llvm-svn: 126922

This speeds up the greedy register allocator by 15%.
DenseMap is not as fast as one might hope.

llvm-svn: 126921

Patch by Jyun-Yan You, with some minor adjustments and a testcase from me.

llvm-svn: 126915

llvm-svn: 126912

When only a single value has been seen, new PHIDefs are never needed.

llvm-svn: 126911

We need to wait until we meet a PHIDef in its defining block before resurrecting
PHIKills in the predecessors.

This should unbreak the llvm-gcc-build-x86_64-darwin10-x-mingw32-x-armeabi bot.

llvm-svn: 126905

David Greene changed CannotYetSelect() to print the full DAG including multiple
copies of operands reached through different paths in the DAG.  Unfortunately
this blows up exponentially in some cases.  The depth limit of 100 is way too
high to prevent this -- I'm seeing a message string of 150MB with a depth of
only 40 in one particularly bad case, even though the DAG has less than 200
nodes.  Part of the problem is that the printing code is following chain
operands, so if you fail to select an operation with a chain, the printer will
follow all the chained operations back to the entry node.

llvm-svn: 126899

No functional change.

llvm-svn: 126898

Patch by Ted Kremenek!

llvm-svn: 126895

Values that map to a single new value in a new interval after splitting don't
need new PHIDefs, and if the parent value was never rematerialized the live
range will be the same.

llvm-svn: 126894

llvm-svn: 126893

llvm-svn: 126882

llvm-svn: 126871

Radar 9056407.

llvm-svn: 126864

llvm-svn: 126862

      missing patterns for them.

      Add a SIMD test subdirectory to hold tests for SIMD instruction
      selection correctness and quality.
'

llvm-svn: 126845

llvm-svn: 126838

- Add '64bit' sub-target option.
- Select 32-bit/64-bit loads/stores based on '64bit' option.
- Fix function parameter order.

Patch by Justin Holewinski

llvm-svn: 126837

uses.

The result produced by the streamer is used to give the linker more accurate
information and to add to llvm.compiler.used. The second improvement removes
the need for the user to add __attribute__((used)) to functions only used in
inline asm. The first one lets us build firefox with LTO on Darwin :-)

llvm-svn: 126830

llvm-svn: 126829

llvm-svn: 126826

llvm-svn: 126825

- Allow i16, i32, i64, float, and double types, using the native .u16,
  .u32, .u64, .f32, and .f64 PTX types.
- Allow loading/storing of all primitive types.
- Allow primitive types to be passed as parameters.
- Allow selection of PTX Version and Shader Model as sub-target attributes.
- Merge integer/floating-point test cases for load/store.
- Use .u32 instead of .s32 to conform to output from NVidia nvcc compiler.

Patch by Justin Holewinski

llvm-svn: 126824

Extract the updateSSA() method from the too long extendRange().

LiveOutCache can be shared among all the new intervals since there is at most
one of the new ranges live out from each basic block.

llvm-svn: 126818

llvm-svn: 126815

This fixes PR9259.

llvm-svn: 126812

llvm-svn: 126811

Simplify the signature - The return value and ParentVNI are no longer needed.

llvm-svn: 126809

llvm-svn: 126806

llvm-svn: 126805

This method could probably be used by LiveIntervalAnalysis::shrinkToUses, and
now it can use extendIntervalEndTo() which coalesces ranges.

llvm-svn: 126803

llvm-svn: 126801

The value map is currently not used, all values are 'complex mapped' and
LiveIntervalMap::mapValue is used to dig them out.

This is the first step in a series changes leading to the removal of
LiveIntervalMap. Its data structures can be shared among all the live intervals
created by a split, so it is wasteful to create a copy for each.

llvm-svn: 126800

Local live range splitting is better driven by interference. This code was just
guessing.

llvm-svn: 126799

This is a waste of time since we already know how to evict all interferences
which is a better approach anyway.

llvm-svn: 126798

Patch by Erik Olofsson!

llvm-svn: 126796

llvm-svn: 126794

Today, the language front ends produces llvm.dbg.* intrinsics, used to encode arguments' debug info, in order any way, most of the times. However, if a front end mix-n-matches llvm.dbg.declare and llvm.dbg.value intrinsics to encode debug info for arguments then code generator needs a way to find argument order.

Use 8 bits from line number field to keep track of argument ordering while encoding debug info for an argument. That leaves 24 bit for line no, DebugLoc also allocates 24 bit for line numbers. If a function has more than 255 arguments then rest of the arguments will be ordered by llvm.dbg.* intrinsics' ordering in IR.

llvm-svn: 126793

addressing code. On 403.gcc this almost halves CodeGenPrepare time and reduces
total llc time by 9.5%. Unfortunately, getNumUses() is still the hottest function
in llc.

llvm-svn: 126782

This effectively disables the 'turbo' functionality of the greedy register
allocator where all new live ranges created by splitting would be reconsidered
as if they were originals.

There are two reasons for doing this, 1. It guarantees that the algorithm
terminates. Early versions were prone to infinite looping in certain corner
cases. 2. It is a 2x speedup. We can skip a lot of unnecessary interference
checks that won't lead to good splitting anyway.

The problem is that region splitting only gets one shot, so it should probably
be changed to target multiple physical registers at once.

Local live range splitting is still 'turbo' enabled. It only accounts for a
small fraction of compile time, so it is probably not necessary to do anything
about that.

llvm-svn: 126781