[Libre-soc-dev] [libre-soc-dev] Alex Oliva's intro, and RFC on mission

Luke Kenneth Casson Leighton lkcl at lkcl.net
Fri Nov 20 13:36:26 GMT 2020

On 11/20/20, Alexandre Oliva <oliva at gnu.org> wrote:

> If you mean https://libre-soc.org/charter/ version 0, dated 2019-08-11,
> (just making sure I haven't run into another outdated link ;-) I have,
> and I hereby confirm my agreement and commitment to abide by it.

star.  it's an experiment on my part, after running lots of different
projects, as much as anything.  a combination of an incredibly simple
code plus some organisational laws which help identify hotspots known
to trip up progress.


first answering your question at

you're basically not comfortable with the possibility that OEMs would
use a processor for unethical purposes.

unfortunately, *all* processors and all ISAs fall into the category of
their *usage* being outside of the control of the designers.  to
illustrate: should we actually add instructions which state "the OEM
Certifies that they will act responsibly with this processor and

doesn't work, does it? :)

should we abandon OpenPOWER, design our own ISA, with a special
license, "omly to be used for ethical purposes, DRM prohibited"?

we will be done with that somewhere in 20 years time after 100+
man-years of effort.

likewise should we build in tracking and monitoring so that *we* can
tell if the processors are being used unethically?

that doesn't lead down an acceptable path either, does it?

the best that we can do is: apply the Bill of Ethics.  if approached
by an unethical company that wishes to do harm by producing products
with DRM is: not support them, by not providing them with resources.

we can encourage companies to not add DRM by creating Reference
Designs tgat don't have it.  given that it takes between USD 50,000
and USD 250,000 to create a Reference Design any OEM is more likely to
use that Reference Design than risk creating their own.

hardware is indeed different: it is the physical world and that means
creating hardware has a cost and resources barrier, which we can
leverage to ensure ethical designs get a boost and unethical ones do

make any sense? :)

jumping straight into practical matters: we're in the middle of
getting SimpleV redone on top of OpenPOWER, and a first step there is
Compressed Instructions.

even though it is very early i'd like to look at how we can get
statistical feedback in order to iterate on the encoding allocation.


More information about the Libre-soc-dev mailing list