[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: gcc && signal 6/11

On May 6, Pasi Pirhonen wrote:
> When i try to compile the latest pgp-package from
> http://www.cs.helsinki.fi/~wirzeniu/debian/index.html
> i repeatedly get signal 6 or 11 from crypto.c (-O2 or -O)
> Without optimisation it compiles just fine.

I think that's the same problem that was previously reported by Michael
Alan Dorman (signal 6 when compiling qmail).  Signal 6 is raised by
abort() during some optimisation passes when a condition that should
never happen is detected.  Signal 11 (SIGSEGV) happens when such a
condition is not detected...  :-(
Thanks for reporting, I'll take a look (I've already almost tracked down
this problem in the qmail case, but the more information the better).

BTW, that `wrong' condition that occurs here is not only exposed by the
optimiser, but introduced there as well, so the workaround is to compile
without -O.

> I really don't think that it's hw-problem as the box can easily

It's a (non-debian-related) gcc problem.


TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-alpha-request@lists.debian.org . 
Trouble?  e-mail to templin@bucknell.edu .

Reply to: