auto-tiling opt2: update condition for divisible tile size · Pull Request !785 · MindSpore/akg

<!--  Thanks for sending a pull request!  Here are some tips for you:

If this is your first time, please read our contributor guidelines: https://gitee.com/mindspore/akg/blob/master/CONTRIBUTING.md
-->

**What type of PR is this?**
> Uncomment only one ` /kind <>` line, hit enter to put that in a new line, and remove leading whitespaces from that line:
>
> /kind bug
> /kind task

kind feature

**What does this PR do / why do we need it**:
This PR optimizes auto-tiling on GPU platform for reduction operators.

The original strategy was to make the tile size align with the shape so that we will not produce any tail part. However, for some irregular shape, prime number for example, this strategy will have very bad performance -- the tile size will be restricted to be very small.

Therefore, we now consider to allow tail part for common reduction operations to improve performance.

**Which issue(s) this PR fixes**:

Fixes https://gitee.com/mindspore/akg/issues/I4NRR9?from=project-issue

**Special notes for your reviewers**:

MindSpore/akg关闭 .gitee-modal { width: 500px !important; }

搜索帮助

MindSpore/akg
关闭