gcc-2.95.1 (Intel) & Stepanov Benchmark failures

Juergen Heinzl juergen@monocerus.demon.co.uk
Sat Oct 16 13:50:00 GMT 1999


Hello,

using gcc-2.95.1 on an Intel machine the results of running the
Stepanov benchmark ...
ftp://ftp.kai.com/pub/benchmarks/stepanov_v1p2.C
.. are slightly pear shaped.

In short, failures with optimizations levels 0,1 and 2 but not 3
due to NaN's.

On a Sun machine, same compiler, the ration between test0 and the
others are less than zero, not quite reasonable but I need to have
a look.

Attached are the intermediate files for the Intel based compiler
and optimizations -O0 and -O3. Might be this is related to Intels'
floating point format though.

Ta',
Juergen

-- 
\ Real name     : Jürgen Heinzl                 \       no flames      /
 \ EMail Private : juergen@monocerus.demon.co.uk \ send money instead /
-------------- next part --------------
A non-text attachment was scrubbed...
Name: opt_O0_stepanov_v1p2.ii.gz
Type: application/x-gzip
Size: 9120 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/gcc-bugs/attachments/19991016/bc5f3f18/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: opt_O0_stepanov_v1p2.s.gz
Type: application/x-gzip
Size: 17282 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/gcc-bugs/attachments/19991016/bc5f3f18/attachment-0001.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: opt_O3_stepanov_v1p2.ii.gz
Type: application/x-gzip
Size: 13018 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/gcc-bugs/attachments/19991016/bc5f3f18/attachment-0002.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: opt_O3_stepanov_v1p2.s.gz
Type: application/x-gzip
Size: 6510 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/gcc-bugs/attachments/19991016/bc5f3f18/attachment-0003.bin>


More information about the Gcc-bugs mailing list