[arcmt] Make arcmt-test accept cc1 options to make it more portable and hopefully fix MSVC failures.

llvm-svn: 133119

llvm-svn: 133118

llvm-svn: 133116

llvm-svn: 133115

on all platforms in non-ARC mode.

llvm-svn: 133114

Function::getNumBlockIDs() should be used instead of Function::size() to set the upper limit on the block IDs since basic blocks might get removed (simplified away) after being initially numbered. Plus the test case, in which SelectionDAGBuilder::visitBr() calls llvm::MachineFunction::removeFromMBBNumbering(), which introduces the hole in numbering leading to an assert in llc (prior to the fix).

llvm-svn: 133113

them only on Darwin tool chains.

llvm-svn: 133112

Add a limit to the number of instructions memdep will scan in a single block.  This prevents (at least in some cases) O(N^2) runtime in passes like DSE.

The limit in this patch is probably too high, but it is enough to stop DSE from going completely insane on a testcase I have (which has a single block with around 50,000 non-aliasing stores in it).

rdar://9471075

llvm-svn: 133111

llvm-svn: 133110

llvm-svn: 133109

llvm-svn: 133108

Follow up to r133032.

llvm-svn: 133107

Add a new MVT::untyped.  This will be used in future work for modelling ISA features like register pairs and lists with "interesting" constraints (such as ARM NEON contiguous register lists or even-odd paired registers).  We need to be able to generate these instructions (often from intrinsics), but don't want to have to assign a legal type to them.  Instead, we'll use an "untyped" edge to bypass the type-checking and simply ensure that the register classes match.

llvm-svn: 133106

This simplifies many of the target description files since it is common
for register classes to be related or contain sequences of numbered
registers.

I have verified that this doesn't change the files generated by TableGen
for ARM and X86. It alters the allocation order of MBlaze GPR and Mips
FGR32 registers, but I believe the change is benign.

llvm-svn: 133105

for this.

llvm-svn: 133104

Language-design credit goes to a lot of people, but I particularly want
to single out Blaine Garst and Patrick Beard for their contributions.

Compiler implementation credit goes to Argyrios, Doug, Fariborz, and myself,
in no particular order.

llvm-svn: 133103

include a specific variable.

llvm-svn: 133102

in order to have its process cleaned up (terminated) upon tearDown is gone for good.
Let's simplify a bunch of Python API test cases.

llvm-svn: 133097

llvm-svn: 133096

llvm-svn: 133095

Depends on LLVM r133093.

llvm-svn: 133094

Patch by Argyrios Kyrtzidis.

llvm-svn: 133093

remove the self.runStarted attribute since the automatic shutdown of processes associated
with the targets are now performed automatically.

llvm-svn: 133092

    bool SBDebugger::DeleteTarget(lldb::SBTarget &target);

which is used in the test tearDown() phase to cleanup the debugger's target list
so that it won't grow larger and larger as test cases are executed.  This is also
a good opportunity to get rid of the arcane requirement that test cases exercising
the Python API must assign the process object to self.process so that it gets
shutdown gracefully.  Instead, the shutdown of the process associated with each
target is now being now automatically.

Also get rid of an API from SBTarget class:

    SBTarget::DeleteTargetFromList(lldb_private::TargetList *list);

llvm-svn: 133091

feature.

Implementation to follow. :)

llvm-svn: 133090

llvm-svn: 133089

like the doc's say it should.  Make sure we have a condition before we set up a test whether
we have one, so we only present a "could not parse condition" error if we actually have a condition.

llvm-svn: 133088

llvm-svn: 133087

llvm-svn: 133086

llvm-svn: 133085

llvm-svn: 133084

llvm-svn: 133083

llvm-svn: 133081

optimizations when emitting calls to the function;  instead those calls may
use faster relocations which require the function to be immediately resolved
upon loading the dynamic object featuring the call.  This is useful when it
is known that the function will be called frequently and pervasively and
therefore there is no merit in delaying binding of the function.

Currently only implemented for x86-64, where it turns into a call through
the global offset table.

Patch by Dan Gohman, who assures me that he's going to add LangRef documentation
for this once it's committed.

llvm-svn: 133080

llvm-svn: 133079

llvm-svn: 133078

llvm-svn: 133077

Replace direct uses of the Debugger's output stream with
uses of the asynchronous stream.

llvm-svn: 133076

llvm-svn: 133075

llvm-svn: 133074