[llvm] Revert "[APInt] Remove multiplicativeInverse with explicit modulus (#… (PR #87812)

Jakub Kuderski via llvm-commits llvm-commits at lists.llvm.org
Fri Apr 5 13:04:39 PDT 2024


kuhar wrote:

> …87644)"
> 
> This reverts commit [0b293e8](https://github.com/llvm/llvm-project/commit/0b293e8c36d97bbd7f85ed5b67ce510ff7fd86ee).
> 
> There are out-of-tree uses of this method, and it is planned to be used as part of a new polynomial dialect in MLIR, a starting PR of which is #72081 (later PRs will add lowerings that need the removed functionality)

I'm generally sympathetic to downstream maintainers, but this seems like an unusual reason for revert. Can't these users implement it on their own as a free function? If we have it upstream with no uses in the codebase, this will lead to eventual bitrot.

https://github.com/llvm/llvm-project/pull/87812


More information about the llvm-commits mailing list