[PATCH] D122900: [x86] Disable optimization involved in an infinite loop inside DAG combiner.

Fri Apr 1 07:39:45 PDT 2022

lebedev.ri added a comment.

In D122900#3422365 <https://reviews.llvm.org/D122900#3422365>, @spatel wrote:

> In D122900#3422344 <https://reviews.llvm.org/D122900#3422344>, @pgousseau wrote:
>
>> In D122900#3422266 <https://reviews.llvm.org/D122900#3422266>, @lebedev.ri wrote:
>>
>>> Do you know which commit introduced the problem, or which transforms conflict?
>>> This looks like a pretty heavy hammer.
>>
>> Thank you Roman for reviewing!
>> The regression appeared after https://github.com/llvm/llvm-project/commit/8a156d1c2795189389fadbf33702384f522f2506 (thank you @wristow  for finding it)
>> As @RKSimon  found out, the IR does not get generated anymore from the original C++ test case but the issue is still present.
>> I am hopeful we can reintroduce the optimization (at a later codegen stage maybe?) but I would like to do some performance analysis first to see if it is still a win.
>
> That change might have exposed the bug from source/IR, but the backend loop has existed since at least the 10.0 release based on a godbolt check:
> https://godbolt.org/z/Tfnaevboz

Then i think just fixing it may be better than removing some chunks of code.

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D122900/new/

https://reviews.llvm.org/D122900