Re: Random corruption in atlas computations

2004-11-02 Thread Camm Maguire
Greetings! Just a followup here. All is well with -O2. I've filed a gcc bug describing the behavior with -O3. Take care, Roman Zippel <[EMAIL PROTECTED]> writes: > Hi, > > Camm Maguire wrote: > > > Greetings! The atlas testers fail at random intervals on m68k, > > It would help a lot to k

Re: Random corruption in atlas computations

2004-10-16 Thread Rick Younie
Roman Zippel wrote: > Hi, > > Camm Maguire wrote: > > >Greetings! The atlas testers fail at random intervals on m68k, > > It would help a lot to know on which machine it fails (esp. type of cpu). Hi. I think Camm was working on crest.debian.org - 060/50. Rick --

Re: Random corruption in atlas computations

2004-10-16 Thread Roman Zippel
Hi, Camm Maguire wrote: Greetings! The atlas testers fail at random intervals on m68k, It would help a lot to know on which machine it fails (esp. type of cpu). bye, Roman

Re: Random corruption in atlas computations

2004-10-16 Thread Finn Thain
On Fri, 15 Oct 2004, Camm Maguire wrote: > Greetings! The atlas testers fail at random intervals on m68k, > preventing the package build to complete. I'm suspecting some cache > or register flushing issue. When I try to stop in gdb at the point > the corrupted value is supposedly written, the

Random corruption in atlas computations

2004-10-15 Thread Camm Maguire
Greetings! The atlas testers fail at random intervals on m68k, preventing the package build to complete. I'm suspecting some cache or register flushing issue. When I try to stop in gdb at the point the corrupted value is supposedly written, the problem never appears. Here are the symptoms: ===