« Return to Thread: catchedge detection error

Re: catchedge detection error

by Clint Whaley :: Rate this Message:

Reply to Author | View in Thread

Jess,

I take it this was the problem where you were doing make -j 16, which never
works for ATLAS?  AFAIK, I have answered all the outstanding suppport
questions there (I've been sick, and am behind on everything).

Cheers,
Clint

I was trying to compile the latest (3.8.3) ATLAS on a 16-core (4x
quad-core Intel Xeon) machine running Centos 5.2, with a custom built
gcc 4.2.3 that passed all build tests before being used.

ATLAS gave the cryptic error I've pasted below (logs attached).

Here's the configure command that I used just beforehand - the 2400
comes from /proc/cpuinfo for one processor, and I built wiht "make -j 16":
  /tmp/cs1jrj/ATLAS/configure --prefix=/opt/packages/gnu/atlas/3.8.3 \
-b 64 -D c -DPentiumCPS=2400  -Fa alg -fPIC \
--with-netlib-lapack=/opt/packages/gnu/lapack/3.2/lib64/liblapack.a

Here's the error message, with some of the surrounding output:
...
make[3]: *** [res/atlas_cacheedge.h] Error 2
make[3]: Leaving directory `/tmp/cs1jrj/ATLAS/build/tune/blas/gemm'
make[2]: *** [/tmp/cs1jrj/ATLAS/build/tune/blas/gemm/res/atlas_cachedge.h]
Error 2
make[2]: Leaving directory `/tmp/cs1jrj/ATLAS/build/bin'
ERROR 639 DURING CACHE EDGE DETECTION!!.
make[2]: Entering directory `/tmp/cs1jrj/ATLAS/build/bin'
cd /tmp/cs1jrj/ATLAS/build ; make error_report
make[3]: Entering directory `/tmp/cs1jrj/ATLAS/build'
make -f Make.top error_report
make[4]: Entering directory `/tmp/cs1jrj/ATLAS/build'
uname -a 2>&1 >> bin/INSTALL_LOG/ERROR.LOG
gcc -v 2>&1  >> bin/INSTALL_LOG/ERROR.LOG

...

I wondered if anyone has seen this error before?  I can't find it in the
errata.

Thanks

Jess
>From math-atlas-devel-bounces@...  Mon Mar 30 05:51:37 2009
Return-Path: <math-atlas-devel-bounces@...>
X-Original-To: whaley@...
Delivered-To: whaley@...
X-ACL-Warn:
Date: Mon, 30 Mar 2009 11:50:50 +0100
From: Jessica Jones <j.r.jones@...>
Organization: Bath University Computing Services
User-Agent: Thunderbird 2.0.0.19 (X11/20090105)
To: "List for developer discussion,
        NOT SUPPORT." <math-atlas-devel@...>
X-kerberosV-authenticator: cs1jrj@...
X-BUCS: cs1jrj
X-Scanner: 3200dbac1a5436572f297700d70b83b6f6a54ea2
X-Headers-End: 1LoF5J-0006p2-1I
Subject: Re: [atlas-devel] catchedge detection error
X-BeenThere: math-atlas-devel@...
X-Mailman-Version: 2.1.9
Precedence: list
Reply-To: "List for developer discussion,
        NOT SUPPORT." <math-atlas-devel@...>
List-Id: "List for developer discussion,
        NOT SUPPORT." <math-atlas-devel.lists.sourceforge.net>
List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/math-atlas-devel>,
        <mailto:math-atlas-devel-request@...?subject=unsubscribe>
List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum_name=math-atlas-devel>
List-Post: <mailto:math-atlas-devel@...>
List-Help: <mailto:math-atlas-devel-request@...?subject=help>
List-Subscribe: <https://lists.sourceforge.net/lists/listinfo/math-atlas-devel>,
        <mailto:math-atlas-devel-request@...?subject=subscribe>
Content-Type: multipart/mixed; boundary="===============4518060735646612051=="
Errors-To: math-atlas-devel-bounces@...
X-Spam-Checker-Version: SpamAssassin 3.0.2 (2004-11-16) on mail0.cs.utsa.edu
X-Spam-Level:
X-Spam-Status: No, score=-2.4 required=5.0 tests=AWL,BAYES_00,HTML_20_30,
        HTML_MESSAGE autolearn=no version=3.0.2
Status: RO

Content-Type: multipart/alternative;
        boundary="------------070203010806050403090609"

Content-Type: text/plain; charset=ISO-8859-1; format=flowed

I assume the lack of response means no one has seen it before, so I've
added it to the tracker.

Thanks

Jess

On 29/03/09 18:23, Jess Jones wrote:

> I was trying to compile the latest (3.8.3) ATLAS on a 16-core (4x
> quad-core Intel Xeon) machine running Centos 5.2, with a custom built
> gcc 4.2.3 that passed all build tests before being used.
>
> ATLAS gave the cryptic error I've pasted below (logs attached).
>
> Here's the configure command that I used just beforehand - the 2400
> comes from /proc/cpuinfo for one processor, and I built wiht "make -j 16":
>   /tmp/cs1jrj/ATLAS/configure --prefix=/opt/packages/gnu/atlas/3.8.3 \
> -b 64 -D c -DPentiumCPS=2400  -Fa alg -fPIC \
> --with-netlib-lapack=/opt/packages/gnu/lapack/3.2/lib64/liblapack.a
>
> Here's the error message, with some of the surrounding output:
> ...
> make[3]: *** [res/atlas_cacheedge.h] Error 2
> make[3]: Leaving directory `/tmp/cs1jrj/ATLAS/build/tune/blas/gemm'
> make[2]: *** [/tmp/cs1jrj/ATLAS/build/tune/blas/gemm/res/atlas_cachedge.h]
> Error 2
> make[2]: Leaving directory `/tmp/cs1jrj/ATLAS/build/bin'
> ERROR 639 DURING CACHE EDGE DETECTION!!.
> make[2]: Entering directory `/tmp/cs1jrj/ATLAS/build/bin'
> cd /tmp/cs1jrj/ATLAS/build ; make error_report
> make[3]: Entering directory `/tmp/cs1jrj/ATLAS/build'
> make -f Make.top error_report
> make[4]: Entering directory `/tmp/cs1jrj/ATLAS/build'
> uname -a 2>&1 >> bin/INSTALL_LOG/ERROR.LOG
> gcc -v 2>&1  >> bin/INSTALL_LOG/ERROR.LOG
>
> ...
>
> I wondered if anyone has seen this error before?  I can't find it in the
> errata.
>
> Thanks
>
> Jess
> ------------------------------------------------------------------------
>
> ------------------------------------------------------------------------------
> ------------------------------------------------------------------------
>
> _______________________________________________
> Math-atlas-devel mailing list
> Math-atlas-devel@...
> https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

**************************************************************************
** R. Clint Whaley, PhD ** Assist Prof, UTSA ** www.cs.utsa.edu/~whaley **
**************************************************************************

------------------------------------------------------------------------------
_______________________________________________
Math-atlas-devel mailing list
Math-atlas-devel@...
https://lists.sourceforge.net/lists/listinfo/math-atlas-devel

 « Return to Thread: catchedge detection error