summaryrefslogtreecommitdiff
path: root/main/openssl/crypto/bn/asm
diff options
context:
space:
mode:
Diffstat (limited to 'main/openssl/crypto/bn/asm')
-rw-r--r--main/openssl/crypto/bn/asm/bn-586.S441
-rw-r--r--main/openssl/crypto/bn/asm/x86-gf2m.S12
-rw-r--r--main/openssl/crypto/bn/asm/x86-mont.S174
3 files changed, 455 insertions, 172 deletions
diff --git a/main/openssl/crypto/bn/asm/bn-586.S b/main/openssl/crypto/bn/asm/bn-586.S
index fe873ce9..66695e26 100644
--- a/main/openssl/crypto/bn/asm/bn-586.S
+++ b/main/openssl/crypto/bn/asm/bn-586.S
@@ -5,6 +5,103 @@
.align 16
bn_mul_add_words:
.L_bn_mul_add_words_begin:
+ call .L000PIC_me_up
+.L000PIC_me_up:
+ popl %eax
+ leal _GLOBAL_OFFSET_TABLE_+[.-.L000PIC_me_up](%eax),%eax
+ movl OPENSSL_ia32cap_P@GOT(%eax),%eax
+ btl $26,(%eax)
+ jnc .L001maw_non_sse2
+ movl 4(%esp),%eax
+ movl 8(%esp),%edx
+ movl 12(%esp),%ecx
+ movd 16(%esp),%mm0
+ pxor %mm1,%mm1
+ jmp .L002maw_sse2_entry
+.align 16
+.L003maw_sse2_unrolled:
+ movd (%eax),%mm3
+ paddq %mm3,%mm1
+ movd (%edx),%mm2
+ pmuludq %mm0,%mm2
+ movd 4(%edx),%mm4
+ pmuludq %mm0,%mm4
+ movd 8(%edx),%mm6
+ pmuludq %mm0,%mm6
+ movd 12(%edx),%mm7
+ pmuludq %mm0,%mm7
+ paddq %mm2,%mm1
+ movd 4(%eax),%mm3
+ paddq %mm4,%mm3
+ movd 8(%eax),%mm5
+ paddq %mm6,%mm5
+ movd 12(%eax),%mm4
+ paddq %mm4,%mm7
+ movd %mm1,(%eax)
+ movd 16(%edx),%mm2
+ pmuludq %mm0,%mm2
+ psrlq $32,%mm1
+ movd 20(%edx),%mm4
+ pmuludq %mm0,%mm4
+ paddq %mm3,%mm1
+ movd 24(%edx),%mm6
+ pmuludq %mm0,%mm6
+ movd %mm1,4(%eax)
+ psrlq $32,%mm1
+ movd 28(%edx),%mm3
+ addl $32,%edx
+ pmuludq %mm0,%mm3
+ paddq %mm5,%mm1
+ movd 16(%eax),%mm5
+ paddq %mm5,%mm2
+ movd %mm1,8(%eax)
+ psrlq $32,%mm1
+ paddq %mm7,%mm1
+ movd 20(%eax),%mm5
+ paddq %mm5,%mm4
+ movd %mm1,12(%eax)
+ psrlq $32,%mm1
+ paddq %mm2,%mm1
+ movd 24(%eax),%mm5
+ paddq %mm5,%mm6
+ movd %mm1,16(%eax)
+ psrlq $32,%mm1
+ paddq %mm4,%mm1
+ movd 28(%eax),%mm5
+ paddq %mm5,%mm3
+ movd %mm1,20(%eax)
+ psrlq $32,%mm1
+ paddq %mm6,%mm1
+ movd %mm1,24(%eax)
+ psrlq $32,%mm1
+ paddq %mm3,%mm1
+ movd %mm1,28(%eax)
+ leal 32(%eax),%eax
+ psrlq $32,%mm1
+ subl $8,%ecx
+ jz .L004maw_sse2_exit
+.L002maw_sse2_entry:
+ testl $4294967288,%ecx
+ jnz .L003maw_sse2_unrolled
+.align 4
+.L005maw_sse2_loop:
+ movd (%edx),%mm2
+ movd (%eax),%mm3
+ pmuludq %mm0,%mm2
+ leal 4(%edx),%edx
+ paddq %mm3,%mm1
+ paddq %mm2,%mm1
+ movd %mm1,(%eax)
+ subl $1,%ecx
+ psrlq $32,%mm1
+ leal 4(%eax),%eax
+ jnz .L005maw_sse2_loop
+.L004maw_sse2_exit:
+ movd %mm1,%eax
+ emms
+ ret
+.align 16
+.L001maw_non_sse2:
pushl %ebp
pushl %ebx
pushl %esi
@@ -17,9 +114,9 @@ bn_mul_add_words:
andl $4294967288,%ecx
movl 32(%esp),%ebp
pushl %ecx
- jz .L000maw_finish
+ jz .L006maw_finish
.align 16
-.L001maw_loop:
+.L007maw_loop:
movl (%ebx),%eax
mull %ebp
@@ -96,13 +193,13 @@ bn_mul_add_words:
subl $8,%ecx
leal 32(%ebx),%ebx
leal 32(%edi),%edi
- jnz .L001maw_loop
-.L000maw_finish:
+ jnz .L007maw_loop
+.L006maw_finish:
movl 32(%esp),%ecx
andl $7,%ecx
- jnz .L002maw_finish2
- jmp .L003maw_end
-.L002maw_finish2:
+ jnz .L008maw_finish2
+ jmp .L009maw_end
+.L008maw_finish2:
movl (%ebx),%eax
mull %ebp
@@ -113,7 +210,7 @@ bn_mul_add_words:
decl %ecx
movl %eax,(%edi)
movl %edx,%esi
- jz .L003maw_end
+ jz .L009maw_end
movl 4(%ebx),%eax
mull %ebp
@@ -124,7 +221,7 @@ bn_mul_add_words:
decl %ecx
movl %eax,4(%edi)
movl %edx,%esi
- jz .L003maw_end
+ jz .L009maw_end
movl 8(%ebx),%eax
mull %ebp
@@ -135,7 +232,7 @@ bn_mul_add_words:
decl %ecx
movl %eax,8(%edi)
movl %edx,%esi
- jz .L003maw_end
+ jz .L009maw_end
movl 12(%ebx),%eax
mull %ebp
@@ -146,7 +243,7 @@ bn_mul_add_words:
decl %ecx
movl %eax,12(%edi)
movl %edx,%esi
- jz .L003maw_end
+ jz .L009maw_end
movl 16(%ebx),%eax
mull %ebp
@@ -157,7 +254,7 @@ bn_mul_add_words:
decl %ecx
movl %eax,16(%edi)
movl %edx,%esi
- jz .L003maw_end
+ jz .L009maw_end
movl 20(%ebx),%eax
mull %ebp
@@ -168,7 +265,7 @@ bn_mul_add_words:
decl %ecx
movl %eax,20(%edi)
movl %edx,%esi
- jz .L003maw_end
+ jz .L009maw_end
movl 24(%ebx),%eax
mull %ebp
@@ -178,7 +275,7 @@ bn_mul_add_words:
adcl $0,%edx
movl %eax,24(%edi)
movl %edx,%esi
-.L003maw_end:
+.L009maw_end:
movl %esi,%eax
popl %ecx
popl %edi
@@ -192,6 +289,34 @@ bn_mul_add_words:
.align 16
bn_mul_words:
.L_bn_mul_words_begin:
+ call .L010PIC_me_up
+.L010PIC_me_up:
+ popl %eax
+ leal _GLOBAL_OFFSET_TABLE_+[.-.L010PIC_me_up](%eax),%eax
+ movl OPENSSL_ia32cap_P@GOT(%eax),%eax
+ btl $26,(%eax)
+ jnc .L011mw_non_sse2
+ movl 4(%esp),%eax
+ movl 8(%esp),%edx
+ movl 12(%esp),%ecx
+ movd 16(%esp),%mm0
+ pxor %mm1,%mm1
+.align 16
+.L012mw_sse2_loop:
+ movd (%edx),%mm2
+ pmuludq %mm0,%mm2
+ leal 4(%edx),%edx
+ paddq %mm2,%mm1
+ movd %mm1,(%eax)
+ subl $1,%ecx
+ psrlq $32,%mm1
+ leal 4(%eax),%eax
+ jnz .L012mw_sse2_loop
+ movd %mm1,%eax
+ emms
+ ret
+.align 16
+.L011mw_non_sse2:
pushl %ebp
pushl %ebx
pushl %esi
@@ -203,8 +328,8 @@ bn_mul_words:
movl 28(%esp),%ebp
movl 32(%esp),%ecx
andl $4294967288,%ebp
- jz .L004mw_finish
-.L005mw_loop:
+ jz .L013mw_finish
+.L014mw_loop:
movl (%ebx),%eax
mull %ecx
@@ -265,14 +390,14 @@ bn_mul_words:
addl $32,%ebx
addl $32,%edi
subl $8,%ebp
- jz .L004mw_finish
- jmp .L005mw_loop
-.L004mw_finish:
+ jz .L013mw_finish
+ jmp .L014mw_loop
+.L013mw_finish:
movl 28(%esp),%ebp
andl $7,%ebp
- jnz .L006mw_finish2
- jmp .L007mw_end
-.L006mw_finish2:
+ jnz .L015mw_finish2
+ jmp .L016mw_end
+.L015mw_finish2:
movl (%ebx),%eax
mull %ecx
@@ -281,7 +406,7 @@ bn_mul_words:
movl %eax,(%edi)
movl %edx,%esi
decl %ebp
- jz .L007mw_end
+ jz .L016mw_end
movl 4(%ebx),%eax
mull %ecx
@@ -290,7 +415,7 @@ bn_mul_words:
movl %eax,4(%edi)
movl %edx,%esi
decl %ebp
- jz .L007mw_end
+ jz .L016mw_end
movl 8(%ebx),%eax
mull %ecx
@@ -299,7 +424,7 @@ bn_mul_words:
movl %eax,8(%edi)
movl %edx,%esi
decl %ebp
- jz .L007mw_end
+ jz .L016mw_end
movl 12(%ebx),%eax
mull %ecx
@@ -308,7 +433,7 @@ bn_mul_words:
movl %eax,12(%edi)
movl %edx,%esi
decl %ebp
- jz .L007mw_end
+ jz .L016mw_end
movl 16(%ebx),%eax
mull %ecx
@@ -317,7 +442,7 @@ bn_mul_words:
movl %eax,16(%edi)
movl %edx,%esi
decl %ebp
- jz .L007mw_end
+ jz .L016mw_end
movl 20(%ebx),%eax
mull %ecx
@@ -326,7 +451,7 @@ bn_mul_words:
movl %eax,20(%edi)
movl %edx,%esi
decl %ebp
- jz .L007mw_end
+ jz .L016mw_end
movl 24(%ebx),%eax
mull %ecx
@@ -334,7 +459,7 @@ bn_mul_words:
adcl $0,%edx
movl %eax,24(%edi)
movl %edx,%esi
-.L007mw_end:
+.L016mw_end:
movl %esi,%eax
popl %edi
popl %esi
@@ -347,6 +472,29 @@ bn_mul_words:
.align 16
bn_sqr_words:
.L_bn_sqr_words_begin:
+ call .L017PIC_me_up
+.L017PIC_me_up:
+ popl %eax
+ leal _GLOBAL_OFFSET_TABLE_+[.-.L017PIC_me_up](%eax),%eax
+ movl OPENSSL_ia32cap_P@GOT(%eax),%eax
+ btl $26,(%eax)
+ jnc .L018sqr_non_sse2
+ movl 4(%esp),%eax
+ movl 8(%esp),%edx
+ movl 12(%esp),%ecx
+.align 16
+.L019sqr_sse2_loop:
+ movd (%edx),%mm0
+ pmuludq %mm0,%mm0
+ leal 4(%edx),%edx
+ movq %mm0,(%eax)
+ subl $1,%ecx
+ leal 8(%eax),%eax
+ jnz .L019sqr_sse2_loop
+ emms
+ ret
+.align 16
+.L018sqr_non_sse2:
pushl %ebp
pushl %ebx
pushl %esi
@@ -356,8 +504,8 @@ bn_sqr_words:
movl 24(%esp),%edi
movl 28(%esp),%ebx
andl $4294967288,%ebx
- jz .L008sw_finish
-.L009sw_loop:
+ jz .L020sw_finish
+.L021sw_loop:
movl (%edi),%eax
mull %eax
@@ -402,59 +550,59 @@ bn_sqr_words:
addl $32,%edi
addl $64,%esi
subl $8,%ebx
- jnz .L009sw_loop
-.L008sw_finish:
+ jnz .L021sw_loop
+.L020sw_finish:
movl 28(%esp),%ebx
andl $7,%ebx
- jz .L010sw_end
+ jz .L022sw_end
movl (%edi),%eax
mull %eax
movl %eax,(%esi)
decl %ebx
movl %edx,4(%esi)
- jz .L010sw_end
+ jz .L022sw_end
movl 4(%edi),%eax
mull %eax
movl %eax,8(%esi)
decl %ebx
movl %edx,12(%esi)
- jz .L010sw_end
+ jz .L022sw_end
movl 8(%edi),%eax
mull %eax
movl %eax,16(%esi)
decl %ebx
movl %edx,20(%esi)
- jz .L010sw_end
+ jz .L022sw_end
movl 12(%edi),%eax
mull %eax
movl %eax,24(%esi)
decl %ebx
movl %edx,28(%esi)
- jz .L010sw_end
+ jz .L022sw_end
movl 16(%edi),%eax
mull %eax
movl %eax,32(%esi)
decl %ebx
movl %edx,36(%esi)
- jz .L010sw_end
+ jz .L022sw_end
movl 20(%edi),%eax
mull %eax
movl %eax,40(%esi)
decl %ebx
movl %edx,44(%esi)
- jz .L010sw_end
+ jz .L022sw_end
movl 24(%edi),%eax
mull %eax
movl %eax,48(%esi)
movl %edx,52(%esi)
-.L010sw_end:
+.L022sw_end:
popl %edi
popl %esi
popl %ebx
@@ -488,8 +636,8 @@ bn_add_words:
movl 32(%esp),%ebp
xorl %eax,%eax
andl $4294967288,%ebp
- jz .L011aw_finish
-.L012aw_loop:
+ jz .L023aw_finish
+.L024aw_loop:
movl (%esi),%ecx
movl (%edi),%edx
@@ -567,11 +715,11 @@ bn_add_words:
addl $32,%edi
addl $32,%ebx
subl $8,%ebp
- jnz .L012aw_loop
-.L011aw_finish:
+ jnz .L024aw_loop
+.L023aw_finish:
movl 32(%esp),%ebp
andl $7,%ebp
- jz .L013aw_end
+ jz .L025aw_end
movl (%esi),%ecx
movl (%edi),%edx
@@ -582,7 +730,7 @@ bn_add_words:
adcl $0,%eax
decl %ebp
movl %ecx,(%ebx)
- jz .L013aw_end
+ jz .L025aw_end
movl 4(%esi),%ecx
movl 4(%edi),%edx
@@ -593,7 +741,7 @@ bn_add_words:
adcl $0,%eax
decl %ebp
movl %ecx,4(%ebx)
- jz .L013aw_end
+ jz .L025aw_end
movl 8(%esi),%ecx
movl 8(%edi),%edx
@@ -604,7 +752,7 @@ bn_add_words:
adcl $0,%eax
decl %ebp
movl %ecx,8(%ebx)
- jz .L013aw_end
+ jz .L025aw_end
movl 12(%esi),%ecx
movl 12(%edi),%edx
@@ -615,7 +763,7 @@ bn_add_words:
adcl $0,%eax
decl %ebp
movl %ecx,12(%ebx)
- jz .L013aw_end
+ jz .L025aw_end
movl 16(%esi),%ecx
movl 16(%edi),%edx
@@ -626,7 +774,7 @@ bn_add_words:
adcl $0,%eax
decl %ebp
movl %ecx,16(%ebx)
- jz .L013aw_end
+ jz .L025aw_end
movl 20(%esi),%ecx
movl 20(%edi),%edx
@@ -637,7 +785,7 @@ bn_add_words:
adcl $0,%eax
decl %ebp
movl %ecx,20(%ebx)
- jz .L013aw_end
+ jz .L025aw_end
movl 24(%esi),%ecx
movl 24(%edi),%edx
@@ -647,7 +795,7 @@ bn_add_words:
addl %edx,%ecx
adcl $0,%eax
movl %ecx,24(%ebx)
-.L013aw_end:
+.L025aw_end:
popl %edi
popl %esi
popl %ebx
@@ -670,8 +818,8 @@ bn_sub_words:
movl 32(%esp),%ebp
xorl %eax,%eax
andl $4294967288,%ebp
- jz .L014aw_finish
-.L015aw_loop:
+ jz .L026aw_finish
+.L027aw_loop:
movl (%esi),%ecx
movl (%edi),%edx
@@ -749,11 +897,11 @@ bn_sub_words:
addl $32,%edi
addl $32,%ebx
subl $8,%ebp
- jnz .L015aw_loop
-.L014aw_finish:
+ jnz .L027aw_loop
+.L026aw_finish:
movl 32(%esp),%ebp
andl $7,%ebp
- jz .L016aw_end
+ jz .L028aw_end
movl (%esi),%ecx
movl (%edi),%edx
@@ -764,7 +912,7 @@ bn_sub_words:
adcl $0,%eax
decl %ebp
movl %ecx,(%ebx)
- jz .L016aw_end
+ jz .L028aw_end
movl 4(%esi),%ecx
movl 4(%edi),%edx
@@ -775,7 +923,7 @@ bn_sub_words:
adcl $0,%eax
decl %ebp
movl %ecx,4(%ebx)
- jz .L016aw_end
+ jz .L028aw_end
movl 8(%esi),%ecx
movl 8(%edi),%edx
@@ -786,7 +934,7 @@ bn_sub_words:
adcl $0,%eax
decl %ebp
movl %ecx,8(%ebx)
- jz .L016aw_end
+ jz .L028aw_end
movl 12(%esi),%ecx
movl 12(%edi),%edx
@@ -797,7 +945,7 @@ bn_sub_words:
adcl $0,%eax
decl %ebp
movl %ecx,12(%ebx)
- jz .L016aw_end
+ jz .L028aw_end
movl 16(%esi),%ecx
movl 16(%edi),%edx
@@ -808,7 +956,7 @@ bn_sub_words:
adcl $0,%eax
decl %ebp
movl %ecx,16(%ebx)
- jz .L016aw_end
+ jz .L028aw_end
movl 20(%esi),%ecx
movl 20(%edi),%edx
@@ -819,7 +967,7 @@ bn_sub_words:
adcl $0,%eax
decl %ebp
movl %ecx,20(%ebx)
- jz .L016aw_end
+ jz .L028aw_end
movl 24(%esi),%ecx
movl 24(%edi),%edx
@@ -829,7 +977,7 @@ bn_sub_words:
subl %edx,%ecx
adcl $0,%eax
movl %ecx,24(%ebx)
-.L016aw_end:
+.L028aw_end:
popl %edi
popl %esi
popl %ebx
@@ -852,8 +1000,8 @@ bn_sub_part_words:
movl 32(%esp),%ebp
xorl %eax,%eax
andl $4294967288,%ebp
- jz .L017aw_finish
-.L018aw_loop:
+ jz .L029aw_finish
+.L030aw_loop:
movl (%esi),%ecx
movl (%edi),%edx
@@ -931,11 +1079,11 @@ bn_sub_part_words:
addl $32,%edi
addl $32,%ebx
subl $8,%ebp
- jnz .L018aw_loop
-.L017aw_finish:
+ jnz .L030aw_loop
+.L029aw_finish:
movl 32(%esp),%ebp
andl $7,%ebp
- jz .L019aw_end
+ jz .L031aw_end
movl (%esi),%ecx
movl (%edi),%edx
@@ -949,7 +1097,7 @@ bn_sub_part_words:
addl $4,%edi
addl $4,%ebx
decl %ebp
- jz .L019aw_end
+ jz .L031aw_end
movl (%esi),%ecx
movl (%edi),%edx
@@ -963,7 +1111,7 @@ bn_sub_part_words:
addl $4,%edi
addl $4,%ebx
decl %ebp
- jz .L019aw_end
+ jz .L031aw_end
movl (%esi),%ecx
movl (%edi),%edx
@@ -977,7 +1125,7 @@ bn_sub_part_words:
addl $4,%edi
addl $4,%ebx
decl %ebp
- jz .L019aw_end
+ jz .L031aw_end
movl (%esi),%ecx
movl (%edi),%edx
@@ -991,7 +1139,7 @@ bn_sub_part_words:
addl $4,%edi
addl $4,%ebx
decl %ebp
- jz .L019aw_end
+ jz .L031aw_end
movl (%esi),%ecx
movl (%edi),%edx
@@ -1005,7 +1153,7 @@ bn_sub_part_words:
addl $4,%edi
addl $4,%ebx
decl %ebp
- jz .L019aw_end
+ jz .L031aw_end
movl (%esi),%ecx
movl (%edi),%edx
@@ -1019,7 +1167,7 @@ bn_sub_part_words:
addl $4,%edi
addl $4,%ebx
decl %ebp
- jz .L019aw_end
+ jz .L031aw_end
movl (%esi),%ecx
movl (%edi),%edx
@@ -1032,20 +1180,20 @@ bn_sub_part_words:
addl $4,%esi
addl $4,%edi
addl $4,%ebx
-.L019aw_end:
+.L031aw_end:
cmpl $0,36(%esp)
- je .L020pw_end
+ je .L032pw_end
movl 36(%esp),%ebp
cmpl $0,%ebp
- je .L020pw_end
- jge .L021pw_pos
+ je .L032pw_end
+ jge .L033pw_pos
movl $0,%edx
subl %ebp,%edx
movl %edx,%ebp
andl $4294967288,%ebp
- jz .L022pw_neg_finish
-.L023pw_neg_loop:
+ jz .L034pw_neg_finish
+.L035pw_neg_loop:
movl $0,%ecx
movl (%edi),%edx
@@ -1122,13 +1270,13 @@ bn_sub_part_words:
addl $32,%edi
addl $32,%ebx
subl $8,%ebp
- jnz .L023pw_neg_loop
-.L022pw_neg_finish:
+ jnz .L035pw_neg_loop
+.L034pw_neg_finish:
movl 36(%esp),%edx
movl $0,%ebp
subl %edx,%ebp
andl $7,%ebp
- jz .L020pw_end
+ jz .L032pw_end
movl $0,%ecx
movl (%edi),%edx
@@ -1139,7 +1287,7 @@ bn_sub_part_words:
adcl $0,%eax
decl %ebp
movl %ecx,(%ebx)
- jz .L020pw_end
+ jz .L032pw_end
movl $0,%ecx
movl 4(%edi),%edx
@@ -1150,7 +1298,7 @@ bn_sub_part_words:
adcl $0,%eax
decl %ebp
movl %ecx,4(%ebx)
- jz .L020pw_end
+ jz .L032pw_end
movl $0,%ecx
movl 8(%edi),%edx
@@ -1161,7 +1309,7 @@ bn_sub_part_words:
adcl $0,%eax
decl %ebp
movl %ecx,8(%ebx)
- jz .L020pw_end
+ jz .L032pw_end
movl $0,%ecx
movl 12(%edi),%edx
@@ -1172,7 +1320,7 @@ bn_sub_part_words:
adcl $0,%eax
decl %ebp
movl %ecx,12(%ebx)
- jz .L020pw_end
+ jz .L032pw_end
movl $0,%ecx
movl 16(%edi),%edx
@@ -1183,7 +1331,7 @@ bn_sub_part_words:
adcl $0,%eax
decl %ebp
movl %ecx,16(%ebx)
- jz .L020pw_end
+ jz .L032pw_end
movl $0,%ecx
movl 20(%edi),%edx
@@ -1194,7 +1342,7 @@ bn_sub_part_words:
adcl $0,%eax
decl %ebp
movl %ecx,20(%ebx)
- jz .L020pw_end
+ jz .L032pw_end
movl $0,%ecx
movl 24(%edi),%edx
@@ -1204,181 +1352,182 @@ bn_sub_part_words:
subl %edx,%ecx
adcl $0,%eax
movl %ecx,24(%ebx)
- jmp .L020pw_end
-.L021pw_pos:
+ jmp .L032pw_end
+.L033pw_pos:
andl $4294967288,%ebp
- jz .L024pw_pos_finish
-.L025pw_pos_loop:
+ jz .L036pw_pos_finish
+.L037pw_pos_loop:
movl (%esi),%ecx
subl %eax,%ecx
movl %ecx,(%ebx)
- jnc .L026pw_nc0
+ jnc .L038pw_nc0
movl 4(%esi),%ecx
subl %eax,%ecx
movl %ecx,4(%ebx)
- jnc .L027pw_nc1
+ jnc .L039pw_nc1
movl 8(%esi),%ecx
subl %eax,%ecx
movl %ecx,8(%ebx)
- jnc .L028pw_nc2
+ jnc .L040pw_nc2
movl 12(%esi),%ecx
subl %eax,%ecx
movl %ecx,12(%ebx)
- jnc .L029pw_nc3
+ jnc .L041pw_nc3
movl 16(%esi),%ecx
subl %eax,%ecx
movl %ecx,16(%ebx)
- jnc .L030pw_nc4
+ jnc .L042pw_nc4
movl 20(%esi),%ecx
subl %eax,%ecx
movl %ecx,20(%ebx)
- jnc .L031pw_nc5
+ jnc .L043pw_nc5
movl 24(%esi),%ecx
subl %eax,%ecx
movl %ecx,24(%ebx)
- jnc .L032pw_nc6
+ jnc .L044pw_nc6
movl 28(%esi),%ecx
subl %eax,%ecx
movl %ecx,28(%ebx)
- jnc .L033pw_nc7
+ jnc .L045pw_nc7
addl $32,%esi
addl $32,%ebx
subl $8,%ebp
- jnz .L025pw_pos_loop
-.L024pw_pos_finish:
+ jnz .L037pw_pos_loop
+.L036pw_pos_finish:
movl 36(%esp),%ebp
andl $7,%ebp
- jz .L020pw_end
+ jz .L032pw_end
movl (%esi),%ecx
subl %eax,%ecx
movl %ecx,(%ebx)
- jnc .L034pw_tail_nc0
+ jnc .L046pw_tail_nc0
decl %ebp
- jz .L020pw_end
+ jz .L032pw_end
movl 4(%esi),%ecx
subl %eax,%ecx
movl %ecx,4(%ebx)
- jnc .L035pw_tail_nc1
+ jnc .L047pw_tail_nc1
decl %ebp
- jz .L020pw_end
+ jz .L032pw_end
movl 8(%esi),%ecx
subl %eax,%ecx
movl %ecx,8(%ebx)
- jnc .L036pw_tail_nc2
+ jnc .L048pw_tail_nc2
decl %ebp
- jz .L020pw_end
+ jz .L032pw_end
movl 12(%esi),%ecx
subl %eax,%ecx
movl %ecx,12(%ebx)
- jnc .L037pw_tail_nc3
+ jnc .L049pw_tail_nc3
decl %ebp
- jz .L020pw_end
+ jz .L032pw_end
movl 16(%esi),%ecx
subl %eax,%ecx
movl %ecx,16(%ebx)
- jnc .L038pw_tail_nc4
+ jnc .L050pw_tail_nc4
decl %ebp
- jz .L020pw_end
+ jz .L032pw_end
movl 20(%esi),%ecx
subl %eax,%ecx
movl %ecx,20(%ebx)
- jnc .L039pw_tail_nc5
+ jnc .L051pw_tail_nc5
decl %ebp
- jz .L020pw_end
+ jz .L032pw_end
movl 24(%esi),%ecx
subl %eax,%ecx
movl %ecx,24(%ebx)
- jnc .L040pw_tail_nc6
+ jnc .L052pw_tail_nc6
movl $1,%eax
- jmp .L020pw_end
-.L041pw_nc_loop:
+ jmp .L032pw_end
+.L053pw_nc_loop:
movl (%esi),%ecx
movl %ecx,(%ebx)
-.L026pw_nc0:
+.L038pw_nc0:
movl 4(%esi),%ecx
movl %ecx,4(%ebx)
-.L027pw_nc1:
+.L039pw_nc1:
movl 8(%esi),%ecx
movl %ecx,8(%ebx)
-.L028pw_nc2:
+.L040pw_nc2:
movl 12(%esi),%ecx
movl %ecx,12(%ebx)
-.L029pw_nc3:
+.L041pw_nc3:
movl 16(%esi),%ecx
movl %ecx,16(%ebx)
-.L030pw_nc4:
+.L042pw_nc4:
movl 20(%esi),%ecx
movl %ecx,20(%ebx)
-.L031pw_nc5:
+.L043pw_nc5:
movl 24(%esi),%ecx
movl %ecx,24(%ebx)
-.L032pw_nc6:
+.L044pw_nc6:
movl 28(%esi),%ecx
movl %ecx,28(%ebx)
-.L033pw_nc7:
+.L045pw_nc7:
addl $32,%esi
addl $32,%ebx
subl $8,%ebp
- jnz .L041pw_nc_loop
+ jnz .L053pw_nc_loop
movl 36(%esp),%ebp
andl $7,%ebp
- jz .L042pw_nc_end
+ jz .L054pw_nc_end
movl (%esi),%ecx
movl %ecx,(%ebx)
-.L034pw_tail_nc0:
+.L046pw_tail_nc0:
decl %ebp
- jz .L042pw_nc_end
+ jz .L054pw_nc_end
movl 4(%esi),%ecx
movl %ecx,4(%ebx)
-.L035pw_tail_nc1:
+.L047pw_tail_nc1:
decl %ebp
- jz .L042pw_nc_end
+ jz .L054pw_nc_end
movl 8(%esi),%ecx
movl %ecx,8(%ebx)
-.L036pw_tail_nc2:
+.L048pw_tail_nc2:
decl %ebp
- jz .L042pw_nc_end
+ jz .L054pw_nc_end
movl 12(%esi),%ecx
movl %ecx,12(%ebx)
-.L037pw_tail_nc3:
+.L049pw_tail_nc3:
decl %ebp
- jz .L042pw_nc_end
+ jz .L054pw_nc_end
movl 16(%esi),%ecx
movl %ecx,16(%ebx)
-.L038pw_tail_nc4:
+.L050pw_tail_nc4:
decl %ebp
- jz .L042pw_nc_end
+ jz .L054pw_nc_end
movl 20(%esi),%ecx
movl %ecx,20(%ebx)
-.L039pw_tail_nc5:
+.L051pw_tail_nc5:
decl %ebp
- jz .L042pw_nc_end
+ jz .L054pw_nc_end
movl 24(%esi),%ecx
movl %ecx,24(%ebx)
-.L040pw_tail_nc6:
-.L042pw_nc_end:
+.L052pw_tail_nc6:
+.L054pw_nc_end:
movl $0,%eax
-.L020pw_end:
+.L032pw_end:
popl %edi
popl %esi
popl %ebx
popl %ebp
ret
.size bn_sub_part_words,.-.L_bn_sub_part_words_begin
+.comm OPENSSL_ia32cap_P,8,4
diff --git a/main/openssl/crypto/bn/asm/x86-gf2m.S b/main/openssl/crypto/bn/asm/x86-gf2m.S
index 9403a5aa..9ed29ae0 100644
--- a/main/openssl/crypto/bn/asm/x86-gf2m.S
+++ b/main/openssl/crypto/bn/asm/x86-gf2m.S
@@ -249,6 +249,18 @@ bn_GF2m_mul_2x2:
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
diff --git a/main/openssl/crypto/bn/asm/x86-mont.S b/main/openssl/crypto/bn/asm/x86-mont.S
index 2bbb0e3a..c701e9e3 100644
--- a/main/openssl/crypto/bn/asm/x86-mont.S
+++ b/main/openssl/crypto/bn/asm/x86-mont.S
@@ -42,6 +42,127 @@ bn_mul_mont:
movl %esi,20(%esp)
leal -3(%edi),%ebx
movl %ebp,24(%esp)
+ call .L001PIC_me_up
+.L001PIC_me_up:
+ popl %eax
+ leal _GLOBAL_OFFSET_TABLE_+[.-.L001PIC_me_up](%eax),%eax
+ movl OPENSSL_ia32cap_P@GOT(%eax),%eax
+ btl $26,(%eax)
+ jnc .L002non_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
+.L0031st:
+ 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 .L0031st
+ 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
+.L004outer:
+ 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
+.L005inner:
+ 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 .L005inner
+ 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 .L004outer
+ emms
+ jmp .L006common_tail
+.align 16
+.L002non_sse2:
movl 8(%esp),%esi
leal 1(%ebx),%ebp
movl 12(%esp),%edi
@@ -52,12 +173,12 @@ bn_mul_mont:
leal 4(%edi,%ebx,4),%eax
orl %edx,%ebp
movl (%edi),%edi
- jz .L001bn_sqr_mont
+ jz .L007bn_sqr_mont
movl %eax,28(%esp)
movl (%esi),%eax
xorl %edx,%edx
.align 16
-.L002mull:
+.L008mull:
movl %edx,%ebp
mull %edi
addl %eax,%ebp
@@ -66,7 +187,7 @@ bn_mul_mont:
movl (%esi,%ecx,4),%eax
cmpl %ebx,%ecx
movl %ebp,28(%esp,%ecx,4)
- jl .L002mull
+ jl .L008mull
movl %edx,%ebp
mull %edi
movl 20(%esp),%edi
@@ -84,9 +205,9 @@ bn_mul_mont:
movl 4(%esi),%eax
adcl $0,%edx
incl %ecx
- jmp .L0032ndmadd
+ jmp .L0092ndmadd
.align 16
-.L0041stmadd:
+.L0101stmadd:
movl %edx,%ebp
mull %edi
addl 32(%esp,%ecx,4),%ebp
@@ -97,7 +218,7 @@ bn_mul_mont:
adcl $0,%edx
cmpl %ebx,%ecx
movl %ebp,28(%esp,%ecx,4)
- jl .L0041stmadd
+ jl .L0101stmadd
movl %edx,%ebp
mull %edi
addl 32(%esp,%ebx,4),%eax
@@ -120,7 +241,7 @@ bn_mul_mont:
adcl $0,%edx
movl $1,%ecx
.align 16
-.L0032ndmadd:
+.L0092ndmadd:
movl %edx,%ebp
mull %edi
addl 32(%esp,%ecx,4),%ebp
@@ -131,7 +252,7 @@ bn_mul_mont:
adcl $0,%edx
cmpl %ebx,%ecx
movl %ebp,24(%esp,%ecx,4)
- jl .L0032ndmadd
+ jl .L0092ndmadd
movl %edx,%ebp
mull %edi
addl 32(%esp,%ebx,4),%ebp
@@ -147,16 +268,16 @@ bn_mul_mont:
movl %edx,32(%esp,%ebx,4)
cmpl 28(%esp),%ecx
movl %eax,36(%esp,%ebx,4)
- je .L005common_tail
+ je .L006common_tail
movl (%ecx),%edi
movl 8(%esp),%esi
movl %ecx,12(%esp)
xorl %ecx,%ecx
xorl %edx,%edx
movl (%esi),%eax
- jmp .L0041stmadd
+ jmp .L0101stmadd
.align 16
-.L001bn_sqr_mont:
+.L007bn_sqr_mont:
movl %ebx,(%esp)
movl %ecx,12(%esp)
movl %edi,%eax
@@ -167,7 +288,7 @@ bn_mul_mont:
andl $1,%ebx
incl %ecx
.align 16
-.L006sqr:
+.L011sqr:
movl (%esi,%ecx,4),%eax
movl %edx,%ebp
mull %edi
@@ -179,7 +300,7 @@ bn_mul_mont:
cmpl (%esp),%ecx
movl %eax,%ebx
movl %ebp,28(%esp,%ecx,4)
- jl .L006sqr
+ jl .L011sqr
movl (%esi,%ecx,4),%eax
movl %edx,%ebp
mull %edi
@@ -203,7 +324,7 @@ bn_mul_mont:
movl 4(%esi),%eax
movl $1,%ecx
.align 16
-.L0073rdmadd:
+.L0123rdmadd:
movl %edx,%ebp
mull %edi
addl 32(%esp,%ecx,4),%ebp
@@ -222,7 +343,7 @@ bn_mul_mont:
adcl $0,%edx
cmpl %ebx,%ecx
movl %ebp,24(%esp,%ecx,4)
- jl .L0073rdmadd
+ jl .L0123rdmadd
movl %edx,%ebp
mull %edi
addl 32(%esp,%ebx,4),%ebp
@@ -238,7 +359,7 @@ bn_mul_mont:
movl %edx,32(%esp,%ebx,4)
cmpl %ebx,%ecx
movl %eax,36(%esp,%ebx,4)
- je .L005common_tail
+ je .L006common_tail
movl 4(%esi,%ecx,4),%edi
leal 1(%ecx),%ecx
movl %edi,%eax
@@ -250,12 +371,12 @@ bn_mul_mont:
xorl %ebp,%ebp
cmpl %ebx,%ecx
leal 1(%ecx),%ecx
- je .L008sqrlast
+ je .L013sqrlast
movl %edx,%ebx
shrl $1,%edx
andl $1,%ebx
.align 16
-.L009sqradd:
+.L014sqradd:
movl (%esi,%ecx,4),%eax
movl %edx,%ebp
mull %edi
@@ -271,13 +392,13 @@ bn_mul_mont:
cmpl (%esp),%ecx
movl %ebp,28(%esp,%ecx,4)
movl %eax,%ebx
- jle .L009sqradd
+ jle .L014sqradd
movl %edx,%ebp
addl %edx,%edx
shrl $31,%ebp
addl %ebx,%edx
adcl $0,%ebp
-.L008sqrlast:
+.L013sqrlast:
movl 20(%esp),%edi
movl 16(%esp),%esi
imull 32(%esp),%edi
@@ -292,9 +413,9 @@ bn_mul_mont:
adcl $0,%edx
movl $1,%ecx
movl 4(%esi),%eax
- jmp .L0073rdmadd
+ jmp .L0123rdmadd
.align 16
-.L005common_tail:
+.L006common_tail:
movl 16(%esp),%ebp
movl 4(%esp),%edi
leal 32(%esp),%esi
@@ -302,13 +423,13 @@ bn_mul_mont:
movl %ebx,%ecx
xorl %edx,%edx
.align 16
-.L010sub:
+.L015sub:
sbbl (%ebp,%edx,4),%eax
movl %eax,(%edi,%edx,4)
decl %ecx
movl 4(%esi,%edx,4),%eax
leal 1(%edx),%edx
- jge .L010sub
+ jge .L015sub
sbbl $0,%eax
andl %eax,%esi
notl %eax
@@ -316,12 +437,12 @@ bn_mul_mont:
andl %eax,%ebp
orl %ebp,%esi
.align 16
-.L011copy:
+.L016copy:
movl (%esi,%ebx,4),%eax
movl %eax,(%edi,%ebx,4)
movl %ecx,32(%esp,%ebx,4)
decl %ebx
- jge .L011copy
+ jge .L016copy
movl 24(%esp),%esp
movl $1,%eax
.L000just_leave:
@@ -336,3 +457,4 @@ bn_mul_mont:
.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,8,4