On Friday, February 23, 2018 at 11:08:23 AM UTC-5, David K. Storrs wrote:
> On Thu, Feb 22, 2018 at 11:14 AM, Alexander McLin <alex....@gmail.com 
> <javascript:>> wrote:
>> As one of those who have been following RISC-V progress for several years 
>> and also interested in seeing Racket being ported to that architecture I 
>> want to drop a note to let you know you have my support!
> As someone who doesn't know a lot about hardware, I'm curious:  what 
> effect would runningthe new architecture have?  Would it enable new 
> functionality, provide performance boosts...?

In the near-future, none. 

RISC-V is designed to be an open-source ISA free of any royalties or 
licensing concerns whatsoever. The idea is to provide a flexible family of 
ISAs which can be combined as needed for one's purposes and any company or 
foundry can manufacture RISC-V CPUs without restrictions. RSIC-V's long 
term goal is to become a universal ISA anyone can use in contrast to X86 or 
ARM which come with hefty licensing fees and restrictions on who can 
manufacture chips which makes it difficult for anyone else to innovate in 
those spaces.

As for enabling new functionality, one major goal for RISC-V is to allow 
the ISA to be extended in well-defined ways which preserve backward 
compatibility but allow innovative features to be enabled for specific 
applications. It remains to see how successful that would be. My 
expectation is better support for hardware-based security features and 
audit of hardware designs by independent parties which is significant given 
how the Meltdown and Spectre vulnerabilities have highlighted the opaque 
nature of the closed-source X86 hardware.

Once anyone who's interested can at relatively low costs explore the design 
space of possible RISC-V hardware, who knows what performance benefits may 
become possible.

You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to