it's depressing to think that the exact thing I want has been attempted in its most powerful form once (that I know of), it was 47 years ago, no one has tried to repeat it since, and the vendor who did it originally survives solely as a few footnotes in Wikipedia
-
-
Replying to @whitequark
I was digging around and found that
@bunniestudios worked on something like this at MIT: https://bunniestudios.com/bunnie/proj/rerisc/rerisc.html …1 reply 3 retweets 14 likes -
"With the ReRISC, compilers can now analyze programs and determine the optimal instruction set architecture (ISA) for that particular program. The code can then be compiled into a binary for that ISA, and executed on the ReRISC."
1 reply 2 retweets 13 likes -
whoah, this is so close that it directly matches some of the notes I've made while trying to work out the implementation :D
1 reply 0 retweets 7 likes -
(strictly speaking it details an implementation I rejected as probably too inefficient to design in, but ... details :P)
1 reply 0 retweets 3 likes
heh that work was done in the 90's as part of a VLSI design class. :P The profs back then didn't like the idea, so I abandoned it: at the time SIMD/VLIW and the Pentium 4's projected 10GHz clock speeds were hot topics - who needed customized instructions with GP CPUs like that?
-
-
i was really thrilled when the Pentium 4 10 GHz model came out!
0 replies 0 retweets 0 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
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.