EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  <20072008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024  Index 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  <20072008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
<== Date ==> <== Thread ==>

Subject: RE: bug in Red Hat Linux g++ 3.2.3
From: "Jeff Hill" <[email protected]>
To: "'Dirk Zimoch'" <[email protected]>
Cc: "'EPICS Tech Talk'" <[email protected]>
Date: Mon, 27 Aug 2007 09:34:29 -0600
Hi Dirk,

Ernest indicated that this might be resolved by patching the compiler. If
that is possible then I am disinclined to create a workaround, but if not
then I will see what can be done.

Jeff

-----Original Message-----
From: Dirk Zimoch [mailto:[email protected]] 
Sent: Monday, August 27, 2007 2:36 AM
To: Jeff Hill
Cc: 'EPICS Tech Talk'
Subject: Re: bug in Red Hat Linux g++ 3.2.3

Hi

The bug also occurs with -O2, only -O1 is safe.
gcc 3.2.3 is the standard version on Scientific Linux 3, which we still 
have in use at the SLS. I would very much appreciate a workaround in the 
EPICS code.

Dirk

Jeff Hill wrote:
>  
> 
> All,
> 
>  
> 
> There is a bug in Red Hat Linux g++ 3.2.3 when -O3 is specified that 
> causes CA R3.14.9 to fail. The bug is known not to occur with several 
> other g++ versions, but many versions have not been tested. I also don't 
> know what versions of Linux g++ are most commonly used at the various 
> EPICS sites.
> 
>  
> 
> So I am curious if this is a substantial issue for the EPICS community 
> or if it can be safely ignored? If defective compilers are still in 
> widespread use I could look for a workaround in the EPICS source code.
> 
>  
> 
> Jeff
> 
> During testing of R3.14.9 I see a bug showing up only in optimized 
> builds on Linux using g++ 3.2.3 that I don't see with other compilers. 
> This bug appears to be related to failure of this particular compiler to 
> follow the strict alias rules specified in the standard when optimizing 
> the byte swapping templates installed for R3.14.9.
> 
> *Problems*
> 
> Red hat Linux g++ 3.2.3 when -O3 is specified
> 
> *No problems*
> 
> mingw g++ 3.4.2 -O3
> 
> Tornado 2.02 g++ -O2
> 
> Microsoft C++ 2005 full optimization builds
> 
> 
> ______________________________________________________
> Jeffrey O. Hill           Internet     [email protected] 
> <mailto:[email protected]>
> LANL MS H820              Voice        505 665 1831
> Los Alamos NM 87545 USA   FAX          505 665 5107
> 
>  
> 

-- 
Dr. Dirk Zimoch
Paul Scherrer Institut, WBGB/006
5232 Villigen PSI, Switzerland
Phone +41 56 310 5182


Replies:
Re: bug in Red Hat Linux g++ 3.2.3 Eric Norum
References:
bug in Red Hat Linux g++ 3.2.3 Jeff Hill
Re: bug in Red Hat Linux g++ 3.2.3 Dirk Zimoch

Navigate by Date:
Prev: Re: Port of EPICS 3.14.9 to ETRAX CRIS architecture - Strange Data Dirk Zimoch
Next: Re: Port of EPICS 3.14.9 to ETRAX CRIS architecture - Strange Data Andrew Johnson
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  <20072008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
Navigate by Thread:
Prev: Re: bug in Red Hat Linux g++ 3.2.3 Dirk Zimoch
Next: Re: bug in Red Hat Linux g++ 3.2.3 Eric Norum
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  <20072008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
ANJ, 10 Nov 2011 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·