diff options
Diffstat (limited to 'sysdeps/sparc/sparc32/udiv.S')
| -rw-r--r-- | sysdeps/sparc/sparc32/udiv.S | 186 |
1 files changed, 90 insertions, 96 deletions
diff --git a/sysdeps/sparc/sparc32/udiv.S b/sysdeps/sparc/sparc32/udiv.S index 4a7d1526b0..0dd48f32b5 100644 --- a/sysdeps/sparc/sparc32/udiv.S +++ b/sysdeps/sparc/sparc32/udiv.S @@ -37,16 +37,8 @@ -#include "sysdep.h" -#ifdef __linux__ -#include <asm/traps.h> -#else -#ifdef __svr4__ +#include <sysdep.h> #include <sys/trap.h> -#else -#include <machine/trap.h> -#endif -#endif ENTRY(.udiv) @@ -63,11 +55,11 @@ ENTRY(.udiv) 1: cmp %o3, %o5 ! if %o1 exceeds %o0, done - blu Lgot_result ! (and algorithm fails otherwise) + blu LOC(got_result) ! (and algorithm fails otherwise) clr %o2 sethi %hi(1 << (32 - 4 - 1)), %g1 cmp %o3, %g1 - blu Lnot_really_big + blu LOC(not_really_big) clr %o4 ! Here the dividend is >= 2**(31-N) or so. We must be careful here, @@ -78,15 +70,15 @@ ENTRY(.udiv) 1: cmp %o5, %g1 bgeu 3f - mov 1, %g7 + mov 1, %g2 sll %o5, 4, %o5 b 1b add %o4, 1, %o4 - ! Now compute %g7. + ! Now compute %g2. 2: addcc %o5, %o5, %o5 - bcc Lnot_too_big - add %g7, 1, %g7 + bcc LOC(not_too_big) + add %g2, 1, %g2 ! We get here if the %o1 overflowed while shifting. ! This means that %o3 has the high-order bit set. @@ -94,20 +86,20 @@ ENTRY(.udiv) 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 + b LOC(do_single_div) + sub %g2, 1, %g2 - Lnot_too_big: + LOC(not_too_big): 3: cmp %o5, %o3 blu 2b nop - be Ldo_single_div + be LOC(do_single_div) 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 ! srl %o5, 1, %o5 - ! dec %g7 + ! dec %g2 ! do single-bit divide steps ! ! We have to be careful here. We know that %o3 >= %o5, so we can do the @@ -116,15 +108,15 @@ ENTRY(.udiv) ! order bit set in the first step, just falling into the regular ! division loop will mess up the first time around. ! So we unroll slightly... - Ldo_single_div: - subcc %g7, 1, %g7 - bl Lend_regular_divide + LOC(do_single_div): + subcc %g2, 1, %g2 + bl LOC(end_regular_divide) nop sub %o3, %o5, %o3 mov 1, %o2 - b Lend_single_divloop + b LOC(end_single_divloop) nop - Lsingle_divloop: + LOC(single_divloop): sll %o2, 1, %o2 bl 1f srl %o5, 1, %o5 @@ -136,217 +128,219 @@ ENTRY(.udiv) add %o3, %o5, %o3 sub %o2, 1, %o2 2: - Lend_single_divloop: - subcc %g7, 1, %g7 - bge Lsingle_divloop + LOC(end_single_divloop): + subcc %g2, 1, %g2 + bge LOC(single_divloop) tst %o3 - b,a Lend_regular_divide + b,a LOC(end_regular_divide) -Lnot_really_big: +LOC(not_really_big): 1: sll %o5, 4, %o5 cmp %o5, %o3 bleu 1b addcc %o4, 1, %o4 - be Lgot_result + be LOC(got_result) sub %o4, 1, %o4 tst %o3 ! set up for initial iteration -Ldivloop: +LOC(divloop): sll %o2, 4, %o2 ! depth 1, accumulated bits 0 - bl L.1.16 + bl LOC(1.16) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 2, accumulated bits 1 - bl L.2.17 + bl LOC(2.17) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 3, accumulated bits 3 - bl L.3.19 + bl LOC(3.19) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 4, accumulated bits 7 - bl L.4.23 + bl LOC(4.23) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 b 9f add %o2, (7*2+1), %o2 - -L.4.23: + +LOC(4.23): ! remainder is negative addcc %o3,%o5,%o3 b 9f add %o2, (7*2-1), %o2 - - -L.3.19: + + +LOC(3.19): ! remainder is negative addcc %o3,%o5,%o3 ! depth 4, accumulated bits 5 - bl L.4.21 + bl LOC(4.21) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 b 9f add %o2, (5*2+1), %o2 - -L.4.21: + +LOC(4.21): ! remainder is negative addcc %o3,%o5,%o3 b 9f add %o2, (5*2-1), %o2 - - - -L.2.17: + + + +LOC(2.17): ! remainder is negative addcc %o3,%o5,%o3 ! depth 3, accumulated bits 1 - bl L.3.17 + bl LOC(3.17) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 4, accumulated bits 3 - bl L.4.19 + bl LOC(4.19) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 b 9f add %o2, (3*2+1), %o2 - -L.4.19: + +LOC(4.19): ! remainder is negative addcc %o3,%o5,%o3 b 9f add %o2, (3*2-1), %o2 - - -L.3.17: + + +LOC(3.17): ! remainder is negative addcc %o3,%o5,%o3 ! depth 4, accumulated bits 1 - bl L.4.17 + bl LOC(4.17) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 b 9f add %o2, (1*2+1), %o2 - -L.4.17: + +LOC(4.17): ! remainder is negative addcc %o3,%o5,%o3 b 9f add %o2, (1*2-1), %o2 - - - - -L.1.16: + + + + +LOC(1.16): ! remainder is negative addcc %o3,%o5,%o3 ! depth 2, accumulated bits -1 - bl L.2.15 + bl LOC(2.15) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 3, accumulated bits -1 - bl L.3.15 + bl LOC(3.15) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 4, accumulated bits -1 - bl L.4.15 + bl LOC(4.15) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 b 9f add %o2, (-1*2+1), %o2 - -L.4.15: + +LOC(4.15): ! remainder is negative addcc %o3,%o5,%o3 b 9f add %o2, (-1*2-1), %o2 - - -L.3.15: + + +LOC(3.15): ! remainder is negative addcc %o3,%o5,%o3 ! depth 4, accumulated bits -3 - bl L.4.13 + bl LOC(4.13) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 b 9f add %o2, (-3*2+1), %o2 - -L.4.13: + +LOC(4.13): ! remainder is negative addcc %o3,%o5,%o3 b 9f add %o2, (-3*2-1), %o2 - - - -L.2.15: + + + +LOC(2.15): ! remainder is negative addcc %o3,%o5,%o3 ! depth 3, accumulated bits -3 - bl L.3.13 + bl LOC(3.13) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 ! depth 4, accumulated bits -5 - bl L.4.11 + bl LOC(4.11) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 b 9f add %o2, (-5*2+1), %o2 - -L.4.11: + +LOC(4.11): ! remainder is negative addcc %o3,%o5,%o3 b 9f add %o2, (-5*2-1), %o2 - - -L.3.13: + + +LOC(3.13): ! remainder is negative addcc %o3,%o5,%o3 ! depth 4, accumulated bits -7 - bl L.4.9 + bl LOC(4.9) srl %o5,1,%o5 ! remainder is positive subcc %o3,%o5,%o3 b 9f add %o2, (-7*2+1), %o2 - -L.4.9: + +LOC(4.9): ! remainder is negative addcc %o3,%o5,%o3 b 9f add %o2, (-7*2-1), %o2 - - - - + + + + 9: -Lend_regular_divide: +LOC(end_regular_divide): subcc %o4, 1, %o4 - bge Ldivloop + bge LOC(divloop) tst %o3 - bl,a Lgot_result + bl,a LOC(got_result) ! non-restoring fixup here (one instruction only!) sub %o2, 1, %o2 -Lgot_result: +LOC(got_result): retl mov %o2, %o0 + +END(.udiv) |