Owen no longer has time to tend to the yaks in SelectionDAG.

llvm-svn: 270253

This patch doesn't affect D19878's context.  So D19878 still cleanly applies.

llvm-svn: 270252

Summary:
As this optimization converts two loads into one load with two shift instructions,
it could potentially hurt performance if a loop is arithmetic operation intensive.

Reviewers: t.p.northover, mcrosier, jmolloy

Subscribers: evandro, jmolloy, aemerson, rengolin, mcrosier, llvm-commits

Differential Revision: http://reviews.llvm.org/D20172

llvm-svn: 270251

Check that the incoming blocks of phi nodes are identical, and block
function merging if they are not.

rdar://problem/26255167

Differential Revision: http://reviews.llvm.org/D20462

llvm-svn: 270250

The Fast mode takes the first mapping, the greedy mode loops over all
the possible mapping for an instruction and choose the cheaper one.
Test case will come with target specific code, since we currently do not
have instructions that have several mappings.

llvm-svn: 270249

The uuid_command was duplicating the load_command.cmdsize field. This removes the duplicate from the YAML mapping and from the test cases.

llvm-svn: 270248

This is now encapsulated in the RepairingPlacement class.

llvm-svn: 270247

We performed a number of memory allocations each time getTTI was called,
remove them by using SmallString.
No functionality change intended.

llvm-svn: 270246

computeMapping.

Computing the cost of a mapping takes some time.
Since in Fast mode, the cost is irrelevant, just spare some cycles by not
computing it.
In Greedy mode, we need to choose the best cost, that means that when
the local cost gets more expensive than the best cost, we can stop
computing the repairing and cost for the current mapping.

llvm-svn: 270245

more precise cost in Greedy mode.

In Fast mode the cost is irrelevant so do not bother requiring that
those passes get scheduled.

llvm-svn: 270244

This adds direct support for computing offsets from the thread pointer
for both variants. Of the architectures we support, variant 1 is used
only by aarch64 (but that doesn't seem to be documented anywhere.)

llvm-svn: 270243

llvm-svn: 270242

There is no need to apply kindof on an unqualified id type.

rdar://24753825

llvm-svn: 270241

The `FreeBSDTargetInfo` class has always set the `__FreeBSD_cc_version`
predefined macro to a rather static value, calculated from the major OS
version.

In the FreeBSD base system, we will start incrementing the value of this
macro whenever we make any signifant change to clang, so we need a way
to configure the macro's value at build time.

Use `FREEBSD_CC_VERSION` for this, which we can define in the FreeBSD
build system using either the `-D` command line option, or an include
file.  Stock builds will keep the earlier value.

Differential Revision: http://reviews.llvm.org/D20037

llvm-svn: 270240

This adds support for handling unknown load commands, and a bogus_load_command tests.

Unknown or unsupported load commands can be specified in YAML by their hex value.

llvm-svn: 270239

Patch by Aaron Enye Shi.

Differential Revision: http://reviews.llvm.org/D17578

llvm-svn: 270238

llvm-svn: 270237

llvm-svn: 270236

The mode should be choose by the target when instantiating the pass.

llvm-svn: 270235

llvm-svn: 270234

- Do not store Twine objects.
- Remove report_fatal_error, since llvm_unreachable does terminate the
  program in release mode.

llvm-svn: 270233

The previous choice of the insertion points for repairing was
straightfoward but may introduce some basic block or edge splitting. In
some situation this is something we can avoid.
For instance, when repairing a phi argument, instead of placing the
repairing on the related incoming edge, we may move it to the previous
block, before the terminators. This is only possible when the argument
is not defined by one of the terminator.

llvm-svn: 270232

Inline getAnalysisUsage() while I'm here.

llvm-svn: 270231

This patch is a first step towards a more extendible method of matching combined target shuffle masks.

Initially this just pulls out the existing basic mask matches and adds support for some 256/512 bit equivalents. Future patterns will require a number of features to be added but I wanted to keep this patch simple.

I hope we can avoid duplication between shuffle lowering and combining and share more complex pattern match functions in future commits.

Differential Revision: http://reviews.llvm.org/D19198

llvm-svn: 270230

llvm-svn: 270229

llvm-svn: 270228

llvm-svn: 270227

Differential Revision: http://reviews.llvm.org/D20454

llvm-svn: 270226

[PM/PartiallyInlineLibCalls] Convert to static function in preparation for porting this pass to the new PM.

llvm-svn: 270225

This is matching what trunk gcc is accepting. Also adds a missing ssse3
case. PR27779. The amount of duplication here is annoying, maybe it
should be factored into a separate .def file?

llvm-svn: 270224

An upcoming change for ld in binutils 2.26 causes this test to
always fail.

llvm-svn: 270223

This was noted in PR24766:
https://llvm.org/bugs/show_bug.cgi?id=24766#c2

We may not know whether the sign bit(s) are zero or one, but we can still
optimize based on knowing that the sign bit is repeated.

Differential Revision: http://reviews.llvm.org/D20275

llvm-svn: 270222

llvm-svn: 270220

llvm-svn: 270219

I accidentally exposed a bug in MCExpr::evaluateAsRelocatableImpl() with the test file added in:
http://reviews.llvm.org/rL269977

Differential Revision: http://reviews.llvm.org/D20434

llvm-svn: 270218

llvm-svn: 270217

llvm-svn: 270216

This should have been done this way from the start, however I somehow missed
r257177.

llvm-svn: 270215

The CL causes a build breakage on platforms where sizeof(double) == sizeof(long double)
and it incorrectly assumes that sizeof(double) and sizeof(long double) is the same
on the host and the target.

llvm-svn: 270214

The various _l locale extension functions originate from very
different places.  Some come from POSIX, some are BSD extensions,
and some are shared BSD and GLIBC extensions. This patch tries to
group the local extension reimplementations by source. This should
make it easier to make libcxx work with POSIX compliant C libraries
that lack these extensions.

The fallback locale functions are also useful on their own for other
lightweight platforms. Putting these fallback implementations in
support/xlocale should enable code sharing.

I have no access to a newlib system or an android system to build
and test with. I _do_ have access to a system without any of the _l
locale extensions though, and I was able to ensure that the new
__posix_l_fallback.h and __strtonum_fallback.h didn't have any massive
problems.

http://reviews.llvm.org/D17416

llvm-svn: 270213