如何正确地将两个 long long int 相乘?
我想乘以 2^32 为基础给出的长数字。我已经想到了一个很好的算法来做到这一点,但不幸的是我被困住了。我遇到的情况是,我如何将两个长整型相乘并在 2^32 的基础上表示它。
#include <stdio.h>
#include <stdlib.h>
#include <limits.h>
typedef unsigned int uint32;
typedef unsigned long long uint64;
int main(int argc, char* argv[] )
{
uint64 a = (uint64)ULONG_MAX;
printf("%llu\n", a);
uint64 b = (uint64)ULONG_MAX;
printf("%llu\n", b);
uint64 c = (uint64)(a*b);
printf("%llu\n", c); // prints 1. that would be to lower 32 bits of the results. the upper half is 0xFFFFFFFE
printf("%llu\n", ULLONG_MAX);
system("pause");
}
为什么 ULLONG_MAX 与 ULONG_MAX 相同?根据 http://en.wikipedia.org/wiki/Limits.h#Member_constants< /a> 应该是 18,446,744,073,709,551,615 我
正如你从我的评论中看到的,我想要两个 uint32 相乘的结果。 下半部分为 0x1,上半部分为 0xFFFFFFFE。我如何获得这些值?
(我在SO上发现了这个问题,但这对我的情况没有帮助,因为给出的答案与我的想法相似:将两个 long long int 相乘 C)
编辑: 我的系统是Windows XP 32位。我正在使用 gcc 3.4.2 (mingw-special)
运行代码时得到的输出:
4294967295
4294967295
1
4294967295
Edit2:
printf("%i\n", sizeof(unsigned long));
printf("%i\n", sizeof(unsigned long long));
返回
4
8
Edit 3: 感谢 Petesh 我找到了解决方案:
printf("%lu\n", c & 0xFFFFFFFF);
printf("%lu\n", (c >> 32));
I want to multiply long numbers which are given in a 2^32 basis. I already thought of an nice algorithm to do that, but unfortunatly I'm stuck. The situation I'm stuck at, is how I do multiply two long ints and represent it on the 2^32 basis.
#include <stdio.h>
#include <stdlib.h>
#include <limits.h>
typedef unsigned int uint32;
typedef unsigned long long uint64;
int main(int argc, char* argv[] )
{
uint64 a = (uint64)ULONG_MAX;
printf("%llu\n", a);
uint64 b = (uint64)ULONG_MAX;
printf("%llu\n", b);
uint64 c = (uint64)(a*b);
printf("%llu\n", c); // prints 1. that would be to lower 32 bits of the results. the upper half is 0xFFFFFFFE
printf("%llu\n", ULLONG_MAX);
system("pause");
}
Why is ULLONG_MAX the same as ULONG_MAX ? According to http://en.wikipedia.org/wiki/Limits.h#Member_constants it should be 18,446,744,073,709,551,615 I
As you can see from my comments, I want the result of the the multiplikation in the two uint32.
The lowerhalf would be 0x1 and the upper half 0xFFFFFFFE. How do I get these values?
(I found this question on SO, but it's not helpful in my situation because the answers given ar similiar to my ideas: Multiplying two long long ints C)
Edit:
My system is Windows XP 32 Bit. I'm using gcc 3.4.2 (mingw-special)
The output I do get while running the code:
4294967295
4294967295
1
4294967295
Edit2:
printf("%i\n", sizeof(unsigned long));
printf("%i\n", sizeof(unsigned long long));
returns
4
8
Edit 3:
Thanks to Petesh I was able to find the solution:
printf("%lu\n", c & 0xFFFFFFFF);
printf("%lu\n", (c >> 32));
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(2)
提示在系统中(“暂停”) - 你在 Windows 上?使用 Microsoft Visual C 运行时打印 long long 需要使用“%I64u”(即大写 i)。
这是基于SO问题 How do you printf an unsigned long long int(unsigned long long int 的格式说明符)?
The hint is in the system("pause") - you're on windows? Printing a long long using the Microsoft visual c runtime requires using '%I64u' (that's a capital i).
This is based on SO question How do you printf an unsigned long long int(the format specifier for unsigned long long int)?
不知道为什么你用你的(未指定的)编译器得到这些结果,但 Ubuntu 10 下的
gcc
给出:最后两个是
0xfffffffe00000001
和 (264< /sup>-1) 分别如您所愿。因此,也许可以考虑切换到更新的编译器。您可能使用的是 C99 之前的编译器。
只是出于兴趣,
sizeof (unsigned long)
和sizeof (unsigned long long)
在您的系统上为您提供了什么。这将有助于解释你的问题。由于您的
sizeof
似乎表明数据类型本身没问题(尽管这些可能无法解决问题 - 它们是通过 相当浅的网络搜索):"%I64u" 作为格式字符串,而不是
"%llu"
。如果 MinGW 使用 MSVCRT 库,则可能需要真正的 64 位 printf 支持。-std=c99
进行编译。Not sure why you're getting those results with your (unspecified) compiler but
gcc
under Ubuntu 10 gives:with those last two being
0xfffffffe00000001
and (264-1) respectively, as you desire.So maybe consider switching to a more up-to-date compiler. It may be you're using a pre-C99 compiler.
Just out of interest, what does
sizeof (unsigned long)
andsizeof (unsigned long long)
give you on your system. This would go a long way towards explaining your problem.A couple of other things to check since your
sizeof
s seem to indicate the data types themselves are okay (though these may not fix the problem - they were found with a fairly shallow web search):"%I64u"
as the format string instead of"%llu"
. If MinGW is using the MSVCRT libs, that might be required for real 64-bitprintf
support.-std=c99
.