Commit Graph

10 Commits

Author SHA1 Message Date
Jingyue Wu
4015c5e2e4 [NaryReassociate] enhances nsw by leveraging @llvm.assume
Summary:
nsw are flaky and can often be removed by optimizations. This patch enhances
nsw by leveraging @llvm.assume in the IR. Specifically, NaryReassociate now
understands that

    assume(a + b >= 0) && assume(a >= 0) ==> a +nsw b

As a result, it can split more sext(a + b) into sext(a) + sext(b) for CSE.

Test Plan: nary-gep.ll

Reviewers: broune, meheff

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D10822

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@241139 91177308-0d34-0410-b5e6-96231b3b80d8
2015-07-01 03:38:49 +00:00
Jingyue Wu
4977e92629 [NaryReassociate] Run EarlyCSE after NaryReassociate
Summary:
This patch made two improvements to NaryReassociate and the NVPTX pipeline

1. Run EarlyCSE/GVN after NaryReassociate to get rid of redundant common
expressions.

2. When adding an instruction to SeenExprs, maps both the SCEV before and after
reassociation to that instruction.

Test Plan: updated @reassociate_gep_nsw in nary-gep.ll

Reviewers: meheff, broune

Reviewed By: broune

Subscribers: dberlin, jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D9947

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@238396 91177308-0d34-0410-b5e6-96231b3b80d8
2015-05-28 04:56:52 +00:00
Jingyue Wu
41cf9ae1b8 [NaryReassoc] reassociate GEP for CSE
Summary:
x = &a[i];
y = &a[i + j];

=>

y = x + j;

along with some refactoring work such as extracting method
findClosestMatchingDominator.

Depends on D9786 which provides the ScalarEvolution::getGEPExpr interface.

Test Plan: nary-gep.ll

Reviewers: meheff, broune

Reviewed By: broune

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D9802

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@237971 91177308-0d34-0410-b5e6-96231b3b80d8
2015-05-21 23:17:30 +00:00
Jingyue Wu
12b5fa5f6e [NaryReassociate] avoid running forever
Avoid running forever by checking we are not reassociating an expression into
the same form.

Tested with @avoid_infinite_loops in nary-add.ll


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@237269 91177308-0d34-0410-b5e6-96231b3b80d8
2015-05-13 18:12:24 +00:00
Jingyue Wu
728ad0157c Resurrect r235688
We should skip vector types which are not SCEVable.

test/CodeGen/NVPTX/sched2.ll passes


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@235695 91177308-0d34-0410-b5e6-96231b3b80d8
2015-04-24 04:22:39 +00:00
Jingyue Wu
f42450abb6 Revert r235688
Seems breaking builds


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@235690 91177308-0d34-0410-b5e6-96231b3b80d8
2015-04-24 03:26:11 +00:00
Jingyue Wu
d83b3b1a8d [NVPTX] enable NaryReassociate in NVPTX
Summary:
We run NaryReassociate right after SLSR because SLSR enables many
opportunities for NaryReassociate. For example, in nary-slsr.ll

  foo((a + b) + c);
  foo((a + b * 2) + c);
  foo((a + b * 3) + c);   // 2 muls and 6 adds

after SLSR:

  ab = a + b;
  foo(ab + c);
  ab2 = ab + b;
  foo(ab2 + c);
  ab3 = ab2 + b;
  foo(ab3 + c);           // 6 adds

after NaryReassociate:

  abc = (a + b) + c;
  foo(abc);
  ab2c = abc + b;
  foo(ab2c);
  ab3c = ab2c + b;
  foo(ab3c);              // 4 adds

Test Plan: nary-slsr.ll

Reviewers: jholewinski, eliben

Reviewed By: eliben

Subscribers: jholewinski, llvm-commits

Differential Revision: http://reviews.llvm.org/D9066

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@235688 91177308-0d34-0410-b5e6-96231b3b80d8
2015-04-24 02:54:06 +00:00
Jingyue Wu
f182c81444 [NaryReassociate] run NaryReassociate iteratively
Summary:
An alternative is to use a worklist approach. However, that approach
would break the traversing order so that we couldn't lookup SeenExprs
efficiently. I don't see a clear winner here, so I picked the easier approach.

Along with two minor improvements:
1. preserves ScalarEvolution by forgetting instructions replaced
2. removes dead code locally avoiding the need of running DCE afterwards

Test Plan: add to slsr-add.ll a test that requires multiple iterations

Reviewers: broune, dberlin, atrick, meheff

Reviewed By: atrick

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D9058

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@235151 91177308-0d34-0410-b5e6-96231b3b80d8
2015-04-17 00:25:10 +00:00
Jingyue Wu
feecc904c4 [NaryReassociate] speeds up candidate searching
Summary:
This fixes a left-over efficiency issue in D8950.

As Andrew and Daniel suggested, we can store the candidates in a stack
and pop the top element when it does not dominate the current
instruction. This reduces the worst-case time complexity to O(n).

Test Plan: a new test in nary-add.ll that exercises this optimization.

Reviewers: broune, dberlin, meheff, atrick

Reviewed By: atrick

Subscribers: llvm-commits, sanjoy

Differential Revision: http://reviews.llvm.org/D9055

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@235129 91177308-0d34-0410-b5e6-96231b3b80d8
2015-04-16 18:42:31 +00:00
Jingyue Wu
9cecacd16a Simplify n-ary adds by reassociation
Summary:
This transformation reassociates a n-ary add so that the add can partially reuse
existing instructions. For example, this pass can simplify

  void foo(int a, int b) {
    bar(a + b);
    bar((a + 2) + b);
  }

to

  void foo(int a, int b) {
    int t = a + b;
    bar(t);
    bar(t + 2);
  }

saving one add instruction.

Fixes PR22357 (https://llvm.org/bugs/show_bug.cgi?id=22357).

Test Plan: nary-add.ll

Reviewers: broune, dberlin, hfinkel, meheff, sanjoy, atrick

Reviewed By: sanjoy, atrick

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D8950

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@234855 91177308-0d34-0410-b5e6-96231b3b80d8
2015-04-14 04:59:22 +00:00