Hi Mark,
I believe the difference should come more from the processor and the pgi
compiler itself than the Linux distribution.
I would still validate your LAPACK build on 64 bits.
Those testings are really torture tests and it shouldn't affect the result on
your program.
Regards,
Julie
On Apr 14, 2011, at 12:26 PM, Mark Dixon wrote:
On Wed, 13 Apr 2011, julie langou wrote:
Hi Mark,
I finally found a machine with the PGI compiler.
I tried both 32 and 64-bit version and everything was fine. I did not use
any special flags for compilation. Output is below
I am using pgf90 11.3-0 64-bit target on x86-64 Linux -tp gh
Julie
Hi Julie,
Firstly, my sincere thanks in persevering this far with my report.
However, I'm at a loss to explain what is going on. Your result is similar
to what I saw under 32-bit mode, but not 64-bit. Attempting to reproduce
your 64-bit compile as much as possible resulted 843 numerical errors.
* "pgf90 -V" gives:
pgf90 11.3-0 64-bit target on x86-64 Linux -tp nehalem
* Relevant bit of BLAS include file:
FORTRAN = pgf90
OPTS =
DRVOPTS =
NOOPT =
LOADER = pgf90
LOADOPTS =
* Relevant bit of LAPACK 3.3.0 include file:
FORTRAN = pgf90
OPTS =
DRVOPTS =
NOOPT =
LOADER = pgf90
LOADOPTS =
* Output of your testing check script appended below - 843 failures.
Could it be related to Linux distribution? I'm running RHEL 5.6 (but with
a 5.5 kernel).
Mark
--
-----------------------------------------------------------------
Mark Dixon Email : m.c.dixon@Domain.Removed
HPC/Grid Systems Support Tel (int): 35429
Information Systems Services Tel (ext): +44(0)113 343 5429
University of Leeds, LS2 9JT, UK
-----------------------------------------------------------------
******************************************************
RESULTS
- numerical failure:
CST drivers: 4 out of 11664 tests failed to pass the threshold
CPB drivers: 2 out of 4750 tests failed to pass the threshold
CLS drivers: 1 out of 65268 tests failed to pass the threshold
DXV drivers: 200 out of 5000 tests failed to pass the threshold
DBD drivers: 56 out of 8360 tests failed to pass the threshold
DBD drivers: 56 out of 8360 tests failed to pass the threshold
DBD drivers: 56 out of 8360 tests failed to pass the threshold
DBD drivers: 56 out of 8360 tests failed to pass the threshold
DBD drivers: 56 out of 8360 tests failed to pass the threshold
SXV drivers: 37 out of 5000 tests failed to pass the threshold
SBD drivers: 56 out of 8360 tests failed to pass the threshold
SBD drivers: 56 out of 8360 tests failed to pass the threshold
SBD drivers: 56 out of 8360 tests failed to pass the threshold
SBD drivers: 56 out of 8360 tests failed to pass the threshold
SBD drivers: 56 out of 8360 tests failed to pass the threshold
SLS drivers: 3 out of 65268 tests failed to pass the threshold
ZXV drivers: 24 out of 5000 tests failed to pass the threshold
CPB: 11 out of 3458 tests failed to pass the threshold
SPB: 1 out of 3458 tests failed to pass the threshold
value of largest test error = 0.189E+01
value of largest test error = 0.160D+01
value of largest test error = 0.197E+01
value of largest test error = 0.189D+01
value of largest test error = 0.148D+01
- illegal error: 0
- info error: 0
- severe error 0
_______________________________________________
Lapack mailing list
Lapack@Domain.Removed
http://lists.eecs.utk.edu/mailman/listinfo/lapack
|