public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c++/65273] New: Incorrect output in SIMPLE program compiled from high version gcc with O2
@ 2015-03-02  4:08 codeeply at foxmail dot com
  2015-03-02  4:25 ` [Bug c++/65273] " pinskia at gcc dot gnu.org
  0 siblings, 1 reply; 2+ messages in thread
From: codeeply at foxmail dot com @ 2015-03-02  4:08 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65273

            Bug ID: 65273
           Summary: Incorrect output in SIMPLE program compiled from high
                    version gcc with O2
           Product: gcc
           Version: 4.8.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: codeeply at foxmail dot com

#include <iostream>
#include <cstring>

using namespace std;

const int N = 2000017;

int hash(const char *s) {
    volatile int ret = 1;
    int i, len = strlen(s);
    for (i = 0; i < len; ++i) {
        ret = ret * 131 + s[i];
    }
    ret %= N;
    if (ret < 0) ret += N;
    return ret;
}

int hash1(const char *s) {
    int ret = 1;
    int i, len = strlen(s);
    for (i = 0; i < len; ++i) {
        ret = ret * 131 + s[i];
    }
    ret %= N;
    if (ret < 0) ret += N;
    return ret;
}

int main() {
    cout << hash("sfihsifef") << endl;
    cout << hash1("sfihsifef") << endl;
    return 0;
}

---------------------------------------------------------------------
:~/test>g++ -O2 g.cpp 
:~/test>./a.out 
602373
-1397644
---------------------------------------------------------------------
It's found that "if (ret < 0) ret += N;" was "optimized" without volatile.
And this bug has been seen in 4.7.2 and 4.8.2, but not in 4.1.2.


^ permalink raw reply	[flat|nested] 2+ messages in thread

* [Bug c++/65273] Incorrect output in SIMPLE program compiled from high version gcc with O2
  2015-03-02  4:08 [Bug c++/65273] New: Incorrect output in SIMPLE program compiled from high version gcc with O2 codeeply at foxmail dot com
@ 2015-03-02  4:25 ` pinskia at gcc dot gnu.org
  0 siblings, 0 replies; 2+ messages in thread
From: pinskia at gcc dot gnu.org @ 2015-03-02  4:25 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65273

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|---                         |INVALID

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Why do you think this is a bug when you do a multiplication and then an
addition which will always be positive if there was no such thing as an
overflow. Note overflow for signed intergers is undefined in c/c++.


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2015-03-02  4:25 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-03-02  4:08 [Bug c++/65273] New: Incorrect output in SIMPLE program compiled from high version gcc with O2 codeeply at foxmail dot com
2015-03-02  4:25 ` [Bug c++/65273] " pinskia at gcc dot gnu.org

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).