Upload Kmake

deps/openssl/config/archs/linux-elf/asm_avx2/crypto/bn/x86-gf2m.S

@@ -0,0 +1,377 @@
+.text
+.type	_mul_1x1_mmx,@function
+.align	16
+_mul_1x1_mmx:
+	%ifdef __CET__
+
+.byte	243,15,30,251
+	%endif
+
+	subl	$36,%esp
+	movl	%eax,%ecx
+	leal	(%eax,%eax,1),%edx
+	andl	$1073741823,%ecx
+	leal	(%edx,%edx,1),%ebp
+	movl	$0,(%esp)
+	andl	$2147483647,%edx
+	movd	%eax,%mm2
+	movd	%ebx,%mm3
+	movl	%ecx,4(%esp)
+	xorl	%edx,%ecx
+	pxor	%mm5,%mm5
+	pxor	%mm4,%mm4
+	movl	%edx,8(%esp)
+	xorl	%ebp,%edx
+	movl	%ecx,12(%esp)
+	pcmpgtd	%mm2,%mm5
+	paddd	%mm2,%mm2
+	xorl	%edx,%ecx
+	movl	%ebp,16(%esp)
+	xorl	%edx,%ebp
+	pand	%mm3,%mm5
+	pcmpgtd	%mm2,%mm4
+	movl	%ecx,20(%esp)
+	xorl	%ecx,%ebp
+	psllq	$31,%mm5
+	pand	%mm3,%mm4
+	movl	%edx,24(%esp)
+	movl	$7,%esi
+	movl	%ebp,28(%esp)
+	movl	%esi,%ebp
+	andl	%ebx,%esi
+	shrl	$3,%ebx
+	movl	%ebp,%edi
+	psllq	$30,%mm4
+	andl	%ebx,%edi
+	shrl	$3,%ebx
+	movd	(%esp,%esi,4),%mm0
+	movl	%ebp,%esi
+	andl	%ebx,%esi
+	shrl	$3,%ebx
+	movd	(%esp,%edi,4),%mm2
+	movl	%ebp,%edi
+	psllq	$3,%mm2
+	andl	%ebx,%edi
+	shrl	$3,%ebx
+	pxor	%mm2,%mm0
+	movd	(%esp,%esi,4),%mm1
+	movl	%ebp,%esi
+	psllq	$6,%mm1
+	andl	%ebx,%esi
+	shrl	$3,%ebx
+	pxor	%mm1,%mm0
+	movd	(%esp,%edi,4),%mm2
+	movl	%ebp,%edi
+	psllq	$9,%mm2
+	andl	%ebx,%edi
+	shrl	$3,%ebx
+	pxor	%mm2,%mm0
+	movd	(%esp,%esi,4),%mm1
+	movl	%ebp,%esi
+	psllq	$12,%mm1
+	andl	%ebx,%esi
+	shrl	$3,%ebx
+	pxor	%mm1,%mm0
+	movd	(%esp,%edi,4),%mm2
+	movl	%ebp,%edi
+	psllq	$15,%mm2
+	andl	%ebx,%edi
+	shrl	$3,%ebx
+	pxor	%mm2,%mm0
+	movd	(%esp,%esi,4),%mm1
+	movl	%ebp,%esi
+	psllq	$18,%mm1
+	andl	%ebx,%esi
+	shrl	$3,%ebx
+	pxor	%mm1,%mm0
+	movd	(%esp,%edi,4),%mm2
+	movl	%ebp,%edi
+	psllq	$21,%mm2
+	andl	%ebx,%edi
+	shrl	$3,%ebx
+	pxor	%mm2,%mm0
+	movd	(%esp,%esi,4),%mm1
+	movl	%ebp,%esi
+	psllq	$24,%mm1
+	andl	%ebx,%esi
+	shrl	$3,%ebx
+	pxor	%mm1,%mm0
+	movd	(%esp,%edi,4),%mm2
+	pxor	%mm4,%mm0
+	psllq	$27,%mm2
+	pxor	%mm2,%mm0
+	movd	(%esp,%esi,4),%mm1
+	pxor	%mm5,%mm0
+	psllq	$30,%mm1
+	addl	$36,%esp
+	pxor	%mm1,%mm0
+	ret
+.size	_mul_1x1_mmx,.-_mul_1x1_mmx
+.type	_mul_1x1_ialu,@function
+.align	16
+_mul_1x1_ialu:
+	%ifdef __CET__
+
+.byte	243,15,30,251
+	%endif
+
+	subl	$36,%esp
+	movl	%eax,%ecx
+	leal	(%eax,%eax,1),%edx
+	leal	(,%eax,4),%ebp
+	andl	$1073741823,%ecx
+	leal	(%eax,%eax,1),%edi
+	sarl	$31,%eax
+	movl	$0,(%esp)
+	andl	$2147483647,%edx
+	movl	%ecx,4(%esp)
+	xorl	%edx,%ecx
+	movl	%edx,8(%esp)
+	xorl	%ebp,%edx
+	movl	%ecx,12(%esp)
+	xorl	%edx,%ecx
+	movl	%ebp,16(%esp)
+	xorl	%edx,%ebp
+	movl	%ecx,20(%esp)
+	xorl	%ecx,%ebp
+	sarl	$31,%edi
+	andl	%ebx,%eax
+	movl	%edx,24(%esp)
+	andl	%ebx,%edi
+	movl	%ebp,28(%esp)
+	movl	%eax,%edx
+	shll	$31,%eax
+	movl	%edi,%ecx
+	shrl	$1,%edx
+	movl	$7,%esi
+	shll	$30,%edi
+	andl	%ebx,%esi
+	shrl	$2,%ecx
+	xorl	%edi,%eax
+	shrl	$3,%ebx
+	movl	$7,%edi
+	andl	%ebx,%edi
+	shrl	$3,%ebx
+	xorl	%ecx,%edx
+	xorl	(%esp,%esi,4),%eax
+	movl	$7,%esi
+	andl	%ebx,%esi
+	shrl	$3,%ebx
+	movl	(%esp,%edi,4),%ebp
+	movl	$7,%edi
+	movl	%ebp,%ecx
+	shll	$3,%ebp
+	andl	%ebx,%edi
+	shrl	$29,%ecx
+	xorl	%ebp,%eax
+	shrl	$3,%ebx
+	xorl	%ecx,%edx
+	movl	(%esp,%esi,4),%ecx
+	movl	$7,%esi
+	movl	%ecx,%ebp
+	shll	$6,%ecx
+	andl	%ebx,%esi
+	shrl	$26,%ebp
+	xorl	%ecx,%eax
+	shrl	$3,%ebx
+	xorl	%ebp,%edx
+	movl	(%esp,%edi,4),%ebp
+	movl	$7,%edi
+	movl	%ebp,%ecx
+	shll	$9,%ebp
+	andl	%ebx,%edi
+	shrl	$23,%ecx
+	xorl	%ebp,%eax
+	shrl	$3,%ebx
+	xorl	%ecx,%edx
+	movl	(%esp,%esi,4),%ecx
+	movl	$7,%esi
+	movl	%ecx,%ebp
+	shll	$12,%ecx
+	andl	%ebx,%esi
+	shrl	$20,%ebp
+	xorl	%ecx,%eax
+	shrl	$3,%ebx
+	xorl	%ebp,%edx
+	movl	(%esp,%edi,4),%ebp
+	movl	$7,%edi
+	movl	%ebp,%ecx
+	shll	$15,%ebp
+	andl	%ebx,%edi
+	shrl	$17,%ecx
+	xorl	%ebp,%eax
+	shrl	$3,%ebx
+	xorl	%ecx,%edx
+	movl	(%esp,%esi,4),%ecx
+	movl	$7,%esi
+	movl	%ecx,%ebp
+	shll	$18,%ecx
+	andl	%ebx,%esi
+	shrl	$14,%ebp
+	xorl	%ecx,%eax
+	shrl	$3,%ebx
+	xorl	%ebp,%edx
+	movl	(%esp,%edi,4),%ebp
+	movl	$7,%edi
+	movl	%ebp,%ecx
+	shll	$21,%ebp
+	andl	%ebx,%edi
+	shrl	$11,%ecx
+	xorl	%ebp,%eax
+	shrl	$3,%ebx
+	xorl	%ecx,%edx
+	movl	(%esp,%esi,4),%ecx
+	movl	$7,%esi
+	movl	%ecx,%ebp
+	shll	$24,%ecx
+	andl	%ebx,%esi
+	shrl	$8,%ebp
+	xorl	%ecx,%eax
+	shrl	$3,%ebx
+	xorl	%ebp,%edx
+	movl	(%esp,%edi,4),%ebp
+	movl	%ebp,%ecx
+	shll	$27,%ebp
+	movl	(%esp,%esi,4),%edi
+	shrl	$5,%ecx
+	movl	%edi,%esi
+	xorl	%ebp,%eax
+	shll	$30,%edi
+	xorl	%ecx,%edx
+	shrl	$2,%esi
+	xorl	%edi,%eax
+	xorl	%esi,%edx
+	addl	$36,%esp
+	ret
+.size	_mul_1x1_ialu,.-_mul_1x1_ialu
+.globl	bn_GF2m_mul_2x2
+.type	bn_GF2m_mul_2x2,@function
+.align	16
+bn_GF2m_mul_2x2:
+.L_bn_GF2m_mul_2x2_begin:
+	%ifdef __CET__
+
+.byte	243,15,30,251
+	%endif
+
+	call	.L000PIC_me_up
+.L000PIC_me_up:
+	popl	%edx
+	leal	OPENSSL_ia32cap_P-.L000PIC_me_up(%edx),%edx
+	movl	(%edx),%eax
+	movl	4(%edx),%edx
+	testl	$8388608,%eax
+	jz	.L001ialu
+	testl	$16777216,%eax
+	jz	.L002mmx
+	testl	$2,%edx
+	jz	.L002mmx
+	movups	8(%esp),%xmm0
+	shufps	$177,%xmm0,%xmm0
+.byte	102,15,58,68,192,1
+	movl	4(%esp),%eax
+	movups	%xmm0,(%eax)
+	ret
+.align	16
+.L002mmx:
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	movl	24(%esp),%eax
+	movl	32(%esp),%ebx
+	call	_mul_1x1_mmx
+	movq	%mm0,%mm7
+	movl	28(%esp),%eax
+	movl	36(%esp),%ebx
+	call	_mul_1x1_mmx
+	movq	%mm0,%mm6
+	movl	24(%esp),%eax
+	movl	32(%esp),%ebx
+	xorl	28(%esp),%eax
+	xorl	36(%esp),%ebx
+	call	_mul_1x1_mmx
+	pxor	%mm7,%mm0
+	movl	20(%esp),%eax
+	pxor	%mm6,%mm0
+	movq	%mm0,%mm2
+	psllq	$32,%mm0
+	popl	%edi
+	psrlq	$32,%mm2
+	popl	%esi
+	pxor	%mm6,%mm0
+	popl	%ebx
+	pxor	%mm7,%mm2
+	movq	%mm0,(%eax)
+	popl	%ebp
+	movq	%mm2,8(%eax)
+	emms
+	ret
+.align	16
+.L001ialu:
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	subl	$20,%esp
+	movl	44(%esp),%eax
+	movl	52(%esp),%ebx
+	call	_mul_1x1_ialu
+	movl	%eax,8(%esp)
+	movl	%edx,12(%esp)
+	movl	48(%esp),%eax
+	movl	56(%esp),%ebx
+	call	_mul_1x1_ialu
+	movl	%eax,(%esp)
+	movl	%edx,4(%esp)
+	movl	44(%esp),%eax
+	movl	52(%esp),%ebx
+	xorl	48(%esp),%eax
+	xorl	56(%esp),%ebx
+	call	_mul_1x1_ialu
+	movl	40(%esp),%ebp
+	movl	(%esp),%ebx
+	movl	4(%esp),%ecx
+	movl	8(%esp),%edi
+	movl	12(%esp),%esi
+	xorl	%edx,%eax
+	xorl	%ecx,%edx
+	xorl	%ebx,%eax
+	movl	%ebx,(%ebp)
+	xorl	%edi,%edx
+	movl	%esi,12(%ebp)
+	xorl	%esi,%eax
+	addl	$20,%esp
+	xorl	%esi,%edx
+	popl	%edi
+	xorl	%edx,%eax
+	popl	%esi
+	movl	%edx,8(%ebp)
+	popl	%ebx
+	movl	%eax,4(%ebp)
+	popl	%ebp
+	ret
+.size	bn_GF2m_mul_2x2,.-.L_bn_GF2m_mul_2x2_begin
+.byte	71,70,40,50,94,109,41,32,77,117,108,116,105,112,108,105
+.byte	99,97,116,105,111,110,32,102,111,114,32,120,56,54,44,32
+.byte	67,82,89,80,84,79,71,65,77,83,32,98,121,32,60,97
+.byte	112,112,114,111,64,111,112,101,110,115,115,108,46,111,114,103
+.byte	62,0
+.comm	OPENSSL_ia32cap_P,16,4
+
+	.section ".note.gnu.property", "a"
+	.p2align 2
+	.long 1f - 0f
+	.long 4f - 1f
+	.long 5
+0:
+	.asciz "GNU"
+1:
+	.p2align 2
+	.long 0xc0000002
+	.long 3f - 2f
+2:
+	.long 3
+3:
+	.p2align 2
+4:

deps/openssl/config/archs/linux-elf/asm_avx2/crypto/bn/x86-mont.S

@@ -0,0 +1,497 @@
+.text
+.globl	bn_mul_mont
+.type	bn_mul_mont,@function
+.align	16
+bn_mul_mont:
+.L_bn_mul_mont_begin:
+	%ifdef __CET__
+
+.byte	243,15,30,251
+	%endif
+
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	xorl	%eax,%eax
+	movl	40(%esp),%edi
+	cmpl	$4,%edi
+	jl	.L000just_leave
+	leal	20(%esp),%esi
+	leal	24(%esp),%edx
+	addl	$2,%edi
+	negl	%edi
+	leal	-32(%esp,%edi,4),%ebp
+	negl	%edi
+	movl	%ebp,%eax
+	subl	%edx,%eax
+	andl	$2047,%eax
+	subl	%eax,%ebp
+	xorl	%ebp,%edx
+	andl	$2048,%edx
+	xorl	$2048,%edx
+	subl	%edx,%ebp
+	andl	$-64,%ebp
+	movl	%esp,%eax
+	subl	%ebp,%eax
+	andl	$-4096,%eax
+	movl	%esp,%edx
+	leal	(%ebp,%eax,1),%esp
+	movl	(%esp),%eax
+	cmpl	%ebp,%esp
+	ja	.L001page_walk
+	jmp	.L002page_walk_done
+.align	16
+.L001page_walk:
+	leal	-4096(%esp),%esp
+	movl	(%esp),%eax
+	cmpl	%ebp,%esp
+	ja	.L001page_walk
+.L002page_walk_done:
+	movl	(%esi),%eax
+	movl	4(%esi),%ebx
+	movl	8(%esi),%ecx
+	movl	12(%esi),%ebp
+	movl	16(%esi),%esi
+	movl	(%esi),%esi
+	movl	%eax,4(%esp)
+	movl	%ebx,8(%esp)
+	movl	%ecx,12(%esp)
+	movl	%ebp,16(%esp)
+	movl	%esi,20(%esp)
+	leal	-3(%edi),%ebx
+	movl	%edx,24(%esp)
+	call	.L003PIC_me_up
+.L003PIC_me_up:
+	popl	%eax
+	leal	OPENSSL_ia32cap_P-.L003PIC_me_up(%eax),%eax
+	btl	$26,(%eax)
+	jnc	.L004non_sse2
+	movl	$-1,%eax
+	movd	%eax,%mm7
+	movl	8(%esp),%esi
+	movl	12(%esp),%edi
+	movl	16(%esp),%ebp
+	xorl	%edx,%edx
+	xorl	%ecx,%ecx
+	movd	(%edi),%mm4
+	movd	(%esi),%mm5
+	movd	(%ebp),%mm3
+	pmuludq	%mm4,%mm5
+	movq	%mm5,%mm2
+	movq	%mm5,%mm0
+	pand	%mm7,%mm0
+	pmuludq	20(%esp),%mm5
+	pmuludq	%mm5,%mm3
+	paddq	%mm0,%mm3
+	movd	4(%ebp),%mm1
+	movd	4(%esi),%mm0
+	psrlq	$32,%mm2
+	psrlq	$32,%mm3
+	incl	%ecx
+.align	16
+.L0051st:
+	pmuludq	%mm4,%mm0
+	pmuludq	%mm5,%mm1
+	paddq	%mm0,%mm2
+	paddq	%mm1,%mm3
+	movq	%mm2,%mm0
+	pand	%mm7,%mm0
+	movd	4(%ebp,%ecx,4),%mm1
+	paddq	%mm0,%mm3
+	movd	4(%esi,%ecx,4),%mm0
+	psrlq	$32,%mm2
+	movd	%mm3,28(%esp,%ecx,4)
+	psrlq	$32,%mm3
+	leal	1(%ecx),%ecx
+	cmpl	%ebx,%ecx
+	jl	.L0051st
+	pmuludq	%mm4,%mm0
+	pmuludq	%mm5,%mm1
+	paddq	%mm0,%mm2
+	paddq	%mm1,%mm3
+	movq	%mm2,%mm0
+	pand	%mm7,%mm0
+	paddq	%mm0,%mm3
+	movd	%mm3,28(%esp,%ecx,4)
+	psrlq	$32,%mm2
+	psrlq	$32,%mm3
+	paddq	%mm2,%mm3
+	movq	%mm3,32(%esp,%ebx,4)
+	incl	%edx
+.L006outer:
+	xorl	%ecx,%ecx
+	movd	(%edi,%edx,4),%mm4
+	movd	(%esi),%mm5
+	movd	32(%esp),%mm6
+	movd	(%ebp),%mm3
+	pmuludq	%mm4,%mm5
+	paddq	%mm6,%mm5
+	movq	%mm5,%mm0
+	movq	%mm5,%mm2
+	pand	%mm7,%mm0
+	pmuludq	20(%esp),%mm5
+	pmuludq	%mm5,%mm3
+	paddq	%mm0,%mm3
+	movd	36(%esp),%mm6
+	movd	4(%ebp),%mm1
+	movd	4(%esi),%mm0
+	psrlq	$32,%mm2
+	psrlq	$32,%mm3
+	paddq	%mm6,%mm2
+	incl	%ecx
+	decl	%ebx
+.L007inner:
+	pmuludq	%mm4,%mm0
+	pmuludq	%mm5,%mm1
+	paddq	%mm0,%mm2
+	paddq	%mm1,%mm3
+	movq	%mm2,%mm0
+	movd	36(%esp,%ecx,4),%mm6
+	pand	%mm7,%mm0
+	movd	4(%ebp,%ecx,4),%mm1
+	paddq	%mm0,%mm3
+	movd	4(%esi,%ecx,4),%mm0
+	psrlq	$32,%mm2
+	movd	%mm3,28(%esp,%ecx,4)
+	psrlq	$32,%mm3
+	paddq	%mm6,%mm2
+	decl	%ebx
+	leal	1(%ecx),%ecx
+	jnz	.L007inner
+	movl	%ecx,%ebx
+	pmuludq	%mm4,%mm0
+	pmuludq	%mm5,%mm1
+	paddq	%mm0,%mm2
+	paddq	%mm1,%mm3
+	movq	%mm2,%mm0
+	pand	%mm7,%mm0
+	paddq	%mm0,%mm3
+	movd	%mm3,28(%esp,%ecx,4)
+	psrlq	$32,%mm2
+	psrlq	$32,%mm3
+	movd	36(%esp,%ebx,4),%mm6
+	paddq	%mm2,%mm3
+	paddq	%mm6,%mm3
+	movq	%mm3,32(%esp,%ebx,4)
+	leal	1(%edx),%edx
+	cmpl	%ebx,%edx
+	jle	.L006outer
+	emms
+	jmp	.L008common_tail
+.align	16
+.L004non_sse2:
+	movl	8(%esp),%esi
+	leal	1(%ebx),%ebp
+	movl	12(%esp),%edi
+	xorl	%ecx,%ecx
+	movl	%esi,%edx
+	andl	$1,%ebp
+	subl	%edi,%edx
+	leal	4(%edi,%ebx,4),%eax
+	orl	%edx,%ebp
+	movl	(%edi),%edi
+	jz	.L009bn_sqr_mont
+	movl	%eax,28(%esp)
+	movl	(%esi),%eax
+	xorl	%edx,%edx
+.align	16
+.L010mull:
+	movl	%edx,%ebp
+	mull	%edi
+	addl	%eax,%ebp
+	leal	1(%ecx),%ecx
+	adcl	$0,%edx
+	movl	(%esi,%ecx,4),%eax
+	cmpl	%ebx,%ecx
+	movl	%ebp,28(%esp,%ecx,4)
+	jl	.L010mull
+	movl	%edx,%ebp
+	mull	%edi
+	movl	20(%esp),%edi
+	addl	%ebp,%eax
+	movl	16(%esp),%esi
+	adcl	$0,%edx
+	imull	32(%esp),%edi
+	movl	%eax,32(%esp,%ebx,4)
+	xorl	%ecx,%ecx
+	movl	%edx,36(%esp,%ebx,4)
+	movl	%ecx,40(%esp,%ebx,4)
+	movl	(%esi),%eax
+	mull	%edi
+	addl	32(%esp),%eax
+	movl	4(%esi),%eax
+	adcl	$0,%edx
+	incl	%ecx
+	jmp	.L0112ndmadd
+.align	16
+.L0121stmadd:
+	movl	%edx,%ebp
+	mull	%edi
+	addl	32(%esp,%ecx,4),%ebp
+	leal	1(%ecx),%ecx
+	adcl	$0,%edx
+	addl	%eax,%ebp
+	movl	(%esi,%ecx,4),%eax
+	adcl	$0,%edx
+	cmpl	%ebx,%ecx
+	movl	%ebp,28(%esp,%ecx,4)
+	jl	.L0121stmadd
+	movl	%edx,%ebp
+	mull	%edi
+	addl	32(%esp,%ebx,4),%eax
+	movl	20(%esp),%edi
+	adcl	$0,%edx
+	movl	16(%esp),%esi
+	addl	%eax,%ebp
+	adcl	$0,%edx
+	imull	32(%esp),%edi
+	xorl	%ecx,%ecx
+	addl	36(%esp,%ebx,4),%edx
+	movl	%ebp,32(%esp,%ebx,4)
+	adcl	$0,%ecx
+	movl	(%esi),%eax
+	movl	%edx,36(%esp,%ebx,4)
+	movl	%ecx,40(%esp,%ebx,4)
+	mull	%edi
+	addl	32(%esp),%eax
+	movl	4(%esi),%eax
+	adcl	$0,%edx
+	movl	$1,%ecx
+.align	16
+.L0112ndmadd:
+	movl	%edx,%ebp
+	mull	%edi
+	addl	32(%esp,%ecx,4),%ebp
+	leal	1(%ecx),%ecx
+	adcl	$0,%edx
+	addl	%eax,%ebp
+	movl	(%esi,%ecx,4),%eax
+	adcl	$0,%edx
+	cmpl	%ebx,%ecx
+	movl	%ebp,24(%esp,%ecx,4)
+	jl	.L0112ndmadd
+	movl	%edx,%ebp
+	mull	%edi
+	addl	32(%esp,%ebx,4),%ebp
+	adcl	$0,%edx
+	addl	%eax,%ebp
+	adcl	$0,%edx
+	movl	%ebp,28(%esp,%ebx,4)
+	xorl	%eax,%eax
+	movl	12(%esp),%ecx
+	addl	36(%esp,%ebx,4),%edx
+	adcl	40(%esp,%ebx,4),%eax
+	leal	4(%ecx),%ecx
+	movl	%edx,32(%esp,%ebx,4)
+	cmpl	28(%esp),%ecx
+	movl	%eax,36(%esp,%ebx,4)
+	je	.L008common_tail
+	movl	(%ecx),%edi
+	movl	8(%esp),%esi
+	movl	%ecx,12(%esp)
+	xorl	%ecx,%ecx
+	xorl	%edx,%edx
+	movl	(%esi),%eax
+	jmp	.L0121stmadd
+.align	16
+.L009bn_sqr_mont:
+	movl	%ebx,(%esp)
+	movl	%ecx,12(%esp)
+	movl	%edi,%eax
+	mull	%edi
+	movl	%eax,32(%esp)
+	movl	%edx,%ebx
+	shrl	$1,%edx
+	andl	$1,%ebx
+	incl	%ecx
+.align	16
+.L013sqr:
+	movl	(%esi,%ecx,4),%eax
+	movl	%edx,%ebp
+	mull	%edi
+	addl	%ebp,%eax
+	leal	1(%ecx),%ecx
+	adcl	$0,%edx
+	leal	(%ebx,%eax,2),%ebp
+	shrl	$31,%eax
+	cmpl	(%esp),%ecx
+	movl	%eax,%ebx
+	movl	%ebp,28(%esp,%ecx,4)
+	jl	.L013sqr
+	movl	(%esi,%ecx,4),%eax
+	movl	%edx,%ebp
+	mull	%edi
+	addl	%ebp,%eax
+	movl	20(%esp),%edi
+	adcl	$0,%edx
+	movl	16(%esp),%esi
+	leal	(%ebx,%eax,2),%ebp
+	imull	32(%esp),%edi
+	shrl	$31,%eax
+	movl	%ebp,32(%esp,%ecx,4)
+	leal	(%eax,%edx,2),%ebp
+	movl	(%esi),%eax
+	shrl	$31,%edx
+	movl	%ebp,36(%esp,%ecx,4)
+	movl	%edx,40(%esp,%ecx,4)
+	mull	%edi
+	addl	32(%esp),%eax
+	movl	%ecx,%ebx
+	adcl	$0,%edx
+	movl	4(%esi),%eax
+	movl	$1,%ecx
+.align	16
+.L0143rdmadd:
+	movl	%edx,%ebp
+	mull	%edi
+	addl	32(%esp,%ecx,4),%ebp
+	adcl	$0,%edx
+	addl	%eax,%ebp
+	movl	4(%esi,%ecx,4),%eax
+	adcl	$0,%edx
+	movl	%ebp,28(%esp,%ecx,4)
+	movl	%edx,%ebp
+	mull	%edi
+	addl	36(%esp,%ecx,4),%ebp
+	leal	2(%ecx),%ecx
+	adcl	$0,%edx
+	addl	%eax,%ebp
+	movl	(%esi,%ecx,4),%eax
+	adcl	$0,%edx
+	cmpl	%ebx,%ecx
+	movl	%ebp,24(%esp,%ecx,4)
+	jl	.L0143rdmadd
+	movl	%edx,%ebp
+	mull	%edi
+	addl	32(%esp,%ebx,4),%ebp
+	adcl	$0,%edx
+	addl	%eax,%ebp
+	adcl	$0,%edx
+	movl	%ebp,28(%esp,%ebx,4)
+	movl	12(%esp),%ecx
+	xorl	%eax,%eax
+	movl	8(%esp),%esi
+	addl	36(%esp,%ebx,4),%edx
+	adcl	40(%esp,%ebx,4),%eax
+	movl	%edx,32(%esp,%ebx,4)
+	cmpl	%ebx,%ecx
+	movl	%eax,36(%esp,%ebx,4)
+	je	.L008common_tail
+	movl	4(%esi,%ecx,4),%edi
+	leal	1(%ecx),%ecx
+	movl	%edi,%eax
+	movl	%ecx,12(%esp)
+	mull	%edi
+	addl	32(%esp,%ecx,4),%eax
+	adcl	$0,%edx
+	movl	%eax,32(%esp,%ecx,4)
+	xorl	%ebp,%ebp
+	cmpl	%ebx,%ecx
+	leal	1(%ecx),%ecx
+	je	.L015sqrlast
+	movl	%edx,%ebx
+	shrl	$1,%edx
+	andl	$1,%ebx
+.align	16
+.L016sqradd:
+	movl	(%esi,%ecx,4),%eax
+	movl	%edx,%ebp
+	mull	%edi
+	addl	%ebp,%eax
+	leal	(%eax,%eax,1),%ebp
+	adcl	$0,%edx
+	shrl	$31,%eax
+	addl	32(%esp,%ecx,4),%ebp
+	leal	1(%ecx),%ecx
+	adcl	$0,%eax
+	addl	%ebx,%ebp
+	adcl	$0,%eax
+	cmpl	(%esp),%ecx
+	movl	%ebp,28(%esp,%ecx,4)
+	movl	%eax,%ebx
+	jle	.L016sqradd
+	movl	%edx,%ebp
+	addl	%edx,%edx
+	shrl	$31,%ebp
+	addl	%ebx,%edx
+	adcl	$0,%ebp
+.L015sqrlast:
+	movl	20(%esp),%edi
+	movl	16(%esp),%esi
+	imull	32(%esp),%edi
+	addl	32(%esp,%ecx,4),%edx
+	movl	(%esi),%eax
+	adcl	$0,%ebp
+	movl	%edx,32(%esp,%ecx,4)
+	movl	%ebp,36(%esp,%ecx,4)
+	mull	%edi
+	addl	32(%esp),%eax
+	leal	-1(%ecx),%ebx
+	adcl	$0,%edx
+	movl	$1,%ecx
+	movl	4(%esi),%eax
+	jmp	.L0143rdmadd
+.align	16
+.L008common_tail:
+	movl	16(%esp),%ebp
+	movl	4(%esp),%edi
+	leal	32(%esp),%esi
+	movl	(%esi),%eax
+	movl	%ebx,%ecx
+	xorl	%edx,%edx
+.align	16
+.L017sub:
+	sbbl	(%ebp,%edx,4),%eax
+	movl	%eax,(%edi,%edx,4)
+	decl	%ecx
+	movl	4(%esi,%edx,4),%eax
+	leal	1(%edx),%edx
+	jge	.L017sub
+	sbbl	$0,%eax
+	movl	$-1,%edx
+	xorl	%eax,%edx
+	jmp	.L018copy
+.align	16
+.L018copy:
+	movl	32(%esp,%ebx,4),%esi
+	movl	(%edi,%ebx,4),%ebp
+	movl	%ecx,32(%esp,%ebx,4)
+	andl	%eax,%esi
+	andl	%edx,%ebp
+	orl	%esi,%ebp
+	movl	%ebp,(%edi,%ebx,4)
+	decl	%ebx
+	jge	.L018copy
+	movl	24(%esp),%esp
+	movl	$1,%eax
+.L000just_leave:
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	popl	%ebp
+	ret
+.size	bn_mul_mont,.-.L_bn_mul_mont_begin
+.byte	77,111,110,116,103,111,109,101,114,121,32,77,117,108,116,105
+.byte	112,108,105,99,97,116,105,111,110,32,102,111,114,32,120,56
+.byte	54,44,32,67,82,89,80,84,79,71,65,77,83,32,98,121
+.byte	32,60,97,112,112,114,111,64,111,112,101,110,115,115,108,46
+.byte	111,114,103,62,0
+.comm	OPENSSL_ia32cap_P,16,4
+
+	.section ".note.gnu.property", "a"
+	.p2align 2
+	.long 1f - 0f
+	.long 4f - 1f
+	.long 5
+0:
+	.asciz "GNU"
+1:
+	.p2align 2
+	.long 0xc0000002
+	.long 3f - 2f
+2:
+	.long 3
+3:
+	.p2align 2
+4: