Bruce Forstall [Wed, 4 May 2016 16:41:57 +0000 (09:41 -0700)]
RyuJIT/x86: convert SBCG to NYI for multi-level expression trees
In general, we decompose longs by walking the tree bottom-up. So, e.g.,
we'll decompose 'add' as follows:
+ (gt_long(lo1, hi1), gt_long(lo2, hi2))
=>
gt_long (+ (lo1, lo2), +Hi (hi1, hi2))
But what if lo1 is '+' and hi1 is '+Hi'? So,
+ (gt_long( + (lo3, lo4), +Hi (hi3, hi4) ), gt_long(lo2, hi2))
=>
gt_long (+ ( + (lo3, lo4), lo2), +Hi ( +Hi (hi3, hi4), hi2))
But we can't put any instructions between '+ (lo3, lo4)' and '+Hi (hi3, hi4)'
or we'll lose the carry bit.
We need to model this differently, or introduce temps. Until this is
done, convert the silent bad codegen into NYI fallback.
Kyungwoo Lee [Wed, 4 May 2016 20:19:52 +0000 (13:19 -0700)]
Merge pull request #4723 from kyulee1/tailcall
ARM64: Enable Tail Call with Vararg
Michelle McDaniel [Wed, 4 May 2016 20:06:50 +0000 (13:06 -0700)]
Merge pull request #4727 from adiaaida/overflowLongArithmetic
Set overflow flag for add/sub hi on x86 TYP_LONG
Sean Gillespie [Wed, 4 May 2016 17:25:03 +0000 (10:25 -0700)]
Merge pull request #4757 from swgillespie/gcstress-disable-tests
Disable two tests under GCStress for #4748 and #4708
Bruce Forstall [Wed, 4 May 2016 17:13:32 +0000 (10:13 -0700)]
Merge pull request #4760 from dotnet-bot/from-tfs
Merge changes from TFS
Kyungwoo Lee [Tue, 26 Apr 2016 21:41:05 +0000 (14:41 -0700)]
ARM64: Enable Tail Call with Vararg
Fixes https://github.com/dotnet/coreclr/issues/4475
I've run into `IMPL_LIMITATION("varags + CEE_JMP doesn't work yet")` in
importer.cpp. This change enables ARM64 tail call path same as other
targets.
1. Similar to amd64 `genFnEpilog`, I made the similar code under
`!FEATURE_FASTTAILCALL`. Since `EC_FUNC_TOKEN_INDIR` is not defined for
arm64, I've made NYI for such case.
2. Added two pseudo branch instructions 'b_tail' and 'br_tail' which form
jmp instruction encodings but follow call instruction semantics since
they are used for tail-call.
3. `GenJmpMethod` is enabled. Code is slightly changed to reflect correct
float argument handlings and multi-reg support.
Rahul Kumar [Wed, 4 May 2016 05:21:57 +0000 (22:21 -0700)]
Merge pull request #4759 from rahku/crossgen
Fix checking of correct errorlevel for crossgen
Kyungwoo Lee [Wed, 4 May 2016 04:45:14 +0000 (21:45 -0700)]
Merge pull request #4771 from myungjoo/fix/4746
ildasm: incorrect signess usage with char
MyungJoo Ham [Wed, 4 May 2016 04:28:31 +0000 (13:28 +0900)]
Unix/Build: use -fsigned-char
Some architectures including ARM assumes that plain char is
unsigned unlike x64.
Because CoreCLR has been assuming char is signed, we need to
explictly tell the toolchain the signedness of char type.
Fix #4746
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Pat Gavlin [Wed, 4 May 2016 04:11:55 +0000 (21:11 -0700)]
Merge pull request #4478 from pgavlin/JitPackaging
Create JIT packages during the build.
Aditya Mandaleeka [Wed, 4 May 2016 03:58:05 +0000 (20:58 -0700)]
Merge pull request #4768 from adityamandaleeka/fix_ci_corefxbins
Add quotes around coreFxBinDir example in runtest.sh
Yi Zhang (CLR) [Wed, 4 May 2016 01:54:22 +0000 (18:54 -0700)]
Merge pull request #4740 from yizhang82/interop_x86_test
Fix interop x86 tests by fixing mismatching calling convention, and enable [NativeCallable] for every architecture
Aditya Mandaleeka [Wed, 4 May 2016 01:32:40 +0000 (18:32 -0700)]
Add quotes around coreFxBinDir example.
Andy Ayers [Wed, 4 May 2016 00:59:13 +0000 (17:59 -0700)]
Merge pull request #4753 from AndyAyersMS/InlineForSize
Inliner: initial version of SizePolicy
Russ Keldorph [Wed, 4 May 2016 00:07:57 +0000 (17:07 -0700)]
Merge pull request #4705 from RussKeldorph/vector3interop
Disable Vector3Interop until failures are addressed
MyungJoo Ham [Tue, 3 May 2016 23:09:10 +0000 (08:09 +0900)]
ARM: fix stack frame management (#4641)
ARM: fix stack frame management
Fix #4638
This commit revises the implementation of CallEHFunclet to establish
stack frame appropriately.
This commit allows libunwind to work correctly for the frame created by
CallEHFunclet.
Aditya Mandaleeka [Tue, 3 May 2016 22:49:38 +0000 (15:49 -0700)]
Merge pull request #4693 from adityamandaleeka/delegateinvoke_platnotsup
Make Delegate BeginInvoke/EndInvoke throw PlatformNotSupportedException
Aditya Mandaleeka [Tue, 3 May 2016 19:33:28 +0000 (12:33 -0700)]
Add test for Begin/EndInvoke throwing PlatformNotSupportedException.
Rahul Kumar [Tue, 3 May 2016 22:20:29 +0000 (15:20 -0700)]
Fix checking of correct errorlevel for crossgen
Bruce Forstall [Tue, 3 May 2016 22:09:53 +0000 (15:09 -0700)]
Merge pull request #4738 from dotnet-bot/from-tfs
Merge changes from TFS
Sean Gillespie [Tue, 3 May 2016 22:09:06 +0000 (15:09 -0700)]
Merge pull request #4721 from swgillespie/gcsimulator
Add a "GCSimulator" CI run
Sean Gillespie [Tue, 3 May 2016 21:27:34 +0000 (14:27 -0700)]
Disable two tests under GCStress for #4748 and #4708
Pat Gavlin [Wed, 20 Apr 2016 21:03:56 +0000 (14:03 -0700)]
Create JIT packages during the build.
Michelle McDaniel [Mon, 2 May 2016 20:52:39 +0000 (13:52 -0700)]
Set overflow flag for add/sub hi on x86 TYP_LONG
When we create the hi operation for add and sub on TYP_LONG, we don't
carry the overflow flag to the hi operation. This change sets the overflow
flag on hiResult in lower if it was set on loResult, and adds GT_ADD_HI
and GT_SUB_HI to the operations that can have overflow. We also need to
pass the unsigned flag to the high part in the instance that we are
dealing with an add or subtract with overflow. Fixes #4596.
Pat Gavlin [Tue, 3 May 2016 19:58:16 +0000 (12:58 -0700)]
Remove SEH interactions between the JIT and the EE.
This change replaces the final uses of SEH between the JIT and the EE with a new method
on ICorJitInfo, `ICorJitInfo::runWithErrorTrap`. This method runs a given function under an
error trap that prevents recoverable errors from propagating out of the call. This allows
the JIT to make calls to JIT/EE interface functions that may throw exceptions without
needing to be aware of the EH ABI, exception types, etc. The method returns true if the
given function completed successfully and false otherwise. This change deprecates
all other SEH-related functionality on the JIT/EE interface (e.g. `ICJI::FilterException` and
`ICJI::HandleException`).
This change does not completely eliminate SEH over the JIT/EE interface: there are still a
handful of places where the JIT expects to be able to catch exceptions thrown by
SuperPMI. These uses of SEH will be removed later on.
Fixes #3058 and #4130.
[tfs-changeset: 1601288]
Sean Gillespie [Fri, 29 Apr 2016 00:03:51 +0000 (17:03 -0700)]
GCSimulator test run
Andy Ayers [Sat, 30 Apr 2016 00:27:40 +0000 (17:27 -0700)]
Inliner: initial version of SizePolicy
Add a new inline policy that tries to inline as much as possible
without increasing method size.
Fix one test that had an expected inline that doesn't happen under
the SizePolicy, by marking a method with AggressiveInline.
Pat Gavlin [Tue, 3 May 2016 16:18:51 +0000 (09:18 -0700)]
Merge pull request #4728 from pgavlin/FixTRTContracts
Fix contracts in `tryResolveToken`.
Deepak Shankargouda [Tue, 3 May 2016 16:17:27 +0000 (09:17 -0700)]
Fixing issue with perf targets restoring test runtime (#4686)
Yi Zhang (CLR) [Tue, 3 May 2016 04:00:21 +0000 (21:00 -0700)]
Fix calling convention mismatch in 7 interop tests and enable [NativeCallable] in all architectures.
Bruce Forstall [Tue, 3 May 2016 14:59:25 +0000 (07:59 -0700)]
Merge pull request #4732 from BruceForstall/SignedIntToLongCastNYI
Convert signed int=>long cast SBCG to NYI
Bruce Forstall [Tue, 3 May 2016 00:58:01 +0000 (17:58 -0700)]
Convert signed int=>long cast SBCG to NYI
This causes many tests that fail with silent bad codegen to pass
with NYI fallback. Many tests are removed from the issues.targets
failing tests exclusion list, but added to the
ryujit_x86_no_fallback_issues.targets exclusion list.
Pat Gavlin [Mon, 2 May 2016 23:40:26 +0000 (16:40 -0700)]
Fix contracts in `tryResolveToken` et. al.
`isValidToken`'s contract was too conservative: it was needlessly marked
`THROWS`, `GC_TRIGGERS`, and `MODE_PREEMPTIVE`. Its contract and those
of its callers have been tightened to `NOTHROW`, `GC_NOTRIGGER`, and
`MODE_ANY`.
Jan Vorlicek [Tue, 3 May 2016 08:42:40 +0000 (10:42 +0200)]
Merge pull request #4726 from adityamandaleeka/array_alloc_bounds_check
Fix array allocation bounds checking on Unix
James Ko [Tue, 3 May 2016 07:56:17 +0000 (03:56 -0400)]
Generalize Buffer.BlockCopy optimization for either src or dst (#4736)
* Generalize Buffer.BlockCopy optimization for either src or dst
Gaurav Khanna [Tue, 3 May 2016 04:46:40 +0000 (21:46 -0700)]
Fix for buildstatus/Issues/Issues.aspx?iid=797159
[tfs-changeset: 1601059]
Jan Vorlicek [Tue, 3 May 2016 04:22:11 +0000 (06:22 +0200)]
Fix handling of FP args on stack in the ArgIterator (#4729)
This change fixes a bug in the ArgIterator that resulted in assigning
floating point arguments that didn't fit to the xmm registers to general
purpose registers in the ArgIterator if there were some general purpose
registers available.
Rama krishnan Raghupathy [Tue, 3 May 2016 03:04:57 +0000 (20:04 -0700)]
Merge pull request #4694 from ramarag/thisptrret
Arm64:Implemeting ThisPtrRetBufPrecode:
Lubomir Litchev [Tue, 3 May 2016 01:23:43 +0000 (18:23 -0700)]
Merge pull request #4616 from LLITCHEV/MultiCoreJit-1
Add ability to AutoStartProfiling for CoreCLR.
Kyungwoo Lee [Tue, 3 May 2016 01:00:47 +0000 (18:00 -0700)]
Merge pull request #4731 from kyulee1/onlypush
ARM64: Trigger Failure Email for Push Only
Kyungwoo Lee [Tue, 3 May 2016 00:56:13 +0000 (17:56 -0700)]
ARM64: Trigger Failure Email for Push Only
Rama Krishnan Raghupathy [Fri, 22 Apr 2016 02:51:49 +0000 (19:51 -0700)]
Arm64:Implemeting ThisPtrRetBufPrecode:
Precode to shuffle this and retbuf for closed delegates over static methods with return buffer
Aditya Mandaleeka [Mon, 2 May 2016 22:29:11 +0000 (15:29 -0700)]
Fix array allocation bounds checking.
Bruce Forstall [Mon, 2 May 2016 22:00:55 +0000 (15:00 -0700)]
Merge pull request #4719 from dotnet-bot/from-tfs
Merge changes from TFS
Gaurav Khanna [Mon, 2 May 2016 21:12:15 +0000 (14:12 -0700)]
Merge pull request #4684 from gkhanna79/CrossgenJIT
Load JIT from custom location for Crossgen
Kyungwoo Lee [Mon, 2 May 2016 21:04:28 +0000 (14:04 -0700)]
Merge pull request #4718 from kyulee1/nativevararg
ARM64: NativeVarArg
Aditya Mandaleeka [Mon, 2 May 2016 18:53:13 +0000 (11:53 -0700)]
Merge pull request #4637 from adityamandaleeka/synch_cleanup
Remove SYNCHMGR_PIPE_BASED_THREAD_BLOCKING and clean up synchmanager
Lubomir Litchev [Tue, 26 Apr 2016 20:11:02 +0000 (13:11 -0700)]
Add ability to AutoStartProfiling for CoreCLR.
Add ability to do MultiCoreJit profiling in the CoreCLR. This is used for
testing of the MultiCoreJit functionality.
Jan Kotas [Mon, 2 May 2016 18:05:17 +0000 (11:05 -0700)]
Merge pull request #4709 from myungjoo/fix/4682
Match Inconsistent CMake/MSBuild Compiler Definitions
Jan Kotas [Mon, 2 May 2016 17:49:32 +0000 (10:49 -0700)]
Merge pull request #4688 from dotnet-bot/from-tfs
Merge changes from TFS
Bruce Forstall [Mon, 2 May 2016 17:26:23 +0000 (10:26 -0700)]
Merge pull request #4692 from BruceForstall/FuncInfo
Add support for easily collecting JitFuncInfoLogFile for the tests
Mike McLaughlin [Mon, 2 May 2016 17:06:21 +0000 (10:06 -0700)]
Merge pull request #4543 from krytarowski/netbsd-support-73
NetBSD limits name of a semaphore to 15 characters (inc. null)
Kyungwoo Lee [Wed, 27 Apr 2016 14:40:19 +0000 (07:40 -0700)]
ARM64: NativeVarArg
In PInovke stub call with vararg that has float value,
we did not properly set up argumnent by passing float registers whith do
not conform ARM64 ABI.
The fix is to set the right calling convention in VM to use
integer registers to pass such float values.
Kyungwoo Lee [Mon, 2 May 2016 16:53:33 +0000 (09:53 -0700)]
Merge pull request #4621 from kyulee1/floatarg
ARM64: Enable Varargs with float argument.
Kyungwoo Lee [Wed, 27 Apr 2016 14:40:19 +0000 (07:40 -0700)]
ARM64: Enable Varargs with float argument.
Fixes https://github.com/dotnet/coreclr/issues/2989
ARM64 ABI does not allow float register arguments for vararg calls.
All such float arguments should be copied to integer registers.
In Lower, I explicitly add a GT_COPY for such case and support the code-gen
to expand it properly using `fmov`.
I also skip genFnPrologCalleeRegArgs() for vararg since we've already
saved all integer registers in the prolog.
Kyungwoo Lee [Mon, 2 May 2016 16:32:09 +0000 (09:32 -0700)]
Merge pull request #4699 from kyulee1/testlst
ARM64: Tests.lst Update
Andy Ayers [Mon, 2 May 2016 16:06:16 +0000 (09:06 -0700)]
Merge pull request #4689 from AndyAyersMS/NoInlineRange
Implement JitNoInlineRange
Gaurav Khanna [Fri, 29 Apr 2016 15:00:17 +0000 (08:00 -0700)]
Enable crossgen to load JIT from a custom location
Continue to link the JIT in for Windows Arm64 builds until 4717 is fixed.
Bruce Forstall [Fri, 29 Apr 2016 22:52:43 +0000 (15:52 -0700)]
Add support for easily collecting JitFuncInfoLogFile for the tests
Simply uncomment one of the three methods for doing the collection in
the ryujit_x86_testenv.cmd file.
Yi Zhang (CLR) [Mon, 2 May 2016 07:42:57 +0000 (00:42 -0700)]
Merge pull request #4710 from yizhang82/array_warn_fix
Fix warnings in ByValArray
Russ Keldorph [Mon, 2 May 2016 04:49:44 +0000 (21:49 -0700)]
Merge pull request #4707 from RussKeldorph/gcincomp
Mark some GC tests incompatible with GCStress
Russ Keldorph [Mon, 2 May 2016 04:44:09 +0000 (21:44 -0700)]
Merge pull request #4704 from RussKeldorph/lotsofinlines
Mark LotsOfInlines tests as optimization-sensitive
Yi Zhang (CLR) [Mon, 2 May 2016 04:03:03 +0000 (21:03 -0700)]
Fix warnings
Jonghyun Park [Mon, 2 May 2016 02:09:50 +0000 (11:09 +0900)]
Fix #4496 (#4503)
* Fix #4496
This commit revises legecy codegen to inserts a bkpt instruction if the current BB
and next BB is not in the same exception handling region.
This code is backported from ARM64 codegen.
* Update ControlPcIsUnwound in UnwindManagedExceptionPassX
* Enable only for ARM32
Enable the patch only for ARM32 in order to fix build break in other
architecture.
MyungJoo Ham [Mon, 2 May 2016 01:58:49 +0000 (10:58 +0900)]
Disable WIN32_REGISTRY for UNIX platforms
Unlike managed build scripts, FEATURE_WIN32_REGISTRY has been
enabled for UNIX-like systems in CMake native build scripts.
Fix #4682 part 3.
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
John Chen [Sun, 1 May 2016 08:15:42 +0000 (01:15 -0700)]
Merge pull request #4556 from JohnChen0/r2r
Ready-to-Run support for generic methods without generic dictionary
Russ Keldorph [Sun, 1 May 2016 06:16:41 +0000 (23:16 -0700)]
Mark some GC tests incompatible with GCStress
These tests rely on counting numbers of collections, which is unreliable
(or at best different than what the test expects) under GCStress.
John Chen (CLR) [Mon, 25 Apr 2016 08:55:00 +0000 (01:55 -0700)]
Support generic methods without generic dictionary
Russ Keldorph [Sun, 1 May 2016 04:35:04 +0000 (21:35 -0700)]
Disable Vector3Interop until failures are addressed
This test is turning several lab runs red. Disabling for now.
Russ Keldorph [Sun, 1 May 2016 04:24:18 +0000 (21:24 -0700)]
Mark LotsOfInlines tests as optimization-sensitive
The LotsOfInlines test times out under COMPlus_JitStress=1.
(http://dotnet-ci.cloudapp.net/job/dotnet_coreclr/job/master/job/x64_checked_windows_nt_jitstress1/48/testReport/).
This change disables it from running under JitStress modes that might
affect optimizations required for it to run in reasonable time.
Mike McLaughlin [Sat, 30 Apr 2016 18:55:55 +0000 (11:55 -0700)]
Merge pull request #4653 from mikem8361/shimrace
Fix issue #4298 "SIGSEGV_libcoreclr.so!Debugger::GetArgCount"
Kyungwoo Lee [Sat, 30 Apr 2016 14:07:54 +0000 (07:07 -0700)]
ARM64: Test.lst Update
Fixes https://github.com/dotnet/coreclr/issues/3970
Fixes https://github.com/dotnet/coreclr/issues/4411
Since https://github.com/dotnet/coreclr/pull/4655 is merged, these tests
seem to be passed in the lab.
Kyungwoo Lee [Sat, 30 Apr 2016 14:02:54 +0000 (07:02 -0700)]
Merge pull request #4687 from kyulee1/zap
ARM64: Enabling Crossgen End-to-End Mscorlib
Aditya Mandaleeka [Sat, 30 Apr 2016 02:46:17 +0000 (19:46 -0700)]
Merge pull request #4698 from adityamandaleeka/fix_ci_corefxbins
Fix CI: Update directory check for case with multiple CoreFX directories
Aditya Mandaleeka [Sat, 30 Apr 2016 02:20:54 +0000 (19:20 -0700)]
Use Unix version of WORKSPACE.
Aditya Mandaleeka [Sat, 30 Apr 2016 02:15:30 +0000 (19:15 -0700)]
Update directory check for case with multiple CoreFX directories.
Michelle McDaniel [Sat, 30 Apr 2016 01:24:09 +0000 (18:24 -0700)]
Merge pull request #4691 from adiaaida/longOperators
Separate NYIs for arithmetic operators on longs
Andy Ayers [Wed, 27 Apr 2016 21:22:35 +0000 (14:22 -0700)]
Implement JitNoInlineRange
Rework and cleanup ConfigMethodRange. Give it a configurable size. Do
some error detection when parsing the range string. Update comments
and add notes at existing usage sites about the behavior when the
range string is not specified.
Use ConfigMethodRange to implement a JitNoInlineRange option that
suppresses inlining in a specified set of methods. Set this up so
inlning still happens if the range string is empty. Choose capacity so
it can hold the entire range string's set of ranges. Add a new
observation for the cases where enabling this JIT option disables
inlines.
Make compMethodHash available for INLINE_DATA builds. Report this hash
in the Xml inline dumps, so it can provide hash values to feed into
JitNoInlineRange strings.
Aditya Mandaleeka [Sat, 30 Apr 2016 01:18:50 +0000 (18:18 -0700)]
Merge pull request #4695 from adityamandaleeka/fix_ci_corefxbins
Modify CI to pass in multiple CoreFX bin directories
Aditya Mandaleeka [Sat, 30 Apr 2016 00:26:38 +0000 (17:26 -0700)]
Add cascading CoreFX directories to Unix CI run
Aditya Mandaleeka [Sat, 30 Apr 2016 00:15:40 +0000 (17:15 -0700)]
Allow multiple CoreFX bin directories to be specified in runtest.sh.
Bruce Forstall [Sat, 30 Apr 2016 01:02:50 +0000 (18:02 -0700)]
Fix desktop build break introduced by CS#1600134
[tfs-changeset: 1600494]
Aditya Mandaleeka [Fri, 29 Apr 2016 23:05:26 +0000 (16:05 -0700)]
Make Delegate BeginInvoke/EndInvoke throw PlatformNotSupportedException.
Michelle McDaniel [Fri, 8 Apr 2016 17:10:42 +0000 (10:10 -0700)]
Separate NYIs for arithmetic operators on longs
To make it more obvious which long operators are causing the NYI to hit,
split out the different operators with their own NYI.
Bruce Forstall [Fri, 29 Apr 2016 22:26:33 +0000 (15:26 -0700)]
Merge pull request #4676 from BruceForstall/x86Testing
Change RyuJIT/x86 testing
Bruce Forstall [Fri, 29 Apr 2016 03:03:03 +0000 (20:03 -0700)]
Change RyuJIT/x86 testing
Change the RyuJIT/x86 test runs so we don't set COMPLUS_AltJitAssertOnNYI=1
by default. This means when the JIT hits an NYI, it falls back to
the legacy x86 back-end instead of raising an assert. The issues.targets
test exclusion file is changed to only exclude those tests that fail,
due to assert or silent bad codegen, in this mode, rather than the
(currently) much, much larger number of exclusions if COMPLUS_AltJitAssertOnNYI=1
is set.
The version of issues.targets is renamed as ryujit_x86_no_fallback_issues.targets.
This is a issues.targets file that contains exclusions for running
with RyuJIT/x86 and COMPLUS_AltJitAssertOnNYI=1, so NYI turn into
asserts.
Kyungwoo Lee [Fri, 29 Apr 2016 17:29:28 +0000 (10:29 -0700)]
ARM64: Enabling Crossgen End-to-End Mscorlib
Fixes https://github.com/dotnet/coreclr/issues/4350
Fixes https://github.com/dotnet/coreclr/issues/4615
This is a bit large change across VM/Zap/JIT to properly support crossgen
scenario.
1. Fix incorrect `ldr` encoding with size.
2. Enforce JIT data following JIT code per method by allocating them together.
This guarantees correct PC-relative encoding for such constant data access
without fix-up.
3. For the general fix-up data acceess, use `adrp/add` instruction pairs with fix-ups.
Two more relocations types are implemented in all sides.
4. Interface dispatch stub is now implemented which is needed for
interface call for crossgen.
I've verified hello world runs with mscorlib.ni.dll.
Kamil Rytarowski [Fri, 29 Apr 2016 21:35:19 +0000 (23:35 +0200)]
Treat NetBSD like FreeBSD and OSX in PAL_VirtualUnwind() (#4551)
This fixes hangs in the runtime observed while running CoreFX managed tests.
Thanks @janvorli and @myungjoo
Fix #4380
shion [Fri, 29 Apr 2016 21:33:40 +0000 (16:33 -0500)]
Add query handling of IUnknown for MergeTokenManager and CMapToken (#4563)
shion [Fri, 29 Apr 2016 21:33:03 +0000 (16:33 -0500)]
Add query handling of ICLRPrivResource for Assembly::CLRPrivResourceAssembly (#4562)
Jonghyun Park [Fri, 29 Apr 2016 21:27:47 +0000 (06:27 +0900)]
Specify the signedness of 'char' in HFA testcases (#4680)
In C#, sbyte always represents 8-bit signed interger.
However, the signedness of char in C is undefined, and the default
signedness of char in ARM is unsigned.
The native code for HFA test does not specify the signedness of char,
and thus the native code assumes that it is unsigned, but the managed
code assumes that it is signed, which leads to unittest failure in ARM.
This commit specifies the signedness of 'char' in order to fix #4639.
Pat Gavlin [Fri, 29 Apr 2016 20:23:52 +0000 (13:23 -0700)]
Eliminate SEH interactions between the JIT and the EE during verification.
For compatibility reasons, `BadImageFormatException`s that are thrown when resolving tokens
during verification are transformed instead into verification exceptions. This transformation
currently takes place in the JIT: the JIT wraps relevant calls to `ICorJitInfo::resolveToken` with
a `__try`/`__except` that handles any exceptions that arise due to invalid metadata. This method
cannot be relied on when consuming the JIT as a dynamic library on non-Windows platforms
due to type identity problems in the runtime's SEH emulation. Furthermore, these filters cannot
use the proposed solution to the JIT's other uses of SEH across the JIT/EE interface, since they
are more complex than a call to `ICorJitInfo::FilterException`. Instead, this change introduces a
new method, `ICorJitInfo::tryResolveToken`, which is only used during verification. This method
implements filtering that is equivalent to the current implementation, but does so in the EE rather
than in the JIT.
This is part of GitHub bugs #3058 and #4130.
[tfs-changeset: 1600443]
Swaroop Sridhar [Fri, 29 Apr 2016 20:11:48 +0000 (13:11 -0700)]
Merge pull request #4655 from swaroop-sridhar/aagc3
ARM64: Update FP-context-pointer in ExceptionFrame
Lakshmi Priya [Fri, 29 Apr 2016 19:47:00 +0000 (12:47 -0700)]
Merge pull request #4662 from Priya91/ubuntu1604
Remove Ubuntu15.10 support.
Mike McLaughlin [Tue, 26 Apr 2016 23:57:31 +0000 (16:57 -0700)]
Fix issue #4298 "SIGSEGV_libcoreclr.so!Debugger::GetArgCount"
The fix is to remove the call to TerminateDebugger in the EE shutdown
path. The reason was to clean up the transport pipe files but that
still happens in coreclr_uninitialize called by the host.
Also added code to clean up the transport named pipes and semaphores
on the debugger side when it detects that the target process has
terminated before it sends the ExitProcess notification.
Plumbed the cleanup call from dbi's ExitProcess code through the
native pipe line to the transport and then to pipe code.
Add PAL_CleanupTargetProcess for the "continue" named semaphore cleanup.
Found and fixed a minor race in dbgshim register runtime startup.
Bruce Forstall [Fri, 29 Apr 2016 18:01:57 +0000 (11:01 -0700)]
Merge pull request #4670 from BruceForstall/Fix3596
Fix 3596: attempt to spill a returned fp local w/o a frame location
Bruce Forstall [Fri, 29 Apr 2016 18:01:21 +0000 (11:01 -0700)]
Merge pull request #4656 from BruceForstall/x86fp
Update RyuJIT/x86 XMM register definitions and usage
Sean Gillespie [Fri, 29 Apr 2016 16:46:46 +0000 (09:46 -0700)]
Merge pull request #4658 from swgillespie/gc-perf-typo
Fix a typo in GCMetrics that provides the wrong generation's statistics
MyungJoo Ham [Fri, 29 Apr 2016 10:38:58 +0000 (19:38 +0900)]
ARM Enable MulticoreJIT for mscorlib.dll build
The native parts of CoreCLR has kept FEATURE_MULTICOREJIT
for ARM devices from the beginning while MSCORLIB.DLL's
build scripts has kept it disabled for ARM.
Because many ARM devices are multicore and
the native part is using FEATURE_MULTICOREJIT anyway,
in order to remove the inconsistency, we enable
FEATURE_MULTICOREJIT for mscorlib build as well.
Fix #4682 part 2.
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
MyungJoo Ham [Fri, 29 Apr 2016 10:34:36 +0000 (19:34 +0900)]
Ensure FEATURE_MANAGED_ETW isn't set for TargetUnix 2
The commit
82492f7fe7d91331508cb1bf9b33e2e9725d9898
has disabled FEATURE_MANAGED_ETW for UNIX builds.
However, the commit somehow omitted updating
CMakeList accordingly while updating msbuild scripts.
This patch does the same for CMake scripts.
Fixes #4682 part 1.
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>