[Libre-soc-dev] planning next OPF ISA WG RFC: ls003, biginteger?

lkcl luke.leighton at gmail.com
Thu Mar 2 22:27:00 GMT 2023

On Thu, Mar 2, 2023 at 5:38 PM Jacob Lifshay <programmerjake at gmail.com> wrote:

> dsld/dsrd are no longer hard to use,


> however they still have the issue that they unnecessarily write to `overflow`,

no, they don't. it's conditional. Rc=1.

> making them much slower than necessary

no, they're not.

> because of having to read/write SO,

no, they don't. read the RFC.

> which means you can't run multiple instructions in parallel unless
> you have special dependency hardware for SO.

which is optional in the SFFS subset. and therefore not
our problem.

> also, their overflow output is inconsistent with other powerisa shift instructions.

luckily we're not doing "other power isa shift instructions".

> i think we should just remove their overflow output, since that matches other powerisa unsigned shift instructions.

a large portion of the objections you raise are invalid, so i'm
leaving it in.

the ISA WG can always object and give reasons why it should
not be in.


More information about the Libre-soc-dev mailing list