I think it'd be nicer to just abort() like we do here, but
cffi is actually testing this behavior (that loading an invalid
library causes an ImportError).

This is being exposed now because before, the library would fail
to compile due to -Werror=implicit-function-declaration and not even
get to the loading step.

I'm still hoping that we can keep this on in some way, because it continues to find issues.
So let's try keeping it on for debug builds.

I'm not sure if this is weakening the check enough (there may be other cases like scipy),
but it's sounding like we need to have this check turned off in our releases regardless,
such as issue #1344.

Undo all the places we changed installed paths

The only things left in there were some old stubs.  We had added the stubs and then later
added the real files; with the old directory structure the real files took precedence,
but then with the new structure the stubs took precedence.

Once those were deleted there wasn't anything left in lib_pyston

We used to have a installed directory structure that mirrors
our source structure.  This required changing quite a few pieces of CPython
code to look in the new places (virtualenv, distutils, getpath.c, sysconfig).
So, undo all those changes, and change our CMake configuration to install into
the paths that CPython expects.

Note: "the paths that CPython expects" is platform-dependent.  The paths I added
here are based on the "posix" style that it supports.  The CPython code has platform-specific
code to look in different places, and I guess the build system must have platform-specific
knowledge of where to put it as well, but our CMake system will always put it in the linux variant.

Fix pointer reuse problems

while this removes some disabled features I'm not a big fan of having disabled stuff around
because it will break and makes development harder.

this is a duplicate - we will allready emit a call to recordType inside the bjit

solves a pointer reuse problem

Remove libpypa parser

After remove libpypa, the function getMagic only return one string. So
just use a char array constant instead a function.

MCJIT: free loaded objects

This fixes the llvm tier memory consumption until we migrate to the ORC JIT which has better memory management.

Misc memory size reductions

this does not hurt perf but still saves some memory

Threading fixes to get cffi-1.7 working

Instead of having it always point to the thread-local threadstate object,
it's a global that gets updated during context switches, like CPython does it.

There are still a number of differences that should be refactored out, but
I think this should bring us quite a bit closer to CPython's implementation.

It's time :(

This new version is causing some issues for us

switch from the SSE prefetch to __builtin_prefetch

string interning: make it slightly more space efficient

hidden classes: free classes of singleton type

this fixes the huge (~350MB) leak in:
for i in xrange(1000000):
    class C(object):
        pass

We used to store the string content twice.
This implementation makes use of DenseSet::find_as functionality where one can search using a different type.
I was not sure if the special DenseMapInfo I had to create causes any problems so I choose to create a new source file
for it so that it does not get picked up somewhere else.