Field G. Van Zee

@jeffhammond In principle, I agree with you. However, this is the sort of thing that is not as practical now that our group is so small. (It also doesn't help...

> **and** arrange for the reference kernel to be compiled with architecture-appropriate flags. The second issue means that e.g. a configuration without an optimized kernel would possibly run faster because...

@devinamatthews Ah, makes sense. Thanks for clarifying. Yeah, `generic` doesn't do jack except use `-O3`, which I'm guessing in our world doesn't do much either.

> Yes, it doesn't make much difference experimentally (on x86_64), but you might expect it to help by including vectorization. I might be willing to add such a flag or...

According to [the netlib documentation for level-1 BLAS](http://www.netlib.org/blas/#_level_1), the `i?amin()` routines are not defined. I also checked the source code for the latest version of the reference BLAS (3.8.0) and...

> We are trying to port our code from MKL to BLIS, so I thought compatibility layer would make BLIS a drop-in replacement for MKL. Yes, sorry for the inconvenience....

> starting from `syr2k` and whacking some stuff off seems like the same thing and pretty easy to do. Agreed. This is how I would approach it. the `syrk` macrokernel...

@poulson @chriscoey Can you elaborate more on what you have in mind, since I'm unfamiliar with this operation? Specifically, I'm interested in knowing the structure of A in `C +=...

@rvdg Thanks, Robert. So is a real domain LDL factorization specified as `A = L D L^T`?

I still haven't seen anyone relate `C += alpha A (D A)^T` and `A -= L D L^T`. Even after you nix the alpha, harmonize the +/-, and assume A...