Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Vexiiriscv instructions #12

Open
ztachip opened this issue Apr 6, 2024 · 4 comments
Open

Vexiiriscv instructions #12

ztachip opened this issue Apr 6, 2024 · 4 comments

Comments

@ztachip
Copy link

ztachip commented Apr 6, 2024

Could you provide some instructions on how to install and use this new version.
Is it similar to VexRiscv?
Thx

@Dolu1990
Copy link
Member

Dolu1990 commented Apr 6, 2024

Hi,

So, currently, the only SoC i got using Vexii is ported to Litex.
It can run linux in single core config, i'm now working on memory coherency / multi core, which works in sim, but i need to test on hardware + cleaning.

It is still WIP. What would you be interrested into (exactly) ?

Is it similar to VexRiscv?

Yes, but better <3, but i'm still working on reaching feature parity.

@ztachip
Copy link
Author

ztachip commented Apr 6, 2024 via email

@Dolu1990
Copy link
Member

Dolu1990 commented Apr 6, 2024

So far, you can generate VexiiRiscv that way :
https://spinalhdl.github.io/VexiiRiscv-RTD/master/VexiiRiscv/HowToUse/index.html#generate-verilog
There is a lot of arguments possible.
The interface generated will be the native one. Currently the only bridge implemented are toward tilelink (but not available there as an option yet, WIP)

Ultimately, bridge toward popular memory bus (wishbone axi ... ) should be added (WIP)

VexRiscv is perfect for what I was using for, but I would like to upgrade from VexRiscv for performance

That is one of the main goal of VexiiRiscv.
I'm realy looking forward for the software / hardware D$ prefetcher, that could reaaaaly help a lot in memory intensive cases.

@Dolu1990
Copy link
Member

Dolu1990 commented Apr 6, 2024

Note that for now, the dual issue do a bit too much in the decode / dispatch stage (all together in one cycle), there is some work there planned to split things in two stages.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants