Comment 2 for bug 1899874

Could this ever cause a serious problem? If the compiler assumes (1+ (length some-sequence)) will be a fixnum, I assume it would transform to a fixnum=>fixnum VOP, which could lead to a fault somewhere, right? Maybe this would only cause serious issues in 0 safety code, and if you're working with extended sequences you should probably be prepared for the worst. I'm still not familiar enough with the compiler to know how cautious it is when transforming based on function types.