You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the code base, the word "table" refers to two things. In both cases it's a 2-d array of base or extension field elements, so the word "table" does seem justified; but I think the two senses may benefit from some disambiguation none the less.
The isolated trace of a chiplet such as processor or hash or u32 etc.
The concatenation of all chiplet-traces along with table-linking columns, degree-lowering columns, and randomizer columns.
Most of the time when sense (2) is used it is distinguished from (1) through the qualifier "master", i.e., master_main_table, master_aux_table, etc. I think the word "master" is a poor choice because it sounds like just another chiplet, one that comes with a "slave" counterpart.
I suggest using the word "aet" for sense (2). So any "table" is always in reference to a chiplet in isolation of the rest of the architecture; and "aet" is always what the prover accesses.
The text was updated successfully, but these errors were encountered:
In the code base, the word "table" refers to two things. In both cases it's a 2-d array of base or extension field elements, so the word "table" does seem justified; but I think the two senses may benefit from some disambiguation none the less.
Most of the time when sense (2) is used it is distinguished from (1) through the qualifier "master", i.e.,
master_main_table
,master_aux_table
, etc. I think the word "master" is a poor choice because it sounds like just another chiplet, one that comes with a "slave" counterpart.I suggest using the word "aet" for sense (2). So any "table" is always in reference to a chiplet in isolation of the rest of the architecture; and "aet" is always what the prover accesses.
The text was updated successfully, but these errors were encountered: