seems like something I should have realized sooner: the effective performance range of RISC style ISA's is not particularly variable...
Unless you either have a real compiler (i.e. gcc, llvm, or firm backend) or write asm by hand, you can't really compare ISAs.
-
-
there are limits, but at least within the ISA I can measure whether or not enabling/disabling features is effective (if both gen sane code).
-
You can only measure whether the changes improve things for a naive compiler not for optimal use of the ISA.
-
fair enough. things could be better, but the codegen could also be a lot worse (like, it has reg-alloc and strength reduction and similar).
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.