mirror of
https://github.com/yuzu-emu/mbedtls.git
synced 2024-11-22 17:25:37 +01:00
Change asm format to \n\t in aesni.c too
This commit is contained in:
Manuel Pégourié-Gonnard
Paul Bakker
parent
03576887c2
commit
0534fd4c1a
378
library/aesni.c
378
library/aesni.c
@ -51,8 +51,8 @@ int aesni_supports( unsigned int what )
|
|||||||
|
|
||||||
if( ! done )
|
if( ! done )
|
||||||
{
|
{
|
||||||
asm( "movl $1, %%eax \n"
|
asm( "movl $1, %%eax \n\t"
|
||||||
"cpuid \n"
|
"cpuid \n\t"
|
||||||
: "=c" (c)
|
: "=c" (c)
|
||||||
:
|
:
|
||||||
: "eax", "ebx", "edx" );
|
: "eax", "ebx", "edx" );
|
||||||
@ -96,35 +96,35 @@ int aesni_crypt_ecb( aes_context *ctx,
|
|||||||
const unsigned char input[16],
|
const unsigned char input[16],
|
||||||
unsigned char output[16] )
|
unsigned char output[16] )
|
||||||
{
|
{
|
||||||
asm( "movdqu (%3), %%xmm0 \n" // load input
|
asm( "movdqu (%3), %%xmm0 \n\t" // load input
|
||||||
"movdqu (%1), %%xmm1 \n" // load round key 0
|
"movdqu (%1), %%xmm1 \n\t" // load round key 0
|
||||||
"pxor %%xmm1, %%xmm0 \n" // round 0
|
"pxor %%xmm1, %%xmm0 \n\t" // round 0
|
||||||
"addq $16, %1 \n" // point to next round key
|
"addq $16, %1 \n\t" // point to next round key
|
||||||
"subl $1, %0 \n" // normal rounds = nr - 1
|
"subl $1, %0 \n\t" // normal rounds = nr - 1
|
||||||
"test %2, %2 \n" // mode?
|
"test %2, %2 \n\t" // mode?
|
||||||
"jz 2f \n" // 0 = decrypt
|
"jz 2f \n\t" // 0 = decrypt
|
||||||
|
|
||||||
"1: \n" // encryption loop
|
"1: \n\t" // encryption loop
|
||||||
"movdqu (%1), %%xmm1 \n" // load round key
|
"movdqu (%1), %%xmm1 \n\t" // load round key
|
||||||
AESENC xmm1_xmm0 "\n" // do round
|
AESENC xmm1_xmm0 "\n\t" // do round
|
||||||
"addq $16, %1 \n" // point to next round key
|
"addq $16, %1 \n\t" // point to next round key
|
||||||
"subl $1, %0 \n" // loop
|
"subl $1, %0 \n\t" // loop
|
||||||
"jnz 1b \n"
|
"jnz 1b \n\t"
|
||||||
"movdqu (%1), %%xmm1 \n" // load round key
|
"movdqu (%1), %%xmm1 \n\t" // load round key
|
||||||
AESENCLAST xmm1_xmm0 "\n" // last round
|
AESENCLAST xmm1_xmm0 "\n\t" // last round
|
||||||
"jmp 3f \n"
|
"jmp 3f \n\t"
|
||||||
|
|
||||||
"2: \n" // decryption loop
|
"2: \n\t" // decryption loop
|
||||||
"movdqu (%1), %%xmm1 \n"
|
"movdqu (%1), %%xmm1 \n\t"
|
||||||
AESDEC xmm1_xmm0 "\n" // do round
|
AESDEC xmm1_xmm0 "\n\t" // do round
|
||||||
"addq $16, %1 \n"
|
"addq $16, %1 \n\t"
|
||||||
"subl $1, %0 \n"
|
"subl $1, %0 \n\t"
|
||||||
"jnz 2b \n"
|
"jnz 2b \n\t"
|
||||||
"movdqu (%1), %%xmm1 \n" // load round key
|
"movdqu (%1), %%xmm1 \n\t" // load round key
|
||||||
AESDECLAST xmm1_xmm0 "\n" // last round
|
AESDECLAST xmm1_xmm0 "\n\t" // last round
|
||||||
|
|
||||||
"3: \n"
|
"3: \n\t"
|
||||||
"movdqu %%xmm0, (%4) \n" // export output
|
"movdqu %%xmm0, (%4) \n\t" // export output
|
||||||
:
|
:
|
||||||
: "r" (ctx->nr), "r" (ctx->rk), "r" (mode), "r" (input), "r" (output)
|
: "r" (ctx->nr), "r" (ctx->rk), "r" (mode), "r" (input), "r" (output)
|
||||||
: "memory", "cc", "xmm0", "xmm1" );
|
: "memory", "cc", "xmm0", "xmm1" );
|
||||||
@ -151,44 +151,44 @@ void aesni_gcm_mult( unsigned char c[16],
|
|||||||
bb[i] = b[15 - i];
|
bb[i] = b[15 - i];
|
||||||
}
|
}
|
||||||
|
|
||||||
asm( "movdqu (%0), %%xmm0 \n" // a1:a0
|
asm( "movdqu (%0), %%xmm0 \n\t" // a1:a0
|
||||||
"movdqu (%1), %%xmm1 \n" // b1:b0
|
"movdqu (%1), %%xmm1 \n\t" // b1:b0
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Caryless multiplication xmm2:xmm1 = xmm0 * xmm1
|
* Caryless multiplication xmm2:xmm1 = xmm0 * xmm1
|
||||||
* using [CLMUL-WP] algorithm 1 (p. 13).
|
* using [CLMUL-WP] algorithm 1 (p. 13).
|
||||||
*/
|
*/
|
||||||
"movdqa %%xmm1, %%xmm2 \n" // copy of b1:b0
|
"movdqa %%xmm1, %%xmm2 \n\t" // copy of b1:b0
|
||||||
"movdqa %%xmm1, %%xmm3 \n" // same
|
"movdqa %%xmm1, %%xmm3 \n\t" // same
|
||||||
"movdqa %%xmm1, %%xmm4 \n" // same
|
"movdqa %%xmm1, %%xmm4 \n\t" // same
|
||||||
PCLMULQDQ xmm0_xmm1 ",0x00 \n" // a0*b0 = c1:c0
|
PCLMULQDQ xmm0_xmm1 ",0x00 \n\t" // a0*b0 = c1:c0
|
||||||
PCLMULQDQ xmm0_xmm2 ",0x11 \n" // a1*b1 = d1:d0
|
PCLMULQDQ xmm0_xmm2 ",0x11 \n\t" // a1*b1 = d1:d0
|
||||||
PCLMULQDQ xmm0_xmm3 ",0x10 \n" // a0*b1 = e1:e0
|
PCLMULQDQ xmm0_xmm3 ",0x10 \n\t" // a0*b1 = e1:e0
|
||||||
PCLMULQDQ xmm0_xmm4 ",0x01 \n" // a1*b0 = f1:f0
|
PCLMULQDQ xmm0_xmm4 ",0x01 \n\t" // a1*b0 = f1:f0
|
||||||
"pxor %%xmm3, %%xmm4 \n" // e1+f1:e0+f0
|
"pxor %%xmm3, %%xmm4 \n\t" // e1+f1:e0+f0
|
||||||
"movdqa %%xmm4, %%xmm3 \n" // same
|
"movdqa %%xmm4, %%xmm3 \n\t" // same
|
||||||
"psrldq $8, %%xmm4 \n" // 0:e1+f1
|
"psrldq $8, %%xmm4 \n\t" // 0:e1+f1
|
||||||
"pslldq $8, %%xmm3 \n" // e0+f0:0
|
"pslldq $8, %%xmm3 \n\t" // e0+f0:0
|
||||||
"pxor %%xmm4, %%xmm2 \n" // d1:d0+e1+f1
|
"pxor %%xmm4, %%xmm2 \n\t" // d1:d0+e1+f1
|
||||||
"pxor %%xmm3, %%xmm1 \n" // c1+e0+f1:c0
|
"pxor %%xmm3, %%xmm1 \n\t" // c1+e0+f1:c0
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Now shift the result one bit to the left,
|
* Now shift the result one bit to the left,
|
||||||
* taking advantage of [CLMUL-WP] eq 27 (p. 20)
|
* taking advantage of [CLMUL-WP] eq 27 (p. 20)
|
||||||
*/
|
*/
|
||||||
"movdqa %%xmm1, %%xmm3 \n" // r1:r0
|
"movdqa %%xmm1, %%xmm3 \n\t" // r1:r0
|
||||||
"movdqa %%xmm2, %%xmm4 \n" // r3:r2
|
"movdqa %%xmm2, %%xmm4 \n\t" // r3:r2
|
||||||
"psllq $1, %%xmm1 \n" // r1<<1:r0<<1
|
"psllq $1, %%xmm1 \n\t" // r1<<1:r0<<1
|
||||||
"psllq $1, %%xmm2 \n" // r3<<1:r2<<1
|
"psllq $1, %%xmm2 \n\t" // r3<<1:r2<<1
|
||||||
"psrlq $63, %%xmm3 \n" // r1>>63:r0>>63
|
"psrlq $63, %%xmm3 \n\t" // r1>>63:r0>>63
|
||||||
"psrlq $63, %%xmm4 \n" // r3>>63:r2>>63
|
"psrlq $63, %%xmm4 \n\t" // r3>>63:r2>>63
|
||||||
"movdqa %%xmm3, %%xmm5 \n" // r1>>63:r0>>63
|
"movdqa %%xmm3, %%xmm5 \n\t" // r1>>63:r0>>63
|
||||||
"pslldq $8, %%xmm3 \n" // r0>>63:0
|
"pslldq $8, %%xmm3 \n\t" // r0>>63:0
|
||||||
"pslldq $8, %%xmm4 \n" // r2>>63:0
|
"pslldq $8, %%xmm4 \n\t" // r2>>63:0
|
||||||
"psrldq $8, %%xmm5 \n" // 0:r1>>63
|
"psrldq $8, %%xmm5 \n\t" // 0:r1>>63
|
||||||
"por %%xmm3, %%xmm1 \n" // r1<<1|r0>>63:r0<<1
|
"por %%xmm3, %%xmm1 \n\t" // r1<<1|r0>>63:r0<<1
|
||||||
"por %%xmm4, %%xmm2 \n" // r3<<1|r2>>62:r2<<1
|
"por %%xmm4, %%xmm2 \n\t" // r3<<1|r2>>62:r2<<1
|
||||||
"por %%xmm5, %%xmm2 \n" // r3<<1|r2>>62:r2<<1|r1>>63
|
"por %%xmm5, %%xmm2 \n\t" // r3<<1|r2>>62:r2<<1|r1>>63
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Now reduce modulo the GCM polynomial x^128 + x^7 + x^2 + x + 1
|
* Now reduce modulo the GCM polynomial x^128 + x^7 + x^2 + x + 1
|
||||||
@ -196,44 +196,44 @@ void aesni_gcm_mult( unsigned char c[16],
|
|||||||
* Currently xmm2:xmm1 holds x3:x2:x1:x0 (already shifted).
|
* Currently xmm2:xmm1 holds x3:x2:x1:x0 (already shifted).
|
||||||
*/
|
*/
|
||||||
/* Step 2 (1) */
|
/* Step 2 (1) */
|
||||||
"movdqa %%xmm1, %%xmm3 \n" // x1:x0
|
"movdqa %%xmm1, %%xmm3 \n\t" // x1:x0
|
||||||
"movdqa %%xmm1, %%xmm4 \n" // same
|
"movdqa %%xmm1, %%xmm4 \n\t" // same
|
||||||
"movdqa %%xmm1, %%xmm5 \n" // same
|
"movdqa %%xmm1, %%xmm5 \n\t" // same
|
||||||
"psllq $63, %%xmm3 \n" // x1<<63:x0<<63 = stuff:a
|
"psllq $63, %%xmm3 \n\t" // x1<<63:x0<<63 = stuff:a
|
||||||
"psllq $62, %%xmm4 \n" // x1<<62:x0<<62 = stuff:b
|
"psllq $62, %%xmm4 \n\t" // x1<<62:x0<<62 = stuff:b
|
||||||
"psllq $57, %%xmm5 \n" // x1<<57:x0<<57 = stuff:c
|
"psllq $57, %%xmm5 \n\t" // x1<<57:x0<<57 = stuff:c
|
||||||
|
|
||||||
/* Step 2 (2) */
|
/* Step 2 (2) */
|
||||||
"pxor %%xmm4, %%xmm3 \n" // stuff:a+b
|
"pxor %%xmm4, %%xmm3 \n\t" // stuff:a+b
|
||||||
"pxor %%xmm5, %%xmm3 \n" // stuff:a+b+c
|
"pxor %%xmm5, %%xmm3 \n\t" // stuff:a+b+c
|
||||||
"pslldq $8, %%xmm3 \n" // a+b+c:0
|
"pslldq $8, %%xmm3 \n\t" // a+b+c:0
|
||||||
"pxor %%xmm3, %%xmm1 \n" // x1+a+b+c:x0 = d:x0
|
"pxor %%xmm3, %%xmm1 \n\t" // x1+a+b+c:x0 = d:x0
|
||||||
|
|
||||||
/* Steps 3 and 4 */
|
/* Steps 3 and 4 */
|
||||||
"movdqa %%xmm1,%%xmm0 \n" // d:x0
|
"movdqa %%xmm1,%%xmm0 \n\t" // d:x0
|
||||||
"movdqa %%xmm1,%%xmm4 \n" // same
|
"movdqa %%xmm1,%%xmm4 \n\t" // same
|
||||||
"movdqa %%xmm1,%%xmm5 \n" // same
|
"movdqa %%xmm1,%%xmm5 \n\t" // same
|
||||||
"psrlq $1, %%xmm0 \n" // e1:x0>>1 = e1:e0'
|
"psrlq $1, %%xmm0 \n\t" // e1:x0>>1 = e1:e0'
|
||||||
"psrlq $2, %%xmm4 \n" // f1:x0>>2 = f1:f0'
|
"psrlq $2, %%xmm4 \n\t" // f1:x0>>2 = f1:f0'
|
||||||
"psrlq $7, %%xmm5 \n" // g1:x0>>7 = g1:g0'
|
"psrlq $7, %%xmm5 \n\t" // g1:x0>>7 = g1:g0'
|
||||||
"pxor %%xmm4, %%xmm0 \n" // e1+f1:e0'+f0'
|
"pxor %%xmm4, %%xmm0 \n\t" // e1+f1:e0'+f0'
|
||||||
"pxor %%xmm5, %%xmm0 \n" // e1+f1+g1:e0'+f0'+g0'
|
"pxor %%xmm5, %%xmm0 \n\t" // e1+f1+g1:e0'+f0'+g0'
|
||||||
// e0'+f0'+g0' is almost e0+f0+g0, except for some missing
|
// e0'+f0'+g0' is almost e0+f0+g0, ex\tcept for some missing
|
||||||
// bits carried from d. Now get those bits back in.
|
// bits carried from d. Now get those\t bits back in.
|
||||||
"movdqa %%xmm1,%%xmm3 \n" // d:x0
|
"movdqa %%xmm1,%%xmm3 \n\t" // d:x0
|
||||||
"movdqa %%xmm1,%%xmm4 \n" // same
|
"movdqa %%xmm1,%%xmm4 \n\t" // same
|
||||||
"movdqa %%xmm1,%%xmm5 \n" // same
|
"movdqa %%xmm1,%%xmm5 \n\t" // same
|
||||||
"psllq $63, %%xmm3 \n" // d<<63:stuff
|
"psllq $63, %%xmm3 \n\t" // d<<63:stuff
|
||||||
"psllq $62, %%xmm4 \n" // d<<62:stuff
|
"psllq $62, %%xmm4 \n\t" // d<<62:stuff
|
||||||
"psllq $57, %%xmm5 \n" // d<<57:stuff
|
"psllq $57, %%xmm5 \n\t" // d<<57:stuff
|
||||||
"pxor %%xmm4, %%xmm3 \n" // d<<63+d<<62:stuff
|
"pxor %%xmm4, %%xmm3 \n\t" // d<<63+d<<62:stuff
|
||||||
"pxor %%xmm5, %%xmm3 \n" // missing bits of d:stuff
|
"pxor %%xmm5, %%xmm3 \n\t" // missing bits of d:stuff
|
||||||
"psrldq $8, %%xmm3 \n" // 0:missing bits of d
|
"psrldq $8, %%xmm3 \n\t" // 0:missing bits of d
|
||||||
"pxor %%xmm3, %%xmm0 \n" // e1+f1+g1:e0+f0+g0
|
"pxor %%xmm3, %%xmm0 \n\t" // e1+f1+g1:e0+f0+g0
|
||||||
"pxor %%xmm1, %%xmm0 \n" // h1:h0
|
"pxor %%xmm1, %%xmm0 \n\t" // h1:h0
|
||||||
"pxor %%xmm2, %%xmm0 \n" // x3+h1:x2+h0
|
"pxor %%xmm2, %%xmm0 \n\t" // x3+h1:x2+h0
|
||||||
|
|
||||||
"movdqu %%xmm0, (%2) \n" // done
|
"movdqu %%xmm0, (%2) \n\t" // done
|
||||||
:
|
:
|
||||||
: "r" (aa), "r" (bb), "r" (cc)
|
: "r" (aa), "r" (bb), "r" (cc)
|
||||||
: "memory", "cc", "xmm0", "xmm1", "xmm2", "xmm3", "xmm4", "xmm5" );
|
: "memory", "cc", "xmm0", "xmm1", "xmm2", "xmm3", "xmm4", "xmm5" );
|
||||||
@ -257,9 +257,9 @@ void aesni_inverse_key( unsigned char *invkey,
|
|||||||
memcpy( ik, fk, 16 );
|
memcpy( ik, fk, 16 );
|
||||||
|
|
||||||
for( fk -= 16, ik += 16; fk > fwdkey; fk -= 16, ik += 16 )
|
for( fk -= 16, ik += 16; fk > fwdkey; fk -= 16, ik += 16 )
|
||||||
asm( "movdqu (%0), %%xmm0 \n"
|
asm( "movdqu (%0), %%xmm0 \n\t"
|
||||||
AESIMC xmm0_xmm0 "\n"
|
AESIMC xmm0_xmm0 "\n\t"
|
||||||
"movdqu %%xmm0, (%1) \n"
|
"movdqu %%xmm0, (%1) \n\t"
|
||||||
:
|
:
|
||||||
: "r" (fk), "r" (ik)
|
: "r" (fk), "r" (ik)
|
||||||
: "memory", "xmm0" );
|
: "memory", "xmm0" );
|
||||||
@ -273,9 +273,9 @@ void aesni_inverse_key( unsigned char *invkey,
|
|||||||
static void aesni_setkey_enc_128( unsigned char *rk,
|
static void aesni_setkey_enc_128( unsigned char *rk,
|
||||||
const unsigned char *key )
|
const unsigned char *key )
|
||||||
{
|
{
|
||||||
asm( "movdqu (%1), %%xmm0 \n" // copy the original key
|
asm( "movdqu (%1), %%xmm0 \n\t" // copy the original key
|
||||||
"movdqu %%xmm0, (%0) \n" // as round key 0
|
"movdqu %%xmm0, (%0) \n\t" // as round key 0
|
||||||
"jmp 2f \n" // skip auxiliary routine
|
"jmp 2f \n\t" // skip auxiliary routine
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Finish generating the next round key.
|
* Finish generating the next round key.
|
||||||
@ -287,31 +287,31 @@ static void aesni_setkey_enc_128( unsigned char *rk,
|
|||||||
* with r4 = X + r0, r5 = r4 + r1, r6 = r5 + r2, r7 = r6 + r3
|
* with r4 = X + r0, r5 = r4 + r1, r6 = r5 + r2, r7 = r6 + r3
|
||||||
* and those are written to the round key buffer.
|
* and those are written to the round key buffer.
|
||||||
*/
|
*/
|
||||||
"1: \n"
|
"1: \n\t"
|
||||||
"pshufd $0xff, %%xmm1, %%xmm1 \n" // X:X:X:X
|
"pshufd $0xff, %%xmm1, %%xmm1 \n\t" // X:X:X:X
|
||||||
"pxor %%xmm0, %%xmm1 \n" // X+r3:X+r2:X+r1:r4
|
"pxor %%xmm0, %%xmm1 \n\t" // X+r3:X+r2:X+r1:r4
|
||||||
"pslldq $4, %%xmm0 \n" // r2:r1:r0:0
|
"pslldq $4, %%xmm0 \n\t" // r2:r1:r0:0
|
||||||
"pxor %%xmm0, %%xmm1 \n" // X+r3+r2:X+r2+r1:r5:r4
|
"pxor %%xmm0, %%xmm1 \n\t" // X+r3+r2:X+r2+r1:r5:r4
|
||||||
"pslldq $4, %%xmm0 \n" // etc
|
"pslldq $4, %%xmm0 \n\t" // etc
|
||||||
"pxor %%xmm0, %%xmm1 \n"
|
"pxor %%xmm0, %%xmm1 \n\t"
|
||||||
"pslldq $4, %%xmm0 \n"
|
"pslldq $4, %%xmm0 \n\t"
|
||||||
"pxor %%xmm1, %%xmm0 \n" // update xmm0 for next time!
|
"pxor %%xmm1, %%xmm0 \n\t" // update xmm0 for next time!
|
||||||
"add $16, %0 \n" // point to next round key
|
"add $16, %0 \n\t" // point to next round key
|
||||||
"movdqu %%xmm0, (%0) \n" // write it
|
"movdqu %%xmm0, (%0) \n\t" // write it
|
||||||
"ret \n"
|
"ret \n\t"
|
||||||
|
|
||||||
/* Main "loop" */
|
/* Main "loop" */
|
||||||
"2: \n"
|
"2: \n\t"
|
||||||
AESKEYGENA xmm0_xmm1 ",0x01 \ncall 1b \n"
|
AESKEYGENA xmm0_xmm1 ",0x01 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm0_xmm1 ",0x02 \ncall 1b \n"
|
AESKEYGENA xmm0_xmm1 ",0x02 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm0_xmm1 ",0x04 \ncall 1b \n"
|
AESKEYGENA xmm0_xmm1 ",0x04 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm0_xmm1 ",0x08 \ncall 1b \n"
|
AESKEYGENA xmm0_xmm1 ",0x08 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm0_xmm1 ",0x10 \ncall 1b \n"
|
AESKEYGENA xmm0_xmm1 ",0x10 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm0_xmm1 ",0x20 \ncall 1b \n"
|
AESKEYGENA xmm0_xmm1 ",0x20 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm0_xmm1 ",0x40 \ncall 1b \n"
|
AESKEYGENA xmm0_xmm1 ",0x40 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm0_xmm1 ",0x80 \ncall 1b \n"
|
AESKEYGENA xmm0_xmm1 ",0x80 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm0_xmm1 ",0x1B \ncall 1b \n"
|
AESKEYGENA xmm0_xmm1 ",0x1B \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm0_xmm1 ",0x36 \ncall 1b \n"
|
AESKEYGENA xmm0_xmm1 ",0x36 \n\tcall 1b \n\t"
|
||||||
:
|
:
|
||||||
: "r" (rk), "r" (key)
|
: "r" (rk), "r" (key)
|
||||||
: "memory", "cc", "0" );
|
: "memory", "cc", "0" );
|
||||||
@ -323,13 +323,13 @@ static void aesni_setkey_enc_128( unsigned char *rk,
|
|||||||
static void aesni_setkey_enc_192( unsigned char *rk,
|
static void aesni_setkey_enc_192( unsigned char *rk,
|
||||||
const unsigned char *key )
|
const unsigned char *key )
|
||||||
{
|
{
|
||||||
asm( "movdqu (%1), %%xmm0 \n" // copy original round key
|
asm( "movdqu (%1), %%xmm0 \n\t" // copy original round key
|
||||||
"movdqu %%xmm0, (%0) \n"
|
"movdqu %%xmm0, (%0) \n\t"
|
||||||
"add $16, %0 \n"
|
"add $16, %0 \n\t"
|
||||||
"movq 16(%1), %%xmm1 \n"
|
"movq 16(%1), %%xmm1 \n\t"
|
||||||
"movq %%xmm1, (%0) \n"
|
"movq %%xmm1, (%0) \n\t"
|
||||||
"add $8, %0 \n"
|
"add $8, %0 \n\t"
|
||||||
"jmp 2f \n" // skip auxiliary routine
|
"jmp 2f \n\t" // skip auxiliary routine
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Finish generating the next 6 quarter-keys.
|
* Finish generating the next 6 quarter-keys.
|
||||||
@ -340,34 +340,34 @@ static void aesni_setkey_enc_192( unsigned char *rk,
|
|||||||
* On exit, xmm0 is r9:r8:r7:r6 and xmm1 is stuff:stuff:r11:r10
|
* On exit, xmm0 is r9:r8:r7:r6 and xmm1 is stuff:stuff:r11:r10
|
||||||
* and those are written to the round key buffer.
|
* and those are written to the round key buffer.
|
||||||
*/
|
*/
|
||||||
"1: \n"
|
"1: \n\t"
|
||||||
"pshufd $0x55, %%xmm2, %%xmm2 \n" // X:X:X:X
|
"pshufd $0x55, %%xmm2, %%xmm2 \n\t" // X:X:X:X
|
||||||
"pxor %%xmm0, %%xmm2 \n" // X+r3:X+r2:X+r1:r4
|
"pxor %%xmm0, %%xmm2 \n\t" // X+r3:X+r2:X+r1:r4
|
||||||
"pslldq $4, %%xmm0 \n" // etc
|
"pslldq $4, %%xmm0 \n\t" // etc
|
||||||
"pxor %%xmm0, %%xmm2 \n"
|
"pxor %%xmm0, %%xmm2 \n\t"
|
||||||
"pslldq $4, %%xmm0 \n"
|
"pslldq $4, %%xmm0 \n\t"
|
||||||
"pxor %%xmm0, %%xmm2 \n"
|
"pxor %%xmm0, %%xmm2 \n\t"
|
||||||
"pslldq $4, %%xmm0 \n"
|
"pslldq $4, %%xmm0 \n\t"
|
||||||
"pxor %%xmm2, %%xmm0 \n" // update xmm0 = r9:r8:r7:r6
|
"pxor %%xmm2, %%xmm0 \n\t" // update xmm0 = r9:r8:r7:r6
|
||||||
"movdqu %%xmm0, (%0) \n"
|
"movdqu %%xmm0, (%0) \n\t"
|
||||||
"add $16, %0 \n"
|
"add $16, %0 \n\t"
|
||||||
"pshufd $0xff, %%xmm0, %%xmm2 \n" // r9:r9:r9:r9
|
"pshufd $0xff, %%xmm0, %%xmm2 \n\t" // r9:r9:r9:r9
|
||||||
"pxor %%xmm1, %%xmm2 \n" // stuff:stuff:r9+r5:r10
|
"pxor %%xmm1, %%xmm2 \n\t" // stuff:stuff:r9+r5:r10
|
||||||
"pslldq $4, %%xmm1 \n" // r2:r1:r0:0
|
"pslldq $4, %%xmm1 \n\t" // r2:r1:r0:0
|
||||||
"pxor %%xmm2, %%xmm1 \n" // update xmm1 = stuff:stuff:r11:r10
|
"pxor %%xmm2, %%xmm1 \n\t" // xmm1 = stuff:stuff:r11:r10
|
||||||
"movq %%xmm1, (%0) \n"
|
"movq %%xmm1, (%0) \n\t"
|
||||||
"add $8, %0 \n"
|
"add $8, %0 \n\t"
|
||||||
"ret \n"
|
"ret \n\t"
|
||||||
|
|
||||||
"2: \n"
|
"2: \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x01 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x01 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x02 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x02 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x04 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x04 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x08 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x08 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x10 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x10 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x20 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x20 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x40 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x40 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x80 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x80 \n\tcall 1b \n\t"
|
||||||
|
|
||||||
:
|
:
|
||||||
: "r" (rk), "r" (key)
|
: "r" (rk), "r" (key)
|
||||||
@ -380,12 +380,12 @@ static void aesni_setkey_enc_192( unsigned char *rk,
|
|||||||
static void aesni_setkey_enc_256( unsigned char *rk,
|
static void aesni_setkey_enc_256( unsigned char *rk,
|
||||||
const unsigned char *key )
|
const unsigned char *key )
|
||||||
{
|
{
|
||||||
asm( "movdqu (%1), %%xmm0 \n"
|
asm( "movdqu (%1), %%xmm0 \n\t"
|
||||||
"movdqu %%xmm0, (%0) \n"
|
"movdqu %%xmm0, (%0) \n\t"
|
||||||
"add $16, %0 \n"
|
"add $16, %0 \n\t"
|
||||||
"movdqu 16(%1), %%xmm1 \n"
|
"movdqu 16(%1), %%xmm1 \n\t"
|
||||||
"movdqu %%xmm1, (%0) \n"
|
"movdqu %%xmm1, (%0) \n\t"
|
||||||
"jmp 2f \n" // skip auxiliary routine
|
"jmp 2f \n\t" // skip auxiliary routine
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Finish generating the next two round keys.
|
* Finish generating the next two round keys.
|
||||||
@ -396,45 +396,45 @@ static void aesni_setkey_enc_256( unsigned char *rk,
|
|||||||
* On exit, xmm0 is r11:r10:r9:r8 and xmm1 is r15:r14:r13:r12
|
* On exit, xmm0 is r11:r10:r9:r8 and xmm1 is r15:r14:r13:r12
|
||||||
* and those have been written to the output buffer.
|
* and those have been written to the output buffer.
|
||||||
*/
|
*/
|
||||||
"1: \n"
|
"1: \n\t"
|
||||||
"pshufd $0xff, %%xmm2, %%xmm2 \n"
|
"pshufd $0xff, %%xmm2, %%xmm2 \n\t"
|
||||||
"pxor %%xmm0, %%xmm2 \n"
|
"pxor %%xmm0, %%xmm2 \n\t"
|
||||||
"pslldq $4, %%xmm0 \n"
|
"pslldq $4, %%xmm0 \n\t"
|
||||||
"pxor %%xmm0, %%xmm2 \n"
|
"pxor %%xmm0, %%xmm2 \n\t"
|
||||||
"pslldq $4, %%xmm0 \n"
|
"pslldq $4, %%xmm0 \n\t"
|
||||||
"pxor %%xmm0, %%xmm2 \n"
|
"pxor %%xmm0, %%xmm2 \n\t"
|
||||||
"pslldq $4, %%xmm0 \n"
|
"pslldq $4, %%xmm0 \n\t"
|
||||||
"pxor %%xmm2, %%xmm0 \n"
|
"pxor %%xmm2, %%xmm0 \n\t"
|
||||||
"add $16, %0 \n"
|
"add $16, %0 \n\t"
|
||||||
"movdqu %%xmm0, (%0) \n"
|
"movdqu %%xmm0, (%0) \n\t"
|
||||||
|
|
||||||
/* Set xmm2 to stuff:Y:stuff:stuff with Y = subword( r11 )
|
/* Set xmm2 to stuff:Y:stuff:stuff with Y = subword( r11 )
|
||||||
* and proceed to generate next round key from there */
|
* and proceed to generate next round key from there */
|
||||||
AESKEYGENA xmm0_xmm2 ",0x00 \n"
|
AESKEYGENA xmm0_xmm2 ",0x00 \n\t"
|
||||||
"pshufd $0xaa, %%xmm2, %%xmm2 \n"
|
"pshufd $0xaa, %%xmm2, %%xmm2 \n\t"
|
||||||
"pxor %%xmm1, %%xmm2 \n"
|
"pxor %%xmm1, %%xmm2 \n\t"
|
||||||
"pslldq $4, %%xmm1 \n"
|
"pslldq $4, %%xmm1 \n\t"
|
||||||
"pxor %%xmm1, %%xmm2 \n"
|
"pxor %%xmm1, %%xmm2 \n\t"
|
||||||
"pslldq $4, %%xmm1 \n"
|
"pslldq $4, %%xmm1 \n\t"
|
||||||
"pxor %%xmm1, %%xmm2 \n"
|
"pxor %%xmm1, %%xmm2 \n\t"
|
||||||
"pslldq $4, %%xmm1 \n"
|
"pslldq $4, %%xmm1 \n\t"
|
||||||
"pxor %%xmm2, %%xmm1 \n"
|
"pxor %%xmm2, %%xmm1 \n\t"
|
||||||
"add $16, %0 \n"
|
"add $16, %0 \n\t"
|
||||||
"movdqu %%xmm1, (%0) \n"
|
"movdqu %%xmm1, (%0) \n\t"
|
||||||
"ret \n"
|
"ret \n\t"
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Main "loop" - Generating one more key than necessary,
|
* Main "loop" - Generating one more key than necessary,
|
||||||
* see definition of aes_context.buf
|
* see definition of aes_context.buf
|
||||||
*/
|
*/
|
||||||
"2: \n"
|
"2: \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x01 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x01 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x02 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x02 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x04 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x04 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x08 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x08 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x10 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x10 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x20 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x20 \n\tcall 1b \n\t"
|
||||||
AESKEYGENA xmm1_xmm2 ",0x40 \ncall 1b \n"
|
AESKEYGENA xmm1_xmm2 ",0x40 \n\tcall 1b \n\t"
|
||||||
:
|
:
|
||||||
: "r" (rk), "r" (key)
|
: "r" (rk), "r" (key)
|
||||||
: "memory", "cc", "0" );
|
: "memory", "cc", "0" );
|
||||||
|
|||||||
Loading…
Reference in New Issue
Block a user