llvm/lib/Transforms
Arnold Schwaighofer a47aa4b4ef LoopVectorize: Support conditional stores by scalarizing
The vectorizer takes a loop like this and widens all instructions except for the
store. The stores are scalarized/unrolled and hidden behind an "if" block.

  for (i = 0; i < 128; ++i) {
    if (a[i] < 10)
      a[i] += val;
  }

  for (i = 0; i < 128; i+=2) {
    v = a[i:i+1];
    v0 = (extract v, 0) + 10;
    v1 = (extract v, 1) + 10;
    if (v0 < 10)
      a[i] = v0;
    if (v1 < 10)
      a[i] = v1;
  }

The vectorizer relies on subsequent optimizations to sink instructions into the
conditional block where they are anticipated.

The flag "vectorize-num-stores-pred" controls whether and how many stores to
handle this way. Vectorization of conditional stores is disabled per default for
now.

This patch also adds a change to the heuristic when the flag
"enable-loadstore-runtime-unroll" is enabled (off by default). It unrolls small
loops until load/store ports are saturated. This heuristic uses TTI's
getMaxUnrollFactor as a measure for load/store ports.

I also added a second flag -enable-cond-stores-vec. It will enable vectorization
of conditional stores. But there is no cost model for vectorization of
conditional stores in place yet so this will not do good at the moment.

rdar://15892953

Results for x86-64 -O3 -mavx +/- -mllvm -enable-loadstore-runtime-unroll
-vectorize-num-stores-pred=1 (before the BFI change):

 Performance Regressions:
   Benchmarks/Ptrdist/yacr2/yacr2 7.35% (maze3() is identical but 10% slower)
   Applications/siod/siod         2.18%
 Performance improvements:
   mesa                          -4.42%
   libquantum                    -4.15%

 With a patch that slightly changes the register heuristics (by subtracting the
 induction variable on both sides of the register pressure equation, as the
 induction variable is probably not really unrolled):

 Performance Regressions:
   Benchmarks/Ptrdist/yacr2/yacr2  7.73%
   Applications/siod/siod          1.97%

 Performance Improvements:
   libquantum                    -13.05% (we now also unroll quantum_toffoli)
   mesa                           -4.27%

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@200270 91177308-0d34-0410-b5e6-96231b3b80d8
2014-01-28 01:01:53 +00:00
..
Hello Set LLVM_EXPORTED_SYMBOL_FILE in CMakeLists whose corresponding Makefiles do so. 2013-12-29 23:06:49 +00:00
InstCombine InstCombine: Don't try to use aggregate elements of ConstantExprs. 2014-01-24 19:02:37 +00:00
Instrumentation Fix known typos 2014-01-24 17:20:08 +00:00
IPO Fix known typos 2014-01-24 17:20:08 +00:00
ObjCARC Fix known typos 2014-01-24 17:20:08 +00:00
Scalar ConstantHoisting: We can't insert instructions directly in front of a PHI node. 2014-01-27 13:11:43 +00:00
Utils PGO branch weight: keep halving the weights until they can fit into 2014-01-27 23:39:03 +00:00
Vectorize LoopVectorize: Support conditional stores by scalarizing 2014-01-28 01:01:53 +00:00
CMakeLists.txt Extracted ObjCARC.cpp into its own library libLLVMObjCARCOpts in preparation for refactoring the ARC Optimizer. 2013-01-28 01:35:51 +00:00
LLVMBuild.txt Extracted ObjCARC.cpp into its own library libLLVMObjCARCOpts in preparation for refactoring the ARC Optimizer. 2013-01-28 01:35:51 +00:00
Makefile Extracted ObjCARC.cpp into its own library libLLVMObjCARCOpts in preparation for refactoring the ARC Optimizer. 2013-01-28 01:35:51 +00:00