Excellent news, I am copying upstream with your findings. And I
appologies for reporting that there was a new upstream release; I read
the numbers wrong, of course (the maintainer lowered the severity of
the bug, which allowed to to enter testing, thusly confusing me).
Thanks,
Justin
On Thu, Jan 27, 2005 at 07:23:54PM +0100, fellow wrote:
> > You might also try running it under valgrind ("valgrind torsmo") and
> > see if it will spot any invalid memory references. (valgrind is for
> > x86 only, but that's okay:)
> >
> > Justin
>
> Hi!
> First of all: sorry, but I have some exams and other activities in the
> past days, but now I'm ready for bug-hunting ;-)
>
> (...)
>
> And after some hours, I found out what the problem is (original log
> is mia, but I can reproduce it, when it's necessary):
>
> ==5830== Process terminating with default action of signal 8 (SIGFPE)
> ==5830== Integer divide by zero at address 0xB0025206
> ==5830== at 0x804E802: generate_text (torsmo.c:1129)
> ==5830== by 0x8050A01: main (torsmo.c:2289)
>
> The 1129th line:
> snprintf(p, 255, "%*u", pad_percents, (cur->swap*100) / cur->swapmax);
>
> I use a custom kernel (http://fellow.linuxforum.hu/patch/pear.html)
> without kernel module and swap support (because I don't need them).
> Now I turned on swap, make a swapfile, reboot... and torsmo-0.18 works
> great!
>
> Please send a (better) bugreport to the author about this bug.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
