Implements loading challenges from the transcript and computing powers of α, which is later used in widgets. More...

#include <transition_widget.hpp>

Inheritance diagram for proof_system::plonk::widget::getters::BaseGetter< Field, Transcript, Settings, num_widget_relations >:

Static Public Member Functions
static challenge_array	get_challenges (const Transcript &transcript, const Field &alpha_base, uint8_t required_challenges)

static Field	update_alpha (const challenge_array &challenges, const size_t num_independent_relations)

Protected Types
typedef containers::challenge_array< Field, num_widget_relations >	challenge_array

Detailed Description

template<class Field, class Transcript, class Settings, size_t num_widget_relations>
class proof_system::plonk::widget::getters::BaseGetter< Field, Transcript, Settings, num_widget_relations >

Implements loading challenges from the transcript and computing powers of α, which is later used in widgets.

Template Parameters

Field	Base field
Transcript	Transcript class
Settings	Configuration
num_widget_relations	How many powers of α are needed

Member Function Documentation

◆ get_challenges()

template<class Field , class Transcript , class Settings , size_t num_widget_relations>

static challenge_array proof_system::plonk::widget::getters::BaseGetter< Field, Transcript, Settings, num_widget_relations >::get_challenges	(	const Transcript &	transcript,
		const Field &	alpha_base,
		uint8_t	required_challenges
	)

inlinestatic

Create a challenge array from transcript. Loads alpha, beta, gamma, eta, zeta and nu and calculates powers of alpha.

Parameters

transcript	Transcript to get challenges from.
alpha_base	alpha to some power (depends on previously used widgets).
required_challenges	Challenge bitmask, which shows when the function should fail.

Returns: A structure with an array of challenge values and powers of alpha.

There are several issues we need to address here:

We can't just set the value to 0. In case of a typo this could lead to a vulnerability
We can't fail when there is no challenge, because getters get activated at various phases
There is no way for us to check accesses in the challenge_array (it would degrade speed)

One of the mitigations we use is we force the transition widget kernel to have members that describe the necessary challenges for quotient polynomial construction and for kate update. We then take them and submit to the get_challenges function. This allows us to catch misuse, but only if the developer is prudent.

Since we can't enforce anything really we introduced a simple mitigation: The challenges that aren't in the transcript are replaced by random values.

The value each of the widget uses and the value the verifier uses will differ. As a result, proof will fail if some widget uses an uninitialized challenge. *

The documentation for this class was generated from the following file:

src/barretenberg/plonk/proof_system/widgets/transition_widgets/transition_widget.hpp

Static Public Member Functions

Protected Types

Detailed Description

Member Function Documentation

◆ get_challenges()