[PATCH] D20474: when calculating RegUsages, ignore instructions which are uniformed after vectorization

[Wei Mi via llvm-commits]

wmi created this revision.
wmi added reviewers: hfinkel, jmolloy.
wmi added subscribers: llvm-commits, davidxl, congh, mkuper.
wmi set the repository for this revision to rL LLVM.
Herald added a subscriber: mzolotukhin.

This is following patch for http://reviews.llvm.org/D15177.

Just as Hal's comment in http://reviews.llvm.org/D15177?id=41809#inline-126094,
For induction variable only used in GetElementPtr and ICmp, D15177 knows it will not have vectorized version and can be added into VecValuesToIgnore, but it didn't consider the case that induction variable may be used by a add/sub before used in GetElementPtr.

for loop like below:

char a[1000];
char b[1000];
for (long i = 0; i < N; i++)
  a[i] = b[i] * 6 + (b[i] + b[i + 1]) * 4 + b[i - 2] + b[i + 2];

When we are computing RegUsages for VF==8 and VF==16,
it is important for the register usages estimation component to know array index exprs like i, i+1, i+2 and i+3 will not have vectorized version after the loop being vectorized, and their live ranges should not be counted as vector register usages, or else it is likely to exaggerate the vector register pressure.

The patch adds instructions for which isUniformAfterVectorization returns true into the VecValuesToIgnore set. A special case is that PHI instructions are never included into the Uniforms set in collectLoopUniforms(), so a special handling for PHI is that when the result of PHI is only used in GetElementPtr or Uniform instructions, the PHI will be added into VecValuesToIgnore set too.

Another following patch in plan is if estimated vector register usage is more than the number of available hardware vector registers for certain VF, don't simply give up the VF. Just add the extra spill cost into the VectorCost. If the total VectorCost of the VF is the lowest, the VF is still worthy to try.

Repository:
  rL LLVM

http://reviews.llvm.org/D20474

Files:
  lib/Transforms/Vectorize/LoopVectorize.cpp
  test/Transforms/LoopVectorize/X86/reg-usage.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D20474.57942.patch
Type: text/x-patch
Size: 4306 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160520/793c56c7/attachment.bin>