Hello,
Long story short: I noticed some "*** Error: Time for operations was
zero" messages when running the testdsecnd binary so I checked the code
(secondtst.f and dsecndtst.f as well).
I am not a fortran expert, but I noticed that 1,000,000 DAXPY / SAXPY
operations should be performed but NMAX is set to 100 and ITS is set to
5000 ... Is it a typo, or did I miss something ?
I don't get any error messages with the following patches:
[mbozzore at stakhanov FC]$ diff -U4 lapack-3.1.1/INSTALL/secondtst.f
/usr/src/redhat/BUILD/lapack-3.1.1/INSTALL/secondtst.f
--- lapack-3.1.1/INSTALL/secondtst.f 2006-11-12 16:55:56.000000000
-0500
+++ /usr/src/redhat/BUILD/lapack-3.1.1/INSTALL/secondtst.f
2008-09-23 13:05:23.000000000 -0400
@@ -5,9 +5,9 @@
* November 2006
*
* .. Parameters ..
INTEGER NMAX, ITS
- PARAMETER ( NMAX = 100, ITS = 5000 )
+ PARAMETER ( NMAX = 100, ITS = 10000 )
* ..
* .. Local Scalars ..
INTEGER I, J
REAL ALPHA, AVG, T1, T2, TNOSEC
[mbozzore at stakhanov FC]$ diff -U4 lapack-3.1.1/INSTALL/dsecndtst.f
/usr/src/redhat/BUILD/lapack-3.1.1/INSTALL/dsecndtst.f
--- lapack-3.1.1/INSTALL/dsecndtst.f 2006-11-12 16:55:56.000000000
-0500
+++ /usr/src/redhat/BUILD/lapack-3.1.1/INSTALL/dsecndtst.f
2008-09-23 12:47:09.000000000 -0400
@@ -5,9 +5,9 @@
* November 2006
*
* .. Parameters ..
INTEGER NMAX, ITS
- PARAMETER ( NMAX = 100, ITS = 5000 )
+ PARAMETER ( NMAX = 100, ITS = 10000 )
* ..
* .. Local Scalars ..
INTEGER I, J
DOUBLE PRECISION ALPHA, AVG, T1, T2, TNOSEC
[mbozzore at stakhanov FC]$
Best regards,
Mehdi
Mehdi Bozzo-Rey <mailto:mbozzore@Domain.Removed>
HPC Solution Developer
Platform OCS5
<http://www.platform.com/Products/platform-open-cluster-stack5>
Platform computing
Phone: +1 905 948 4649
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://lists.cs.utk.edu/private/lapack/attachments/20080923/095dd09c/attachment.html
|