Move the flags around slightly.

llvm-svn: 152929

We currently do not support pointer types in affine expressions. Hence, we
disallow in the SCoP detection. Later we may decide to add support for them.

This fixes PR12277

Reported-By: Sebastian Pop <sebpop@gmail.com>
llvm-svn: 152928

No functionality change.

llvm-svn: 152927

Alter the terminal streams such that they do not get added to the atexit chain, and thus never get destructed.

llvm-svn: 152926

llvm-svn: 152925

llvm-svn: 152924

This reverts commit 7dd9b6327b54b08ece32a4607d5ac093b518b79a.

llvm-svn: 152923

llvm-svn: 152921

  so lit find Polly's library in off-tree build.

llvm-svn: 152920

This fixes PR 4307.

Patch by Eitan Adler!

llvm-svn: 152918

This solves the 'isl_ctx freed, but some objects still reference it' problem
reported in PR12276.

llvm-svn: 152917

lit/TestRunner.py: [Win32] Check all opened_files[] released, rather than (obsoleted) written_files[].

In previous case,
RUN: foo -o %t
RUN: FileCheck < %t
RUN: bar -o %t

2nd read handle might prevent manipulation of 3rd %t in bar, to remove and rename.

llvm-svn: 152916

We can simply confirm the handle released to open it with EXCLUSIVE. Attempting renaming was bad.

llvm-svn: 152915

Enable incremental parsing by the Preprocessor,
where more code can be provided after an EOF.
It mainly prevents the tearing down of the topmost lexer.
To be used like this:
PP.enableIncrementalProcessing();
while (getMoreSource()) {
  while (Parser.ParseTopLevelDecl(ADecl)) {...}
}
PP.enableIncrementalProcessing(false);

llvm-svn: 152914

This also fixes UMax where we did not correctly keep track of the parameters.
Fixes PR12275.

Reported-By: Sebastian Pop <sebpop@gmail.com>
llvm-svn: 152913

llvm-svn: 152912

 this option in the configure step of Polly's builder to get more readable
 output from the stdio log.

llvm-svn: 152910

llvm-svn: 152909

llvm-svn: 152908

alignment. If that's the case, then we want to make sure that we don't increase
the alignment of the store instruction. Because if we increase it to be "more
aligned" than the pointer, code-gen may use instructions which require a greater
alignment than the pointer guarantees.
<rdar://problem/11043589>

llvm-svn: 152907

llvm-svn: 152906

Reintroduce lazy name lookup table building, ensuring that the lazy building step
produces the same lookup table that would be built by the eager step.

Avoid building a lookup table for the translation unit outside C++, even in cases
where we can't recover the contents of the table from the declaration chain on
the translation unit, since we're not going to perform qualified lookup into it
anyway. Continue to support lazily building such lookup tables for now, though,
since ASTMerge uses them.

In my tests, this performs very similarly to ToT with r152608 backed out, for C,
Obj-C and C++, and does not suffer from PR10447.

llvm-svn: 152905

It was added in 2007 as the first cut at supporting no-inline
attributes, but we didn't have function attributes of any form at the
time. However, it was added without any mention in the LangRef or other
documentation.

Later on, in 2008, Devang added function notes for 'inline=never' and
then turned them into proper function attributes. From that point
onward, as far as I can tell, the world moved on, and no one has touched
'llvm.noinline' in any meaningful way since.

It's time has now come. We have had better mechanisms for doing this for
a long time, all the frontends I'm aware of use them, and this is just
holding back progress. Given that it was never a documented feature of
the IR, I've provided no auto-upgrade support. If people know of real,
in-the-wild bitcode that relies on this, yell at me and I'll add it, but
I *seriously* doubt anyone cares.

llvm-svn: 152904

directly query the function information which this set was representing.
This simplifies the interface of the inline cost analysis, and makes the
always-inline pass significantly more efficient.

Previously, always-inline would first make a single set of every
function in the module *except* those marked with the always-inline
attribute. It would then query this set at every call site to see if the
function was a member of the set, and if so, refuse to inline it. This
is quite wasteful. Instead, simply check the function attribute directly
when looking at the callsite.

The normal inliner also had similar redundancy. It added every function
in the module with the noinline attribute to its set to ignore, even
though inside the cost analysis function we *already tested* the
noinline attribute and produced the same result.

The only tricky part of removing this is that we have to be able to
correctly remove only the functions inlined by the always-inline pass
when finalizing, which requires a bit of a hack. Still, much less of
a hack than the set of all non-always-inline functions was. While I was
touching this function, I switched a heavy-weight set to a vector with
sort+unique. The algorithm already had a two-phase insert and removal
pattern, we were just needlessly paying the uniquing cost on every
insert.

This probably speeds up some compiles by a small amount (-O0 compiles
with lots of always-inline, so potentially heavy libc++ users), but I've
not tried to measure it.

I believe there is no functional change here, but yell if you spot one.
None are intended.

Finally, the direction this is going in is to greatly simplify the
inline cost query interface so that we can replace its implementation
with a much more clever one. Along the way, all the APIs get simplified,
so it seems incrementally good.

llvm-svn: 152903

llvm-svn: 152901

llvm-svn: 152900

Const-correct the FixedLenDecoderEmitter. Pass a few things by const reference instead of value to avoid some copying.

llvm-svn: 152899

analysis implementation. The header was already separated. Also cleanup
all the comments in the header to follow a nice modern doxygen form.

There is still plenty of cruft here, but some of that will fall out in
subsequent refactorings and this was an easy step in the right
direction. No functionality changed here.

llvm-svn: 152898

- Clang now completes all Objective-C objects (if
  they are not already complete, and they have
  external lexical sources) during structure
  layout, avoiding a LLDB crash.

- The Clang Decl printer handles reference types
  correctly.  This prevents LLDB from crashing
  when expression logging is enabled.

llvm-svn: 152897

test/Driver/linker-opts.c: Mark as XFAIL:cygming, using gcc linker driver. It should be aware of LIBRARY_PATH.

llvm-svn: 152896

These edges are not really necessary, but it is consistent with the
way we currently create physreg edges. Scheduler heuristics that
expect a DAG edge to the block terminator could benefit from this
change. Although in the future I hope we have a better mechanism for
modeling latency across scheduling regions.

llvm-svn: 152895

llvm-svn: 152894

llvm-svn: 152893

Only record IVUsers that are dominated by simplified loop
headers. Otherwise SCEVExpander will crash while looking for a
preheader.

I previously tried to work around this in LSR itself, but that was
insufficient. This way, LSR can continue to run if some uses are not
in simple loops, as long as we don't attempt to analyze those users.

Fixes <rdar://problem/11049788> Segmentation fault: 11 in LoopStrengthReduce

llvm-svn: 152892

Don't try to create "store atomic" instructions of non-integer types; they aren't supported at the moment.  PR12040.

llvm-svn: 152891

llvm-svn: 152890

Spacing fixes. Mostly aligning arguments that spilled onto next line with the opening parenthese instead of 2 spaces in.

llvm-svn: 152889

llvm-svn: 152888

on our internal nightly testers.  So, basically revert r152486 again.

Abbreviated original commit message:
Implement a more intelligent way of spilling uses across an invoke boundary.

It looks as if Chander's inlining work, r152737, exposed an issue.

llvm-svn: 152887

Don't know if it is an anathema to include '#' in the warning name.  I'm fine
with changing it, but this was the most obvious name.

llvm-svn: 152886