Re: must_malloc

From: Erick Gallesio <eg_at_kaolin.unice.fr>
Date: Thu, 22 Dec 1994 11:32:06 +0000

> in STk-2.1.5 [OSF1-V2.1-alpha] report error is still "buggy"
> In fact what happen is just that during the err() processing
> there is a call to must_malloc that is not successful.
> (malloc return a Null pointer). I haven't fond why, but ps report:
> S UID PID PPID %CPU PRI NI RSS WCHAN TT TIME COMMAND
> S 10061 21675 19193 12.0 41 0 1.2M event p7 0:00.51 stk-bin
>
> could it be that the process cannot allocate memory anymore (doubt it but
> you never know) ?
>
> Any similar experience ? Anyone running successfully Stk on a alpha box ?
>

I have compiled STk on a Alpha box and don't seen this problem. I have not
make a lot of tests on this architecture apart testing the error procedure
(which I knew as "buggy"). In particular, I knew that an undefined variable
leads to
an error (core dump?). This not the cas in 2.1.5

The only problem I had with Alpha is that gcc cannot compile STk (it seems
that the problem comes from GMP which cannot be compiled with gcc). So I have
configured the system with
        CC=cc configure
and it (seems to) work.

If you can give me a piece of code which conduct to the problem, I can try to
fix the problem.

                -- Erick
Received on Thu Dec 22 1994 - 11:32:07 CET

This archive was generated by hypermail 2.3.0 : Mon Jul 21 2014 - 19:38:59 CEST