cc1plus 3.4 segfault.

[Karel Gardas]

On Sun, 15 Jun 2003, Michael Ritzert wrote:

> Hi Karel,
>
> On Sat, Jun 07, 2003 at 04:55:58PM +0200, Karel Gardas wrote:
> > On Fri, 6 Jun 2003, Jim Wilson wrote:
> >
> > > Karel Gardas wrote:>
> > > > Is this well known bug, or should I enter bugzilla report including
> > > > preprocessed sources?
> > >
> > > Submit a bug report.  It doesn't look familiar to me, and apparently not
> > > to anyone else today.
> >
> > ACK. BTW: I have found that it only happens when I use PCH, w/o PCH the
> > problematic file compiles well. I will try to post more info next week.
>
> Do you have any news on this?

I'm not able to duplicate this problem with gcc 3.4 20030617.

Cheers,

Karel

> I am stumbling over the same error with my project, too:
>
> Program received signal SIGSEGV, Segmentation fault.
> ggc_set_mark (p=0x4) at /home/ritzert/gcc/HEAD/gcc/gcc/ggc-page.c:1165
> 1165      if (entry->in_use_p[word] & mask)
> (gdb) p word
> $1 = 38487494
> (gdb) up
> #1  0x08086dfa in gt_ggc_mx_lang_tree_node (x_p=0x59c0aab4) at gtype-cp.h:98
> 98         xlimit = ((union lang_tree_node *)TREE_CHAIN (&(*xlimit).generic));
> (gdb) up
> #2  0x08086f4c in gt_ggc_mx_lang_tree_node (x_p=0x5b25c8b4) at gtype-cp.h:324
> 324               gt_ggc_m_9tree_node ((*x).identifier.error_locus);
> (gdb) up
> #3  0x082760c9 in gt_ggc_m_P9tree_node4htab (x_p=0x57850180) at gtype-desc.c:1093
> 1093              gt_ggc_m_9tree_node ((*x).entries[i0]);
>
> Unfortunately I can't provide a testcase, since it's 200k of confidential code.
>
> My platform is i686-pc-linux-gnu.
>
> Michael
>
>

--
Karel Gardas                  kgardas@objectsecurity.com
ObjectSecurity Ltd.           http://www.objectsecurity.com