# Formal specs and optimizations in general **URL:** \ **Category:** language design\ **Created:** [December 15, 2020, 2:54pm UTC](https://internals.rust-lang.org/t/formal-specs-and-optimizations-in-general/13635 "2020-12-15T14:54:57Z")\ **Posts on this page:** 1\ **Showing post:** 6
**Author:** ![steffahn](https://sea2.discourse-cdn.com/flex002/user_avatar/internals.rust-lang.org/steffahn/32/13288_2.png) [@steffahn](https://internals.rust-lang.org/u/steffahn)\ **Post date:** [December 17, 2020, 12:21am UTC](https://internals.rust-lang.org/t/formal-specs-and-optimizations-in-general/13635/6 "2020-12-17T00:21:32Z")
> [@Pointers Are Complicated II, or: We need better language specs](https://internals.rust-lang.org/t/pointers-are-complicated-ii-or-we-need-better-language-specs/13562/29): > > Basically you are saying there are two LLVM IRs, IR1 and IR2, with the same syntax but different semantics: IR1 has provenance and IR2 doesn't. The given optimization can only be correct in IR2, so at some point in the optimization pipeline there needs to be a clear transition from working with IR1 to working with IR2. Yep, that was what I meant. > [@Pointers Are Complicated II, or: We need better language specs](https://internals.rust-lang.org/t/pointers-are-complicated-ii-or-we-need-better-language-specs/13562/29): > > Who wants to create a bugreport? > [@Pointers Are Complicated II, or: We need better language specs](https://internals.rust-lang.org/t/pointers-are-complicated-ii-or-we-need-better-language-specs/13562/29): > > Looks like we need to clarify the docs. This would be two bug reports then, I guess.. I can do those tomorrow. --- _[View the full topic](https://internals.rust-lang.org/t/formal-specs-and-optimizations-in-general/13635)._