News for hackers. Stuff that matters.

A comment on Harvey – An effort to provide a modern, distributed, 64-bit operating system. Parent comment

Re:Harvey – An effort to provide a modern, distributed, 64-b... by jstewartmobile on Saturday December 23, 2017 @12:21AM (#15993049)
When it's time to speak directly to hardware, Rust and Haskell aren't exactly silver bullets: so much of the state ends up being in the chips instead of the program--thus outside the compiler's scope.

If we want to get away from C-level twiddling, we will also need to get away from C-level hardware.

[ Parent | on HN ]
Re:Harvey – An effort to provide a modern, distributed, 64-b... by tome on Saturday December 23, 2017 @05:06AM (#15993631)
It's not outside the compiler's scope at all. See for example http://www.clash-lang.org/.
[ Parent | on HN ]
Re:Harvey – An effort to provide a modern, distributed, 64-b... by jstewartmobile on Saturday December 23, 2017 @07:26AM (#15993939)
How does an HDL address the limitation of software compilers at the software-hardware boundary?

By scope, I mean that the compiler can really only see as far as the source code it is compiling. In something like an OS--especially a non-microkernel--an awful lot ends up having to be read and written across the boundary, and in an unsafe context.

[ Parent | on HN ]
Re:Harvey – An effort to provide a modern, distributed, 64-b... by tome on Sunday December 24, 2017 @07:08AM (#15999204)
Oh I see. You were talking about a compiler for software talking to hardware. Well, I don't see why talking to hardware is any different to talking to a CPU or main memory or a disk.
[ Parent | on HN ]
Re:Harvey – An effort to provide a modern, distributed, 64-b... by jstewartmobile on Sunday December 24, 2017 @09:09AM (#15999582)
Not so much "talking" as directly reading/writing memory outside of your source scope that is also mutated by other chips--like a DMA buffer.

Or for a traditional kernel, reading/writing directly into the address space of a user process. Wouldn't have to worry so much about race conditions there since the process could be suspended, but there is still an opportunity to fudge the indexing.

[ Parent | on HN ]
Re:Harvey – An effort to provide a modern, distributed, 64-b... by nickelbox on Saturday December 23, 2017 @06:06AM (#15993767)
It looks like something that just gets complied to VHDL and friends? Is there some way to make the description available to user code/GHC? Now that would be interesting.
[ Parent | on HN ]