&asm_init($ARGV[0],$0);
+$sse2=0;
+for (@ARGV) { $sse2=1 if (/-DOPENSSL_IA32_SSE2/); }
+
+&external_label("OPENSSL_ia32cap_P") if ($sse2);
+
&bn_mul_add_words("bn_mul_add_words");
&bn_mul_words("bn_mul_words");
&bn_sqr_words("bn_sqr_words");
{
local($name)=@_;
- &function_begin($name,"");
+ &function_begin_B($name,$sse2?"EXTRN\t_OPENSSL_ia32cap_P:DWORD":"");
+
+ $r="eax";
+ $a="edx";
+ $c="ecx";
+
+ if ($sse2) {
+ &picmeup("eax","OPENSSL_ia32cap_P");
+ &bt(&DWP(0,"eax"),26);
+ &jnc(&label("maw_non_sse2"));
+
+ &mov($r,&wparam(0));
+ &mov($a,&wparam(1));
+ &mov($c,&wparam(2));
+ &movd("mm0",&wparam(3)); # mm0 = w
+ &pxor("mm1","mm1"); # mm1 = carry_in
+ &jmp(&label("maw_sse2_entry"));
+
+ &set_label("maw_sse2_unrolled",16);
+ &movd("mm3",&DWP(0,$r,"",0)); # mm3 = r[0]
+ &paddq("mm1","mm3"); # mm1 = carry_in + r[0]
+ &movd("mm2",&DWP(0,$a,"",0)); # mm2 = a[0]
+ &pmuludq("mm2","mm0"); # mm2 = w*a[0]
+ &movd("mm4",&DWP(4,$a,"",0)); # mm4 = a[1]
+ &pmuludq("mm4","mm0"); # mm4 = w*a[1]
+ &movd("mm6",&DWP(8,$a,"",0)); # mm6 = a[2]
+ &pmuludq("mm6","mm0"); # mm6 = w*a[2]
+ &movd("mm7",&DWP(12,$a,"",0)); # mm7 = a[3]
+ &pmuludq("mm7","mm0"); # mm7 = w*a[3]
+ &paddq("mm1","mm2"); # mm1 = carry_in + r[0] + w*a[0]
+ &movd("mm3",&DWP(4,$r,"",0)); # mm3 = r[1]
+ &paddq("mm3","mm4"); # mm3 = r[1] + w*a[1]
+ &movd("mm5",&DWP(8,$r,"",0)); # mm5 = r[2]
+ &paddq("mm5","mm6"); # mm5 = r[2] + w*a[2]
+ &movd("mm4",&DWP(12,$r,"",0)); # mm4 = r[3]
+ &paddq("mm7","mm4"); # mm7 = r[3] + w*a[3]
+ &movd(&DWP(0,$r,"",0),"mm1");
+ &movd("mm2",&DWP(16,$a,"",0)); # mm2 = a[4]
+ &pmuludq("mm2","mm0"); # mm2 = w*a[4]
+ &psrlq("mm1",32); # mm1 = carry0
+ &movd("mm4",&DWP(20,$a,"",0)); # mm4 = a[5]
+ &pmuludq("mm4","mm0"); # mm4 = w*a[5]
+ &paddq("mm1","mm3"); # mm1 = carry0 + r[1] + w*a[1]
+ &movd("mm6",&DWP(24,$a,"",0)); # mm6 = a[6]
+ &pmuludq("mm6","mm0"); # mm6 = w*a[6]
+ &movd(&DWP(4,$r,"",0),"mm1");
+ &psrlq("mm1",32); # mm1 = carry1
+ &movd("mm3",&DWP(28,$a,"",0)); # mm3 = a[7]
+ &add($a,32);
+ &pmuludq("mm3","mm0"); # mm3 = w*a[7]
+ &paddq("mm1","mm5"); # mm1 = carry1 + r[2] + w*a[2]
+ &movd("mm5",&DWP(16,$r,"",0)); # mm5 = r[4]
+ &paddq("mm2","mm5"); # mm2 = r[4] + w*a[4]
+ &movd(&DWP(8,$r,"",0),"mm1");
+ &psrlq("mm1",32); # mm1 = carry2
+ &paddq("mm1","mm7"); # mm1 = carry2 + r[3] + w*a[3]
+ &movd("mm5",&DWP(20,$r,"",0)); # mm5 = r[5]
+ &paddq("mm4","mm5"); # mm4 = r[5] + w*a[5]
+ &movd(&DWP(12,$r,"",0),"mm1");
+ &psrlq("mm1",32); # mm1 = carry3
+ &paddq("mm1","mm2"); # mm1 = carry3 + r[4] + w*a[4]
+ &movd("mm5",&DWP(24,$r,"",0)); # mm5 = r[6]
+ &paddq("mm6","mm5"); # mm6 = r[6] + w*a[6]
+ &movd(&DWP(16,$r,"",0),"mm1");
+ &psrlq("mm1",32); # mm1 = carry4
+ &paddq("mm1","mm4"); # mm1 = carry4 + r[5] + w*a[5]
+ &movd("mm5",&DWP(28,$r,"",0)); # mm5 = r[7]
+ &paddq("mm3","mm5"); # mm3 = r[7] + w*a[7]
+ &movd(&DWP(20,$r,"",0),"mm1");
+ &psrlq("mm1",32); # mm1 = carry5
+ &paddq("mm1","mm6"); # mm1 = carry5 + r[6] + w*a[6]
+ &movd(&DWP(24,$r,"",0),"mm1");
+ &psrlq("mm1",32); # mm1 = carry6
+ &paddq("mm1","mm3"); # mm1 = carry6 + r[7] + w*a[7]
+ &movd(&DWP(28,$r,"",0),"mm1");
+ &lea($r,&DWP(32,$r));
+ &psrlq("mm1",32); # mm1 = carry_out
+
+ &sub($c,8);
+ &jz(&label("maw_sse2_exit"));
+ &set_label("maw_sse2_entry");
+ &test($c,0xfffffff8);
+ &jnz(&label("maw_sse2_unrolled"));
+
+ &set_label("maw_sse2_loop",4);
+ &movd("mm2",&DWP(0,$a)); # mm2 = a[i]
+ &movd("mm3",&DWP(0,$r)); # mm3 = r[i]
+ &pmuludq("mm2","mm0"); # a[i] *= w
+ &lea($a,&DWP(4,$a));
+ &paddq("mm1","mm3"); # carry += r[i]
+ &paddq("mm1","mm2"); # carry += a[i]*w
+ &movd(&DWP(0,$r),"mm1"); # r[i] = carry_low
+ &sub($c,1);
+ &psrlq("mm1",32); # carry = carry_high
+ &lea($r,&DWP(4,$r));
+ &jnz(&label("maw_sse2_loop"));
+ &set_label("maw_sse2_exit");
+ &movd("eax","mm1"); # c = carry_out
+ &emms();
+ &ret();
+
+ &set_label("maw_non_sse2",16);
+ }
+
+ # function_begin prologue
+ &push("ebp");
+ &push("ebx");
+ &push("esi");
+ &push("edi");
&comment("");
$Low="eax";
&jz(&label("maw_finish"));
- &set_label("maw_loop",0);
-
- &mov(&swtmp(0),"ecx"); #
+ &set_label("maw_loop",16);
for ($i=0; $i<32; $i+=4)
{
&comment("Round $i");
- &mov("eax",&DWP($i,$a,"",0)); # *a
+ &mov("eax",&DWP($i,$a)); # *a
&mul($w); # *a * w
- &add("eax",$c); # L(t)+= *r
- &mov($c,&DWP($i,$r,"",0)); # L(t)+= *r
+ &add("eax",$c); # L(t)+= c
&adc("edx",0); # H(t)+=carry
- &add("eax",$c); # L(t)+=c
+ &add("eax",&DWP($i,$r)); # L(t)+= *r
&adc("edx",0); # H(t)+=carry
- &mov(&DWP($i,$r,"",0),"eax"); # *r= L(t);
+ &mov(&DWP($i,$r),"eax"); # *r= L(t);
&mov($c,"edx"); # c= H(t);
}
&comment("");
- &mov("ecx",&swtmp(0)); #
- &add($a,32);
- &add($r,32);
&sub("ecx",8);
+ &lea($a,&DWP(32,$a));
+ &lea($r,&DWP(32,$r));
&jnz(&label("maw_loop"));
&set_label("maw_finish",0);
for ($i=0; $i<7; $i++)
{
&comment("Tail Round $i");
- &mov("eax",&DWP($i*4,$a,"",0));# *a
+ &mov("eax",&DWP($i*4,$a)); # *a
&mul($w); # *a * w
&add("eax",$c); # L(t)+=c
- &mov($c,&DWP($i*4,$r,"",0)); # L(t)+= *r
&adc("edx",0); # H(t)+=carry
- &add("eax",$c);
+ &add("eax",&DWP($i*4,$r)); # L(t)+= *r
&adc("edx",0); # H(t)+=carry
&dec("ecx") if ($i != 7-1);
- &mov(&DWP($i*4,$r,"",0),"eax"); # *r= L(t);
- &mov($c,"edx"); # c= H(t);
+ &mov(&DWP($i*4,$r),"eax"); # *r= L(t);
+ &mov($c,"edx"); # c= H(t);
&jz(&label("maw_end")) if ($i != 7-1);
}
&set_label("maw_end",0);
{
local($name)=@_;
- &function_begin($name,"");
+ &function_begin_B($name,$sse2?"EXTRN\t_OPENSSL_ia32cap_P:DWORD":"");
+
+ $r="eax";
+ $a="edx";
+ $c="ecx";
+
+ if ($sse2) {
+ &picmeup("eax","OPENSSL_ia32cap_P");
+ &bt(&DWP(0,"eax"),26);
+ &jnc(&label("mw_non_sse2"));
+
+ &mov($r,&wparam(0));
+ &mov($a,&wparam(1));
+ &mov($c,&wparam(2));
+ &movd("mm0",&wparam(3)); # mm0 = w
+ &pxor("mm1","mm1"); # mm1 = carry = 0
+
+ &set_label("mw_sse2_loop",16);
+ &movd("mm2",&DWP(0,$a)); # mm2 = a[i]
+ &pmuludq("mm2","mm0"); # a[i] *= w
+ &lea($a,&DWP(4,$a));
+ &paddq("mm1","mm2"); # carry += a[i]*w
+ &movd(&DWP(0,$r),"mm1"); # r[i] = carry_low
+ &sub($c,1);
+ &psrlq("mm1",32); # carry = carry_high
+ &lea($r,&DWP(4,$r));
+ &jnz(&label("mw_sse2_loop"));
+
+ &movd("eax","mm1"); # return carry
+ &emms();
+ &ret();
+ &set_label("mw_non_sse2",16);
+ }
+
+ # function_begin prologue
+ &push("ebp");
+ &push("ebx");
+ &push("esi");
+ &push("edi");
&comment("");
$Low="eax";
{
local($name)=@_;
- &function_begin($name,"");
+ &function_begin_B($name,$sse2?"EXTRN\t_OPENSSL_ia32cap_P:DWORD":"");
+
+ $r="eax";
+ $a="edx";
+ $c="ecx";
+
+ if ($sse2) {
+ &picmeup("eax","OPENSSL_ia32cap_P");
+ &bt(&DWP(0,"eax"),26);
+ &jnc(&label("sqr_non_sse2"));
+
+ &mov($r,&wparam(0));
+ &mov($a,&wparam(1));
+ &mov($c,&wparam(2));
+
+ &set_label("sqr_sse2_loop",16);
+ &movd("mm0",&DWP(0,$a)); # mm0 = a[i]
+ &pmuludq("mm0","mm0"); # a[i] *= a[i]
+ &lea($a,&DWP(4,$a)); # a++
+ &movq(&DWP(0,$r),"mm0"); # r[i] = a[i]*a[i]
+ &sub($c,1);
+ &lea($r,&DWP(8,$r)); # r += 2
+ &jnz(&label("sqr_sse2_loop"));
+
+ &emms();
+ &ret();
+ &set_label("sqr_non_sse2",16);
+ }
+
+ # function_begin prologue
+ &push("ebp");
+ &push("ebx");
+ &push("esi");
+ &push("edi");
&comment("");
$r="esi";
{
local($name)=@_;
- &function_begin($name,"");
+ &function_begin_B($name,"");
&mov("edx",&wparam(0)); #
&mov("eax",&wparam(1)); #
- &mov("ebx",&wparam(2)); #
- &div("ebx");
- &function_end($name);
+ &mov("ecx",&wparam(2)); #
+ &div("ecx");
+ &ret();
+ &function_end_B($name);
}
sub bn_add_words