[llvm] [SLP]Attempt to vectorize long stores, if short one failed. (PR #88563)

[Nikita Popov via llvm-commits]

nikic wrote:

I've reverted this change due to large compile-time regressions (https://llvm-compile-time-tracker.com/compare.php?from=c7657cf7d1ee57f9cb9133164536591a1842b43c&to=7d4e8c1f3bbfe976f4871c9cf953f76d771b0eda&stat=instructions:u).

The regressions go up to 10% on individual files (https://llvm-compile-time-tracker.com/compare.php?from=c7657cf7d1ee57f9cb9133164536591a1842b43c&to=7d4e8c1f3bbfe976f4871c9cf953f76d771b0eda&stat=instructions%3Au&details=on). Though the 10% regressions are in LTO builds on small files, so easier to investigate is probably something like constants.c from pairlocalalign that regresses by 5% in a normal O3 build.



https://github.com/llvm/llvm-project/pull/88563