Eric Lemings wrote:
Also runs successfully when built with Sun C++ 5.8 in
BUILDMODE=threads,shared,debug,wide.
Right, that's what our nightly build results indicate.
What I said was 15S (threads,archive,debug,wide), not 15D.
Note that we're using the base 5.8 release, not the latest
patch (the latest patch installed on marbles is 121018-13
from 2008/01/02). But they both exhibit the same problem.
The stack trace from the 15S build is attached.
Martin
-----Original Message-----
From: Martin Sebor [mailto:mse...@gmail.com] On Behalf Of Martin Sebor
Sent: Wednesday, April 09, 2008 4:30 PM
To: de...@stdcxx.apache.org
Subject: Re: [jira] Commented: (STDCXX-828) [AMD64] SIGABRT
in 27.istream.fmat.arith
Eric Lemings wrote:
-----Original Message-----
From: Martin Sebor [mailto:mse...@gmail.com] On Behalf Of
Martin Sebor
Sent: Wednesday, April 09, 2008 10:09 AM
To: de...@stdcxx.apache.org
Subject: Re: [jira] Commented: (STDCXX-828) [AMD64] SIGABRT
in 27.istream.fmat.arith
The test also fails in 15D builds (on AMD64?) so you should be able
to get a full stack trace.
It doesn't fail in manual testing with this build config. :(
Have you tried Sun C++ 5.8?
I see SIGABRT in 8D, 12D, and 12S with Sun C++ 5.9/Solaris 10/AMD64
and in 12D and 15S with Sun C++ 5.8/Solaris 10/AMD64 in nightly
builds.
6 messages in org.apache.incubator.stdcxx-devRe: [jira] Commented: (STDCXX-828) [A...
.txt
