From my understanding, memory training is/was a closely held secret of memory makers and EDA IP houses who sold memory controller IP to all the chip vendors. This in turn makes fully open motherboard firmware almost impossible as no one can write code for memory training to bring up the chip. That piece of code has to be loaded as a blob - if you can get the blob.

Why do we need training?

It's in the link. https://www.systemverilog.io/design/ddr4-initialization-and-...