Re: [BUG REPORT] gtags segmentation fault

[Hideki IWAMOTO]

On Sat, 14 Jun 2003 20:20:52 +0900, Hideki IWAMOTO wrote...
> On Sat, 14 Jun 2003 17:27:26 +0900, Hideki IWAMOTO wrote...
> > On Sat, 14 Jun 2003 16:16:45 +0900, Hideki IWAMOTO wrote...
> > > While gtags processes the source code of linux-2.4.21, segmentation fault 
> > > occurs.
> > > When global-4.5.2 are used, this does not happen.
> > > 
> > > --------------+------------------------------
> > >               |     version of global
> > >   source code | 4.5.2     4.5.3    cvs HEAD
> > > --------------+------------------------------
> > >  linux-2.4.20 |   OK        OK        OK
> > >  linux-2.4.21 |   OK        NG        NG
> > > --------------+------------------------------
> > > 
> > > I am going to investigate in detail from now.
> > > 
> > 
> > Though regrettable, since I do not understand the source code of db, I 
> > cannot understand what has happened.
> 
> This is another information.
> 
> It is dependent on GTAGSCACHE whether it happens or not.

I investigated at the time of when this problem would come to occur.
It does not occur as of April 16, but happens on April 20.
Only gctags is changed in the meantime and gtags is not changed.

----------------------+--------
 date                 | result
----------------------+--------
 2003-04-01 00:00 JST | OK
 2003-04-15 00:00 JST | OK
 2003-04-16 10:00 JST | OK
 2003-04-20 00:00 JST | NG
 2003-05-01 00:00 JST | NG
----------------------+--------

Next, I used it combining gctags and gtags from which the date differs.
It seems that a cause is in change of gctags.

-----------------------+--------
 gctags     gtags      | result
-----------------------+--------
 2003-04-16 2003-04-16 | OK
 2003-04-20 2003-04-16 | NG
 2003-04-16 2003-04-20 | OK
 2003-04-20 2003-04-20 | NG
-----------------------+--------

----
Hideki IWAMOTO  address@hidden