| 1 | ; $Id: bs3-wc32-U8M.asm 106061 2024-09-16 14:03:52Z vboxsync $
|
|---|
| 2 | ;; @file
|
|---|
| 3 | ; BS3Kit - 32-bit Watcom C/C++, 64-bit integer multiplication.
|
|---|
| 4 | ;
|
|---|
| 5 |
|
|---|
| 6 | ;
|
|---|
| 7 | ; Copyright (C) 2007-2024 Oracle and/or its affiliates.
|
|---|
| 8 | ;
|
|---|
| 9 | ; This file is part of VirtualBox base platform packages, as
|
|---|
| 10 | ; available from https://www.virtualbox.org.
|
|---|
| 11 | ;
|
|---|
| 12 | ; This program is free software; you can redistribute it and/or
|
|---|
| 13 | ; modify it under the terms of the GNU General Public License
|
|---|
| 14 | ; as published by the Free Software Foundation, in version 3 of the
|
|---|
| 15 | ; License.
|
|---|
| 16 | ;
|
|---|
| 17 | ; This program is distributed in the hope that it will be useful, but
|
|---|
| 18 | ; WITHOUT ANY WARRANTY; without even the implied warranty of
|
|---|
| 19 | ; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|---|
| 20 | ; General Public License for more details.
|
|---|
| 21 | ;
|
|---|
| 22 | ; You should have received a copy of the GNU General Public License
|
|---|
| 23 | ; along with this program; if not, see <https://www.gnu.org/licenses>.
|
|---|
| 24 | ;
|
|---|
| 25 | ; The contents of this file may alternatively be used under the terms
|
|---|
| 26 | ; of the Common Development and Distribution License Version 1.0
|
|---|
| 27 | ; (CDDL), a copy of it is provided in the "COPYING.CDDL" file included
|
|---|
| 28 | ; in the VirtualBox distribution, in which case the provisions of the
|
|---|
| 29 | ; CDDL are applicable instead of those of the GPL.
|
|---|
| 30 | ;
|
|---|
| 31 | ; You may elect to license modified versions of this file under the
|
|---|
| 32 | ; terms and conditions of either the GPL or the CDDL or both.
|
|---|
| 33 | ;
|
|---|
| 34 | ; SPDX-License-Identifier: GPL-3.0-only OR CDDL-1.0
|
|---|
| 35 | ;
|
|---|
| 36 |
|
|---|
| 37 | %include "bs3kit-template-header.mac"
|
|---|
| 38 |
|
|---|
| 39 |
|
|---|
| 40 | ;;
|
|---|
| 41 | ; 64-bit unsigned & signed integer multiplication.
|
|---|
| 42 | ;
|
|---|
| 43 | ; @returns EDX:EAX as the product.
|
|---|
| 44 | ; @param EDX:EAX Factor 1 - edx=F1H, eax=F1L.
|
|---|
| 45 | ; @param ECX:EBX Factor 2 - ecx=F2H, ebx=F2L.
|
|---|
| 46 | ;
|
|---|
| 47 | global $??I8M
|
|---|
| 48 | $??I8M:
|
|---|
| 49 | global $??U8M
|
|---|
| 50 | $??U8M:
|
|---|
| 51 | ;
|
|---|
| 52 | ; If both the high dwords are zero, we can get away with
|
|---|
| 53 | ; a simple 32-bit multiplication.
|
|---|
| 54 | ;
|
|---|
| 55 | test ecx, ecx
|
|---|
| 56 | jnz .big
|
|---|
| 57 | test edx, edx
|
|---|
| 58 | jnz .big
|
|---|
| 59 | mul ebx
|
|---|
| 60 | ret
|
|---|
| 61 |
|
|---|
| 62 | .big:
|
|---|
| 63 | ;
|
|---|
| 64 | ; Imagine we use 4294967296-base (2^32), so each factor has two
|
|---|
| 65 | ; digits H and L, thus we have: F1H:F1L * F2H:F1L which we can
|
|---|
| 66 | ; multipy like we learned in primary school. Since the result
|
|---|
| 67 | ; is limited to 64-bit, we can skip F1H*F2H and discard the
|
|---|
| 68 | ; high 32-bit in F1L*F2H and F1H*F2L.
|
|---|
| 69 | ; result = ((F1L*F2H) << 32)
|
|---|
| 70 | ; + ((F1H*F2L) << 32)
|
|---|
| 71 | ; + (F1L*F2L);
|
|---|
| 72 | ;
|
|---|
| 73 | push ecx ; Preserve ECX just to be nice.
|
|---|
| 74 | push eax ; Stash F1L for later.
|
|---|
| 75 | push edx ; Stash F1H for later.
|
|---|
| 76 |
|
|---|
| 77 | ; ECX = F1L*F2H
|
|---|
| 78 | mul ecx
|
|---|
| 79 | mov ecx, eax
|
|---|
| 80 |
|
|---|
| 81 | ; ECX += F1H * F2L
|
|---|
| 82 | pop eax
|
|---|
| 83 | mul ebx
|
|---|
| 84 | add ecx, eax
|
|---|
| 85 |
|
|---|
| 86 | ; EDX:EAX = F1L * F2L
|
|---|
| 87 | pop eax
|
|---|
| 88 | mul ebx
|
|---|
| 89 | add edx, ecx
|
|---|
| 90 |
|
|---|
| 91 | pop ecx
|
|---|
| 92 | ret
|
|---|
| 93 |
|
|---|
