Conversion to UTF-8 where needed

[openssl.git] / crypto / bn / asm / ia64.S

diff --git a/crypto/bn/asm/ia64.S b/crypto/bn/asm/ia64.S
index 7b82b820e62aacdc51278b1d3774314c8b4592d3..c0cee8211c031c2df63aeac0421351eedc829776 100644 (file)
--- a/crypto/bn/asm/ia64.S
+++ b/crypto/bn/asm/ia64.S
@@ -171,21 +171,21 @@
 .skip  32      // makes the loop body aligned at 64-byte boundary
 bn_add_words:
        .prologue
-       .fframe 0
        .save   ar.pfs,r2
 { .mii;        alloc           r2=ar.pfs,4,12,0,16
        cmp4.le         p6,p0=r35,r0    };;
 { .mfb;        mov             r8=r0                   // return value
 (p6)   br.ret.spnt.many        b0      };;
 
-       .save   ar.lc,r3
 { .mib;        sub             r10=r35,r0,1
+       .save   ar.lc,r3
        mov             r3=ar.lc
        brp.loop.imp    .L_bn_add_words_ctop,.L_bn_add_words_cend-16
                                        }
-       .body
 { .mib;        ADDP            r14=0,r32               // rp
+       .save   pr,r9
        mov             r9=pr           };;
+       .body
 { .mii;        ADDP            r15=0,r33               // ap
        mov             ar.lc=r10
        mov             ar.ec=6         }
@@ -224,21 +224,21 @@ bn_add_words:
 .skip  32      // makes the loop body aligned at 64-byte boundary
 bn_sub_words:
        .prologue
-       .fframe 0
        .save   ar.pfs,r2
 { .mii;        alloc           r2=ar.pfs,4,12,0,16
        cmp4.le         p6,p0=r35,r0    };;
 { .mfb;        mov             r8=r0                   // return value
 (p6)   br.ret.spnt.many        b0      };;
 
-       .save   ar.lc,r3
 { .mib;        sub             r10=r35,r0,1
+       .save   ar.lc,r3
        mov             r3=ar.lc
        brp.loop.imp    .L_bn_sub_words_ctop,.L_bn_sub_words_cend-16
                                        }
-       .body
 { .mib;        ADDP            r14=0,r32               // rp
+       .save   pr,r9
        mov             r9=pr           };;
+       .body
 { .mii;        ADDP            r15=0,r33               // ap
        mov             ar.lc=r10
        mov             ar.ec=6         }
@@ -283,7 +283,6 @@ bn_sub_words:
 .skip  32      // makes the loop body aligned at 64-byte boundary
 bn_mul_words:
        .prologue
-       .fframe 0
        .save   ar.pfs,r2
 #ifdef XMA_TEMPTATION
 { .mfi;        alloc           r2=ar.pfs,4,0,0,0       };;
@@ -294,9 +293,10 @@ bn_mul_words:
        cmp4.le         p6,p0=r34,r0
 (p6)   br.ret.spnt.many        b0              };;
 
-       .save   ar.lc,r3
 { .mii;        sub     r10=r34,r0,1
+       .save   ar.lc,r3
        mov     r3=ar.lc
+       .save   pr,r9
        mov     r9=pr                   };;
 
        .body
@@ -397,22 +397,21 @@ bn_mul_words:
 .skip  48      // makes the loop body aligned at 64-byte boundary
 bn_mul_add_words:
        .prologue
-       .fframe 0
        .save   ar.pfs,r2
-       .save   ar.lc,r3
-       .save   pr,r9
 { .mmi;        alloc           r2=ar.pfs,4,4,0,8
        cmp4.le         p6,p0=r34,r0
+       .save   ar.lc,r3
        mov             r3=ar.lc        };;
 { .mib;        mov             r8=r0           // return value
        sub             r10=r34,r0,1
 (p6)   br.ret.spnt.many        b0      };;
 
-       .body
 { .mib;        setf.sig        f8=r35          // w
+       .save   pr,r9
        mov             r9=pr
        brp.loop.imp    .L_bn_mul_add_words_ctop,.L_bn_mul_add_words_cend-16
                                        }
+       .body
 { .mmi;        ADDP            r14=0,r32       // rp
        ADDP            r15=0,r33       // ap
        mov             ar.lc=r10       }
@@ -466,7 +465,6 @@ bn_mul_add_words:
 .skip  32      // makes the loop body aligned at 64-byte boundary 
 bn_sqr_words:
        .prologue
-       .fframe 0
        .save   ar.pfs,r2
 { .mii;        alloc           r2=ar.pfs,3,0,0,0
        sxt4            r34=r34         };;
@@ -476,9 +474,10 @@ bn_sqr_words:
        nop.f           0x0
 (p6)   br.ret.spnt.many        b0      };;
 
-       .save   ar.lc,r3
 { .mii;        sub     r10=r34,r0,1
+       .save   ar.lc,r3
        mov     r3=ar.lc
+       .save   pr,r9
        mov     r9=pr                   };;
 
        .body
@@ -545,7 +544,6 @@ bn_sqr_words:
 .align 64
 bn_sqr_comba8:
        .prologue
-       .fframe 0
        .save   ar.pfs,r2
 #if defined(_HPUX_SOURCE) && !defined(_LP64)
 { .mii;        alloc   r2=ar.pfs,2,1,0,0
@@ -570,7 +568,7 @@ bn_sqr_comba8:
 // I've estimated this routine to run in ~120 ticks, but in reality
 // (i.e. according to ar.itc) it takes ~160 ticks. Are those extra
 // cycles consumed for instructions fetch? Or did I misinterpret some
-// clause in Itanium µ-architecture manual? Comments are welcomed and
+// clause in Itanium µ-architecture manual? Comments are welcomed and
 // highly appreciated.
 //
 // On Itanium 2 it takes ~190 ticks. This is because of stalls on
@@ -617,7 +615,6 @@ bn_sqr_comba8:
 .align 64
 bn_mul_comba8:
        .prologue
-       .fframe 0
        .save   ar.pfs,r2
 #if defined(_HPUX_SOURCE) && !defined(_LP64)
 { .mii;        alloc   r2=ar.pfs,3,0,0,0
@@ -1175,7 +1172,6 @@ bn_mul_comba8:
 .align 64
 bn_sqr_comba4:
        .prologue
-       .fframe 0
        .save   ar.pfs,r2
 #if defined(_HPUX_SOURCE) && !defined(_LP64)
 { .mii;        alloc   r2=ar.pfs,2,1,0,0
@@ -1208,7 +1204,6 @@ bn_sqr_comba4:
 .align 64
 bn_mul_comba4:
        .prologue
-       .fframe 0
        .save   ar.pfs,r2
 #if defined(_HPUX_SOURCE) && !defined(_LP64)
 { .mii;        alloc   r2=ar.pfs,3,0,0,0
@@ -1411,11 +1406,11 @@ equ=p24
 .align 64
 bn_div_words:
        .prologue
-       .fframe 0
        .save   ar.pfs,r2
-       .save   b0,r3
 { .mii;        alloc           r2=ar.pfs,3,5,0,8
+       .save   b0,r3
        mov             r3=b0
+       .save   pr,r10
        mov             r10=pr          };;
 { .mmb;        cmp.eq          p6,p0=r34,r0
        mov             r8=-1