Moritz Imfeld comments

Results 8 comments of


                                            Moritz Imfeld

vzext.vf2 instruction execution problem

Hi @stevobailey AFAIK, to be compliant with `Zve32x` there should be support for `*.vf[2|4]`, where `vf4` is only legal if `SEW` = 32 (because the instruction will extend from elements...

Wrong result generated by multiply unit (probably control logic related)

From further observation I found that **this issue only arises when "commit delay" is enabled**. Usually what causes the result to be wrong is that either the value on the...

narrowing instructions are never popped from the instruction queue

Additionally, this can also happen for **masked** `vwredsum(u).vs`. However, I don't know if the root cause is the same.

RAM_ASIC register file has multiply driven registers

Hi @stevobailey and @michael-platzer This is indeed a very good point. When I implemented the ASIC register file I did not think about this case. And when compiling the code...

Fix for #100

I will check what caused the CI error and push another commit once I fixed it.

Fix for #100

Hi @michael-platzer Thanks a lot for reviewing! I incorporated your feedback and now the implementation looks a lot cleaner. If you want I can squash the commits before merging, let...

x_mem_result_ready needs to be added for subsystem sharing

Hi @Silabs-ArjanB I disagree that it would be difficult to add cv32e40p (and probably also in the cv32e40x). You could add a buffer for x-interface memory result transactions on the...

Applying fusion to a ViT from a different package

Hi, Thanks for your interest in our work! There are three tasks you need to do to add support for a new architecture: - Write a top-level fusion function (like...