more of a hardware problem... if CPU's let you directly address and manage L1, L2, L3 cache memory high performance programmers would love it and languages like C++ would immediately add support.
Anyone can prefetch data into the cache, but letting programmers control cache eviction would open the door to all kinds of user errors which would kill performance.
true but it could also be very efficient, especially in cases where you have the exact hardware to test on like a game console
There are tricks to pull chunks of memory into cache as is, no? Not that they are ideal.
asm { }
yea its a lot of "guesswork" and trial and error trying to optimize cache usage though