DenseMap's hash function uses slightly more entropy and reduces hash collisions
significantly.  I also experimented with Hashing.h, but it didn't gave a lot of
improvement while being much more expensive to compute.

llvm-svn: 154996

backend flags in TargetOptions.h into their own class in r145714.

llvm-svn: 154993

llvm-svn: 154992

llvm-svn: 154990

If the loop contains invoke instructions, whose unwind edge escapes the loop,
then don't try to unswitch the loop. Doing so may cause the unwind edge to be
split, which not only is non-trivial but doesn't preserve loop simplify
information.

Fixes PR12573

llvm-svn: 154987

llvm-svn: 154985

This introduces a threshold of 200 IV Users, which is very
conservative but should be sufficient to avoid serious compile time
sink or stack overflow. The llvm test-suite with LTO never exceeds 190
users per loop.

The bug doesn't relate to a specific type of loop. Checking in an
arbitrary giant loop as a unit test would be silly.

Fixes rdar://11262507.

llvm-svn: 154983

llvm-svn: 154971

Fixes build on MSVC

llvm-svn: 154970

llvm-svn: 154967

llvm-svn: 154962

also fix SimplifyLibCalls to use TLI rather than compile-time conditionals to enable optimizations on floor, ceil, round, rint, and nearbyint

llvm-svn: 154960

llvm-svn: 154959

transformation:

(X op C1) ^ C2 --> (X op C1) & ~C2 iff (C1&C2) == C2

should be done.  

This change has been tested:
 Using a debug+asserts build:
   on the specific test case that brought this bug to light
   make check-all
   lnt nt
   using this clang to build a release version of clang
 Using the release+asserts clang-with-clang build:
   on the specific test case that brought this bug to light
   make check-all
   lnt nt

Checking in because Evan wants it checked in.  Test case forthcoming after
scrubbing.

llvm-svn: 154955

llvm-svn: 154953

llvm-svn: 154951

llvm-svn: 154948

llvm-svn: 154944

llvm-svn: 154936

Patch by Vladimir Medic. 

llvm-svn: 154935

llvm-svn: 154930

llvm-svn: 154927

This made the test fail with -mcpu=generic (when building on a non-x86 host).

llvm-svn: 154926

llvm-svn: 154921

llvm-svn: 154916

llvm-svn: 154915

This isn't right either, reverting for now.

llvm-svn: 154910

llvm-svn: 154908

llvm-svn: 154907

llvm-svn: 154905

for the life of me remember why I wrote it this way, but I can't see any good
reason for it now. This patch replaces the custom linked list with an ilist.

This change should preserve the existing numberings exactly, so no generated code
should change (if it does, file a bug!).

llvm-svn: 154904

llvm-svn: 154895

re-done by wiser souls.

llvm-svn: 154894

llvm-svn: 154892

about what folks should expect w.r.t. the new algorithm.

llvm-svn: 154891

llvm-svn: 154889

instructions with writebacks. And add test a case for all opcodes handed by
DecodeVLD2DupInstruction() in ARMDisassembler.cpp .

llvm-svn: 154884

temporarily XFAIL this test until post RA
live-ins is properly enabled.

llvm-svn: 154882

llvm-svn: 154881

llvm-svn: 154879