mirror/gmsm
1
0
Fork 0
mirror of https://github.com/emmansun/gmsm.git synced 2025-04-26 04:06:18 +08:00
Code Issues Packages Projects Releases Wiki Activity

reduce sm3 arm64 instructions

Browse Source
This commit is contained in:
Emman 2021-12-29 08:37:58 +08:00
parent 4991042efd
commit 875395a593
1 changed files with 37 additions and 37 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

74
sm3/sm3block_arm64.s
View File

@ -26,13 +26,13 @@
MOVW ((index-12)*4)(BP), BX; \ MOVW ((index-12)*4)(BP), BX; \
EORW BX, AX; \ EORW BX, AX; \
MOVW ((index-5)*4)(BP), BX; \ MOVW ((index-5)*4)(BP), BX; \
EORW BX, AX; \ EORW BX, AX; \ // AX = x
MOVW AX, BX; \ //MOVW AX, BX; \ // BX = x
RORW $17, BX; \ RORW $17, AX, BX; \ // BX = ROTL(15, x)
MOVW AX, CX; \ //MOVW AX, CX; \ // CX = x
RORW $9, CX; \ RORW $9, AX, CX; \ // CX = ROTL(23, x)
EORW BX, AX; \ EORW BX, AX; \ // AX = x xor ROTL(15, x)
EORW CX, AX; \ EORW CX, AX; \ // AX = x xor ROTL(15, x) xor ROTL(23, x)
MOVW ((index-9)*4)(BP), BX; \ MOVW ((index-9)*4)(BP), BX; \
RORW $25, BX; \ RORW $25, BX; \
MOVW ((index-2)*4)(BP), CX; \ MOVW ((index-2)*4)(BP), CX; \
@ -44,8 +44,8 @@
// x = ROTL(12, a) + e + ROTL(index, const) // x = ROTL(12, a) + e + ROTL(index, const)
// ret = ROTL(7, x) // ret = ROTL(7, x)
#define SM3SS1(const, a, e) \ #define SM3SS1(const, a, e) \
MOVW a, BX; \ //MOVW a, BX; \
RORW $20, BX; \ RORW $20, a, BX; \
ADDW e, BX; \ ADDW e, BX; \
ADDW $const, BX; \ ADDW $const, BX; \
RORW $25, BX RORW $25, BX
@ -53,16 +53,16 @@
// Calculate tt1 in CX // Calculate tt1 in CX
// ret = (a XOR b XOR c) + d + (ROTL(12, a) XOR ss1) + (Wt XOR Wt+4) // ret = (a XOR b XOR c) + d + (ROTL(12, a) XOR ss1) + (Wt XOR Wt+4)
#define SM3TT10(index, a, b, c, d) \ #define SM3TT10(index, a, b, c, d) \
MOVW a, CX; \ //MOVW a, CX; \
MOVW b, DX; \ //MOVW b, DX; \
EORW CX, DX; \ EORW a, b, DX; \
MOVW c, hlp0; \ //MOVW c, hlp0; \
EORW hlp0, DX; \ // (a XOR b XOR c) EORW c, DX; \ // (a XOR b XOR c)
ADDW d, DX; \ // (a XOR b XOR c) + d ADDW d, DX; \ // (a XOR b XOR c) + d
MOVW ((index)*4)(BP), hlp0; \ //Wt MOVW ((index)*4)(BP), hlp0; \ //Wt
EORW hlp0, AX; \ //Wt XOR Wt+4 EORW hlp0, AX; \ //Wt XOR Wt+4
ADDW AX, DX; \ ADDW AX, DX; \
RORW $20, CX; \ RORW $20, a, CX; \
EORW BX, CX; \ // ROTL(12, a) XOR ss1 EORW BX, CX; \ // ROTL(12, a) XOR ss1
ADDW DX, CX // (a XOR b XOR c) + d + (ROTL(12, a) XOR ss1) ADDW DX, CX // (a XOR b XOR c) + d + (ROTL(12, a) XOR ss1)
@ -71,28 +71,28 @@
#define SM3TT20(e, f, g, h) \ #define SM3TT20(e, f, g, h) \
ADDW h, hlp0; \ //Wt + h ADDW h, hlp0; \ //Wt + h
ADDW BX, hlp0; \ //Wt + h + ss1 ADDW BX, hlp0; \ //Wt + h + ss1
MOVW e, BX; \ //MOVW e, BX; \
MOVW f, DX; \ //MOVW f, DX; \
EORW DX, BX; \ // e XOR f EORW e, f, BX; \ // e XOR f
MOVW g, DX; \ //MOVW g, DX; \
EORW DX, BX; \ // e XOR f XOR g EORW g, BX; \ // e XOR f XOR g
ADDW hlp0, BX // (e XOR f XOR g) + Wt + h + ss1 ADDW hlp0, BX // (e XOR f XOR g) + Wt + h + ss1
// Calculate tt1 in CX, used DX, hlp0 // Calculate tt1 in CX, used DX, hlp0
// ret = ((a AND b) OR (a AND c) OR (b AND c)) + d + (ROTL(12, a) XOR ss1) + (Wt XOR Wt+4) // ret = ((a AND b) OR (a AND c) OR (b AND c)) + d + (ROTL(12, a) XOR ss1) + (Wt XOR Wt+4)
#define SM3TT11(index, a, b, c, d) \ #define SM3TT11(index, a, b, c, d) \
MOVW a, CX; \ //MOVW a, CX; \
MOVW b, DX; \ //MOVW b, DX; \
ANDW CX, DX; \ // a AND b ANDW a, b, DX; \ // a AND b
MOVW c, hlp0; \ //MOVW c, hlp0; \
ANDW hlp0, CX; \ // a AND c ANDW a, c, CX; \ // a AND c
ORRW DX, CX; \ // (a AND b) OR (a AND c) ORRW DX, CX; \ // (a AND b) OR (a AND c)
MOVW b, DX; \ //MOVW b, DX; \
ANDW hlp0, DX; \ // b AND c ANDW b, c, DX; \ // b AND c
ORRW CX, DX; \ // (a AND b) OR (a AND c) OR (b AND c) ORRW CX, DX; \ // (a AND b) OR (a AND c) OR (b AND c)
ADDW d, DX; \ ADDW d, DX; \
MOVW a, CX; \ //MOVW a, CX; \
RORW $20, CX; \ RORW $20, a, CX; \
EORW BX, CX; \ EORW BX, CX; \
ADDW DX, CX; \ // ((a AND b) OR (a AND c) OR (b AND c)) + d + (ROTL(12, a) XOR ss1) ADDW DX, CX; \ // ((a AND b) OR (a AND c) OR (b AND c)) + d + (ROTL(12, a) XOR ss1)
MOVW ((index)*4)(BP), hlp0; \ MOVW ((index)*4)(BP), hlp0; \
@ -104,12 +104,12 @@
#define SM3TT21(e, f, g, h) \ #define SM3TT21(e, f, g, h) \
ADDW h, hlp0; \ // Wt + h ADDW h, hlp0; \ // Wt + h
ADDW BX, hlp0; \ // h + ss1 + Wt ADDW BX, hlp0; \ // h + ss1 + Wt
MOVW e, BX; \ //MOVW e, BX; \
MOVW f, DX; \ //MOVW f, DX; \
ANDW BX, DX; \ // e AND f ANDW e, f, DX; \ // e AND f
MVNW BX, BX; \ // NOT(e) MVNW e, BX; \ // NOT(e)
MOVW g, AX; \ //MOVW g, AX; \
ANDW AX, BX; \ // NOT(e) AND g ANDW g, BX; \ // NOT(e) AND g
ORRW DX, BX; \ ORRW DX, BX; \
ADDW hlp0, BX ADDW hlp0, BX
@ -117,8 +117,8 @@
RORW $23, b; \ RORW $23, b; \
MOVW CX, h; \ // a = ttl MOVW CX, h; \ // a = ttl
RORW $13, f; \ RORW $13, f; \
MOVW BX, CX; \ //MOVW BX, CX; \
RORW $23, CX; \ RORW $23, BX, CX; \
EORW BX, CX; \ // tt2 XOR ROTL(9, tt2) EORW BX, CX; \ // tt2 XOR ROTL(9, tt2)
RORW $15, BX; \ RORW $15, BX; \
EORW BX, CX; \ // tt2 XOR ROTL(9, tt2) XOR ROTL(17, tt2) EORW BX, CX; \ // tt2 XOR ROTL(9, tt2) XOR ROTL(17, tt2)