6 // Copyright (c) 2000, Intel Corporation
7 // All rights reserved.
9 // Contributed 2/15/2000 by Marius Cornea, John Harrison, Cristina Iordache,
10 // Ted Kubaska, Bob Norin, and Shane Story of the Computational Software Lab,
13 // WARRANTY DISCLAIMER
15 // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
16 // "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
17 // LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
18 // A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL INTEL OR ITS
19 // CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
20 // EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
21 // PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
22 // PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
23 // OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY OR TORT (INCLUDING
24 // NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
25 // SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
27 // Intel Corporation is the author of this code, and requests that all
28 // problem reports or change requests be submitted to it directly at
29 // http://developer.intel.com/opensource.
34 // 32-bit signed integer divide
44 alloc r31=ar.pfs,2,0,0,0
50 // 32-BIT SIGNED INTEGER DIVIDE BEGINS HERE
52 // general register used:
53 // r32 - 32-bit signed integer dividend
54 // r33 - 32-bit signed integer divisor
55 // r8 - 32-bit signed integer result
56 // r2 - scratch register
57 // floating-point registers used: f6, f7, f8, f9
58 // predicate registers used: p6
82 (p6) fma.s1 f6=f6,f8,f0
86 // (3) e0 = 1 - b * y0
87 (p6) fnma.s1 f7=f7,f8,f1
91 // (4) q1 = q0 + e0 * q0
92 (p6) fma.s1 f6=f7,f6,f6
96 // (5) e1 = e0 * e0 + 2^-34
97 (p6) fma.s1 f7=f7,f7,f9
101 // (6) q2 = q1 + e1 * q1
102 (p6) fma.s1 f8=f7,f6,f6
107 fcvt.fx.trunc.s1 f8=f8
110 // quotient will be in the least significant 32 bits of r8 (if b != 0)
116 // 32-BIT SIGNED INTEGER DIVIDE ENDS HERE