Brian Robbins [Thu, 21 Apr 2016 04:10:16 +0000 (21:10 -0700)]
Fix empty perf map files generated by crossgen.
Kyungwoo Lee [Tue, 19 Apr 2016 22:12:04 +0000 (15:12 -0700)]
Merge pull request #4414 from kyulee1/cleantest
ARM64: Clean Up Tests.lst
Kyungwoo Lee [Tue, 19 Apr 2016 18:11:03 +0000 (11:11 -0700)]
ARM64: Clean Up Tests.lst
- Remove all the issue tags that are resolved, or correct the issue number.
- Try to move UNSTABLE to either EXPECTED_PASS or EXPECTED_FAIL.
- Delete security transparency tests unavailable in .NET Core -- NonVirtualCall/OverwriterArray
Matt Mitchell [Tue, 19 Apr 2016 19:55:50 +0000 (12:55 -0700)]
Merge pull request #4415 from mmitche/update-ubuntu-deps
Update coreclr dependencies
Matt Mitchell [Tue, 19 Apr 2016 18:26:47 +0000 (11:26 -0700)]
Update coreclr dependencies
Andy Ayers [Tue, 19 Apr 2016 06:21:02 +0000 (23:21 -0700)]
Merge pull request #4374 from AndyAyersMS/CheckToolRestore
Detect if powershell version is suitable for restoring build tools
Rahul Kumar [Tue, 19 Apr 2016 05:18:43 +0000 (22:18 -0700)]
Merge pull request #4349 from rahku/master
enable build of cross target components.
This enables the build of following cross target components:
1. crossgen
2. mscordacore.dll
3. sos.dll
4. mscordbi.dll
This is currently only enabled for arm64 and generates x64 binaries. For cross builds it uses the same hierarchy of cmake files from source root folder. At the end of build, build calls cmake again to generate solution for cross build specifying both host & target arch. This PR adds host & target arch configuration in cmake. I decided to reuse CLR_CMAKE_PLATFORM_ARCH_* to be as the host arch and CLR_CMAKE_TARGET_ARCH_* to be as target arch. For non-cross builds they are one and same and differ in cross builds. I have also refactored root CMakeLists.txt into smaller files. Binaries for cross builds are placed at bin\product<OS.Arch.Type><hostArch> (eg for arm64 they will be at bin\product\OS.arm64.debug\x64)
Rahul Kumar [Tue, 12 Apr 2016 21:40:23 +0000 (14:40 -0700)]
enable build of cross target components.
Currently only enabled for arm64
Jan Kotas [Tue, 19 Apr 2016 04:54:30 +0000 (21:54 -0700)]
Merge pull request #3915 from hughbe/region-info-params
Add parameter names to RegionInfo constructor
Jan Kotas [Tue, 19 Apr 2016 04:53:06 +0000 (21:53 -0700)]
Merge pull request #3535 from hughbe/patch-3
Add parameter names to ArgumentExceptions in Guid
Jan Kotas [Tue, 19 Apr 2016 04:50:56 +0000 (21:50 -0700)]
Merge pull request #4247 from leemgs/upstream-cross-toolchain-unify2
Linux/ARM: Unifying the redundant usages of objcopy
Jan Kotas [Tue, 19 Apr 2016 04:41:33 +0000 (21:41 -0700)]
Merge pull request #4283 from krytarowski/netbsd-support-61
Implement GetEntrypointExecutableAbsolutePath() on NetBSD
Jan Kotas [Tue, 19 Apr 2016 04:38:47 +0000 (21:38 -0700)]
Merge pull request #4371 from krytarowski/netbsd-support-65
Add NetBSD support in dir.props
Jan Kotas [Tue, 19 Apr 2016 04:38:14 +0000 (21:38 -0700)]
Merge pull request #4348 from AndyAyersMS/FixPerfMapAV
Fix a segfault in perf map generation
Jan Kotas [Tue, 19 Apr 2016 04:37:40 +0000 (21:37 -0700)]
Merge pull request #4373 from krytarowski/netbsd-support-67
NetBSD: Assume GNU GCC compatible va_list in Clang/LLVM
Jan Kotas [Tue, 19 Apr 2016 04:33:44 +0000 (21:33 -0700)]
Merge pull request #4401 from dotnet-bot/from-tfs
Merge changes from TFS
noahfalk [Tue, 19 Apr 2016 01:40:33 +0000 (18:40 -0700)]
Merge pull request #4312 from mjsabby/muks/enable_rejit_xplat
Enable ReJIT on all platforms
Kyungwoo Lee [Tue, 19 Apr 2016 01:37:52 +0000 (18:37 -0700)]
Merge pull request #4392 from kyulee1/fixround
ARM64: Fix Round Operation
Kyungwoo Lee [Mon, 18 Apr 2016 17:42:51 +0000 (10:42 -0700)]
ARM64: Fix Round Operation
Math.Round is implemented as a target intrinsic for Arm64.
JIT emitted `frinta` which rounds the ties to away.
As described in MSDN
https://msdn.microsoft.com/en-us/library/system.math.round(v=vs.110).aspx#Round2_Example,
we should round the ties to even.
So, I corrected `frintn` instruction instead for such intrinsic expansion.
I've also identified that `EvalMathFuncUnary` incorrectly folds the round
operation when the argument is constant at the compile time.
The logic itself is not the right semantic as described above in MSDN.
I made a separate helper function which is essentially a duplicate of
classlib. This is not Arm64 specific fix but applies to all other targets.
So, I just fixed it while I'm here.
John Chen [Mon, 18 Apr 2016 23:38:39 +0000 (16:38 -0700)]
Fix nidump to avoid crashing while working with Ready to Run images
[tfs-changeset: 1597064]
Rahul Kumar [Mon, 18 Apr 2016 22:58:44 +0000 (15:58 -0700)]
Merge pull request #4394 from gkhanna79/VCRTArm64Fix
Fix Arm64 break
Russ Keldorph [Mon, 18 Apr 2016 22:24:14 +0000 (15:24 -0700)]
Merge pull request #4391 from RussKeldorph/buildarch
Build break: Replace $(Platform) with $(BuildArch) in arch-specific IL projects
Pat Gavlin [Mon, 18 Apr 2016 20:06:30 +0000 (13:06 -0700)]
Merge pull request #4390 from pgavlin/JitCleanup
Cleanup some minor JIT issues.
Gaurav Khanna [Mon, 18 Apr 2016 19:49:49 +0000 (12:49 -0700)]
Fix Arm64 break
Aditya Mandaleeka [Mon, 18 Apr 2016 19:36:53 +0000 (12:36 -0700)]
Merge pull request #4370 from krytarowski/netbsd-support-64
Print message about detected NetBSD
Aditya Mandaleeka [Mon, 18 Apr 2016 19:36:27 +0000 (12:36 -0700)]
Merge pull request #4378 from krytarowski/netbsd-support-68
Implement GetProcessIdDisambiguationKey() for NetBSD
Bruce Forstall [Mon, 18 Apr 2016 18:07:38 +0000 (11:07 -0700)]
Merge pull request #4323 from BruceForstall/CLRABI
Add CLR ABI document to Book of the Runtime
Bruce Forstall [Wed, 13 Apr 2016 23:07:06 +0000 (16:07 -0700)]
Add CLR ABI document to Book of the Runtime
Russ Keldorph [Sat, 16 Apr 2016 15:03:50 +0000 (08:03 -0700)]
Replace $(Platform) with $(BuildArch) in arch-specific IL projects
I recently fixed some IL projects for non-x64 architectures, but I tested
the wrong property. It appears that $(Platform) is set to AnyCpu, and
$(BuildArch) is what I want.
Pat Gavlin [Mon, 11 Apr 2016 21:12:21 +0000 (14:12 -0700)]
Standardize on `assert` in the JIT.
A small amount of JIT code was using `_ASSERTE` instead of `assert`.
Change that code to use `assert` instead, which operates correctly
w.r.t. the JIT host.
Pat Gavlin [Mon, 11 Apr 2016 21:04:54 +0000 (14:04 -0700)]
Stanardize on `DEBUGARG` in the JIT.
The JIT uses both `DEBUGARG` (defined in jit.h) and `DEBUG_ARG`
(defined in /src/inc/check.h) for debug-only arguments. This
change replaces all uses of `DEBUG_ARG` with `DEBUGARG`.
Pat Gavlin [Thu, 31 Mar 2016 19:44:53 +0000 (12:44 -0700)]
Remove uses of LP{,C}UTF8 in the JIT.
Replace these with the corresponding `char*` type.
Gaurav Khanna [Mon, 18 Apr 2016 17:27:55 +0000 (10:27 -0700)]
Merge pull request #4381 from gkhanna79/VCRTFix
Cleanup VCRuntime140.dll dependency from native components
Carol Eidt [Mon, 18 Apr 2016 17:09:44 +0000 (10:09 -0700)]
Merge pull request #3807 from CarolEidt/FirstClassStructsDoc
Design document for first-class structs
Andy Ayers [Mon, 18 Apr 2016 16:29:59 +0000 (09:29 -0700)]
Merge pull request #4375 from AndyAyersMS/UseBudget
Inliner: use time budget to avoid excessive inlining
Pat Gavlin [Mon, 18 Apr 2016 16:02:17 +0000 (09:02 -0700)]
Merge pull request #4376 from pgavlin/GcInfoRemoveUtilcode
Stop including utilcode.h in the GC info encoder.
Kamil Rytarowski [Sat, 16 Apr 2016 00:22:23 +0000 (02:22 +0200)]
Implement GetProcessIdDisambiguationKey() for NetBSD
Make use of the kvm(3) interface to get "struct kinfo_proc2" of the given
process.
struct kinfo_proc2 {
///....
uint32_t p_ustart_sec; /* STRUCT TIMEVAL: starting time. */
uint32_t p_ustart_usec; /* STRUCT TIMEVAL: starting time. */
///...
};
--- /usr/include/sys/sysctl.h
Gaurav Khanna [Sat, 16 Apr 2016 17:15:08 +0000 (10:15 -0700)]
Cleanup VCRuntime140.dll dependency from native components
Kyungwoo Lee [Sat, 16 Apr 2016 17:05:36 +0000 (10:05 -0700)]
Merge pull request #4360 from kyulee1/fixpsp
ARM64: Fix Storing PSPSym for localloc
Kyungwoo Lee [Fri, 15 Apr 2016 15:45:34 +0000 (08:45 -0700)]
ARM64: Fix Storing PSPSym for localloc
JIT incorrectly saves/restores PSPSym causing AVs in EH tests that have localloc with filter.
The fix is to use a right opcode (Load instead of Store) when saving PSPSym.
Jan Vorlicek [Sat, 16 Apr 2016 01:05:14 +0000 (03:05 +0200)]
Merge pull request #4357 from krytarowski/netbsd-support-63
NetBSD: Add support for PAL_IsDebuggerPresent()
Andy Ayers [Fri, 15 Apr 2016 21:03:46 +0000 (14:03 -0700)]
Detect if powershell version is suitable for restoring build tools
Powershell v3 or higher is required to decompress the build tools
bundle. Windows 7 typically has v2 installed.
Detect earlier versions and fail with a pointer to the
build prerequisites and a download link for an updated version.
Also propagate failures out of init-tools.cmd. The last step
of this is unchecked since the build tool init script can return
nonzero status even on success.
See #3510.
Lubomir Litchev [Sat, 16 Apr 2016 00:01:40 +0000 (17:01 -0700)]
Merge pull request #4346 from LLITCHEV/Issue4276
Set the lvIsMultiRegArgOrRet for a variable containing the result of in-lined multi-register return call.
Jan Kotas [Fri, 15 Apr 2016 23:58:36 +0000 (16:58 -0700)]
Merge pull request #4369 from mikedn/msgbox
Always try to load user32 instead of ntuser api set
Andy Ayers [Fri, 15 Apr 2016 01:49:25 +0000 (18:49 -0700)]
Fix a segfault in perf map generation
Make sure instance vars get initialized to nullptr.
Kamil Rytarowski [Fri, 15 Apr 2016 21:40:59 +0000 (23:40 +0200)]
NetBSD: Assume GNU GCC compatible va_list in Clang/LLVM
Retire unneeded __GNUC_VA_LIST as currently all platforms build with Clang.
Jan Vorlicek [Fri, 15 Apr 2016 23:20:33 +0000 (01:20 +0200)]
Merge pull request #4342 from ZhichengZhu/clrstackr
add sos clrstack -r option
Sean Gillespie [Fri, 15 Apr 2016 23:19:52 +0000 (16:19 -0700)]
Merge pull request #4368 from swgillespie/long-running-gc-fixes
Fix a difference in passing the sequential parameter between Bash and…
Pat Gavlin [Fri, 15 Apr 2016 22:35:33 +0000 (15:35 -0700)]
Stop including utilcode.h in the GC info encoder.
This required two other minor changes:
- A prototype for `ThrowOutOfMemory` is now defined in gcinfoencoder.h
- `BitArray` now defines an overload of `operator new` that accepts an
`IAllocator`.
Aditya Mandaleeka [Fri, 15 Apr 2016 22:45:07 +0000 (15:45 -0700)]
Merge pull request #4372 from krytarowski/netbsd-support-66
NetBSD has RAND_MAX = 0x7fffffff
Kamil Rytarowski [Fri, 15 Apr 2016 21:17:38 +0000 (23:17 +0200)]
The RAND_MAX value can vary by platform
Noted on NetBSD where it is 0x7FFFFFFF.
Andy Ayers [Fri, 15 Apr 2016 18:24:25 +0000 (11:24 -0700)]
Inliner: use time budget to avoid excessive inlining
Use the time budget and time estimates to stop inlining once the
overall jit time increase estimate for the method is 10x the initial
jit time estimate.
This is implemented as part of `LegacyPolicy` and so can impact
the current inline behavior.
The budget is intentionally set quite high so that it only kicks in
for very rare cases where the call tree below the root is deep and wide
with many small methods. In extended testing on desktop this limit
fires in exactly two cases, both HFA tests.
In CoreCLR tests 12 of the HFA tests hit this limit. I've added
a directed test case here that came from the original bug report.
Closes #2472.
Kamil Rytarowski [Fri, 15 Apr 2016 11:23:31 +0000 (13:23 +0200)]
NetBSD: Add support for PAL_IsDebuggerPresent()
Reuse the kvm(3) interface to grab "struct kinfo_proc".
NAME
kvm - kernel memory interface
LIBRARY
Kernel Data Access Library (libkvm, -lkvm)
DESCRIPTION
The kvm library provides a uniform interface for accessing kernel virtual
memory images, including live systems and crash dumps. Access to live
systems is via /dev/mem while crash dumps can be examined via the core
file generated by savecore(8). The interface behaves identically in both
cases. Memory can be read and written, kernel symbol addresses can be
looked up efficiently, and information about user processes can be
gathered.
kvm_open() is first called to obtain a descriptor for all subsequent
calls
Rama krishnan Raghupathy [Fri, 15 Apr 2016 21:52:11 +0000 (14:52 -0700)]
Merge pull request #4347 from ramarag/zapnyi
Removing NYI for ZapUnwindData
Mike Danes [Fri, 15 Apr 2016 20:53:31 +0000 (23:53 +0300)]
Always try to load user32 instead of ntuser api set
Jan Vorlicek [Fri, 15 Apr 2016 21:33:39 +0000 (23:33 +0200)]
Merge pull request #4140 from myungjoo/implement/4040
Implement StartUnwindingNativeFrames Linux/ARM
Jan Kotas [Fri, 15 Apr 2016 21:18:17 +0000 (14:18 -0700)]
Merge pull request #3240 from krytarowski/netbsd-support-50
Fix issue unveiled on NetBSD: Add PAL__vsnprintf shadow in PAL
Kamil Rytarowski [Fri, 15 Apr 2016 21:14:39 +0000 (23:14 +0200)]
Add NetBSD support in dir.props
Kamil Rytarowski [Fri, 15 Apr 2016 21:01:26 +0000 (23:01 +0200)]
Print message about detected NetBSD
Lubomir Litchev [Thu, 14 Apr 2016 22:36:25 +0000 (15:36 -0700)]
Set the lvIsMultiRegArgOrRet for a variable containing the result of
inlined multi-register return call.
This change makes sure the lvIsMultiRegArgOrRet on a variable used to
store the result of inlined multi-register return function is set.
The way the code works today, this is not an issue since the code in
fgAttachStructInlineeToAsg always uses a CopyObj/Blk to asign the return
value from the registers to the var on stack. The CopyBlock/Obj gets the
address of the variable making it address-exposed and that prevents struct
promotion.
In the future (when CopyObj/Blk is not used) not having
lvIsMultiRegArgOrRet set could cause a problem.
Fixes issue 4276.
Sean Gillespie [Fri, 15 Apr 2016 20:44:12 +0000 (13:44 -0700)]
Fix a difference in passing the sequential parameter between Bash and Batch
Zhicheng Zhu [Thu, 14 Apr 2016 21:42:23 +0000 (14:42 -0700)]
add clrstack -r option
Gaurav Khanna [Fri, 15 Apr 2016 19:35:57 +0000 (12:35 -0700)]
Merge pull request #4359 from gkhanna79/ROLibFix2
DelayLoadAndFixWinRTInterop
Jose Perez Rodriguez [Fri, 15 Apr 2016 18:05:40 +0000 (11:05 -0700)]
Merge pull request #4361 from joperezr/AddServiceableAttribute
Adding Serviceable attribute to mscorlib
Koundinya Veluri [Fri, 15 Apr 2016 17:12:07 +0000 (10:12 -0700)]
Merge pull request #4352 from kouvel/ArmAssertFix
Fix assertion failure on ARM
Pat Gavlin [Fri, 15 Apr 2016 16:55:56 +0000 (09:55 -0700)]
Merge pull request #4338 from pgavlin/AddJit32Build
Make it possible to build JIT32 in the OSS tree.
Matt Mitchell [Fri, 15 Apr 2016 16:20:25 +0000 (09:20 -0700)]
Merge pull request #4363 from mmitche/update-corefx-paths
Update the corefx copy steps to point to the new location
Matt Mitchell [Fri, 15 Apr 2016 16:09:41 +0000 (09:09 -0700)]
Update the corefx copy steps to point to the new location
Jose Perez Rodriguez [Fri, 15 Apr 2016 15:52:51 +0000 (08:52 -0700)]
Adding Serviceable attribute to mscorlib
Gaurav Khanna [Fri, 15 Apr 2016 14:26:17 +0000 (07:26 -0700)]
DelayLoadAndFixWinRTInterop
Jan Kotas [Fri, 15 Apr 2016 12:48:39 +0000 (05:48 -0700)]
Merge pull request #4355 from dotnet-bot/from-tfs
Merge changes from TFS
MyungJoo Ham [Tue, 5 Apr 2016 02:25:36 +0000 (11:25 +0900)]
Implement StartUnwindingNativeFrames Linux/ARM
StartUnwindingNativeFrames for Unix/Linux ARM.
The code is ported from AMD64 code of the same function and OS.
CONTEXT_* are moved from context2.S to asmconstants.h so that
the same macros may be sharaed between multiple files.
Fix #4040
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Jan Vorlicek [Fri, 15 Apr 2016 09:47:04 +0000 (11:47 +0200)]
Merge pull request #4260 from myungjoo/fix/4259
ARM: fix build warning for unw usage
Koundinya Veluri [Fri, 15 Apr 2016 02:52:55 +0000 (19:52 -0700)]
Fix assertion failure on ARM
StompWriteBarrierEphemeral may be called on the init path, where the runtime is technically not suspended, but has not started yet. Added a check for g_fEEInit in the call.
Jan Kotas [Fri, 15 Apr 2016 05:41:24 +0000 (22:41 -0700)]
Merge pull request #4340 from jkotas/equality-comparer
Use beforefieldinit field for Comparer/EqualityComparer.Default
Jan Kotas [Fri, 15 Apr 2016 05:37:00 +0000 (22:37 -0700)]
Merge pull request #4353 from myungjoo/fix/4107
ARM/Linux: Mark r0-r3 Saved
MyungJoo Ham [Fri, 15 Apr 2016 04:28:29 +0000 (13:28 +0900)]
ARM/Linux: Mark r0-r3 Saved
Without this, exception handlers (try-catch) that
try to get SP of ThePreStub's caller (managed frame) mislocate
the caller's SP by 4 words (0x10) and get lost while
traversing managed frames.
Fix #4107
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Pat Gavlin [Fri, 15 Apr 2016 04:04:33 +0000 (21:04 -0700)]
Merge pull request #4343 from pgavlin/SeparateBitPosition
Move `BitPosition` into its own file.
Jan Kotas [Fri, 15 Apr 2016 01:46:52 +0000 (18:46 -0700)]
Fix razzle build breaks in bcl.win
[tfs-changeset: 1596379]
MyungJoo Ham [Fri, 15 Apr 2016 01:00:40 +0000 (10:00 +0900)]
Style fix: remove obsolete contents
We no longer have StartUnwindingNativeFrames.
Signed-off-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Jan Kotas [Thu, 14 Apr 2016 21:35:35 +0000 (14:35 -0700)]
Use beforefieldinit field for Comparer/EqualityComparer.Default
Matt Ellis [Thu, 14 Apr 2016 22:59:13 +0000 (15:59 -0700)]
Merge pull request #4326 from dotnet-bot/from-tfs
Merge changes from TFS
Andy Ayers [Thu, 14 Apr 2016 22:57:39 +0000 (15:57 -0700)]
Merge pull request #4335 from AndyAyersMS/InlineStrategy
Inliner: introducing InlineStrategy
Rama Krishnan Raghupathy [Wed, 6 Apr 2016 22:36:42 +0000 (15:36 -0700)]
Removing NYI for ZapUnwindData
Pat Gavlin [Thu, 14 Apr 2016 19:57:41 +0000 (12:57 -0700)]
Make it possible to build JIT32 in the OSS tree.
This change adds a new argument to build.cmd, buildjit32, that configures
the build to build and link JIT32 instead of RyuJIT if the sources are
available in `src/jit32`.
Fadi Hanna [Thu, 14 Apr 2016 21:18:21 +0000 (14:18 -0700)]
Merge pull request #4339 from fadimounir/deltests
Delete fragile tests (will fix tests in a later PR)
Fadi Hanna [Thu, 14 Apr 2016 21:14:52 +0000 (14:14 -0700)]
Delete fragile tests (will fix tests in a later PR)
Pat Gavlin [Mon, 11 Apr 2016 17:24:23 +0000 (10:24 -0700)]
Move `BitPosition` into its own file.
This will allow the GC info encoder and the JIT to use this
function without pulling in the entire utilcode header.
Andy Ayers [Tue, 12 Apr 2016 20:30:07 +0000 (13:30 -0700)]
Inliner: introducing InlineStrategy
Create the `InlineStrategy` class to hold inline-relevant data that
spans multiple inlines. It tracks the number of inline candidates,
attempts, successes, last successful policy, and holds onto the root
context, and so on. The strategy is responsible for creating contexts
and reporting overall results for a method (either tree-based or
data-based).
The strategy also includes a simple jit time estimate. From various
observations, post-inline jit time can be modelled acceptably by simple
linear relationships on the input IL size. The strategy uses these models
to estimate the initial and current jit time. Estimate units are roughly
microseconds.
The strategy creates a time budget to flag cases where the estimated
jit time increase due to inlining is unreasonbly large. The budget
is initially based on the root method size, and may increase if there
are non-discretionary force inlines.
Once we have a bit more vetting of the budgeting mechanism, policies
will use it to limit inlining in a small number of runaway inlining cases
(see for example #2472).
Remove code under MEASURE_INLINING since the strategy plus the context
tree (optionally extended via policies) contains all that data and more.
Likewise, consolidate a number of the compiler's inlining-related member
variables into the strategy.
Jan Vorlicek [Thu, 14 Apr 2016 20:36:08 +0000 (22:36 +0200)]
Merge pull request #4330 from janvorli/fix-sigsegv-deadlock
Fix possibility of deadlock in the SIGSEGV handler
Jan Kotas [Thu, 14 Apr 2016 19:59:17 +0000 (12:59 -0700)]
Fix full framework build break
[tfs-changeset: 1596259]
Pat Gavlin [Thu, 14 Apr 2016 19:57:01 +0000 (12:57 -0700)]
Merge pull request #4322 from pgavlin/TlsLogEnv
Collect the JIT's TLS variables in a single type.
Pat Gavlin [Thu, 14 Apr 2016 17:58:36 +0000 (10:58 -0700)]
Merge pull request #4314 from pgavlin/GcInfoSList
Replace `SList` in the GC info encoders.
Jan Vorlicek [Thu, 14 Apr 2016 17:44:56 +0000 (19:44 +0200)]
Fix potential deadlock when calling ExecutionManager::IsManagedCode
The function can deadlock if the calling thread was holding the
ExecutionManager writer lock.
Jose Perez Rodriguez [Thu, 14 Apr 2016 17:44:29 +0000 (10:44 -0700)]
Merge pull request #4308 from joperezr/RemoveWarnings
Remove native build warnings
Jose Perez Rodriguez [Tue, 12 Apr 2016 22:56:08 +0000 (15:56 -0700)]
Updating BuildTools
Russ Keldorph [Thu, 14 Apr 2016 14:10:24 +0000 (07:10 -0700)]
Merge pull request #4327 from RussKeldorph/structinregs
Fix structinregs test on x86
Jan Vorlicek [Wed, 6 Apr 2016 18:41:51 +0000 (20:41 +0200)]
Fix possibility of deadlock in the SIGSEGV handler
The SIGSEGV handler was transitively calling pthread_attr_get_np to
get current stack limit no matter where the exception happened. The
problem was that if the segmentation violation happened in a C
runtime function, this could lead to deadlock since pthread_attr_get_np
is not signal safe.
The fix is to add callback that will figure out if the failing instruction
was in managed code or in one of the jit helpers and if not, it would
not attempt to handle the signal.
I have also removed non signal safe calls from the signal handlers code
path where we don't know yet whether the interrupted code was in a function
that could collide with a function that we call from the signal handler
or not. This includes TRACE and ASSERT calls that were calling string
formatting functions.
Russ Keldorph [Thu, 14 Apr 2016 04:45:05 +0000 (21:45 -0700)]
Fix structinregs test on x86
Fix a typo in _MSC_VER resulting in a mismatched calling convention on
x86. While I'm in here, I'm also fixing some benign mismatched function
pointer type definitions.
Jan Kotas [Thu, 14 Apr 2016 03:54:21 +0000 (20:54 -0700)]
Add back Path APIs needed by UWP
[tfs-changeset: 1596013]