X-Git-Url: https://git.openssl.org/?p=openssl.git;a=blobdiff_plain;f=crypto%2Fsha%2Fasm%2Fsha512-x86_64.pl;fp=crypto%2Fsha%2Fasm%2Fsha512-x86_64.pl;h=bdf1d28fda29efb516f48bfee9bcd339b5befb76;hp=7a8ed7c7f211f107f648194bafb8f6e6d61b27ae;hb=399976c7ba8dcc9eba3641e03d0bb41e4d137ed1;hpb=ed874fac6399d5064d6eb8fe2022b918aeaf75af

diff --git a/crypto/sha/asm/sha512-x86_64.pl b/crypto/sha/asm/sha512-x86_64.pl
index 7a8ed7c7f2..bdf1d28fda 100755
--- a/crypto/sha/asm/sha512-x86_64.pl
+++ b/crypto/sha/asm/sha512-x86_64.pl
@@ -176,7 +176,7 @@ $Tbl="%rbp";
 $_ctx="16*$SZ+0*8(%rsp)";
 $_inp="16*$SZ+1*8(%rsp)";
 $_end="16*$SZ+2*8(%rsp)";
-$_rsp="16*$SZ+3*8(%rsp)";
+$_rsp="`16*$SZ+3*8`(%rsp)";
 $framesz="16*$SZ+4*8";
 
 
@@ -269,6 +269,7 @@ $code=<<___;
 .type	$func,\@function,3
 .align	16
 $func:
+.cfi_startproc
 ___
 $code.=<<___ if ($SZ==4 || $avx);
 	lea	OPENSSL_ia32cap_P(%rip),%r11
@@ -302,12 +303,19 @@ $code.=<<___ if ($SZ==4);
 ___
 $code.=<<___;
 	mov	%rsp,%rax		# copy %rsp
+.cfi_def_cfa_register	%rax
 	push	%rbx
+.cfi_push	%rbx
 	push	%rbp
+.cfi_push	%rbp
 	push	%r12
+.cfi_push	%r12
 	push	%r13
+.cfi_push	%r13
 	push	%r14
+.cfi_push	%r14
 	push	%r15
+.cfi_push	%r15
 	shl	\$4,%rdx		# num*16
 	sub	\$$framesz,%rsp
 	lea	($inp,%rdx,$SZ),%rdx	# inp+num*16*$SZ
@@ -316,6 +324,7 @@ $code.=<<___;
 	mov	$inp,$_inp		# save inp, 2nd arh
 	mov	%rdx,$_end		# save end pointer, "3rd" arg
 	mov	%rax,$_rsp		# save copy of %rsp
+.cfi_cfa_expression	$_rsp,deref,+8
 .Lprologue:
 
 	mov	$SZ*0($ctx),$A
@@ -382,15 +391,24 @@ $code.=<<___;
 	jb	.Lloop
 
 	mov	$_rsp,%rsi
+.cfi_def_cfa	%rsi,8
 	mov	-48(%rsi),%r15
+.cfi_restore	%r15
 	mov	-40(%rsi),%r14
+.cfi_restore	%r14
 	mov	-32(%rsi),%r13
+.cfi_restore	%r13
 	mov	-24(%rsi),%r12
+.cfi_restore	%r12
 	mov	-16(%rsi),%rbp
+.cfi_restore	%rbp
 	mov	-8(%rsi),%rbx
+.cfi_restore	%rbx
 	lea	(%rsi),%rsp
+.cfi_def_cfa_register	%rsp
 .Lepilogue:
 	ret
+.cfi_endproc
 .size	$func,.-$func
 ___
 
@@ -760,14 +778,22 @@ $code.=<<___;
 .type	${func}_ssse3,\@function,3
 .align	64
 ${func}_ssse3:
+.cfi_startproc
 .Lssse3_shortcut:
 	mov	%rsp,%rax		# copy %rsp
+.cfi_def_cfa_register	%rax
 	push	%rbx
+.cfi_push	%rbx
 	push	%rbp
+.cfi_push	%rbp
 	push	%r12
+.cfi_push	%r12
 	push	%r13
+.cfi_push	%r13
 	push	%r14
+.cfi_push	%r14
 	push	%r15
+.cfi_push	%r15
 	shl	\$4,%rdx		# num*16
 	sub	\$`$framesz+$win64*16*4`,%rsp
 	lea	($inp,%rdx,$SZ),%rdx	# inp+num*16*$SZ
@@ -776,6 +802,7 @@ ${func}_ssse3:
 	mov	$inp,$_inp		# save inp, 2nd arh
 	mov	%rdx,$_end		# save end pointer, "3rd" arg
 	mov	%rax,$_rsp		# save copy of %rsp
+.cfi_cfa_expression	$_rsp,deref,+8
 ___
 $code.=<<___ if ($win64);
 	movaps	%xmm6,16*$SZ+32(%rsp)
@@ -1074,6 +1101,7 @@ $code.=<<___;
 	jb	.Lloop_ssse3
 
 	mov	$_rsp,%rsi
+.cfi_def_cfa	%rsi,8
 ___
 $code.=<<___ if ($win64);
 	movaps	16*$SZ+32(%rsp),%xmm6
@@ -1083,14 +1111,22 @@ $code.=<<___ if ($win64);
 ___
 $code.=<<___;
 	mov	-48(%rsi),%r15
+.cfi_restore	%r15
 	mov	-40(%rsi),%r14
+.cfi_restore	%r14
 	mov	-32(%rsi),%r13
+.cfi_restore	%r13
 	mov	-24(%rsi),%r12
+.cfi_restore	%r12
 	mov	-16(%rsi),%rbp
+.cfi_restore	%rbp
 	mov	-8(%rsi),%rbx
+.cfi_restore	%rbx
 	lea	(%rsi),%rsp
+.cfi_def_cfa_register	%rsp
 .Lepilogue_ssse3:
 	ret
+.cfi_endproc
 .size	${func}_ssse3,.-${func}_ssse3
 ___
 }
@@ -1104,14 +1140,22 @@ $code.=<<___;
 .type	${func}_xop,\@function,3
 .align	64
 ${func}_xop:
+.cfi_startproc
 .Lxop_shortcut:
 	mov	%rsp,%rax		# copy %rsp
+.cfi_def_cfa_register	%rax
 	push	%rbx
+.cfi_push	%rbx
 	push	%rbp
+.cfi_push	%rbp
 	push	%r12
+.cfi_push	%r12
 	push	%r13
+.cfi_push	%r13
 	push	%r14
+.cfi_push	%r14
 	push	%r15
+.cfi_push	%r15
 	shl	\$4,%rdx		# num*16
 	sub	\$`$framesz+$win64*16*($SZ==4?4:6)`,%rsp
 	lea	($inp,%rdx,$SZ),%rdx	# inp+num*16*$SZ
@@ -1120,6 +1164,7 @@ ${func}_xop:
 	mov	$inp,$_inp		# save inp, 2nd arh
 	mov	%rdx,$_end		# save end pointer, "3rd" arg
 	mov	%rax,$_rsp		# save copy of %rsp
+.cfi_cfa_expression	$_rsp,deref,+8
 ___
 $code.=<<___ if ($win64);
 	movaps	%xmm6,16*$SZ+32(%rsp)
@@ -1446,6 +1491,7 @@ $code.=<<___;
 	jb	.Lloop_xop
 
 	mov	$_rsp,%rsi
+.cfi_def_cfa	%rsi,8
 	vzeroupper
 ___
 $code.=<<___ if ($win64);
@@ -1460,14 +1506,22 @@ $code.=<<___ if ($win64 && $SZ>4);
 ___
 $code.=<<___;
 	mov	-48(%rsi),%r15
+.cfi_restore	%r15
 	mov	-40(%rsi),%r14
+.cfi_restore	%r14
 	mov	-32(%rsi),%r13
+.cfi_restore	%r13
 	mov	-24(%rsi),%r12
+.cfi_restore	%r12
 	mov	-16(%rsi),%rbp
+.cfi_restore	%rbp
 	mov	-8(%rsi),%rbx
+.cfi_restore	%rbx
 	lea	(%rsi),%rsp
+.cfi_def_cfa_register	%rsp
 .Lepilogue_xop:
 	ret
+.cfi_endproc
 .size	${func}_xop,.-${func}_xop
 ___
 }
@@ -1480,14 +1534,22 @@ $code.=<<___;
 .type	${func}_avx,\@function,3
 .align	64
 ${func}_avx:
+.cfi_startproc
 .Lavx_shortcut:
 	mov	%rsp,%rax		# copy %rsp
+.cfi_def_cfa_register	%rax
 	push	%rbx
+.cfi_push	%rbx
 	push	%rbp
+.cfi_push	%rbp
 	push	%r12
+.cfi_push	%r12
 	push	%r13
+.cfi_push	%r13
 	push	%r14
+.cfi_push	%r14
 	push	%r15
+.cfi_push	%r15
 	shl	\$4,%rdx		# num*16
 	sub	\$`$framesz+$win64*16*($SZ==4?4:6)`,%rsp
 	lea	($inp,%rdx,$SZ),%rdx	# inp+num*16*$SZ
@@ -1496,6 +1558,7 @@ ${func}_avx:
 	mov	$inp,$_inp		# save inp, 2nd arh
 	mov	%rdx,$_end		# save end pointer, "3rd" arg
 	mov	%rax,$_rsp		# save copy of %rsp
+.cfi_cfa_expression	$_rsp,deref,+8
 ___
 $code.=<<___ if ($win64);
 	movaps	%xmm6,16*$SZ+32(%rsp)
@@ -1754,6 +1817,7 @@ $code.=<<___;
 	jb	.Lloop_avx
 
 	mov	$_rsp,%rsi
+.cfi_def_cfa	%rsi,8
 	vzeroupper
 ___
 $code.=<<___ if ($win64);
@@ -1768,14 +1832,22 @@ $code.=<<___ if ($win64 && $SZ>4);
 ___
 $code.=<<___;
 	mov	-48(%rsi),%r15
+.cfi_restore	%r15
 	mov	-40(%rsi),%r14
+.cfi_restore	%r14
 	mov	-32(%rsi),%r13
+.cfi_restore	%r13
 	mov	-24(%rsi),%r12
+.cfi_restore	%r12
 	mov	-16(%rsi),%rbp
+.cfi_restore	%rbp
 	mov	-8(%rsi),%rbx
+.cfi_restore	%rbx
 	lea	(%rsi),%rsp
+.cfi_def_cfa_register	%rsp
 .Lepilogue_avx:
 	ret
+.cfi_endproc
 .size	${func}_avx,.-${func}_avx
 ___
 
@@ -1831,14 +1903,22 @@ $code.=<<___;
 .type	${func}_avx2,\@function,3
 .align	64
 ${func}_avx2:
+.cfi_startproc
 .Lavx2_shortcut:
 	mov	%rsp,%rax		# copy %rsp
+.cfi_def_cfa_register	%rax
 	push	%rbx
+.cfi_push	%rbx
 	push	%rbp
+.cfi_push	%rbp
 	push	%r12
+.cfi_push	%r12
 	push	%r13
+.cfi_push	%r13
 	push	%r14
+.cfi_push	%r14
 	push	%r15
+.cfi_push	%r15
 	sub	\$`2*$SZ*$rounds+4*8+$win64*16*($SZ==4?4:6)`,%rsp
 	shl	\$4,%rdx		# num*16
 	and	\$-256*$SZ,%rsp		# align stack frame
@@ -1848,6 +1928,7 @@ ${func}_avx2:
 	mov	$inp,$_inp		# save inp, 2nd arh
 	mov	%rdx,$_end		# save end pointer, "3rd" arg
 	mov	%rax,$_rsp		# save copy of %rsp
+.cfi_cfa_expression	$_rsp,deref,+8
 ___
 $code.=<<___ if ($win64);
 	movaps	%xmm6,16*$SZ+32(%rsp)
@@ -2128,6 +2209,7 @@ $code.=<<___;
 .Ldone_avx2:
 	lea	($Tbl),%rsp
 	mov	$_rsp,%rsi
+.cfi_def_cfa	%rsi,8
 	vzeroupper
 ___
 $code.=<<___ if ($win64);
@@ -2142,14 +2224,22 @@ $code.=<<___ if ($win64 && $SZ>4);
 ___
 $code.=<<___;
 	mov	-48(%rsi),%r15
+.cfi_restore	%r15
 	mov	-40(%rsi),%r14
+.cfi_restore	%r14
 	mov	-32(%rsi),%r13
+.cfi_restore	%r13
 	mov	-24(%rsi),%r12
+.cfi_restore	%r12
 	mov	-16(%rsi),%rbp
+.cfi_restore	%rbp
 	mov	-8(%rsi),%rbx
+.cfi_restore	%rbx
 	lea	(%rsi),%rsp
+.cfi_def_cfa_register	%rsp
 .Lepilogue_avx2:
 	ret
+.cfi_endproc
 .size	${func}_avx2,.-${func}_avx2
 ___
 }}