Bug 840 - Formal proofs and unit tests for cryptoprimitives
Summary: Formal proofs and unit tests for cryptoprimitives
Status: RESOLVED FIXED
Alias: None
Product: Libre-SOC's first SoC
Classification: Unclassified
Component: Source Code (show other bugs)
Version: unspecified
Hardware: Other Linux
: High enhancement
Assignee: 2021-02-052
URL: https://libre-soc.org/openpower/sv/bi...
Depends on:
Blocks: 589
  Show dependency treegraph
 
Reported: 2022-05-21 16:29 BST by Luke Kenneth Casson Leighton
Modified: 2024-08-21 20:59 BST (History)
6 users (show)

See Also:
NLnet milestone: NLnet.2021.02A.052.CryptoRouter
total budget (EUR) for completion of task and all subtasks: 9500
budget (EUR) for this task, excluding subtasks' budget: 7750
parent task for budget allocation: 589
child tasks for budget allocation: 967 977 1159 1167
The table of payments (in EUR) for this task; TOML format:
lkcl={amount=3750, submitted=2024-05-22} [jacob] amount = 4000 submitted = 2024-05-21 paid = 2024-08-21


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Luke Kenneth Casson Leighton 2022-05-21 16:29:56 BST
Formal correctness proofs and unit tests for the cryptoprimitives
are needed in the HDL and simulators, demonstrating or proving
that the ISA pseudocode, HDL, and simulator are correctly implemented.
Comment 1 Luke Kenneth Casson Leighton 2023-09-05 05:06:41 BST

    
Comment 2 Luke Kenneth Casson Leighton 2024-05-22 05:56:49 BST
jacob you put the foral proofs for the work you just did under the wrong bug.
can you please put the notes you added recently copied under THIS bug
which is specifically for Formal Proofs.
Comment 3 Jacob Lifshay 2024-05-22 05:59:48 BST
(In reply to Luke Kenneth Casson Leighton from comment #2)
> jacob you put the foral proofs for the work you just did under the wrong bug.
> can you please put the notes you added recently copied under THIS bug
> which is specifically for Formal Proofs.

I'll just link to all the comments:
note about moving them:
https://bugs.libre-soc.org/show_bug.cgi?id=785#c14

completion of gfbinv's formal proof:
https://bugs.libre-soc.org/show_bug.cgi?id=785#c10

completion of gfbmul/gfbmadd's formal proof:
https://bugs.libre-soc.org/show_bug.cgi?id=785#c12
Comment 4 Luke Kenneth Casson Leighton 2024-05-22 06:18:07 BST
(In reply to Jacob Lifshay from comment #3)

> I'll just link to all the comments:

that works. remember to cross-ref (edit those comments, link to here)
Comment 5 Luke Kenneth Casson Leighton 2024-08-10 11:40:27 BST
what questions do you (NLnet) have which allow this 100% completed RFP
to be paid? both jacob and i are waiting for payment on work that is
100% completed and has proof (comment #3) that it is completed.
Comment 6 Jacob Lifshay 2024-08-16 22:25:52 BST
as requested by nlnet, I added some documentation in the form of links on the wiki to where the hardware implementations and tests and formal proofs for gfbinv and gfbmadd can be found:
https://git.libre-soc.org/?p=libreriscv.git;a=commitdiff;h=b215d742df3838addea3b425c16bf139e3d5f3c8;hp=67bec36c57df5d5de2ecc373026cd015d02c6c40
Comment 7 Luke Kenneth Casson Leighton 2024-08-16 23:42:39 BST
(In reply to Jacob Lifshay from comment #6)
> as requested by nlnet,

can you please publish that request. i.e. do not engage in private conversations.
it is critically important to ensure full transparency.

> I added some documentation in the form of links on
> the wiki to where the hardware implementations and tests and formal proofs
> for gfbinv and gfbmadd can be found:

fantastic. i added an href name and updated the URL field.

please get that request (any communication) from nlnet published here.
really: it is *unsafe* to engage in private conversations with them.
it is unfortunately not appropriate to explain why.
Comment 8 Jacob Lifshay 2024-08-16 23:54:46 BST
(In reply to Luke Kenneth Casson Leighton from comment #7)
> (In reply to Jacob Lifshay from comment #6)
> > as requested by nlnet,
> 
> can you please publish that request. i.e. do not engage in private
> conversations.
> it is critically important to ensure full transparency.

I asked for permission, we'll see what they say.
Comment 9 Luke Kenneth Casson Leighton 2024-08-17 00:50:33 BST
(In reply to Jacob Lifshay from comment #8)

> I asked for permission, we'll see what they say.

they should not have been speaking privately and directly to
you in the first place. just fyi: you and i, as the primary
developers, have been accused by nlnet of fabricating work
or failing to do the correct or full work, all these years.
of failing to meet milestone obligations in full (creating
milestones, then submitting an RFP for work that has not
been completed, as listed and stated in the milestone
description). hence why it is so dangerous, now, to interact
privately with them.
Comment 10 Luke Kenneth Casson Leighton 2024-08-19 04:06:31 BST
financial accounting and audit purposes only:

put correct assignment to 2021-02-052 (not 2021-08-071)

and correct submission RFP date as it crossed a day barrier,
due to nlnet's server being in UTC+2 when the submission
was done from the UK (UTC+1) and the records being done between
23:01 and 00:00 on the 21st in the UK, nlnet's server (in holland)
recorded this as the *22nd*.
Comment 11 Luke Kenneth Casson Leighton 2024-08-19 04:22:22 BST
a private and secretive conversation took place between an
agent of nlnet and jacob lifshay (see comment #6 for evidence
of the existence of that private, secretive and non-transparent
conversation)

we have evidence of a review taking place, in the form of a disconnected
message from nlnet's system (no cross-referencing provided). only by chance
was it possible to make the connection that it was this bugreport that had
been reviewed.

https://lists.libre-soc.org/pipermail/libre-soc-dev/2024-August/006407.html
Comment 12 Luke Kenneth Casson Leighton 2024-08-19 04:27:25 BST
in checking the NLnet RFP records (their system) more closely,
it is *jacob's* RFP that has been approved on this bugreport,
but mine has *not* been approved. they are for the exact same
work, the exact same bugreport, but jacob's RFP has been
approved (payment now pending) and mine has not.
Comment 13 Jacob Lifshay 2024-08-21 10:17:04 BST
Michiel, I'll note that if the documentation I added to the wiki was sufficient for me to get paid for this task, I would assume it's also sufficient for Luke to get paid for the exact same task, him getting assigned EUR 3750 of the budget on this task is intentional and was agreed on by both of us.
Comment 14 Luke Kenneth Casson Leighton 2024-08-21 16:55:05 BST
(In reply to Jacob Lifshay from comment #13)
> Michiel, I'll note that if the documentation I added to the wiki was
> sufficient for me to get paid for this task, I would assume it's also
> sufficient for Luke to get paid for the exact same task, him getting
> assigned EUR 3750 of the budget on this task is intentional and was agreed
> on by both of us.

jacob please email them and/or call them directly as the obvious implication
from what they have are saying is, "luke is an untrustworthy abuser whom
we cannot listen to, therefore let's punish him by shutting down ALL
RFPs".

   https://lists.libre-soc.org/pipermail/libre-soc-dev/2024-August/006419.html

HELP.
Comment 15 Jacob Lifshay 2024-08-21 20:59:58 BST
(In reply to Jacob Lifshay from comment #13)

Michiel gave me permission to quote the following reply:

> > I would assume it's also sufficient for Luke to get paid for the exact same task
> 
> That is an understandable but incorrect inference. Even within a single task, different people will have distinct roles and responsibilities. You made an engineering contribution. That part seems finished, and so it was paid. As the project lead, Luke is also responsible for integrating the outputs of this and all other dangling task into the documentation skeleton of this particular project. The project is claimed to be completed, so we are asking where the outputs we were promised by Luke can be found.
> 
> At this moment, there is nothing even remotely shaped like a cryptorouter documented anywhere - not even a failed, incomplete one. We need Luke (or someone else knowledgeable) to provide this skeleton.
> 
> I think this would probably be the right starting point:
> 
> https://libre-soc.org/crypto_router_asic/
> 
> This page was last edited Thu Nov 25 17:13:51 2021, but from there it would seem feasible to get the heavily fragmented information into a presentable shape.