Jan Kotas [Mon, 23 Mar 2015 15:49:02 +0000 (08:49 -0700)]
Merge pull request #527 from jkotas/macunwind
Fix reporting of callee saved registers with nongnu libuwind
Jan Kotas [Mon, 23 Mar 2015 00:52:27 +0000 (17:52 -0700)]
Incorporated codereview feedback
Jan Kotas [Mon, 23 Mar 2015 00:45:46 +0000 (17:45 -0700)]
Merge pull request #535 from Djuffin/stress_log
Make stresslog work for libraries other than CoreClr (I need it for DBI)
Jan Kotas [Mon, 23 Mar 2015 00:10:31 +0000 (17:10 -0700)]
Merge pull request #532 from jkotas/slotassert
Fix invocation of shared generic multicast delegate
Jan Kotas [Mon, 23 Mar 2015 00:09:58 +0000 (17:09 -0700)]
Merge pull request #537 from stephentoub/comparison_fixes
Fixes to string comparison in stubbed out Unix globalization
Eugene Zemtsov [Mon, 23 Mar 2015 00:07:17 +0000 (17:07 -0700)]
Replace PAL_GetCoreClrModuleBase() with PAL_GetPalModuleBase()
Eugene Zemtsov [Fri, 20 Mar 2015 09:22:02 +0000 (02:22 -0700)]
Make stresslog work for libraries other than CoreClr (I need it for DBI)
Introducing of PAL_GetPalModuleBase removes dependency on CoreCLR name for
stress log module address.
Stephen Toub [Sun, 22 Mar 2015 20:04:45 +0000 (16:04 -0400)]
Fix string.ToUpper/ToLower to use the computed result
Amazing the difference a single character can make :)
Stephen Toub [Sun, 22 Mar 2015 19:08:16 +0000 (15:08 -0400)]
Fix CompareStringOrdinalAscii to support !ignoreCase
CompareStringOrdinalAscii is using the ignoreCase argument to decide whether to normalize the input characters to upper case. But the function then mistakenly ignores the result of this and always normalizes. This commit just changes the implementation to use the chars the resulted from deciding whether to normalize or not.
Stephen Toub [Sun, 22 Mar 2015 20:35:48 +0000 (16:35 -0400)]
Merge pull request #533 from jkotas/glob
Initialize number separators in stubbed out globalization
Jan Kotas [Sun, 22 Mar 2015 13:24:37 +0000 (06:24 -0700)]
Merge pull request #534 from jfrijters/botr-typos
Fixed two typos in BoTR clr-code-guide.md.
Jeroen Frijters [Sat, 21 Mar 2015 08:12:31 +0000 (09:12 +0100)]
Fixed two typos.
Jan Kotas [Sun, 22 Mar 2015 00:41:36 +0000 (17:41 -0700)]
Initialize number separators in stubbed out globalization
Jan Kotas [Sat, 21 Mar 2015 23:43:47 +0000 (16:43 -0700)]
Fix invocation of shared generic multicast delegate
JITing of IL stubs for invocation of shared generic multicast delegates
e.g. Action<string> a = .. + ..; a(); crashed in JIT on security check
because of IL stubs have non-standard MethodDesc. The security check is
unnecessary for IL stubs - the fix is to shortcircuit it for IL stubs.
Jan Kotas [Sat, 21 Mar 2015 18:43:19 +0000 (11:43 -0700)]
Merge pull request #530 from stephentoub/fix_indexofordinal
Fix an IndexOutOfRangeException in IndexOfOrdinal on Unix
Stephen Toub [Sat, 21 Mar 2015 17:05:30 +0000 (13:05 -0400)]
Fix an IndexOutOfRangeException in IndexOfOrdinal
In the temporary implementation of CompareInfo.IndexOfOrdinal on Unix, we sometimes hit an IndexOutOfRangeException as the inner loop tries to go past the end of the source string.
Stephen Toub [Sat, 21 Mar 2015 15:03:07 +0000 (11:03 -0400)]
Merge pull request #529 from stephentoub/tweak_glob_stubs
Stub out more globalization with values rather than with exceptions
Stephen Toub [Sat, 21 Mar 2015 13:45:10 +0000 (09:45 -0400)]
Stub out more globalization with values rather than with exceptions
Much of our globalization support is not yet implemented and is appropriately throwing NotImplementedException. However, this is hindering progress in other areas of the system that end up relying on globalization. This commit just changes the common cases where we were throwing NotImplementedException to instead return a default value; the "TODO: Implement this" comments have been left in place.
Jan Kotas [Sat, 21 Mar 2015 04:53:23 +0000 (21:53 -0700)]
Merge pull request #526 from kangaroo/issue-263
Correct #if conditional for WRITE_0_BYTES_HANGS_TTY
Geoff Norton [Sat, 21 Mar 2015 04:17:53 +0000 (21:17 -0700)]
Correct #if conditional for WRITE_0_BYTES_HANGS_TTY
Jan Kotas [Sat, 21 Mar 2015 04:08:31 +0000 (21:08 -0700)]
Fix reporting of callee saved registers with nongnu libuwind
The callee saved registers were not preserved for GC stackwalk with nongnu
libunwind that does not provide context pointers. Fixed by adding an extra
space to MachState to preserve them.
To avoid large amounts of copy&pasted code, introduced macro to enumerate
all callee saved registers, similar to existing macro to enumerate all
argument registers. This macro makes the code to copy the register values
between different structures more compact, and less prone to typos.
Jan Kotas [Fri, 20 Mar 2015 16:32:14 +0000 (09:32 -0700)]
Merge pull request #511 from kangaroo/unwind-fix-two
Fix unwinding past main on OSX
Geoff Norton [Thu, 19 Mar 2015 03:58:12 +0000 (20:58 -0700)]
Fix unwinding past main on OSX
When reaching the bottom of the real stack, OSX does not change the $pc
to 0x0, like llvm libunwind. It returns 0 from unw_step, and leaves the
$pc unchanged, so we will track cur and prev pc and guard for this as well
Fixes Localloc codegen test, and also verified a simple throw from main
works as expected.
Jan Kotas [Fri, 20 Mar 2015 14:05:27 +0000 (07:05 -0700)]
Merge pull request #522 from jkotas/gcstress
Workaround for DECIMAL calling convention in FCalls
Jan Kotas [Fri, 20 Mar 2015 12:18:28 +0000 (05:18 -0700)]
Merge pull request #520 from kangaroo/build-fix
Fix the build on OSX. std::string is used without <string>
Jan Kotas [Fri, 20 Mar 2015 12:15:51 +0000 (05:15 -0700)]
Workaround for DECIMAL calling convention in FCalls
Jan Kotas [Fri, 20 Mar 2015 11:43:16 +0000 (04:43 -0700)]
Merge pull request #521 from dotnet-bot/from-tfs
Merge changes from TFS
Jan Kotas [Fri, 20 Mar 2015 11:42:38 +0000 (04:42 -0700)]
Merge pull request #514 from soimort/fix-glibc-2.18-and-earlier
Fix build on glibc versions prior to 2.19
Jan Kotas [Fri, 20 Mar 2015 11:40:47 +0000 (04:40 -0700)]
Merge pull request #519 from jkotas/gcstress
Correct order of argument registers
Jan Kotas [Fri, 20 Mar 2015 11:16:15 +0000 (04:16 -0700)]
Fix contract for ETW::TypeSystemLog::FlushObjectAllocationEvents
[tfs-changeset: 1436668]
Jan Kotas [Fri, 20 Mar 2015 10:55:36 +0000 (03:55 -0700)]
Merge pull request #518 from janvorli/fix-personality-routines
Fix Unix asm stubs personality routines
Jan Vorlicek [Fri, 20 Mar 2015 00:51:53 +0000 (01:51 +0100)]
Fix Unix asm stubs personality routines
The Unix asm stubs had personality routines that were the windows personality
routines. This was incorrect since the signature of the Unix ones is completely
different. Moreover, some of these personality routines were not necessary
on Unix at all and two of them should dump the stack and terminate the process
instead of the handling that was necessary on Windows.
Finally, I've fixed an issue in the HandleHardwareException that contains
_ASSERTE in case the hardware exception was unhandled. But _ASSERTE contains
a debugger break instruction, which in the absence of the debugger invokes
the HandleHardwareException again. So I've added detection of the
breakpoint.
Jan Kotas [Fri, 20 Mar 2015 04:10:37 +0000 (21:10 -0700)]
Correct order of argument registers
Stephen Toub [Fri, 20 Mar 2015 04:04:05 +0000 (00:04 -0400)]
Merge pull request #515 from stephentoub/pal_startprocess
Add ForkAndExecProcess to PAL to support CoreFX
Geoff Norton [Fri, 20 Mar 2015 03:22:16 +0000 (20:22 -0700)]
Fix the build on OSX. std::string is used without <string>
Stephen Toub [Thu, 19 Mar 2015 15:16:13 +0000 (11:16 -0400)]
Add ForkAndExecProcess to PAL to support CoreFX
System.Diagnostics.Process.Start needs to fork/execve
in order to create the process, but fork is unsafe
in managed code. This commit pushes the core logic
needed to fork/execve into native code; a separate commit
in CoreFX will then take advantage of this function.
For now, I'm putting this in the runtime. Longer-term we
can evaluate whether it makes sense to distribute a separate
library with System.Diagnostics.Process just for this function.
Jan Kotas [Fri, 20 Mar 2015 02:41:56 +0000 (19:41 -0700)]
Merge pull request #517 from jaredpar/fix-build-unixmscorlib
Build unixmscorlib without CMake
Mort Yao [Fri, 20 Mar 2015 01:56:18 +0000 (09:56 +0800)]
Do not use _POSIX_C_SOURCE on Darwin
Mort Yao [Thu, 19 Mar 2015 10:57:48 +0000 (18:57 +0800)]
Explicitly specify POSIX.1-2008 conformance
- Fix build on glibc versions prior to 2.19
Jared Parsons [Thu, 19 Mar 2015 23:59:26 +0000 (16:59 -0700)]
Build unixmscorlib without CMake
The Linux build instructions don't list CMake as a pre-requisite for
building mscorlib yet build.cmd errors out if it is not present on the
machine. CMake isn't actually necessary for this particular build so
instead of requiring users to install CMake when not needed I added code
to skip the CMake detection logic in this case.
Jan Kotas [Thu, 19 Mar 2015 22:32:58 +0000 (15:32 -0700)]
Merge pull request #516 from jkotas/assemblyloadcontextfix
Throw FileNotFoundException in AssemblyLoadContext for unresolved Assemb...
Mike McLaughlin [Thu, 19 Mar 2015 22:27:26 +0000 (15:27 -0700)]
Merge pull request #500 from mikem8361/dumpcmds
Enable the rest of the sos dump commands and supporting functions.
Mike McLaughlin [Thu, 19 Mar 2015 21:55:18 +0000 (14:55 -0700)]
Merge remote-tracking branch 'upstream/master' into dumpcmds
Mike McLaughlin [Mon, 16 Mar 2015 22:02:02 +0000 (15:02 -0700)]
Enable the rest of the Dump commands and supporting functions.
Undo some mistakes made in my previous commits.
Don't create the sync manager, seh threads and seh itself.
Addressed @jkotas, @sergiy-k and @janvorli review feedback. Added a set of PAL_Initialize flags that control what gets initialized and a new PAL_InitializeDLL entry point that the SOS and DAC module use. This new entry point only does the subset of PAL init that is needed for a module outside of the coreclr runtime.
Fixed GetCurrentFrame() to get the current selected stack frame.
Used @janvorli suggestion of installing a PAL hardware exception and having it throw a C++ exception so the DAC can cause invalid pointers.
Jan Kotas [Thu, 19 Mar 2015 21:36:11 +0000 (14:36 -0700)]
Throw FileNotFoundException in AssemblyLoadContext for unresolved AssemblyName
FileLoadException thrown internally by AssemblyLoadContext for unresolved AssemblyName was escaping from Type.GetType(..., throwOnError:false).
The fix is to throw FileNotFoundException instead that is one of the specific types specifically filtered out for throwOnError:false.
Lubomir Litchev [Thu, 19 Mar 2015 21:27:14 +0000 (14:27 -0700)]
Merge pull request #468 from LLITCHEV/work
Enable ETW_EBP_FRAMED flag on System V systems.
Eugene [Thu, 19 Mar 2015 02:03:29 +0000 (19:03 -0700)]
Merge pull request #508 from dotnet-bot/from-tfs
Merge changes from TFS
Eugene Zemtsov [Thu, 19 Mar 2015 01:38:09 +0000 (18:38 -0700)]
Adding one more missing changes for ICastable
[tfs-changeset: 1435505]
Jan Kotas [Thu, 19 Mar 2015 01:28:18 +0000 (18:28 -0700)]
Merge pull request #507 from dotnet-bot/from-tfs
Merge changes from TFS
Jan Kotas [Thu, 19 Mar 2015 01:19:38 +0000 (18:19 -0700)]
Merge pull request #496 from dotnet-bot/from-tfs
Merge changes from TFS
Eugene Zemtsov [Thu, 19 Mar 2015 00:59:02 +0000 (17:59 -0700)]
Activating ICastable for CoreClr
[tfs-changeset: 1435446]
Jan Kotas [Thu, 19 Mar 2015 00:34:21 +0000 (17:34 -0700)]
Merge pull request #495 from kangaroo/issue-452
Fix unwinding through assembly helpers
Jan Kotas [Thu, 19 Mar 2015 00:33:59 +0000 (17:33 -0700)]
Merge pull request #498 from kangaroo/issue-192
Implement hardware exception handling for OSX
Geoff Norton [Wed, 18 Mar 2015 23:14:56 +0000 (16:14 -0700)]
Implement hardware exception handling for OSX
As a first pass this keeps the existing old PAL_DispatchException
machinery, but directs it to the new hardware exception handler
registered by the runtime. A future optimization can probably
eliminate the PAL_DispatchExceptionWrapper as we do not need to
work around the unwinding problems cited anymore.
Eugene Zemtsov [Wed, 18 Mar 2015 21:40:26 +0000 (14:40 -0700)]
Implement runtime support for ICastable interface
The goal of this change is to facilitate an alternative (MCG based) way of doing COM interop, we're going to use it on Unix platforms.
New ICastable interface allows objects to pretend at runtime that they support an interface and to provide an alternative type that is used to resolve actual calls to interface methods.
BE VERY CAREFUL: This is a very dangerous feature, and at this stage it can easily lead to memory corruption without any native code involved.
Reviewers: Yi Zhang, Noah Falk, Jan Kotas. DDR clean.
[tfs-changeset: 1435198]
Lubomir Litchev [Fri, 13 Mar 2015 15:34:17 +0000 (08:34 -0700)]
Enable ETW_EBP_FRAMED flag on System V systems.
Enable the ETW_EBP_FRAMED flag on System V systems to allow for using
frame pointer (RBP) for mostly all functions.
The change includes:
1. Removing RBP from all the lists of registers that the RA uses to assign
to tmps/vars/etc.
2. Enable generation of unwind info for pushing of REG_FPBASE even if not included in callee-save list
of registers.
3 Fixed a conservative assert - for System V systems stack offset for reg
passed argument can be 0 or even positive.
Geoff Norton [Wed, 18 Mar 2015 21:06:02 +0000 (14:06 -0700)]
Fix unwinding through assembly helpers
NESTED_ENTRY/NESTED_END assembly helpers are currently using cfi ops
that are not supported by Apple's compact unwind format. Additionally
the apple linker appears to be doing the wrong thing, and mapping functions
like CallDescrWorkerInternal to an invalid compatc unwind encoding. This
patch manually emits the __compact_unwind section manually for the affected
functions, and points the unwinder to parse the DWARF instead of using the
compact table.
Rich Lander [Wed, 18 Mar 2015 20:53:57 +0000 (13:53 -0700)]
Merge pull request #494 from richlander/rich-introclr
Add heading to intro
Jan Kotas [Wed, 18 Mar 2015 19:20:55 +0000 (12:20 -0700)]
Merge pull request #492 from jkotas/vsprompt
Remove check for VS Developer Command Prompt
Jan Kotas [Wed, 18 Mar 2015 16:09:22 +0000 (09:09 -0700)]
Merge pull request #490 from janvorli/hardware-exceptions
Implement hardware exception handling
Richard Lander [Wed, 18 Mar 2015 15:55:14 +0000 (08:55 -0700)]
Add heading to intro - fix links
Jan Vorlicek [Tue, 17 Mar 2015 16:31:48 +0000 (17:31 +0100)]
Implement hardware exception handling
This change implements handling of hardware exceptions that happened in the
managed code or in one of the JIT helpers. The coreclr registers a callback
with PAL. This callback is invoked by PAL whenever a hardware exception happens.
coreclr code then dispatches that exception using the DispatchManagedException.
Jan Kotas [Wed, 18 Mar 2015 03:33:30 +0000 (20:33 -0700)]
Remove check for VS Developer Command Prompt
Jan Vorlicek [Wed, 18 Mar 2015 08:49:34 +0000 (09:49 +0100)]
Merge pull request #486 from janvorli/handle-null-context-pointers
Fix handling NULL context pointers in one more place
Rich Lander [Wed, 18 Mar 2015 03:37:47 +0000 (20:37 -0700)]
Merge pull request #483 from richlander/rich-docsreadme
Convert document index into more GH-centric format
Lubomir Litchev [Wed, 18 Mar 2015 03:35:12 +0000 (20:35 -0700)]
Merge pull request #484 from LLITCHEV/work2
Fix a long standing bug for AMD64 with very large frames and using a scratch register.
Jan Kotas [Wed, 18 Mar 2015 00:46:02 +0000 (17:46 -0700)]
Merge pull request #491 from dotnet-bot/from-tfs
Merge changes from TFS
Jan Kotas [Tue, 17 Mar 2015 23:52:40 +0000 (16:52 -0700)]
Fix MacOS build break
[tfs-changeset: 1434447]
Jan Kotas [Tue, 17 Mar 2015 22:45:18 +0000 (15:45 -0700)]
Merge pull request #485 from dotnet-bot/from-tfs
Merge changes from TFS
Lubomir Litchev [Tue, 17 Mar 2015 17:49:06 +0000 (10:49 -0700)]
Fix a long standing bug for AMD64 with bad assert.
When having a really large frame the code for generating funclet prolog
expects to have RAX being used as initReg. For AMD64 the RBP is being
pushed and then used as scratch (and passed as initReg.) This was causing
the assert to trigger.
Added code to handle properly use of RBP on AMD64 and also modified the
assert to properly allow RAX and RBP as initReg.
Jan Vorlicek [Tue, 17 Mar 2015 21:42:02 +0000 (22:42 +0100)]
Fix handling NULL context pointers in ExceptionTracker::UpdateNonvolatileRegisters
This change fixes handling of NULL context pointers in ExceptionTracker::UpdateNonvolatileRegisters
on OSX where the NULL value of a context pointer indicates that the unwinder doesn't know the
context pointer.
dotnet-bot [Tue, 17 Mar 2015 18:54:26 +0000 (11:54 -0700)]
Merge changes from parent branch
[tfs-changeset: 1434167]
Jan Kotas [Tue, 17 Mar 2015 18:01:27 +0000 (11:01 -0700)]
Merge pull request #440 from brianrob/AllocationSamplingShutdown
Flush sampled allocations to event stream at runtime shutdown
Richard Lander [Tue, 17 Mar 2015 18:00:44 +0000 (11:00 -0700)]
Convert document index into more GH-centric format
Brian Robbins [Thu, 12 Mar 2015 21:26:35 +0000 (14:26 -0700)]
Flush sampled object allocation on runtime shutdown.
Jan Kotas [Tue, 17 Mar 2015 17:36:02 +0000 (10:36 -0700)]
Merge pull request #482 from richlander/jack-pappas-dotnet-standards
Respond to feedback for .NET Standards PR #413
Richard Lander [Tue, 17 Mar 2015 16:56:37 +0000 (09:56 -0700)]
Update doc based on @CarolEidt feedback.
Richard Lander [Tue, 17 Mar 2015 16:53:21 +0000 (09:53 -0700)]
Merge branch 'dotnet-standards' of https://github.com/jack-pappas/coreclr into jack-pappas-dotnet-standards
Jan Kotas [Tue, 17 Mar 2015 15:59:40 +0000 (08:59 -0700)]
Merge pull request #476 from jkotas/viva
Fix some of the issues found by Viva64 static analyser
Matt Mitchell [Tue, 17 Mar 2015 15:26:57 +0000 (08:26 -0700)]
Merge pull request #469 from mmitche/mac-ci
Remove travis CI and add Mac badges
Jan Kotas [Tue, 17 Mar 2015 14:32:03 +0000 (07:32 -0700)]
Fix some of the issues found by Viva64 static analyser
Jan Kotas [Tue, 17 Mar 2015 12:22:54 +0000 (05:22 -0700)]
Merge pull request #473 from kangaroo/osx-debuggerpresent
Implement IsDebuggerPresent for OSX
Geoff Norton [Tue, 17 Mar 2015 02:48:39 +0000 (19:48 -0700)]
Implement IsDebuggerPresent for OSX
Mike McLaughlin [Tue, 17 Mar 2015 00:34:14 +0000 (17:34 -0700)]
Merge pull request #466 from mikem8361/printexception
Enable sos's PrintException command
Mike McLaughlin [Tue, 17 Mar 2015 00:07:44 +0000 (17:07 -0700)]
Merge remote-tracking branch 'origin/printexception' into printexception
Mike McLaughlin [Fri, 13 Mar 2015 22:36:02 +0000 (15:36 -0700)]
Build the PrintException command and related support functions.
Undef some temporary code that uses the global pointer table and fixed a pre-existing off by one error getting a managed string.
Fixed unnecessary string length addition.
Implement PAL_IsDebuggerPresent for Linux so the managed DebugBreak stops in native debugger.
Finish the off by one string length change.
@janvorli review feedback addressed: fixed buffer off by one error and use '\0' instead of 0.
@jkotas review feedback addressed: increate the file buffer PAL_IsDebuggerPresent uses to 2K.
Jan Kotas [Mon, 16 Mar 2015 23:18:15 +0000 (16:18 -0700)]
Merge pull request #472 from jkotas/IsPrecodeByAsm
Delete PrecodeStubManager::PrecodeIsPrecodeByAsm method
Jan Kotas [Mon, 16 Mar 2015 22:54:49 +0000 (15:54 -0700)]
Merge pull request #465 from janvorli/nonexistent-context-pointers
Make GC work without context pointers on OSX
Jan Kotas [Mon, 16 Mar 2015 22:51:52 +0000 (15:51 -0700)]
Delete PrecodeStubManager::PrecodeIsPrecodeByAsm method
This method was expensive to port. There was just one place remaining that used. I have replaced it with different implementation.
Jan Kotas [Mon, 16 Mar 2015 22:40:12 +0000 (15:40 -0700)]
Merge pull request #470 from qbit/master
Initial support for OpenBSD
Mike McLaughlin [Mon, 16 Mar 2015 22:11:36 +0000 (15:11 -0700)]
Increate the file buffer PAL_IsDebuggerPresent uses to 2K.
Jan Vorlicek [Sat, 14 Mar 2015 00:06:14 +0000 (01:06 +0100)]
Make GC work without context pointers on OSX
On OSX, the libunwind doesn't have support for getting context pointers.
This change modifies the way GC handles object pointers in registers
so that when the context pointer cannot be obtained, the object is
pinned so that GC doesn't move it and thus doesn't need to update
the object reference in the register value stored on the stack.
Aaron Bieber [Mon, 16 Mar 2015 20:50:38 +0000 (14:50 -0600)]
fix build on OpenBSD
fix indent
more indent
fix indenting
Matt Mitchell [Mon, 16 Mar 2015 21:54:06 +0000 (14:54 -0700)]
Merge pull request #433 from mmitche/fix-indepdent-test-build
Fixes for building tests alone
Matt Mitchell [Mon, 16 Mar 2015 20:43:43 +0000 (13:43 -0700)]
Remove travis CI and add Mac badges
Matt Mitchell [Wed, 11 Mar 2015 18:45:35 +0000 (11:45 -0700)]
Fixes for building tests alone
This change fixes an issue where tests cannot be built alone as well as cleans up the paths. The general path cleanup was done so that the path format and variable names generally follow the main build of the coreclr repo. Some highlights are:
* Don't use Cmake as a directory name for intermediates, use intermediates instead.
* Disable 4820 warning (this is about padding) when building tests. Reduces noise in output significantly
* Some fixes in typos and casing in the project files and build scripts
Mike McLaughlin [Mon, 16 Mar 2015 18:50:25 +0000 (11:50 -0700)]
@janvorli review feedback addressed.
Mike McLaughlin [Mon, 16 Mar 2015 17:27:59 +0000 (10:27 -0700)]
Merge remote-tracking branch 'upstream/master' into printexception
Matt Mitchell [Mon, 16 Mar 2015 15:13:30 +0000 (08:13 -0700)]
Merge pull request #464 from mmitche/remove-ci-script
Remove the CI cmd script, it's unused
Matt Mitchell [Mon, 16 Mar 2015 15:12:28 +0000 (08:12 -0700)]
Remove the CI cmd script. Unused.