[Libre-soc-dev] SimdSignal update

lkcl luke.leighton at gmail.com
Thu Oct 28 09:46:14 BST 2021



On October 28, 2021 2:55:10 AM UTC, Jacob Lifshay <programmerjake at gmail.com> wrote:

>If you depended on a width *hint*, XLEN would still have the hint be
>64-bits so SimdSignals of XLEN-bits would *still* be 64-bits wide.

this misses the interaction between fixed width (XLEN, 64Bit) SimdSignals and some temporary (intermediary) signals that will have been specified at elwidth priority (usually by code that is tmp = Signal(4) being converted to tmp = scope.Signal(4))

which is why a flag indicating the priority mode is needed (FIXEDWID, ELWID, BOTHDASHNEITHER)

it's quite, quite mad, but workable.

i'll fill out some examples explaining it, later today.  if you can help also fill out case by case some examples in the wiki page as well that would help both understanding (for you) and also documentation.

when the examples are comprehensively completed it will be easy to post-analyse them all for patterns.

at that point it is the "ahh okaaay" realisation moment.

sometimes the only way to do these things is to literally walk through every damn permutation.  i learned that with PartitionedGt/Eq/Lt and PartitionedShift.

l.



More information about the Libre-soc-dev mailing list