diff options
author | Ralf Baechle <ralf@linux-mips.org> | 1997-01-07 02:33:00 +0000 |
---|---|---|
committer | <ralf@linux-mips.org> | 1997-01-07 02:33:00 +0000 |
commit | beb116954b9b7f3bb56412b2494b562f02b864b1 (patch) | |
tree | 120e997879884e1b9d93b265221b939d2ef1ade1 /arch/sparc/lib/sdiv.S | |
parent | 908d4681a1dc3792ecafbe64265783a86c4cccb6 (diff) |
Import of Linux/MIPS 2.1.14
Diffstat (limited to 'arch/sparc/lib/sdiv.S')
-rw-r--r-- | arch/sparc/lib/sdiv.S | 222 |
1 files changed, 119 insertions, 103 deletions
diff --git a/arch/sparc/lib/sdiv.S b/arch/sparc/lib/sdiv.S index 2fa7a9794..e0ad80b6f 100644 --- a/arch/sparc/lib/sdiv.S +++ b/arch/sparc/lib/sdiv.S @@ -1,4 +1,5 @@ -/* sdiv.S: This routine was taken from glibc-1.09 and is covered +/* $Id: sdiv.S,v 1.6 1996/10/02 17:37:00 davem Exp $ + * sdiv.S: This routine was taken from glibc-1.09 and is covered * by the GNU Library General Public License Version 2. */ @@ -46,13 +47,14 @@ ! compute sign of result; if neither is negative, no problem orcc %o1, %o0, %g0 ! either negative? bge 2f ! no, go do the divide - xor %o1, %o0, %g6 ! compute sign in any case + xor %o1, %o0, %g2 ! compute sign in any case + tst %o1 bge 1f - tst %o0 + tst %o0 ! %o1 is definitely negative; %o0 might also be negative bge 2f ! if %o0 not negative... - sub %g0, %o1, %o1 ! in any case, make %o1 nonneg + sub %g0, %o1, %o1 ! in any case, make %o1 nonneg 1: ! %o0 is negative, %o1 is nonnegative sub %g0, %o0, %o0 ! make %o0 nonnegative 2: @@ -60,22 +62,24 @@ ! Ready to divide. Compute size of quotient; scale comparand. orcc %o1, %g0, %o5 bne 1f - mov %o0, %o3 + mov %o0, %o3 ! Divide by zero trap. If it returns, return 0 (about as ! wrong as possible, but that is what SunOS does...). ta ST_DIV0 retl - clr %o0 + clr %o0 1: cmp %o3, %o5 ! if %o1 exceeds %o0, done blu Lgot_result ! (and algorithm fails otherwise) - clr %o2 + clr %o2 + sethi %hi(1 << (32 - 4 - 1)), %g1 + cmp %o3, %g1 blu Lnot_really_big - clr %o4 + clr %o4 ! Here the dividend is >= 2**(31-N) or so. We must be careful here, ! as our usual N-at-a-shot divide step will cause overflow and havoc. @@ -85,15 +89,18 @@ 1: cmp %o5, %g1 bgeu 3f - mov 1, %g7 + mov 1, %g7 + sll %o5, 4, %o5 + b 1b - add %o4, 1, %o4 + add %o4, 1, %o4 ! Now compute %g7. - 2: addcc %o5, %o5, %o5 + 2: + addcc %o5, %o5, %o5 bcc Lnot_too_big - add %g7, 1, %g7 + add %g7, 1, %g7 ! We get here if the %o1 overflowed while shifting. ! This means that %o3 has the high-order bit set. @@ -101,15 +108,18 @@ sll %g1, 4, %g1 ! high order bit srl %o5, 1, %o5 ! rest of %o5 add %o5, %g1, %o5 + b Ldo_single_div - sub %g7, 1, %g7 + sub %g7, 1, %g7 Lnot_too_big: - 3: cmp %o5, %o3 + 3: + cmp %o5, %o3 blu 2b - nop + nop + be Ldo_single_div - nop + nop /* NB: these are commented out in the V8-Sparc manual as well */ /* (I do not understand this) */ ! %o5 > %o3: went too far: back up 1 step @@ -126,19 +136,23 @@ Ldo_single_div: subcc %g7, 1, %g7 bl Lend_regular_divide - nop + nop + sub %o3, %o5, %o3 mov 1, %o2 + b Lend_single_divloop - nop + nop Lsingle_divloop: sll %o2, 1, %o2 + bl 1f - srl %o5, 1, %o5 + srl %o5, 1, %o5 ! %o3 >= 0 sub %o3, %o5, %o3 + b 2f - add %o2, 1, %o2 + add %o2, 1, %o2 1: ! %o3 < 0 add %o3, %o5, %o3 sub %o2, 1, %o2 @@ -146,7 +160,8 @@ Lend_single_divloop: subcc %g7, 1, %g7 bge Lsingle_divloop - tst %o3 + tst %o3 + b,a Lend_regular_divide Lnot_really_big: @@ -154,83 +169,81 @@ Lnot_really_big: sll %o5, 4, %o5 cmp %o5, %o3 bleu 1b - addcc %o4, 1, %o4 + addcc %o4, 1, %o4 + be Lgot_result - sub %o4, 1, %o4 + sub %o4, 1, %o4 tst %o3 ! set up for initial iteration Ldivloop: sll %o2, 4, %o2 ! depth 1, accumulated bits 0 bl L.1.16 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 2, accumulated bits 1 bl L.2.17 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 3, accumulated bits 3 bl L.3.19 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 4, accumulated bits 7 bl L.4.23 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 - b 9f - add %o2, (7*2+1), %o2 - + b 9f + add %o2, (7*2+1), %o2 + L.4.23: ! remainder is negative addcc %o3,%o5,%o3 - b 9f - add %o2, (7*2-1), %o2 - - + b 9f + add %o2, (7*2-1), %o2 + L.3.19: ! remainder is negative addcc %o3,%o5,%o3 ! depth 4, accumulated bits 5 bl L.4.21 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 - b 9f - add %o2, (5*2+1), %o2 - + b 9f + add %o2, (5*2+1), %o2 + L.4.21: ! remainder is negative addcc %o3,%o5,%o3 - b 9f - add %o2, (5*2-1), %o2 - - - + b 9f + add %o2, (5*2-1), %o2 + L.2.17: ! remainder is negative addcc %o3,%o5,%o3 ! depth 3, accumulated bits 1 bl L.3.17 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 4, accumulated bits 3 bl L.4.19 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 - b 9f - add %o2, (3*2+1), %o2 - + b 9f + add %o2, (3*2+1), %o2 + L.4.19: ! remainder is negative addcc %o3,%o5,%o3 - b 9f - add %o2, (3*2-1), %o2 + b 9f + add %o2, (3*2-1), %o2 L.3.17: @@ -238,126 +251,129 @@ L.3.17: addcc %o3,%o5,%o3 ! depth 4, accumulated bits 1 bl L.4.17 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 - b 9f - add %o2, (1*2+1), %o2 - + b 9f + add %o2, (1*2+1), %o2 + L.4.17: ! remainder is negative addcc %o3,%o5,%o3 - b 9f - add %o2, (1*2-1), %o2 - - - - + b 9f + add %o2, (1*2-1), %o2 + L.1.16: ! remainder is negative addcc %o3,%o5,%o3 ! depth 2, accumulated bits -1 bl L.2.15 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 3, accumulated bits -1 bl L.3.15 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 4, accumulated bits -1 bl L.4.15 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 - b 9f - add %o2, (-1*2+1), %o2 - + b 9f + add %o2, (-1*2+1), %o2 + L.4.15: ! remainder is negative addcc %o3,%o5,%o3 - b 9f - add %o2, (-1*2-1), %o2 - - + b 9f + add %o2, (-1*2-1), %o2 + L.3.15: ! remainder is negative addcc %o3,%o5,%o3 ! depth 4, accumulated bits -3 bl L.4.13 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 - b 9f - add %o2, (-3*2+1), %o2 - + b 9f + add %o2, (-3*2+1), %o2 + L.4.13: ! remainder is negative addcc %o3,%o5,%o3 - b 9f - add %o2, (-3*2-1), %o2 - - - + b 9f + add %o2, (-3*2-1), %o2 + L.2.15: ! remainder is negative addcc %o3,%o5,%o3 ! depth 3, accumulated bits -3 bl L.3.13 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 4, accumulated bits -5 bl L.4.11 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 - b 9f - add %o2, (-5*2+1), %o2 - + b 9f + add %o2, (-5*2+1), %o2 + L.4.11: ! remainder is negative addcc %o3,%o5,%o3 - b 9f - add %o2, (-5*2-1), %o2 - - + b 9f + add %o2, (-5*2-1), %o2 + L.3.13: ! remainder is negative addcc %o3,%o5,%o3 ! depth 4, accumulated bits -7 bl L.4.9 - srl %o5,1,%o5 + srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 - b 9f - add %o2, (-7*2+1), %o2 - + b 9f + add %o2, (-7*2+1), %o2 + L.4.9: ! remainder is negative addcc %o3,%o5,%o3 - b 9f - add %o2, (-7*2-1), %o2 - - - - + b 9f + add %o2, (-7*2-1), %o2 + 9: Lend_regular_divide: subcc %o4, 1, %o4 bge Ldivloop - tst %o3 + tst %o3 + bl,a Lgot_result ! non-restoring fixup here (one instruction only!) sub %o2, 1, %o2 - Lgot_result: ! check to see if answer should be < 0 - tst %g6 + tst %g2 bl,a 1f - sub %g0, %o2, %o2 + sub %g0, %o2, %o2 1: retl - mov %o2, %o0 + mov %o2, %o0 + + .globl .div_patch +.div_patch: + sra %o0, 0x1f, %o2 + wr %o2, 0x0, %y + nop + nop + nop + sdivcc %o0, %o1, %o0 + bvs,a 1f + xnor %o0, %g0, %o0 +1: retl + nop |