logos-storage/plonky2
1
0
Fork 0
mirror of https://github.com/logos-storage/plonky2.git synced 2026-04-18 01:23:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

mul_fp2

Browse Source
This commit is contained in:
Dmitry Vagner 2022-11-16 15:12:42 -08:00
parent c1f7d1ce5a
commit b779b82512
1 changed files with 77 additions and 36 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

113
evm/src/cpu/kernel/asm/fields/field_macros.asm
View File

@ -13,7 +13,33 @@
// stack: ind0, x1 // stack: ind0, x1
%mload_kernel_general %mload_kernel_general
// stack: x0, x1 // stack: x0, x1
%endmacro %endmacro
%macro mul_fp2
// stack: a, b, c, d
DUP4
DUP3
MULFP254
// stack: bd, a, b, c, d
DUP4
DUP3
MULFP254
// stack: ac , bd, a, b, c, d
SUBFP254
// stack: ac - bd, a, b, c, d
SWAP4
// stack: d, a, b, c, ac - bd
MULFP254
// stack: ad, b, c, ac - bd
SWAP2
// stack: c, b, ad, ac - bd
MULFP254
// stack: bc , ad, ac - bd
ADDFP254
// stack: bc + ad, ac - bd
SWAP1
// stack: ac - bd, bc + ad
%endmacro
// cost: 6 loads + 6 dup/swaps + 5 adds = 6*4 + 6*1 + 5*2 = 40 // cost: 6 loads + 6 dup/swaps + 5 adds = 6*4 + 6*1 + 5*2 = 40
%macro load_fp6 %macro load_fp6
@ -354,6 +380,26 @@
// stack: g0, g1, g2, g3, g4, g5, X, f0, f1, f2, f3, f4, f5 // stack: g0, g1, g2, g3, g4, g5, X, f0, f1, f2, f3, f4, f5
%endmacro %endmacro
%macro sh
// stack: f0 , f0_, f1, f1_, f2 , f2_
SWAP2
// stack: f1 , f0_, g0 , f1_, f2 , f2_
SWAP4
// stack: f2 , f0_, g0 , f1_, g1 , f2_
SWAP1
// stack: f0_, f2 , g0 , f1_, g1 , f2_
SWAP3
// stack: f1_, f2 , g0 , g0_, g1 , f2_
SWAP5
// stack: f2_, f2 , g0 , g0_, g1 , g1_
SWAP1
// stack: f2 , f2_, g0 , g0_, g1 , g1_
%i9
// stack: g2_, g2 , g0 , g0_, g1 , g1_
SWAP1
// stack: g2 , g2_, g0 , g0_, g1 , g1_
%endmacro
// cost: 16 // cost: 16
%macro add_fp6 %macro add_fp6
// stack: f0, f1, f2, f3, f4, f5, g0, g1, g2, g3, g4, g5 // stack: f0, f1, f2, f3, f4, f5, g0, g1, g2, g3, g4, g5
@ -472,6 +518,25 @@
// stack: c * f0, c * f1, c * f2, c * f3, c * f4, c * f5 // stack: c * f0, c * f1, c * f2, c * f3, c * f4, c * f5
%endmacro %endmacro
// cost: 9; note this returns y, x for the output x + yi
%macro i9
// stack: a , b
DUP2
// stack: b, a , b
DUP2
// stack: a , b, a , b
PUSH 9 MULFP254
// stack: 9a , b, a , b
SUBFP254
// stack: 9a - b, a , b
SWAP2
// stack: b , a, 9a - b
PUSH 9 MULFP254
// stack 9b , a, 9a - b
ADDFP254
// stack: 9b + a, 9a - b
%endmacro
/// cost: 1 i9 (9) + 16 dups + 15 swaps + 12 muls + 6 adds/subs = 58 /// cost: 1 i9 (9) + 16 dups + 15 swaps + 12 muls + 6 adds/subs = 58
/// ///
/// G0 + G1t + G2t^2 = (a+bi)t * (F0 + F1t + F2t^2) /// G0 + G1t + G2t^2 = (a+bi)t * (F0 + F1t + F2t^2)
@ -660,41 +725,17 @@
// stack: g0, g0_, g1, g1_, g2, g2_ // stack: g0, g0_, g1, g1_, g2, g2_
%endmacro %endmacro
// cost: 9; note this returns y, x for the output x + yi %macro frob_fp2
%macro i9 // stack: n , a, b
// stack: a , b PUSH 2
DUP2 MULFP254
// stack: b, a , b PUSH 1
DUP2
// stack: a , b, a , b
PUSH 9 MULFP254
// stack: 9a , b, a , b
SUBFP254 SUBFP254
// stack: 9a - b, a , b // stack: 1-2n , a, b
SWAP2 DUP3
// stack: b , a, 9a - b MULFP254
PUSH 9 MULFP254 // stack: (1-2n)b, a, b
// stack 9b , a, 9a - b
ADDFP254
// stack: 9b + a, 9a - b
%endmacro
%macro sh
// stack: f0 , f0_, f1, f1_, f2 , f2_
SWAP2 SWAP2
// stack: f1 , f0_, g0 , f1_, f2 , f2_ POP
SWAP4 // stack: a, (1-2n)b
// stack: f2 , f0_, g0 , f1_, g1 , f2_
SWAP1
// stack: f0_, f2 , g0 , f1_, g1 , f2_
SWAP3
// stack: f1_, f2 , g0 , g0_, g1 , f2_
SWAP5
// stack: f2_, f2 , g0 , g0_, g1 , g1_
SWAP1
// stack: f2 , f2_, g0 , g0_, g1 , g1_
%i9
// stack: g2_, g2 , g0 , g0_, g1 , g1_
SWAP1
// stack: g2 , g2_, g0 , g0_, g1 , g1_
%endmacro %endmacro