Daniel Karrasch

Yes, the current status is that you explicitly hook into the `_compositemul!` machinery, which was introduced for exactly that purpose, when users know that they are dealing with `CompositeMap`s and...

Should we even backport this to v1.10?

So, the main idea is not to confuse the compiler unnecessarily with uppercase/lowercase H or S when, for the result type, that distinction is irrelevant anyway, right? Because that distinction...

I love it. I always dreamt of the day when that character stuff be inferred, or constant-propagated far enough. Is this ready to go now? I think we should first...

That's right, and there is also #206 here. The "drawback" of the workaround above is that `(if tA == 'T' transpose(A) elseif tA == 'C' A' else A end) *...

These functions are now owned by LinearAlgebra (and not `LinearAlgebra.BLAS`) and admit generic fallbacks, likely using broadcasting but I don't remember. Another option could be to introduce `_axp[b]y!`, that use...

No, this will be backported to v1.9 for some sorting reason, I believe.

I'm not sure how exactly the pivoting works, but I thought that pivoted Cholesky is rank revealing in the sense that the (almost) zero diagonal entries come at the end:...

What if we specialized `copy_similar` to `HermOrSym`? We could copy then only half of the data from the parent (i.e., without `getindex` passing through the `HermOrSym` branches)?

BTW, should we make use of `copytrito!` in `copyto!(::Hermitian, ::Hermitian)`? We could then try something else: define `eigencopy_oftype(A::Hermitian, S) = copymutable_oftype(A, S)`. If I made no mistake, this reduces the...