Very strange. I suspect that it's a compiler bug, but I don't know for
sure. Are using the same version of gcc on both Solaris and AIX.
Dave
"Laidlaw,
Angus" To: "'[EMAIL PROTECTED]'"
<alaidlaw@axs <[EMAIL PROTECTED]>
one.com> cc: (bcc: David N Bertoni/CAM/Lotus)
Subject: RE: Xalan 1.1 on AIX with g++
07/03/2001
08:29 AM
Please
respond to
xalan-dev
I am blowing up in XalanDOMString::operator=(this, theRHS)
The vector for theRHS is corrupt. _M_finish has a value
that is 265 million greater then it should be. _M_end_of_storage
is correct.
Just a note: our code is working with xalan/xercies on
NT, Solaris (built with g++) and hpux (built with hpux aCC) it is only
on aix that I am having this problem.
Angus
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Monday, July 02, 2001 2:55 PM
To: [EMAIL PROTECTED]
Subject: Re: Xalan 1.1 on AIX with g++
Hmm, we've never built Xalan on AIX using g++. It could be a compiler bug,
a bug in your code, or a bug in the Xalan code. Is the crash in
XalanDOMString's copy constructor? Does this happen inside of Xalan, or in
your own code?
Dave
"Laidlaw,
Angus" To:
"'[EMAIL PROTECTED]'"
<alaidlaw@axs <[EMAIL PROTECTED]>
one.com> cc: (bcc: David N
Bertoni/CAM/Lotus)
Subject: Xalan 1.1 on AIX with
g++
07/02/2001
11:31 AM
Please
respond to
xalan-dev
Is anybody using Xalan-c 1.1 on AIX 4.3.2 built with g++? I am
having problems
when a function returns a XalanDOMSString. Sometimes
XalanDOMSString val = function();
works, sometimes it segmentation faults. If I make &val an
arguement to the function
and return the value through it that way it works.
Angus Laidlaw
- Xalan 1.1 on AIX with g++ Laidlaw, Angus
- Re: Xalan 1.1 on AIX with g++ David_N_Bertoni
- RE: Xalan 1.1 on AIX with g++ Laidlaw, Angus
- RE: Xalan 1.1 on AIX with g++ David_N_Bertoni
- RE: Xalan 1.1 on AIX with g++ Laidlaw, Angus
