* Fix the return Tuple fields names

5 years agoTweaking some APIs according to user feedback. (#23857)
Tanner Gooding [Wed, 10 Apr 2019 17:42:17 +0000 (10:42 -0700)]
Tweaking some APIs according to user feedback. (#23857)

* Updating the Avx2.ConvertToVector256UInt APIs to return a signed type (matching the native signature and Sse41 APIs).

* Adding explicit overloads for `Sse41.ConvertToVector128Int*` and `Avx2.ConvertToVector256Int*` that take a pointer

* Updating the word ordering for the Compare/CompareScalar methods.

5 years agoUpdate dependencies from https://github.com/dotnet/core-setup build 20190409.17 ...
dotnet-maestro[bot] [Wed, 10 Apr 2019 16:38:40 +0000 (12:38 -0400)]
Update dependencies from https://github.com/dotnet/core-setup build 20190409.17 (#23865)

- Microsoft.NETCore.App - 3.0.0-preview5-27609-17

5 years agoUpdate CoreClr to preview5-27610-71 (#23859)
dotnet-maestro-bot [Wed, 10 Apr 2019 16:38:28 +0000 (09:38 -0700)]
Update CoreClr to preview5-27610-71 (#23859)

5 years agoUpdate Windows instructions (#23841)
Jared Parsons [Wed, 10 Apr 2019 13:57:38 +0000 (06:57 -0700)]
Update Windows instructions (#23841)

The instructions for CMake failed to list 3.14.1 when using Visual Studio 2019.

5 years agoFix few issues discovered by testing
Jan Vorlicek [Wed, 10 Apr 2019 13:18:03 +0000 (15:18 +0200)]
Fix few issues discovered by testing

5 years agoUpdate profiling-api-status.md
Noah Falk [Wed, 10 Apr 2019 11:05:03 +0000 (04:05 -0700)]
Update profiling-api-status.md

Clarifying that profiler attach/detach are a known gap

5 years agoFix RecycledLists size on Unix
Jan Vorlicek [Wed, 10 Apr 2019 10:08:27 +0000 (12:08 +0200)]
Fix RecycledLists size on Unix

The list size was set to g_SystemInfo.dwNumberOfProcessors which is a
number of processors the current process is allowed to run on, but not
the total number of processors in the system. Fixed to use
PAL_GetTotalCpuCount.

Also revert a change to the mbind node mask length computation I've
incorrectly made in my last commit and make it clear that the value is
a number of used bits in the node mask, which is the highest numa node
plus 1. And finally, re-reading the mbind doc, I've found that the
maxnode parameter is in fact "number of nodes" in the mask, so fixing
that too.

5 years agoChange optExtractSideEffList to optExtractSideEffList.
Sergey Andreenko [Wed, 10 Apr 2019 08:59:31 +0000 (01:59 -0700)]
Change optExtractSideEffList to optExtractSideEffList.

5 years agoFix newlines for ilegal trees in `gtDispTree`.
Sergey Andreenko [Wed, 10 Apr 2019 08:57:04 +0000 (01:57 -0700)]
Fix newlines for  ilegal trees in `gtDispTree`.

5 years agoExpose new EventCounter APIs (#23808)
Sung Yoon Whang [Wed, 10 Apr 2019 08:19:30 +0000 (01:19 -0700)]
Expose new EventCounter APIs (#23808)

* rename BaseCounter to DiagnosticCounter

* Change MetaData->Metadata

* Make EventSource and Name a property for counter classes

* Make the counter APIs public

* fix build errors

* Change float to double

* Few cleanups, fix test

* fix GetMetadataString

* PR feedback

* More PR feedback

5 years agoFix incorrect Debug.Assert statements
Levi Broderick [Wed, 10 Apr 2019 06:07:05 +0000 (23:07 -0700)]
Fix incorrect Debug.Assert statements

5 years agoTemporarily disable failing CoreFX tests
Levi Broderick [Wed, 10 Apr 2019 04:46:37 +0000 (21:46 -0700)]
Temporarily disable failing CoreFX tests

5 years agoRemove dlclose call to unload CoreCLR (#23863)
Next Turn [Wed, 10 Apr 2019 04:18:24 +0000 (12:18 +0800)]
Remove dlclose call to unload CoreCLR (#23863)

5 years agoAdd back using to EventSourceException
Santiago Fernandez Madero [Fri, 5 Apr 2019 23:32:35 +0000 (16:32 -0700)]
Add back using to EventSourceException

Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
5 years agoUpdate CoreRT for changes in shared partition
Jan Kotas [Fri, 5 Apr 2019 04:49:22 +0000 (21:49 -0700)]
Update CoreRT for changes in shared partition

Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
5 years agoExpand masks in Utf8Utility, fix const in fallback path
Levi Broderick [Wed, 10 Apr 2019 01:48:01 +0000 (18:48 -0700)]
Expand masks in Utf8Utility, fix const in fallback path

5 years agoAdd 'SetNotificationForWaitCompletion' to ILLinkTrim.xml (#23822)
Gregg Miskelly [Wed, 10 Apr 2019 02:17:39 +0000 (19:17 -0700)]
Add 'SetNotificationForWaitCompletion' to ILLinkTrim.xml (#23822)

SetNotificationForWaitCompletion is meant to be func-eval'ed by the debugger to enable async step out. But this method was being removed by the IL linker. This adds it back.

5 years agoPR feedback: Enable SSE2 in Utf16Utility code
Levi Broderick [Wed, 10 Apr 2019 00:31:39 +0000 (17:31 -0700)]
PR feedback: Enable SSE2 in Utf16Utility code