Brian Robbins [Thu, 19 May 2016 23:08:32 +0000 (16:08 -0700)]
Generate perfmap symbol file for System.Private.CoreLib.
Kyungwoo Lee [Thu, 19 May 2016 21:48:42 +0000 (14:48 -0700)]
Merge pull request #5099 from kyulee1/fixr2r
ARM64: Fix R2R EntryPoint for Intrinsic
Russ Keldorph [Thu, 19 May 2016 21:39:49 +0000 (14:39 -0700)]
Merge pull request #5098 from RussKeldorph/objects
Mark doublealign test as incompatible with GCStress
Russ Keldorph [Thu, 19 May 2016 21:39:09 +0000 (14:39 -0700)]
Merge pull request #5100 from RussKeldorph/gcstress
Fix GCStress=0xC when build(tests).cmd called from outside root
Gaurav Khanna [Thu, 19 May 2016 19:54:30 +0000 (12:54 -0700)]
Merge pull request #5091 from gkhanna79/Fix4531
Make Microsoft.NetCore.JIT package dependency of CoreCLR package
Russ Keldorph [Thu, 19 May 2016 17:44:38 +0000 (10:44 -0700)]
Fix GCStress=0xC when build(tests).cmd called from outside root
Use absolute path intead of relative path when invoking
setup-runtime-dependencies.cmd, which is required for GCStress=0xC to
work.
Sean Gillespie [Thu, 19 May 2016 17:17:17 +0000 (10:17 -0700)]
Merge pull request #4991 from swgillespie/gcsimulator-buildandrun
Switch GCSimulator to BuildAndRun so that a shell wrapper gets generated
Mike McLaughlin [Thu, 19 May 2016 17:14:56 +0000 (10:14 -0700)]
Implement source/line number support for Exception.StackTrace. (#5078)
* Implement source/line number support for Exception.StackTrace.
Uses a new helper class (StackTraceSymbols) in System.Diagnostics.StackTrace assembly
to get source/line info for portable PDBs. On Windows both types of pdbs are support;
on xplat only portable PDBs.
The first part of the work is in the GetStackFramesInternal pinvoke to return the neccessary
info needed by the portable pdb reader: MethodToken, assembly path, loaded PE image address
and if in-memory symbol file, the address.
The second part is in the StackFrameHelper in mscorlib. It attempts to load the System.Diagnostics.StackTrace,
gets the new type and creates a delegate to the helper function which is called for each stack
frame.
* Code review feedback.
William Godbe [Thu, 19 May 2016 17:13:10 +0000 (10:13 -0700)]
Merge pull request #5031 from wtgodbe/pipelineFix
Update build scripts to fix small errors
Kyungwoo Lee [Thu, 19 May 2016 16:51:56 +0000 (09:51 -0700)]
ARM64: Fix R2R EntryPoint for Intrinsic
This is fix for missing case that we should handle call that needs to pass
indirect cell address. The call appears in rationalizer, which is
originally an intrinsic when we import it.
I also refactor the code so that when we set entryPoint we ensure
tagging such information on the tree.
Russ Keldorph [Thu, 19 May 2016 17:09:22 +0000 (10:09 -0700)]
Mark doublealign test as incompatible with GCStress
The test makes assumptions about allocation alignment that cannot be
expected to hold under GCStress, particularly on 32-bit platforms.
SaeHie Park [Thu, 19 May 2016 16:25:00 +0000 (01:25 +0900)]
ARM, SOS: correct IP and SP value in ClrStack command (#5077)
This fixes casting 64bit pointer value in 32bit machines
that show wrong IP and SP values in ClrStack command.
Related Issue: #5037
Russ Keldorph [Thu, 19 May 2016 14:55:38 +0000 (07:55 -0700)]
Merge pull request #5057 from RussKeldorph/smallframe
Exclude smallFrame for JIT32
Senthil [Thu, 19 May 2016 06:48:34 +0000 (23:48 -0700)]
Merge pull request #5001 from AlexGhiondea/ReadAppContextOverrides
Introduce a way to override AppContext switches from the project.json
Jan Kotas [Thu, 19 May 2016 04:47:22 +0000 (21:47 -0700)]
Merge pull request #5076 from dotnet-bot/from-tfs
Merge changes from TFS
mkborg [Thu, 19 May 2016 04:44:54 +0000 (08:44 +0400)]
Skip configuration with skipnative option. (#5059)
Fixes #4022.
Gaurav Khanna [Thu, 19 May 2016 04:38:48 +0000 (21:38 -0700)]
Make Microsoft.NetCore.JIT package dependency of CoreCLR package
Kyungwoo Lee [Thu, 19 May 2016 03:36:50 +0000 (20:36 -0700)]
Merge pull request #5073 from kyulee1/newtest
ARM64: New Test Binary Update
Kyungwoo Lee [Thu, 19 May 2016 01:30:55 +0000 (18:30 -0700)]
Merge pull request #5020 from kyulee1/imp
ARM64: Enable End-To-End ReadyToRun (R2R) Crossgen
Kyungwoo Lee [Wed, 18 May 2016 21:10:57 +0000 (14:10 -0700)]
ARM64: New Test Binary Update
Currently 135 of 9661 are failing.
Those are marked as EXPECTED_FAIL.
~55: mcc failed due to differnet struct calling convetion
~13: regression from the prior tests: marked as EXPECTED_FAIL;REGRESS
~22: failures from new tests: marked as NEW;EXPECTED_FAIL
~45: HFA
Tests.lst head now contains git hash number which tells us what version
we use to build such binaries. This is also used to dispatch different test folder in the lab.
sbomer [Wed, 18 May 2016 23:40:12 +0000 (16:40 -0700)]
Port from tfs: onload exception debugger crash fix (#4868)
The ExceptionHijackPersonalityRoutine was retrieving a context from a
particular stack frame, but the calling conventions for the function
executing in that frame allow it to overwrite the context. This was
causing the debugger to crash for exceptions thrown from the onload
method in winforms on x64.
The fix is to instead retrieve the context from the previous stack
frame, whose layout is set up explicitly in ExceptionHijack.
Bruce Forstall [Wed, 18 May 2016 23:18:38 +0000 (16:18 -0700)]
Merge pull request #5070 from dotnet-bot/from-tfs
Merge changes from TFS
Sean Gillespie [Mon, 16 May 2016 18:53:07 +0000 (11:53 -0700)]
Switch GCSimulator to BuildAndRun so that a shell wrapper gets generated
wtgodbe [Tue, 17 May 2016 23:27:35 +0000 (16:27 -0700)]
Update build scripts to fix small errors
Alex Ghiondea [Wed, 18 May 2016 21:57:00 +0000 (14:57 -0700)]
Explicitly reference the file containing the mscorlib reference assembly.
[tfs-changeset: 1606007]
Aditya Mandaleeka [Wed, 18 May 2016 21:51:05 +0000 (14:51 -0700)]
Merge pull request #5039 from adityamandaleeka/sigterm_chaining
Put SIGTERM registration behind a PAL initialization flag
Hugh Bellamy [Wed, 18 May 2016 21:27:37 +0000 (22:27 +0100)]
Cache GetLowerBound in Array.Reverse and Array.Sort (#5019)
- Avoids calling internal native method 4 times per sort: 2 times is
enough
- Avoids calling internal native method 2 times per reverse: 1 time is
enough
Kyungwoo Lee [Sat, 14 May 2016 20:06:57 +0000 (13:06 -0700)]
ARM64: Enable End-To-End ReadyToRun (R2R) Crossgen
Fixes https://github.com/dotnet/coreclr/issues/4649
The immediate issues was NYI on genEmitHelperCalls. The initial
implementation for the missing part was enough to just crossgen System.dll.
But running tests revealed various issues in crossgened binaries (R2R).
Most common user/helper calls in R2R are represented as indirect calls
similar to interface call using virtual stub dispatch cell --
thunk/helper needs a indirect cell address to update the final target
address on the data location. `IsDelayLoadHelper` and `IsLazyHelper` belong to
this case.
Instead of passing such parameter, x64/x86 uses an encoding trick -- it
assumes the call is dispatched like `call [addr]`. So from the return
address, runtime could extract indirect cell address. Unfortunately this is not
an option for arm64 (actually arm as well but I haven't fixed it in this
change) where indirect call on memory is not encodable.
So, I made the following changes:
1. For the call requiring that needs to pass indirect cell address, I
tagged the call tree via `setR2RRelativeIndir`. Tried to be comprehensive,
but I may miss something. Currently, it includes a regular call and
various helpers for (virtual) load function pointer/static data access, etc.
Hopely we change JIT/EE interface somehow that gives us such explicit information.
2. Use the X11 to record indirect cell address for such call tree in
lower similar to VSD.
3. Fixed encodings `ZapIndirectHelperThunk`. In particular the immediate
value/offset for `ldr` should be scaled down 4 times since HW will scale
it 4 times.
4. Implement `genEmitHelperCalls` for indirect case. This is not the case requiring indirect
cell address. This is the case we inlined the indirect helper thunk for
the speed. I'm seeing the case for size opt helper call, we invoke a
direct call to such thunk which actually uses x12 to dispatch the final
target. Likewise, I used x12 for this expansion which seems a trash register that is not
overlapped with arugments with jit helpers like writer barriers.
With this change, I've tested various cases/scenraios locally.
Also I've verified all tests are passed against mscorlib.ni.dll and System.ni.dll.
Rama krishnan Raghupathy [Wed, 18 May 2016 21:17:54 +0000 (14:17 -0700)]
Merge pull request #5069 from ramarag/R2RCleanup
Removing hard coded path Fixes #5065
Yi Zhang (CLR) [Wed, 18 May 2016 20:36:13 +0000 (13:36 -0700)]
Merge pull request #5048 from yizhang82/hostpath-fix
Move NATIVE_DLL_SEARCH_DIRECTORIES probing earlier right before we pr…
Russ Keldorph [Wed, 18 May 2016 20:10:28 +0000 (13:10 -0700)]
Merge pull request #5044 from RussKeldorph/lifetime2
Fix and reenable lifetime2
James Ko [Wed, 18 May 2016 20:09:33 +0000 (16:09 -0400)]
Remove Encoding FEATURE_* defines from mscorlib (#5026)
* Remove Encoding FEATURE_* defines from mscorlib
* Remove strange character from Encoding.cs
Rama Krishnan Raghupathy [Wed, 18 May 2016 19:28:20 +0000 (12:28 -0700)]
Removing hard coded path
Bruce Forstall [Wed, 18 May 2016 19:33:06 +0000 (12:33 -0700)]
Merge pull request #5062 from dotnet-bot/from-tfs
Merge changes from TFS
Rahul Kumar [Wed, 18 May 2016 19:16:50 +0000 (12:16 -0700)]
Merge pull request #5036 from rahku/fixes
Fix for issue# 3651
Venkata Sivaramakrishna Ramadugu [Wed, 18 May 2016 17:59:02 +0000 (10:59 -0700)]
Fix to Jit32 test b46659.exe failure.
Here is the git hub issue: https://github.com/dotnet/coreclr/issues/4547
The repro case is
i = -111
while ( i > 0)
{
// another nested loop goes here
}
return 100
After assertionProp, due to while loop condition lot of blocks/stmts get deleted as unreachable but optLoopCount is never decremented. Later when optLoopCodeMotion() gets called, it iterates through optLoopTable[] and tries to access head
if ((loopRmv & loopBit) ||
tail == 0 ||
head->bbTreeList == 0 )
Since entire method reduces to a single basic block, head will be nullptr and accessing head->bbTreeList is leading to an AV.
Fix: As a safe fix for RTM, also check whether head is nullptr.
[tfs-changeset: 1605919]
Jeremy Kuhne [Wed, 18 May 2016 17:41:27 +0000 (10:41 -0700)]
Merge pull request #4902 from JeremyKuhne/WindowsLongPath
Allow long paths in CoreCLR managed code
Bruce Forstall [Wed, 18 May 2016 17:33:28 +0000 (10:33 -0700)]
Rollback CS#1603899 that led to a JIT assert ngen'ing System.Windows.Forms.dll
====================
007551: Merge pull request #1241 from mikedn/modopt
Extend the DIV/MOD dividend into RDX:RAX only if needed
====================
Assert failure(PID 33656 [0x00008378], Thread: 17792 [0x4580]): Assertion failed 'addr->OperIsAddrMode() || (addr->IsCnsIntOrI() && addr->isContained()) || !addr->isContained()' in 'System.Windows.Forms.CheckedListBox:OnDrawItem(ref):this' (IL size 1216)
File: e:\dd\projectk\src\ndp\clr\src\jit\emitxarch.cpp Line: 2698
Image: C:\Windows\Microsoft.NET\Framework64\v4.0.rb1605209\mscorsvw.exe
The tree:
***** BB41, stmt 82 (embedded)
( 6, 8) [003723] ------------ * stmtExpr void (embedded) (IL 0x109... ???)
N1045 ( 3, 2) [000115] ------------ | /--* lclVar ref V00 this u:2 REG rcx $80
N1047 ( 1, 4) [002642] ------------ | +--* const long 344 field offset Fseq[idealCheckSize] REG NA $10b
N1049 ( 4, 6) [002643] -------N---- | /--* + byref REG NA $356
N1051 ( 6, 8) [000116] ----GO------ | /--* indir int REG rcx <l:$685, c:$2ef>
N1053 ( 6, 8) [003669] DA--GO------ \--* st.lclVar int V172 cse1 rcx REG rcx RV
During codegen:
Generating BB41, stmt 71 Holding variables: [rbx rsi rdi r12-r15]
Generating: N1043 ( 3, 2) [000114] ------------ * lclVar int V05 loc3 u:3 r12 (last use) REG r12 RV $31a
Generating: N1045 ( 3, 2) [000115] ------------ * lclVar ref V00 this u:2 REG rcx $80
IN00db: mov rcx, gword ptr [V00 rbp+10H]
GC regs:
00000040 {rsi} =>
00000042 {rcx rsi}
Generating: N1047 ( 1, 4) [002642] ------------ * const long 344 field offset Fseq[idealCheckSize] REG NA $10b
Generating: N1049 ( 4, 6) [002643] -------N---- * + byref REG NA $356
Generating: N1051 ( 6, 8) [000116] ----GO------ * indir int REG rcx <l:$685, c:$2ef>
... assert ...
(This is rollback #2: the TFS/GitHub mirror unfortunately undid rollback CS#1605814 with CS#1605840. This change should avoid that problem.)
[tfs-changeset: 1605917]
Russ Keldorph [Wed, 18 May 2016 17:22:45 +0000 (10:22 -0700)]
Exclude smallframe for JIT32
This test was flagged as incompatible with COMPlus_JITMinOpts in the
previous test harness. It doesn't fail for RyuJIT, though, even under
MinOpts, so, rather than disabling it everywhere under MinOpts, I'm only
disabling it for JIT32.
Fixes #4548
Fadi Hanna [Wed, 18 May 2016 17:18:53 +0000 (10:18 -0700)]
Porting the changes to support generic dictionaries lookups in R2R images to the JIT32 implementation, to support the x86 architecture.
All of the changes in importer.cpp were ported from JIT\importer.cpp, which was merged to the coreclr master branch last Friday evening (reviewed and signed-off in a PR on GitHub).
The remaining changes include:
1) Adjusting #ifdefs to enable the x86 scenario
2) Implementation of the R2R stub for generic dictionary lookups
3) Small change in prestub.cpp required for x86 to work correctly
[tfs-changeset: 1605906]
Alex Ghiondea [Wed, 18 May 2016 16:21:25 +0000 (09:21 -0700)]
Explicitly specify the location from where to pick up the model.corelib.xml
[tfs-changeset: 1605881]
Dongyun Jin [Wed, 18 May 2016 15:44:36 +0000 (00:44 +0900)]
Fix Interop TCs to use long long instead of long (#5054)
Some TCs were using long for 64bit variable.
While it is correct on x86, long is 32bit on ARM.
Fix TCs to use long long instead so that they can pass on ARM as well.
Fix #5053
Signed-off-by: Dongyun Jin <dongyun.jin@samsung.com>
Bruce Forstall [Wed, 18 May 2016 14:19:23 +0000 (07:19 -0700)]
Merge pull request #5055 from dotnet-bot/from-tfs
Merge changes from TFS
John Chen [Wed, 18 May 2016 09:46:23 +0000 (02:46 -0700)]
Fix crossgen /createpdb when input filename is given without path (#5045)
Fix a bug in "crossgen /createpdb" command that can cause it to fail
if the input filename is given without a path.
Swaroop Sridhar [Wed, 18 May 2016 09:44:44 +0000 (02:44 -0700)]
ARM64: Fix LR update in a few stubs (#5013)
Fix LR capture/restore in a few stubs helpers.
This change fixed failures in 6 GC tests under GCStress.
Stephen Toub [Wed, 18 May 2016 09:43:08 +0000 (05:43 -0400)]
Add private properties for debugger access to Task fields (#5024)
Jan Kotas [Wed, 18 May 2016 09:37:56 +0000 (02:37 -0700)]
Add missing method contract
[tfs-changeset: 1605842]
James Ko [Wed, 18 May 2016 09:19:30 +0000 (05:19 -0400)]
String.Equals: Don't typecast if object refs are the same (#5021)
Jan Kotas [Wed, 18 May 2016 09:18:20 +0000 (02:18 -0700)]
Merge pull request #5050 from dotnet-bot/from-tfs
Merge changes from TFS
Rama krishnan Raghupathy [Wed, 18 May 2016 07:59:10 +0000 (00:59 -0700)]
Merge pull request #5028 from ramarag/R2Rtests
Guard Crossgen Test Runtime logic with RunCrossGen Env
this fixes https://github.com/dotnet/coreclr/issues/3759
Matt Ellis [Wed, 18 May 2016 06:39:18 +0000 (23:39 -0700)]
Add missing .gitmirrorall file
[tfs-changeset: 1605815]
Yi Zhang [Wed, 18 May 2016 06:01:51 +0000 (23:01 -0700)]
Move NATIVE_DLL_SEARCH_DIRECTORIES probing earlier right before we probe for absolute path and assembly path. This is needed to make sure servicing directories (set in NATIVE_DLL_SEARCH_DIRECTORIES by host) are honored correctly before everything else, along with rest of the NATIVE_DLL_SEARCH_DIRECTORIES probing path. In order to preserve compat with UWP host, I kept the rest of the interop probing logic even when NATIVE_DLL_SEARCH_DIRECTORIES probing path is specified.
To my horror I discovered the fact that we actually never set hmod when loading DLL from NATIVE_DLL_SEARCH_DIRECTORIES probing, and the probably the only saving grace is that our ultimate fallback LoadLibrary call after that would "just work" and return the HMODULE we just loaded... This is now fixed.
Gaurav Khanna [Wed, 18 May 2016 05:50:26 +0000 (22:50 -0700)]
Merge pull request #5034 from dotnet/revert-4969-RelaxALC
Revert "Enable overriding assemblies from TPA"
tijoytom [Wed, 18 May 2016 05:36:25 +0000 (22:36 -0700)]
Merge pull request #5016 from tijoytom/master
[Do not Merge]Renaming ZeroFreeAllocMemUTF8 to ZeroFreeCoTaskMemUTF8
tijoytom [Wed, 18 May 2016 05:29:45 +0000 (22:29 -0700)]
Merge branch 'master' of https://github.com/dotnet/coreclr
Kshama Pawar [Wed, 18 May 2016 04:50:56 +0000 (21:50 -0700)]
Merge pull request #5032 from kspawa/r2rngen
Treat R2R images as NI for profiling
Russ Keldorph [Wed, 18 May 2016 04:33:51 +0000 (21:33 -0700)]
Fix and reenable lifetime2
lifetime2 is another questionable test case that relies on lifetime
guarantees that don't exist. Per our convention, force optimization on
and mark it as "optimization sensitive" to make a best effort to get it to
pass.
Fixes #1037
Geunsik Lim [Wed, 18 May 2016 02:44:23 +0000 (11:44 +0900)]
Linux/ARM: Change default clang version from 3.5 to 3.6 (#4928)
In case of the cross compilation on the ubuntu 14.04 x64
for Linux/ARM, the default version of clang/llvm that is
generated by "#> sudo ./cross/build-rootfs.sh arm" command
is 3.6 without 3.5 as following:
u14.04@x64> ls ./coreclr/cross/rootfs/arm/usr/lib/clang/3.6
include lib
u14.04@x64> ls ./coreclr/cross/rootfs/arm/usr/lib/llvm-3.6/
bin build include lib share
Let's keep the clang/llvm version consistently by moving
the default version from 'clang3.5' to 'clang3.6' at build-time
for Linux/ARM.
Signed-off-by: Geunsik Lim <geunsik.lim@samsung.com>
Rahul Kumar [Wed, 18 May 2016 02:26:06 +0000 (19:26 -0700)]
Fix for issue# 3651
Gaurav Khanna [Wed, 18 May 2016 01:46:21 +0000 (18:46 -0700)]
Revert "Enable overriding assemblies from TPA"
Rama Krishnan Raghupathy [Sat, 14 May 2016 02:41:46 +0000 (19:41 -0700)]
Guard Crossgen Test Runtime logic with RunCrossGen Env
Adds Semaphores to prevent race conditions
Pat Gavlin [Wed, 18 May 2016 00:19:37 +0000 (17:19 -0700)]
Merge pull request #4994 from pgavlin/gh4912.PAGE_SIZE.jit
Encapsulate or replace references to CORINFO_PAGE_SIZE.
Pat Gavlin [Wed, 18 May 2016 00:18:07 +0000 (17:18 -0700)]
Merge pull request #5015 from pgavlin/jitOnDllHooks
Remove jitOnDllProcess{Attach,Detach}.
James Ko [Wed, 18 May 2016 00:16:32 +0000 (20:16 -0400)]
Add more help options to corerun (#4973)
James Ko [Wed, 18 May 2016 00:12:15 +0000 (20:12 -0400)]
Remove redundant Rank checks from SZArrayHelper/Enumerator (#5022)
Aditya Mandaleeka [Wed, 18 May 2016 00:06:28 +0000 (17:06 -0700)]
Put SIGTERM registration behind a PAL initialization flag.
Dongyun Jin [Wed, 18 May 2016 00:01:05 +0000 (09:01 +0900)]
Some test cases require more than 2GB RAM (#4778)
Some test cases try to allocate 2GB, which is huge
for some testing environments.
Fix #4714
Signed-off-by: Dongyun Jin <dongyun.jin@samsung.com>
Kshama Pawar [Tue, 17 May 2016 23:42:35 +0000 (16:42 -0700)]
Treat R2R images as NI for profiling
Aditya Mandaleeka [Tue, 17 May 2016 23:37:35 +0000 (16:37 -0700)]
Merge pull request #4995 from adityamandaleeka/reenable_native_callable
Reenable NativeCallable test on x86 legacy backend
Aditya Mandaleeka [Tue, 17 May 2016 23:35:45 +0000 (16:35 -0700)]
Update contributing-workflow.md
Fix broken link to CODE_OWNERS file.
Dmitri-Botcharnikov [Tue, 17 May 2016 23:21:36 +0000 (03:21 +0400)]
Fix breakpoints on jitted code for Linux/ARM (#5014)
noahfalk [Tue, 17 May 2016 23:18:15 +0000 (16:18 -0700)]
Start using CODE_OWNERS.TXT (#5027)
Jose Perez Rodriguez [Tue, 17 May 2016 22:21:08 +0000 (17:21 -0500)]
Enable Checksum csc param for mscorlib (#4996)
Enable Checksum csc param for mscorlib
noahfalk [Tue, 17 May 2016 22:10:20 +0000 (15:10 -0700)]
Merge pull request #4954 from kvochko/sos_tests
Added tests for libsosplugin
tijoytom [Tue, 17 May 2016 22:03:35 +0000 (15:03 -0700)]
updating model.corlib.xml
tijoytom [Tue, 17 May 2016 22:01:38 +0000 (15:01 -0700)]
Merge branch 'master' of https://github.com/dotnet/coreclr
William Godbe [Tue, 17 May 2016 21:07:45 +0000 (14:07 -0700)]
Merge pull request #5023 from wtgodbe/pipeline
Add dev workflow scripts for build pipeline
wtgodbe [Wed, 27 Apr 2016 21:35:45 +0000 (14:35 -0700)]
Add dev workflow scripts for build pipeline
AlexGhiondea [Tue, 17 May 2016 20:45:59 +0000 (13:45 -0700)]
Merge pull request #4874 from gkhanna79/SPC_S1
Add support for System.Private.CoreLib to be core library
Sejong Oh [Tue, 17 May 2016 20:35:32 +0000 (13:35 -0700)]
Merge pull request #4999 from sejongoh/avoid_ti_method_check
Add typeInfo::GetMehtod2 to avoid checking GetType == TI_METHOD.
Pat Gavlin [Mon, 16 May 2016 21:05:17 +0000 (14:05 -0700)]
Encapsulate or replace references to CORINFO_PAGE_SIZE.
This symbol will be replaced with a new field in CORINFO_EE_INFO as part of #4912.
These changes prepare the JIT and EE for that change by (respectively) encapsulating
references to CORINFO_PAGE_SIZE in Compiler::eeGetPageSize and replacing
CORINFO_PAGE_SIZE with OS_PAGE_SIZE.
Pat Gavlin [Tue, 17 May 2016 16:05:04 +0000 (09:05 -0700)]
Remove jitOnDllProcess{Attach,Detach}.
These functions have been obviated by a more clearly-defined JIT
startup and shutdown interface.
This fixes a crash in coreclr.dll during DLL unload: the runtime was
calling jitOnDllProcessDetach (which calls jitShutdown) without having
called jitStartup to initialized the JIT.
Fadi Hanna [Tue, 17 May 2016 18:51:14 +0000 (11:51 -0700)]
Implementing the Unix calling convention for the generic dictionary access stubs in R2R (#4998)
Sejong OH [Mon, 16 May 2016 22:42:07 +0000 (15:42 -0700)]
Add typeInfo::GetMehtod2 to avoid checking GetType == TI_METHOD.
If FEATURE_CORECLR is enabled, the pointer type is not known to be a method pointer type when GetMethod is called in the importer.
Sejong Oh [Tue, 17 May 2016 18:16:27 +0000 (11:16 -0700)]
Merge pull request #4789 from sejongoh/always_download_coredistools
Always download coredistools package
Yi Zhang (CLR) [Tue, 17 May 2016 17:12:04 +0000 (10:12 -0700)]
Merge pull request #5008 from yizhang82/dllimportpath-fix2
Enable DefaultDllImportSearchPathsAttribute in CoreCLR Windows. Host …
Carol Eidt [Tue, 17 May 2016 16:41:30 +0000 (09:41 -0700)]
Merge pull request #5002 from CarolEidt/Long32Doc
Document about supporting long on 32-bit architectures.
James Ko [Tue, 17 May 2016 15:46:08 +0000 (11:46 -0400)]
Optimize Array.GetLength, GetUpperBound, GetLowerBound (#4993)
... if the argument is 0. As the rank will always be greater, this is safe.
Matt Warren [Tue, 17 May 2016 15:42:25 +0000 (16:42 +0100)]
Update StringObject comment in object.h (#5011)
* m_ArrayLength is no longer present (I did a search of the source and couldn't find it), so this comment is misleading.
* Also removing mis-leading comment about high two bits of field
Jan Kotas [Tue, 17 May 2016 15:36:57 +0000 (08:36 -0700)]
Move notion of restricted physical memory behind the GCToOSInterface (#5007)
Russ Keldorph [Tue, 17 May 2016 15:32:15 +0000 (08:32 -0700)]
Merge pull request #4997 from RussKeldorph/fix3379
Rename some IL tests whose names conflicted with C# tests
Yi Zhang (CLR) [Wed, 4 May 2016 06:02:10 +0000 (23:02 -0700)]
Enable DefaultDllImportSearchPathsAttribute in CoreCLR Windows. Host NATIVE_DLL_SEARCH_DIRECTORIES has higher priority.
Jan Kotas [Tue, 17 May 2016 05:46:36 +0000 (22:46 -0700)]
Merge pull request #5003 from jkotas/corert-gc
Update GC from CoreRT
Gaurav Khanna [Mon, 9 May 2016 22:57:44 +0000 (15:57 -0700)]
Initial change to support System.Private.CoreLib.dll as Core Library.
Alex Ghiondea [Mon, 16 May 2016 23:28:23 +0000 (16:28 -0700)]
Introduce a way to override AppContext switches from the project.json file
When switch value is requested, we are going to check and see if we have an
override that we read from the project.json file. If we do, we are going to try
and parse that as a boolean and use as the value for that switch.
Kshama Pawar [Tue, 17 May 2016 02:25:33 +0000 (19:25 -0700)]
Merge pull request #4975 from kspawa/ldiv
Replace ldiv with normal division for PAL
Sejong Oh [Wed, 4 May 2016 21:06:16 +0000 (14:06 -0700)]
Always download coredistools package
runtest.sh/cmd currently downloads the coredistools package if COMPlus_GCStress variable is set by TestEnv or test-env options. This is a bit complicated and this doesn't work with a new option for GC stress testing in build/buildtest.cmd. This changeset moves downloading the coredistools package to buildtest.cmd where it takes gcstress option and downloads other packages.
Linux build doesn't have buildtest.sh, so package donwloading is still in
runtest.sh
Jan Kotas [Mon, 16 May 2016 23:50:07 +0000 (16:50 -0700)]
Update GC from CoreRT
https://github.com/dotnet/corert/tree/master/src/Native/gc
854b9ea7accfbd24423a10382c1a5a2bd6753450
Russ Keldorph [Mon, 16 May 2016 22:23:15 +0000 (15:23 -0700)]
Rename some IL tests whose names conflicted with C# tests
For catchrettoinnertry, we appeared to have some redundant test projects.
I removed the redundant ones and cleaned up the remaining.
Fixes #3379
Carol Eidt [Tue, 17 May 2016 00:03:30 +0000 (17:03 -0700)]
Document about supporting long on 32-bit architectures.