[all-commits] [llvm/llvm-project] eac348: [LoopInterchange] Try to achieve the most optimal ...
CongzheUalberta via All-commits
all-commits at lists.llvm.org
Wed Apr 6 12:32:41 PDT 2022
Branch: refs/heads/main
Home: https://github.com/llvm/llvm-project
Commit: eac34875109898ac01985f4afa937eec30c1c387
https://github.com/llvm/llvm-project/commit/eac34875109898ac01985f4afa937eec30c1c387
Author: Congzhe Cao <congzhe.cao at huawei.com>
Date: 2022-04-06 (Wed, 06 Apr 2022)
Changed paths:
M llvm/lib/Transforms/Scalar/LoopInterchange.cpp
M llvm/test/Transforms/LoopInterchange/phi-ordering.ll
A llvm/test/Transforms/LoopInterchange/pr43326-ideal-access-pattern.ll
Log Message:
-----------
[LoopInterchange] Try to achieve the most optimal access pattern after interchange
Motivated by pr43326 (https://bugs.llvm.org/show_bug.cgi?id=43326), where a slightly
modified case is as follows.
void f(int e[10][10][10], int f[10][10][10]) {
for (int a = 0; a < 10; a++)
for (int b = 0; b < 10; b++)
for (int c = 0; c < 10; c++)
f[c][b][a] = e[c][b][a];
}
The ideal optimal access pattern after running interchange is supposed to be the following
void f(int e[10][10][10], int f[10][10][10]) {
for (int c = 0; c < 10; c++)
for (int b = 0; b < 10; b++)
for (int a = 0; a < 10; a++)
f[c][b][a] = e[c][b][a];
}
Currently loop interchange is limited to picking up the innermost loop and finding an order
that is locally optimal for it. However, the pass failed to produce the globally optimal
loop access order. For more complex examples what we get could be quite far from the
globally optimal ordering.
What is proposed in this patch is to do a "bubble-sort" fashion when doing interchange.
By comparing neighbors in `LoopList` in each iteration, we would be able to move each loop
onto a most appropriate place, hence this is an approach that tries to achieve the
globally optimal ordering.
The motivating example above is added as a test case.
Reviewed By: Meinersbur
Differential Revision: https://reviews.llvm.org/D120386
More information about the All-commits
mailing list