Matteo Perotti
Matteo Perotti
Hi kuopinghsu, If I see correctly, you have run the standard versions of the programs (they are not vectorized). Guess that Ara is not used at all, in this case....
Hey Hossein, thanks for this important contribution and sorry for the delay in the answer ;) I see that the `riscv-tests` are not completely successful, are you still working on...
Hello Hossein, thank you for the PR! I have still not checked precisely the assignments within the `simd_alu` in terms of math correctness, but I started a review process and...
Ehy Hossein, do you have any news? ;)
Thank you @M-Ijaz-10x @hossein1387 for the joint effort! In parallel, I am solving the backend problems we meet after synthesis. I would freeze the merge of this PR until then
We can close timing after place & route, I will check again before merging with the missing fixed-point instructions, but everything seems okay so far! ;-)
@hossein1387 , I have created a PR on your GitHub to fix the problems with the CI https://github.com/PolyMTL-Gr2m/ara/pull/1. If you merge it, this CI should be fixed as well.
Hello zaazad, Thanks for opening this issue: VM support is an important milestone. Currently, we are focusing our efforts on alignment with the specs + verification + benchmarking, would you...
Let's rebase this, modify the changelog, and launch a backend run!
Hello @tangcy98, You can use our software environment and write your own programs. The compiler (you pointed out `riscv64-unknown-elf-gcc`, but we use `clang` (`llvm`) for our programs). We tell the...