llvm-6502/test/Transforms/LoopVectorize/X86
Arnold Schwaighofer 5f0d9dbdf4 X86 cost model: Adjust cost for custom lowered vector multiplies
This matters for example in following matrix multiply:

int **mmult(int rows, int cols, int **m1, int **m2, int **m3) {
  int i, j, k, val;
  for (i=0; i<rows; i++) {
    for (j=0; j<cols; j++) {
      val = 0;
      for (k=0; k<cols; k++) {
        val += m1[i][k] * m2[k][j];
      }
      m3[i][j] = val;
    }
  }
  return(m3);
}

Taken from the test-suite benchmark Shootout.

We estimate the cost of the multiply to be 2 while we generate 9 instructions
for it and end up being quite a bit slower than the scalar version (48% on my
machine).

Also, properly differentiate between avx1 and avx2. On avx-1 we still split the
vector into 2 128bits and handle the subvector muls like above with 9
instructions.
Only on avx-2 will we have a cost of 9 for v4i64.

I changed the test case in test/Transforms/LoopVectorize/X86/avx1.ll to use an
add instead of a mul because with a mul we now no longer vectorize. I did
verify that the mul would be indeed more expensive when vectorized with 3
kernels:

for (i ...)
   r += a[i] * 3;
for (i ...)
  m1[i] = m1[i] * 3; // This matches the test case in avx1.ll
and a matrix multiply.

In each case the vectorized version was considerably slower.

radar://13304919

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@176403 91177308-0d34-0410-b5e6-96231b3b80d8
2013-03-02 04:02:52 +00:00
..
avx1.ll X86 cost model: Adjust cost for custom lowered vector multiplies 2013-03-02 04:02:52 +00:00
conversion-cost.ll ARM Cost model: Use the size of vector registers and widest vectorizable instruction to determine the max vectorization factor. 2013-01-09 22:29:00 +00:00
cost-model.ll Loop Vectorizer: Update the cost model of scatter/gather operations and make 2012-12-23 07:23:55 +00:00
gcc-examples.ll Remove the -licm pass from the loop vectorizer test because the loop vectorizer does it now. 2013-01-09 01:20:59 +00:00
lit.local.cfg
min-trip-count-switch.ll Made the min-trip-count-switch test X86-specific to avoid 2013-01-31 10:33:22 +00:00
no-vector.ll
parallel-loops-after-reg2mem.ll Forgot to 'svn add' the LoopVectorizer tests for the new parallel loop metadata, sorry. 2013-02-15 21:50:19 +00:00
parallel-loops.ll Forgot to 'svn add' the LoopVectorizer tests for the new parallel loop metadata, sorry. 2013-02-15 21:50:19 +00:00
small-size.ll Move test that depends on the x86 target into a target-specific directory. 2013-01-16 13:25:56 +00:00
struct-store.ll Fix typo "Makre" -> "Make". 2012-12-24 19:55:47 +00:00
unroll_selection.ll LoopVectorizer: Implement a new heuristics for selecting the unroll factor. 2013-01-20 05:24:29 +00:00
unroll-small-loops.ll Change the cpu type in the test. 2013-01-20 05:20:56 +00:00
vector_ptr_load_store.ll Formatting. 2013-02-05 15:32:16 +00:00