Egor Chesakov [Tue, 25 Jun 2019 00:09:14 +0000 (17:09 -0700)]
Add /eng/pipelines/*.yml (#25325)
Jarret Shook [Mon, 24 Jun 2019 23:18:14 +0000 (16:18 -0700)]
Throw platform unsupported error for windows arm32 varargs (#25320)
* Throw unsupported platform for native vcarargs on Arm32 Windows
* Re-enable baseservices vararg test
Tanner Gooding [Mon, 24 Jun 2019 22:39:50 +0000 (15:39 -0700)]
Updating Dragon4 to ensure the number buffer always provides a significant digit if one exists. (#25096)
* Updating Dragon4 to ensure the number buffer always provides a significant digit if one exists.
* Changing System.Number.RoundNumber to not round up floating-point numbers.
* Re-enabling the RealFormatterTestsBase CoreFX tests
* Updating Number.RoundNumber to take a isCorrectlyRounded parameter and to use IEEE compliant rounding for floating-point numbers.
* Change SinglePrecisionCustomFormat to 7, ensuring it matches the value used in netcoreapp2.1
Andrii Siriak [Fri, 21 Jun 2019 22:25:09 +0000 (01:25 +0300)]
Switch from IndexOf(string) to Contains(string) for better readability and performance (#38757)
* Fix performance IndexOf(string) -> Contains(string)
* Make comparison ordinal
Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
Ganbarukamo41 [Mon, 24 Jun 2019 19:44:50 +0000 (04:44 +0900)]
Make Vector*.ToString() Culture invariant (#25343)
* Remove IFormattable
* Hardcode characters
* Add tests
José Rivero [Mon, 24 Jun 2019 19:18:54 +0000 (12:18 -0700)]
Multiple bug fixes (#25308)
- Fixes EventPipe does not properly dispose of itself on an active session error #25228
- On EventPipeSession::ThreadProc, if an error occurs on the IPC streaming, then EventPipe::Disable will be invoked.
- Fixes Prevent EventPipe Sessions IDs from being reused on disconnect #25229
Michal Strehovský [Mon, 24 Jun 2019 19:17:01 +0000 (21:17 +0200)]
Remove AggressiveOptimization flag from System.Text (#25356)
Based on conversation with Levi, I don't think this flag is needed.
The flag currently controls two things:
* Disqualifies the method from precompilation
* Disables Tier-0 JIT for the method
Removing the attribute allows us to precompile some of the methods (not all, because of `Vector<T>`). It measurably helps improve startup time because many of the attributed methods are complex and take a while to compile.
Once tier-1 JIT kicks in, the generated code should be identical to what we were getting before.
Andy Ayers [Mon, 24 Jun 2019 18:19:58 +0000 (11:19 -0700)]
JIT: see if jmp offset will fit in 32 bit displacement (#25348)
On x86 we can always reach any IAT_PVALUE entry via a 32 bit indirect
jump, and proper handling of jmp epilogs depends on this. So check if the
target address is reachable in 32 bits, and if so, use the jmp [addr] form
on x86, and the jmp [rip + disp] form for x64.
Fixes #25345
Fixes #25346
Undoes the assertion change from #25302
Fixes #25286
Tomáš Rylek [Mon, 24 Jun 2019 18:06:42 +0000 (11:06 -0700)]
Update R2R format doc to 3.0 (#25297)
I have updated the various enumerations and tables. I didn't try
to go overboard w.r.t. understanding tables I'm not yet familiar
with, most notably the diagnostic, inlining and profiling tables.
dotnet-maestro[bot] [Mon, 24 Jun 2019 14:49:06 +0000 (07:49 -0700)]
Update dependencies from https://dev.azure.com/dnceng/internal/_git/dotnet-optimization build
20190624.1 (#25354)
- optimization.IBC.CoreCLR - 99.99.99-master-
20190624.1
- optimization.PGO.CoreCLR - 99.99.99-master-
20190624.1
dotnet-maestro[bot] [Mon, 24 Jun 2019 14:48:42 +0000 (07:48 -0700)]
Update dependencies from https://github.com/dotnet/arcade build
20190623.4 (#25351)
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19323.4
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19323.4
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19323.4
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19323.4
dotnet-maestro[bot] [Mon, 24 Jun 2019 14:45:06 +0000 (07:45 -0700)]
Update dependencies from https://github.com/dotnet/core-setup build
20190623.05 (#25353)
- Microsoft.NETCore.App - 3.0.0-preview7-27823-05
dotnet-maestro[bot] [Mon, 24 Jun 2019 14:38:15 +0000 (07:38 -0700)]
Update dependencies from https://github.com/dotnet/corefx build
20190623.4 (#25352)
- Microsoft.Bcl.AsyncInterfaces - 1.0.0-preview7.19323.4
- Microsoft.NETCore.Platforms - 3.0.0-preview7.19323.4
- Microsoft.Private.CoreFx.NETCoreApp - 4.6.0-preview7.19323.4
Michal Strehovský [Mon, 24 Jun 2019 14:00:29 +0000 (16:00 +0200)]
Extend the assert for zero-offset fields (#25341)
CoreRT has an [actual field](https://github.com/dotnet/corert/blob/
ba3ba929c88bd80ec5d4dd9cf6917471ad8fa180/src/System.Private.CoreLib/src/System/Object.CoreRT.cs#L30) for the method table table pointer in `System.Object`. CoreCLR does shenanigans when laying out `System.Object` to leave a pointer-sized hole instead.
I think the zero-offset field is triggering these asserts.
Stephen Toub [Mon, 24 Jun 2019 13:27:47 +0000 (09:27 -0400)]
Force secondary await continuations to run asynchronously (#25280)
* Force secondary await continuations to run asynchronously
For performance reasons, await continuations have been invoked synchronously, meaning they're invoked as part of the antecedent task's completion (as long as that task allows it, as long as there's sufficient stack space, etc.) This generally works out well in the case where there's a single await continuation, which is far and away the common case. However, it can cause problems in situations where there are multiple await continuations, as those continuations will end up being serialized, which can lead to slowdowns and deadlocks in niche situations. To address that, this commit backs off a bit. The first await continuation is still invoked synchronously, but subsequent await continuations are invoked asynchronously, such that they are not blocked by a previously registered await continuation.
* Fix nits
dotnet-maestro[bot] [Sun, 23 Jun 2019 16:05:06 +0000 (09:05 -0700)]
Update dependencies from https://github.com/dotnet/core-setup build
20190622.05 (#25338)
- Microsoft.NETCore.App - 3.0.0-preview7-27822-05
Marco Rossignoli [Sun, 23 Jun 2019 15:30:58 +0000 (17:30 +0200)]
MemoryExtensions ToUpper / ToLower throws for overlapping buffer (#25327)
dotnet-maestro[bot] [Sun, 23 Jun 2019 15:30:10 +0000 (08:30 -0700)]
Update dependencies from https://dev.azure.com/dnceng/internal/_git/dotnet-optimization build
20190623.1 (#25339)
- optimization.IBC.CoreCLR - 99.99.99-master-
20190623.1
- optimization.PGO.CoreCLR - 99.99.99-master-
20190623.1
dotnet-maestro[bot] [Sun, 23 Jun 2019 15:29:57 +0000 (08:29 -0700)]
Update dependencies from https://github.com/dotnet/arcade build
20190622.2 (#25336)
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19322.2
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19322.2
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19322.2
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19322.2
dotnet-maestro[bot] [Sun, 23 Jun 2019 15:27:00 +0000 (08:27 -0700)]
Update dependencies from https://github.com/dotnet/corefx build
20190622.7 (#25337)
- Microsoft.Bcl.AsyncInterfaces - 1.0.0-preview7.19322.7
- Microsoft.NETCore.Platforms - 3.0.0-preview7.19322.7
- Microsoft.Private.CoreFx.NETCoreApp - 4.6.0-preview7.19322.7
Roman Marusyk [Sun, 23 Jun 2019 02:55:17 +0000 (05:55 +0300)]
Change exception throwing in Path.GetRelativePath (#25047)
* Fix Path.GetRelativePath throws ArgumentNullException when relativeTo is empty string
* Fix ArgumentException for relativeTo
Herman Eldering [Sun, 23 Jun 2019 02:52:42 +0000 (04:52 +0200)]
ArgumentException: use system-supplied message if message is null (#25115)
Enable using a system-supplied message while also providing an argument/parameter name by passing in a null message. Made to match the behavior of FileNotFoundException.
Andy Ayers [Sun, 23 Jun 2019 02:06:27 +0000 (19:06 -0700)]
Fix ReturnKind classification for byref like types returned in registers (#25333)
These types contain byrefs, and so when returned in registers we may need
to avoid GC stress at the return site.
Addresses part of #24263.
dotnet-maestro[bot] [Sat, 22 Jun 2019 16:43:47 +0000 (09:43 -0700)]
Update dependencies from https://github.com/dotnet/arcade build
20190621.75 (#25328)
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19321.75
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19321.75
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19321.75
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19321.75
dotnet-maestro[bot] [Sat, 22 Jun 2019 16:43:36 +0000 (09:43 -0700)]
Update dependencies from https://github.com/dotnet/corefx build
20190621.8 (#25329)
- Microsoft.Bcl.AsyncInterfaces - 1.0.0-preview7.19321.8
- Microsoft.NETCore.Platforms - 3.0.0-preview7.19321.8
- Microsoft.Private.CoreFx.NETCoreApp - 4.6.0-preview7.19321.8
dotnet-maestro[bot] [Sat, 22 Jun 2019 16:43:24 +0000 (09:43 -0700)]
Update dependencies from https://github.com/dotnet/core-setup build
20190621.07 (#25330)
- Microsoft.NETCore.App - 3.0.0-preview7-27821-07
dotnet-maestro[bot] [Sat, 22 Jun 2019 16:07:02 +0000 (16:07 +0000)]
Update dependencies from https://dev.azure.com/dnceng/internal/_git/dotnet-optimization build
20190622.1 (#25331)
- optimization.IBC.CoreCLR - 99.99.99-master-
20190622.1
- optimization.PGO.CoreCLR - 99.99.99-master-
20190622.1
Leandro A. F. Pereira [Sat, 22 Jun 2019 08:52:48 +0000 (01:52 -0700)]
Fix generation of RID on distros that do not set VERSION_ID (#25034)
Rolling-release distros do not set this variable, leading to invalid
RIDs such as "gentoo.-x64", which causes build failures.
Omit VERSION_ID (and the period preceding it) when unset, so it's
consistent with host and PlatformAbstractions.
Fixes #19769.
dotnet-maestro[bot] [Sat, 22 Jun 2019 08:33:30 +0000 (01:33 -0700)]
[master] Update dependencies from dotnet/corefx (#25120)
* Update dependencies from https://github.com/dotnet/corefx build
20190621.1
- Microsoft.Bcl.AsyncInterfaces - 1.0.0-preview7.19321.1
- Microsoft.NETCore.Platforms - 3.0.0-preview7.19321.1
- Microsoft.Private.CoreFx.NETCoreApp - 4.6.0-preview7.19321.1
* Disable failing tests
Tamar Christina [Sat, 22 Jun 2019 07:47:20 +0000 (08:47 +0100)]
Arm64: Correct encoding of fabd (#25150)
Maoni Stephens [Sat, 22 Jun 2019 01:06:39 +0000 (18:06 -0700)]
don't require seg size to be power of 2 for large pages (#25216)
large pages will have segments aligned to 16mb (the default min seg size for hardlimit)
dotnet-maestro[bot] [Sat, 22 Jun 2019 00:19:33 +0000 (17:19 -0700)]
Update dependencies from https://github.com/dotnet/arcade build
20190620.1 (#25311)
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19320.1
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19320.1
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19320.1
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19320.1
dotnet-maestro[bot] [Sat, 22 Jun 2019 00:19:00 +0000 (17:19 -0700)]
Update dependencies from https://dev.azure.com/dnceng/internal/_git/dotnet-optimization build
20190621.1 (#25314)
- optimization.IBC.CoreCLR - 99.99.99-master-
20190621.1
- optimization.PGO.CoreCLR - 99.99.99-master-
20190621.1
dotnet-maestro[bot] [Sat, 22 Jun 2019 00:18:41 +0000 (17:18 -0700)]
[master] Update dependencies from dotnet/core-setup (#25254)
* Update dependencies from https://github.com/dotnet/core-setup build
20190620.07
- Microsoft.NETCore.App - 3.0.0-preview7-27820-07
Brian Sullivan [Fri, 21 Jun 2019 22:56:36 +0000 (15:56 -0700)]
Merge pull request #25260 from briansull/GitHub_25134
Fix Issue #25134 - AssertionProp incorrectly removes cast from uint
Noah Falk [Fri, 21 Jun 2019 22:27:21 +0000 (15:27 -0700)]
Add EventPipe Processor Number support and make NetTrace the default … (#25276)
* Add EventPipe Processor Number support and make NetTrace the default format
The EventPipe header now has a Processor Number field. On windows we query the correct value, on other OSes we currently have a -1 placeholder, but the field is written to the format regardless.
NetTrace is now the default format when using the environment variable, and the format must be explicitly configured when using the IPC channel or managed API. A parallel change in the diagnostics repo is changing dotnet-trace and dotnet-counter to specify nettrace format which means .NET devs should see nettrace almost exclusively from now on. If for whatever reason it is needed, NetPerf remains available if a scenario explicitly requests to use it.
* PR feedback + attempting to fix broken tests
Sergey Andreenko [Fri, 21 Jun 2019 21:49:57 +0000 (14:49 -0700)]
WorkAround for #25050. (#25316)
* WorkAround for #25050.
A temporary workaround to push the milestone for this issue to 3.next.
* Response review.
Carol Eidt [Fri, 21 Jun 2019 21:26:11 +0000 (14:26 -0700)]
Fix test to use mutual waits (#25317)
Fix #25245
Santiago Fernandez Madero [Fri, 21 Jun 2019 17:54:36 +0000 (10:54 -0700)]
Address nullable feedback for System.Diagnostics.Tracing and StackTrace (#25303)
* Address nullable feedback for System.Diagnostics.Tracing and StackTrace
* Disable corefx tests that need to be updated
Carol Eidt [Fri, 21 Jun 2019 17:40:03 +0000 (10:40 -0700)]
Update instructions to build individual tests
cc @jashook and @sandreenko
Mike McLaughlin [Fri, 21 Jun 2019 16:40:16 +0000 (09:40 -0700)]
ifdef Linux specific readvirtual code (#25299)
Stephen Toub [Fri, 21 Jun 2019 15:56:13 +0000 (11:56 -0400)]
Remove several unnecessary string allocations (#25305)
Andy Ayers [Fri, 21 Jun 2019 15:32:37 +0000 (08:32 -0700)]
JIT: update epilog exit size assert for x86 (#25302)
With the advent of #25251 we can now see the epilog exit sequence on x86 vary
by as much as 6 bytes. Update the assert accordingly.
Fixes #25301.
Ao Qi [Fri, 21 Jun 2019 12:29:01 +0000 (20:29 +0800)]
Fixed mis-matching arguments in src/CMakeLists.txt (#25307)
Jan Vorlicek [Fri, 21 Jun 2019 12:21:34 +0000 (14:21 +0200)]
Disable unloading incompatible test for runincontext (#25126)
* Disable unloading incompatible test for runincontext
These tests do various things that make them incompatible with running
in an unloadable assembly load context using the runincontext tool.
I've attached comments to the UnloadabilityIncompatible element
describing the specific problems of each test.
I've also found a problem in the parameter passing in the
runincontext.sh where it was always passing 7 parameters to each test
and tests that were checking the number of parameters were failing due
to that. This change fixes it by passing just the parameters that were
passed to the runincontext.sh.
* Added printing unhandled exception details if it happened in the test
* Added 10ms wait to the GC collection loop to also give time to async
stuff to completely exit threadpool threads
David Mason [Fri, 21 Jun 2019 09:20:57 +0000 (02:20 -0700)]
Mark the profiler attach thread during attach
Jan Vorlicek [Fri, 21 Jun 2019 08:48:08 +0000 (10:48 +0200)]
Fix OSX floating point state extraction (#25295)
There was a bug reported on a very recent Mac with Intel i9 processor. A
crash in the RtlRestoreContext was happening at the fxrstor instruction
due to the fact that the floating point state data were garbage.
The investigation has shown that sometimes, the x86_FLOAT_STATE64
cannot be obtained using the thread_get_state API. And it was also found
that at the same time, the x86_AVX_STATE64 can be obtained. The state
extracted by the AVX variant contains all the registers that the FLOAT
variant would extract.
However, in some cases, even the x86_AVX_STATE64 cannot be obtained and
there is a third flavor that we can get - x86_AVX512_STATE64.
Unfortunately, there are cases where none of those can be obtained.
It is not clear what causes these cases, it seems only kernel debugging
can give us an answer to that.
This change modifies the way we extract the floating point state. We
first try to get the AVX state, if we fail, we try the AVX512 and
finally we fall back to the FLOAT state. If we fail to get the floating
point state with any of these, we return context without the floating
point state flag set. Also, if only getting the FLOAT state succeeds,
we return context without the XSTATE flag set.
Stephen Toub [Fri, 21 Jun 2019 04:36:42 +0000 (00:36 -0400)]
Reduce string allocation in DiagnosticCounter.GetMetadataString (#25219)
When the _metadata dictionary contains a single key/value pair, the method ends up allocating an unnecessary StringBuilder, char[] beneath it, and input string.
When the _metadata dictionary contains more key/value pairs, the method ends up allocating an unnecessary string per key/value pair.
Vladimir Sadov [Fri, 21 Jun 2019 02:54:34 +0000 (19:54 -0700)]
ensure process-wide fence when updating GC write barrier on ARM64 (#25130)
* ensure process-wide fences when updating GC write barrier on ARM64
dotnet-maestro[bot] [Fri, 21 Jun 2019 01:47:15 +0000 (18:47 -0700)]
[master] Update dependencies from dotnet/arcade (#25253)
* Update dependencies from https://github.com/dotnet/arcade build
20190619.25
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19319.25
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19319.25
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19319.25
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19319.25
dotnet-maestro[bot] [Fri, 21 Jun 2019 01:44:26 +0000 (18:44 -0700)]
[master] Update dependencies from dnceng/internal/dotnet-optimization (#25255)
* Update dependencies from https://dev.azure.com/dnceng/internal/_git/dotnet-optimization build
20190620.1
- optimization.IBC.CoreCLR - 99.99.99-master-
20190620.1
- optimization.PGO.CoreCLR - 99.99.99-master-
20190620.1
Brian Sullivan [Fri, 21 Jun 2019 01:43:32 +0000 (18:43 -0700)]
Added the complementary test cases of going from a negative int to an unsigned.
Jeremy Koritzinsky [Fri, 21 Jun 2019 00:57:58 +0000 (17:57 -0700)]
Revert "ILMarshaler Refactor (#21227)" (#25292)
This reverts commit
70febbad4bbe23326c26480b2848cdfe9fc8450c.
Gleb Balykov [Fri, 21 Jun 2019 00:55:10 +0000 (03:55 +0300)]
Add custom default base address option to crossgen (#25227)
* Add crossgen option to setup default base address for native image
This is enabled only with -DFEATURE_ENABLE_NO_ADDRESS_SPACE_RANDOMIZATION.
* Mmap native images at default base address if env variable COMPlus_UseDefaultBaseAddr=0x1 is setup.
This is enabled only with -DFEATURE_ENABLE_NO_ADDRESS_SPACE_RANDOMIZATION.
Sung Yoon Whang [Thu, 20 Jun 2019 23:33:51 +0000 (16:33 -0700)]
Fix eventpipe filtering (#25221)
* Add DOTNET_TRACE_CONTEXT and change macros to use DOTNET_TRACE_CONTEXT instead of MCGEN_TRACE_CONTEXT
* Fixing macro definitions
* eventing codegen scripts now generates EventPipe trace contexts
* Fix macros to use the EVENTPIPE_TRACE_Context
* Fix linux build
* Fix windows build
* Update Eventpipe provider context at EtwCallbackComon
* break in switch
* Update rundown provider context manually
* PR feedback
* Eventpipe->EventPipe
* cleanup in codegen script
Sung Yoon Whang [Thu, 20 Jun 2019 23:32:20 +0000 (16:32 -0700)]
Update linux-performace-tracing doc to add eventpipe/dotnet-trace (#25212)
* Update linux-performace-tracing doc to add eventpipe/dotnet-trace
* typo
* Add more detailed doc instead of just a link :)
* remove redundant part
* Update based on noah's feedback
Jeremy Koritzinsky [Thu, 20 Jun 2019 23:12:02 +0000 (16:12 -0700)]
Remove dead code related to copy-constructors and intercept stubs (#25264)
* Remove dead code related to copy-constructors and intercept stubs
* PR Feedback.
Jeremy Koritzinsky [Thu, 20 Jun 2019 23:11:17 +0000 (16:11 -0700)]
Disable NativeVarargsTest on Windows Nano Server. (#25284)
Egor Chesakov [Thu, 20 Jun 2019 22:36:25 +0000 (15:36 -0700)]
Move resources/containers out of azure-pipelines.yml to platform-matrix.yml (#25155)
Tomáš Rylek [Thu, 20 Jun 2019 21:49:27 +0000 (14:49 -0700)]
Initial check-in of R2R format documentation (#25267)
In this initial check-in I have converted Jan Kotas' R2R format doc
from an internal Word document to the md format and I removed some
references to other internal docs. I plan to follow up by updating
the description to the current format version (e.g. apparently the
description doesn't yet include the instance entrypoint table) and
by adding specific extensions for the composite R2R format.
Thanks
Tomas
Matt Galbraith [Thu, 20 Jun 2019 21:44:45 +0000 (14:44 -0700)]
Merge pull request #25285 from MattGal/update-arm32-dockerimages
Update ARM32 Helix Images to include Python 3
Mike McLaughlin [Thu, 20 Jun 2019 21:08:55 +0000 (14:08 -0700)]
Remove SOS and lldb plugin (#25220)
* Remove SOS and lldb plugin
Issue: https://github.com/dotnet/coreclr/issues/24092
Egor Chesakov [Thu, 20 Jun 2019 20:00:53 +0000 (13:00 -0700)]
Triage issues in #25098 (#25240)
Andy Ayers [Thu, 20 Jun 2019 19:04:54 +0000 (12:04 -0700)]
JIT: use register indirect for indirect jumps (#25251)
If we are jumping to another method via an indirection cell, the cell address
may be too far from the jump to address via a RIP-relative indirection. So
instead, load the target address into RAX and jump register indirect.
Fixes #22342.
Matt Galbraith [Thu, 20 Jun 2019 19:00:16 +0000 (12:00 -0700)]
missed a spot
Matt Galbraith [Thu, 20 Jun 2019 18:57:09 +0000 (11:57 -0700)]
Update ARM32 Helix Images to include Python 3
Jan Vorlicek [Thu, 20 Jun 2019 16:30:15 +0000 (18:30 +0200)]
Remove test 366085 (#25277)
This test was checking for "Parameter name:" substring in the
ArgumentNullException message. But a recent change #25185 modified
that string to just "Parameter".
We also have a coverage for this in corefx tests, so I am removing
this test.
Stephen Toub [Thu, 20 Jun 2019 16:27:21 +0000 (12:27 -0400)]
Fix visibility on Exception.InnerExceptionPrefix (#25278)
Bruce Forstall [Wed, 19 Jun 2019 23:39:16 +0000 (16:39 -0700)]
Merge pull request #25265 from BruceForstall/FixCorefxTestingScript
Update run-corefx-tests.py for new 3.0.0 directory name
Carol Eidt [Wed, 19 Jun 2019 23:21:03 +0000 (16:21 -0700)]
Restrict copyprop if address-taken (#25247)
* Restrict copyprop if address-taken
Fix #25231
Bruce Forstall [Wed, 19 Jun 2019 22:57:02 +0000 (15:57 -0700)]
Update run-corefx-tests.py for new 3.0.0 directory name
Santiago Fernandez Madero [Wed, 19 Jun 2019 21:33:29 +0000 (14:33 -0700)]
Address nullable feedback for S.T.Overlapped, S.Memory, S.R.InteropServices (#25214)
* Address nullable feedback for System.Threading.Overlapped
* Address nullable feedback for System.Memory
* Address nullable for System.Runtime.InteropServices
* Update xml comments and make CultureInfo nullable to match String.ToLower/Upper feedback
* PR feedback: revert accepting null comparer in BinarySearch
* Disable test that needs to be updated in corefx
Brian Sullivan [Wed, 19 Jun 2019 18:18:02 +0000 (11:18 -0700)]
Fix Issue #25134 - AssertionProp incorrectly removes cast from uint
Add additional check for the GT_UNSIGNED flag
+ Ran clang-format
+ Code review feedback, use IsUnsigned()
John Salem [Wed, 19 Jun 2019 17:29:49 +0000 (10:29 -0700)]
Change Diagnostic Server transport name to align with spec (#25031)
resolves 24859
Dan Moseley [Wed, 19 Jun 2019 08:28:30 +0000 (01:28 -0700)]
Exception format cleanup subset (#25185)
* Change ArgumentException message to one line
* Disable 2 tests
* Remove duplication in AggregateException.ToString()
* Inner exceptions on new lines
* Caps letter
* Typo
Jeremy Koritzinsky [Wed, 19 Jun 2019 00:26:10 +0000 (17:26 -0700)]
ILMarshaler Refactor (#21227)
* Add support for fields to ILStubMarshalHome objects.
* Make the specific code streams in the ILMarshaler private members.
* Make home members private.
* Move the IL code stream members back to being protected for the marshalers that override EmitMarshalArgumentCLRToNative instead of EmitMarshalArgumentContents/Space.
* Convert virtual specifiers in overrides to override specifiers.
* Remove unused and undefined member.
* Cleanup a few missed virtual/override specifiers.
* Refactor setup methods. Refactor some of the overrides to duplicate less code.
* Refactor the setup stream and add some comments around how the ArgumentOverride/ReturnOverride marshalling system works.
* Use the cleanup work list in SafeHandle marshalling since the work list is now implemented entirely in managed code.
* Generalize DelegateCleanupWorkListElement -> KeepAliveCleanupWorkListElement and rename the managed entry point to match.
* Refactor direct usage of the cleanup code stream in non-overridden marshalers to use the cleanup work list instead.
* Refactor AsAny marshalling to fit into the regular ILMarshaler code-paths.
* Move ILArgIteratorMarshaler over to the full ILMarshaler infrastructure.
* Port ILBlittablePtrMarshaler over to not directly reference m_pcsMarshal.
* Make the specific code streams private and have ILNativeArrayMarshaler fetch the one it needs from the NDirectStubLinker.
* Devirtualize a method on ILMarshaler.
* Fix broken metasig
* Revert "Use the cleanup work list in SafeHandle marshalling since the work list is now implemented entirely in managed code."
This reverts commit
aedcdfb4830475ff8f4d582147cdc0eb497dd681.
* Fix ILArgIteratorMarshaler.
* Take 2 on using the cleanup work list for SafeHandles
* Remove unused field
* SafeHandleMarshaler doesn't need to have extra cleanup code when using the CleanupWorkList.
* Move the rest of the SafeHandle marshalling into ArgumentOverride.
* Moved Pinned Native Array special case into an ArgumentOverride.
* Devirtualize
* Remove invalid field hometype.
* Make ILMarshaler::m_pslNDirect private.
* Native Array marshalling fixes.
* Fix STLOC->LDLOC mistakes.
* Add override hook to allow a marshaler to request that byval contents are always converted to native even when not marked as In (i.e. explicitly marked only as Out). Used in AsAny for byval arrays.
* PR Feedback.
* Add explicit pinning path to ilmarshalers.
* Move implementation of ILNativeArrayMarshaler::CanMarshalViaPinning to ilmarshalers.cpp.
* Add missing override specifier.
* Don't create a managed marshaler if we are able to marshal via pinning. Remove extraneous checks against pinning.
* Convert ILWSTRMarshaler to use the MarshalViaPinning functions for its pinning fast-path.
* Enable LPWSTR marshalling to fall back to allocating on the heap for large strings when stack-allocating for native space (byref in-only semantics).
* PR Feedback.
dotnet-maestro[bot] [Tue, 18 Jun 2019 21:17:41 +0000 (14:17 -0700)]
Update dependencies from https://github.com/dotnet/core-setup build
20190618.02 (#25223)
- Microsoft.NETCore.App - 3.0.0-preview7-27818-02
Jan Kotas [Tue, 18 Jun 2019 21:17:18 +0000 (14:17 -0700)]
Relax signed/unsigned checks in Array.Copy to match other places (#25209)
Disable failing tests. Also enable tests for changes that propagated through the system.
Fixes dotnet/corefx#38535
Jeremy Koritzinsky [Tue, 18 Jun 2019 21:05:15 +0000 (14:05 -0700)]
Don't free the loader handle if the LoaderAllocator is already dead. (#25208)
Bruce Forstall [Tue, 18 Jun 2019 20:59:48 +0000 (13:59 -0700)]
Reduce corefx testing load (#25235)
* Add new helixQueueGroup to allow restricting platforms tested
This is used to force corefx testing to only run on one Helix queue
for each platform, instead of multiple.
Jarret Shook [Tue, 18 Jun 2019 20:44:00 +0000 (13:44 -0700)]
Fix 25233 (#25234)
* Fix ibcmerge.csproj to use IbcMergePackageVersion
This fixes a copy/paste bug.
* Fix more copy/paste in comment
* Use ! in place of %
Mike McLaughlin [Tue, 18 Jun 2019 20:41:17 +0000 (13:41 -0700)]
Sign the DAC with MicrosoftSHA2 cert (#25205)
The current Microsoft certificate breaks minidumps/WER. The dbghelp
MiniDumpWriteDump code rejected the default cert. This started
failing when the obsolete SHA1 cert was removed from all of our
binaries.
Issue: https://devdiv.visualstudio.com/DevDiv/_workitems/edit/919061
Carol Eidt [Tue, 18 Jun 2019 20:12:31 +0000 (13:12 -0700)]
Fix contained LEA handling (#25135)
* Fix contained LEA handling
This adds an LEA case to both `LinearScan::BuildOperandUses` and `CodeGen::genConsumeRegs`.
Fix #25039
Sergey Andreenko [Tue, 18 Jun 2019 16:30:59 +0000 (09:30 -0700)]
Start cleaning emitxarch::GetSize estimates. (#25160)
* Add comments for _idCodeSize and a few asserts.
* Delete emitSizeOfJump.
* Delete emitInstCodeSz.
* Response review.
dotnet-maestro[bot] [Tue, 18 Jun 2019 14:46:59 +0000 (14:46 +0000)]
Update dependencies from https://dev.azure.com/dnceng/internal/_git/dotnet-optimization build
20190618.1 (#25224)
- optimization.IBC.CoreCLR - 99.99.99-master-
20190618.1
- optimization.PGO.CoreCLR - 99.99.99-master-
20190618.1
Michal Strehovský [Tue, 18 Jun 2019 07:07:15 +0000 (09:07 +0200)]
Prevent loading byref-like types with invalid layout (#25200)
First approximation of a fix for #25057.
This has two problems:
* We're checking for any byref-like typed fields. Types that don't actually contain interior pointers but were marked as `ref struct` will fail to load when not aligned properly.
* We're not doing the deep validation that we do for reference types to make sure the `ByReference<T>` field doesn't overlap with another non-byreference field.
Question is whether we're okay with those limitations, or whether we need a better fix. Better fix would likely entail inefficiently walking over the fields à la `FindByRefPointerOffsetsInByRefLikeObject` (doing the more efficient thing that we do for object references below would require a GCDesc representation of byrefness).
Contributes to #25057.
Fadi Hanna [Tue, 18 Jun 2019 02:36:24 +0000 (19:36 -0700)]
Allow PEImageLayout to be loadable for R2R images loaded from streams. (#25159)
* Allow PEImageLayout to be loadable for R2R images loaded from streams.
* Add regression test coverage
* Change CheckILOnly to return true for R2R images (similar to the ILOnly API)
* Early out for R2R images
* Remove R2R check
Viktor Hofer [Fri, 14 Jun 2019 13:12:42 +0000 (15:12 +0200)]
Delete dead uapaot code (dotnet/corefx#38330)
* Delete dead uapaot code
Signed-off-by: dotnet-bot <dotnet-bot@microsoft.com>
mvandeberg [Mon, 17 Jun 2019 13:16:24 +0000 (23:16 +1000)]
Remove unused TRACE_LEVEL_NONE (#23711)
Removed define TRACE_LEVEL_NONE from eventtracebase.h.
Removed static_assert(GCEventLevel_None == TRACE_LEVEL_NONE, "GCEventLevel_None value mismatch"); from eventtrace.cpp
Enhancement #23711
Jeremy Koritzinsky [Tue, 18 Jun 2019 00:17:47 +0000 (17:17 -0700)]
Add IJW native varargs tests. (#24983)
* Refactor IJW test infra. Add some tests for some interesting simple native varargs.
* Add missing include of IjwHelper.cs
* Fix Cmake test project name.
* PR Feedback.
* Fix compilation error.
* Deploy msvcp*d.dll as part of the copy-local'd CRT.
Jan Kotas [Tue, 18 Jun 2019 00:15:56 +0000 (17:15 -0700)]
Stop ignoring crashes around IUnknown::Release calls (#25210)
We do not want to be masking any memory corruption and bugs in the runtime.
dotnet-maestro[bot] [Mon, 17 Jun 2019 22:58:30 +0000 (15:58 -0700)]
[master] Update dependencies from dnceng/internal/dotnet-optimization (#25181)
* Update dependencies from https://dev.azure.com/dnceng/internal/_git/dotnet-optimization build
20190615.1
- optimization.IBC.CoreCLR - 99.99.99-master-
20190615.1
- optimization.PGO.CoreCLR - 99.99.99-master-
20190615.1
* Update dependencies from https://dev.azure.com/dnceng/internal/_git/dotnet-optimization build
20190616.1
- optimization.IBC.CoreCLR - 99.99.99-master-
20190616.1
- optimization.PGO.CoreCLR - 99.99.99-master-
20190616.1
* Update dependencies from https://dev.azure.com/dnceng/internal/_git/dotnet-optimization build
20190616.3
- optimization.IBC.CoreCLR - 99.99.99-master-
20190616.3
- optimization.PGO.CoreCLR - 99.99.99-master-
20190616.3
dotnet-maestro[bot] [Mon, 17 Jun 2019 21:52:34 +0000 (14:52 -0700)]
[master] Update dependencies from dotnet/arcade (#25179)
* Update dependencies from https://github.com/dotnet/arcade build
20190614.27
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19314.27
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19314.27
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19314.27
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19314.27
* Update dependencies from https://github.com/dotnet/arcade build
20190615.2
- Microsoft.DotNet.Arcade.Sdk - 1.0.0-beta.19315.2
- Microsoft.DotNet.Build.Tasks.Feed - 2.2.0-beta.19315.2
- Microsoft.DotNet.Build.Tasks.Packaging - 1.0.0-beta.19315.2
- Microsoft.DotNet.Helix.Sdk - 2.0.0-beta.19315.2
Sung Yoon Whang [Mon, 17 Jun 2019 21:31:55 +0000 (14:31 -0700)]
make DiagnosticCounter and its children lock 'this' instead of an arbitrary String (#25173)
Jan Vorlicek [Mon, 17 Jun 2019 21:13:33 +0000 (23:13 +0200)]
Enable running SIGSEGV handler on non-alternate stack (#25196)
When a third party library that someone loads into a coreclr process registers
its own SIGSEGV handler and then chain-calls coreclr SIGSEGV handler on a
non-alternate stack, coreclr would currently crash.
This fix enables it to execute the SIGSEGV handler on the non-alternate stack
(original stack of the interrupted thread) in such case.
The disadvantage is that stack overflow would lead to silent crash in such a
case, but we cannot do anything about it.
Jan Kotas [Mon, 17 Jun 2019 18:01:52 +0000 (11:01 -0700)]
Fix missing MethodEvents for R2R images (#25199)
Fixes #25014
Juan Hoyos [Mon, 17 Jun 2019 17:22:13 +0000 (10:22 -0700)]
Enable createdump on arm and arm64 (#24832)
* Enable build of ARM64 createdump build
- Adds definitions of necessary constructs and aligments for Elf formats in aarch64
- Work around changes in ptrace for aarch64
* Package createdump in runtime transport for arm and arm64
* Use special name register definitions from the PAL
* Add createdump_lib to arm64 sos and package plugin
Egor Chesakov [Mon, 17 Jun 2019 17:13:14 +0000 (10:13 -0700)]
Fix official build - specify platformGroup for internal/Release/build-job.yml (#25204)
dotnet-maestro[bot] [Mon, 17 Jun 2019 15:18:10 +0000 (11:18 -0400)]
Update dependencies from https://github.com/dotnet/core-setup build
20190617.01 (#25197)
- Microsoft.NETCore.App - 3.0.0-preview7-27817-01
Ben Adams [Mon, 17 Jun 2019 15:14:53 +0000 (16:14 +0100)]
IndexOfAny(values.Length == 0) == -1 (#25075)
* IndexOfAny(values.Length == 0) == -1
* Skip tests