lkml.org 
[lkml]   [2009]   [Nov]   [3]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectRe: [PATCH -v4] crypto: Add PCLMULQDQ accelerated GHASH implementation
On Mon, Nov 02, 2009 at 04:46:04PM +0100, Ingo Molnar wrote:
>
> Yeah. Or just a single block of:
>
>
> #ifndef __ASSEMBLY__
> ...
> #endif /* __ASSEMBLY__ */
>
> around the C bits - anything outside that is good for assembly as well.

OK I'll throw this into cryptodev:

commit 3b0d65969b549b796abc6f0230f6142fed365d49
Author: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue Nov 3 09:11:15 2009 -0500

crypto: ghash-intel - Add PSHUFB macros

Add PSHUFB macros instead of repeating byte sequences, suggested
by Ingo.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Ingo Molnar <mingo@elte.hu>

diff --git a/arch/x86/crypto/ghash-clmulni-intel_asm.S b/arch/x86/crypto/ghash-clmulni-intel_asm.S
index 71768d5..5958498 100644
--- a/arch/x86/crypto/ghash-clmulni-intel_asm.S
+++ b/arch/x86/crypto/ghash-clmulni-intel_asm.S
@@ -17,6 +17,7 @@
*/

#include <linux/linkage.h>
+#include <asm/i387.h>

.align 16
.Lbswap_mask:
@@ -101,7 +102,7 @@ ENTRY(clmul_ghash_mul)
movups (%rsi), SHASH
movaps .Lbswap_mask, BSWAP
# pshufb BSWAP, DATA
- .byte 0x66, 0x0f, 0x38, 0x00, 0xc5
+ PSHUFB_XMM5_XMM0
call __clmul_gf128mul_ble
# pshufb BSWAP, DATA
.byte 0x66, 0x0f, 0x38, 0x00, 0xc5
@@ -119,12 +120,12 @@ ENTRY(clmul_ghash_update)
movups (%rdi), DATA
movups (%rcx), SHASH
# pshufb BSWAP, DATA
- .byte 0x66, 0x0f, 0x38, 0x00, 0xc5
+ PSHUFB_XMM5_XMM0
.align 4
.Lupdate_loop:
movups (%rsi), IN1
# pshufb BSWAP, IN1
- .byte 0x66, 0x0f, 0x38, 0x00, 0xf5
+ PSHUFB_XMM5_XMM6
pxor IN1, DATA
call __clmul_gf128mul_ble
sub $16, %rdx
@@ -132,7 +133,7 @@ ENTRY(clmul_ghash_update)
cmp $16, %rdx
jge .Lupdate_loop
# pshufb BSWAP, DATA
- .byte 0x66, 0x0f, 0x38, 0x00, 0xc5
+ PSHUFB_XMM5_XMM0
movups DATA, (%rdi)
.Lupdate_just_ret:
ret
@@ -146,7 +147,7 @@ ENTRY(clmul_ghash_setkey)
movaps .Lbswap_mask, BSWAP
movups (%rsi), %xmm0
# pshufb BSWAP, %xmm0
- .byte 0x66, 0x0f, 0x38, 0x00, 0xc5
+ PSHUFB_XMM5_XMM0
movaps %xmm0, %xmm1
psllq $1, %xmm0
psrlq $63, %xmm1
diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h
index 0b20bbb..ebfb8a9 100644
--- a/arch/x86/include/asm/i387.h
+++ b/arch/x86/include/asm/i387.h
@@ -10,6 +10,8 @@
#ifndef _ASM_X86_I387_H
#define _ASM_X86_I387_H

+#ifndef __ASSEMBLY__
+
#include <linux/sched.h>
#include <linux/kernel_stat.h>
#include <linux/regset.h>
@@ -411,4 +413,9 @@ static inline unsigned short get_fpu_mxcsr(struct task_struct *tsk)
}
}

+#endif /* __ASSEMBLY__ */
+
+#define PSHUFB_XMM5_XMM0 .byte 0x66, 0x0f, 0x38, 0x00, 0xc5
+#define PSHUFB_XMM5_XMM6 .byte 0x66, 0x0f, 0x38, 0x00, 0xf5
+
#endif /* _ASM_X86_I387_H */
Thanks,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

\
 
 \ /
  Last update: 2009-11-03 15:15    [W:0.178 / U:0.692 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site