[llvm-dev] Loop Access Analysis
Min-Yih Hsu via llvm-dev
llvm-dev at lists.llvm.org
Sat Dec 21 18:04:56 PST 2019
The Loop Optimization Working Group had a survey about “canonical” loop required by each loop transformation or analysis. That is, what should we normalize a loop before running a certain loop pass.
Here is the survey result: https://tinyurl.com/rhuzny2
(Note that there are two workspaces: “By precondition” and “Breakdown”)
> On Dec 22, 2019, at 2:40 AM, Shraiysh Vaishay via llvm-dev <llvm-dev at lists.llvm.org> wrote:
> Hi all,
> Just putting this here in case someone else stumbles upon this.
> So, it looks like it is important to rotate the loop for the pass to run properly. Running `-passes="function(mem2reg,loop(rotate))"` started displaying proper analysis.
> A related issue was mentioned earlier [here](https://lists.llvm.org/pipermail/llvm-dev/2013-May/062260.html <https://lists.llvm.org/pipermail/llvm-dev/2013-May/062260.html>)
> On Sat, Dec 21, 2019 at 3:33 PM Shraiysh Vaishay <cs17btech11050 at iith.ac.in <mailto:cs17btech11050 at iith.ac.in>> wrote:
> Hi all,
> What kind of input loop is expected for the loop-access pass to run smoothly?
> Also, what passes (like mem2reg) should be run to ensure the desired form is attained. Is running mem2reg, indvars and loop-simplify enough for that?
> Any help will be appreciated.
> Thanks and regards,
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the llvm-dev