llvm-svn: 147807

The pass is prone to looping, and it is better to crash than loop
forever, even in a -Asserts build.

<rdar://problem/10660175>

llvm-svn: 147806

llvm-svn: 147805

llvm-svn: 147802

After collecting chains, check if any should be materialized. If so,
hide the chained IV users from the LSR solver. LSR will only solve for
the head of the chain. GenerateIVChains will then materialize the
chained IV users by computing the IV relative to its previous value in
the chain.

In theory, chained IV users could be exposed to LSR's solver. This
would be considerably complicated to implement and I'm not aware of a
case where we need it. In practice it's more important to
intelligently prune the search space of nontrivial loops before
running the solver, otherwise the solver is often forced to prune the
most optimal solutions. Hiding the chained users does this well, so
that LSR is more likely to find the best IV for the chain as a whole.

llvm-svn: 147801

llvm-svn: 147799

llvm-svn: 147798

This collects a set of IV uses within the loop whose values can be
computed relative to each other in a sequence. Following checkins will
make use of this information.

llvm-svn: 147797

llvm-svn: 147796

llvm-svn: 147794

llvm-svn: 147793

llvm-svn: 147792

llvm-svn: 147791

it was checked in as:
 virtual bool ABI::FixCodeAddress (lldb::addr_t pc);

when it should have been:

virtual lldb::addr_t ABI::FixCodeAddress (lldb::addr_t pc);


 

llvm-svn: 147790

llvm-svn: 147789

llvm-svn: 147788

AsmParser holds info specific to target parser.
AsmParserVariant holds info specific to asm variants supported by the target.

llvm-svn: 147787

padding insertion will not work if the coerced type is not a structure.

llvm-svn: 147786

llvm-svn: 147785

llvm-svn: 147784

Thanks to David Blaikie for pointing this out.

llvm-svn: 147783

the anonymous namespace to its parent. Semantically, this means that
the anonymous namespaces defined in one module are distinct from the
anonymous namespaces defined in another module.

llvm-svn: 147782

Patch by  Joe Groff!

llvm-svn: 147781

that we can merge, for example, two occurrences of

  namespace N { void f(); }

in two disjoint modules.

llvm-svn: 147780

llvm-svn: 147779

modules. Teach name lookup into namespaces to search in each of the
merged DeclContexts as well as the (now-primary) DeclContext. This
supports the common case where two different modules put something
into the same namespace.

llvm-svn: 147778

This subsumes several other transforms while enabling us to catch more cases.

llvm-svn: 147777

llvm-svn: 147776

is important because it's fairly common for headers (especially system
headers) to want to provide only those typedefs needed for that
particular header, based on some guard macro, e.g.,

#ifndef _SIZE_T
#define _SIZE_T
typedef long size_t;
#endif

which is repeated in a number of headers. The guard macro protects
against duplicate definitions. However, this means that only the first
occurrence of this pattern actually defines size_t, so the submodule
corresponding to this header has the only visible definition. If a
user then imports a different submodule from the same module, size_t
will be known but not visible, and therefore cannot be used.

By allowing redefinition of typedefs, each header that wants to define
size_t can do so independently, so it will be available in the
corresponding submodules.

llvm-svn: 147775

llvm-svn: 147774

this substraction will result in small negative numbers at worst which
become very large positive numbers on assignment and are thus caught by
the <=4 check on the next line. The >0 check clearly intended to catch
these as negative numbers.

Spotted by inspection, and impossible to trigger given the shift widths
that can be used.

llvm-svn: 147773

llvm-svn: 147772

llvm-svn: 147771

Remove AVX hack in X86Subtarget. AVX/AVX2 are now treated as an SSE level. Predicate functions have been altered to maintain previous names and behavior.

llvm-svn: 147770

llvm-svn: 147769

Reorder a bunch of patterns to put the AVX version first thus giving it priority over the SSE version. Another step towards trying to remove the AVX hack that disables SSE from X86Subtarget.

llvm-svn: 147768

Clean up patterns for MOVNT*. Not sure why there were floating point types on MOVNTPS and MOVNTDQ. And v4i64 was completely missing.

llvm-svn: 147767

Mark MOVNTI as being supported in SSE2 OR AVX mode. This instruction has no AVX equivalent so we should use the SSE version.

llvm-svn: 147766

Move SSE2 logical operations PAND/POR/PXOR/PANDN above SSE1 logical operations ANDPS/ORPS/XORPS/ANDNPS. This fixes a pattern ordering issue that meant that the SSE2 instructions could never be directly selected since the SSE1 patterns would always match first. This is largely moot with the ExeDepsFix pass, but I'm trying to audit for all such ordering issues.

llvm-svn: 147765

Change some places that were checking for AVX OR SSE1/2 to use hasXMM/hasXMMInt instead. Also fix one place that checked SSE3, but accidentally excluded AVX to use hasSSE3orAVX. This is a step towards removing the AVX hack from the X86Subtarget.h

llvm-svn: 147764