This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Lapack test suite results for gcc 3.3 on i686-pc-linux-gnu


--- Begin Message ---
As requested, here are lapack test suite results for g77 version 3.3
20030325 on the i686-pc-linux-gnu target. There are indeed differences
when ffloat-store is added to the OPTS flags. Though some tests pass
when float-store is active, there are some additional regressions. I
have no idea whether this is the expected or at least an acceptable
result or if there are serious problems which should be fixed for the
upcoming gcc 3.3 release.

Hope this helps,

Peter Schmid

gcc 3.3
OPTS     = -g -O2 -funroll-loops 
ced.out: CEV:    2 out of   926 tests failed to pass the threshold
ced.out: CVX:   24 out of  5196 tests failed to pass the threshold
cgd.out: CGV drivers:     65 out of   1092 tests failed to pass the
threshold
csep.out: CST:    1 out of  4662 tests failed to pass the threshold
csep.out: CST:    1 out of  4662 tests failed to pass the threshold
csep.out: CST drivers:      1 out of  11664 tests failed to pass the
threshold
csep.out: CST:    1 out of  4662 tests failed to pass the threshold
csep.out: CST drivers:      1 out of  11664 tests failed to pass the
threshold
dgd.out: DXV drivers:    200 out of   5000 tests failed to pass the
threshold
sgd.out: SGV drivers:      6 out of   1092 tests failed to pass the
threshold
sgd.out: SXV drivers:     37 out of   5000 tests failed to pass the
threshold
ssep.out: SST drivers:      1 out of  14256 tests failed to pass the
threshold
zed.out: ZEV:    3 out of   927 tests failed to pass the threshold
zed.out: ZVX:   26 out of  5198 tests failed to pass the threshold
zgd.out: ZGV drivers:     66 out of   1092 tests failed to pass the
threshold
zgd.out: ZXV drivers:     24 out of   5000 tests failed to pass the
threshold

gcc 3.3
OPTS     = -g -O2 -funroll-loops -ffloat-store
ced.out: CEV:    2 out of   926 tests failed to pass the threshold
ced.out: CVX:   22 out of  5194 tests failed to pass the threshold
csep.out: CST:    1 out of  4662 tests failed to pass the threshold
csep.out: CST drivers:      2 out of  11664 tests failed to pass the
threshold
ded.out: DES:    1 out of  3270 tests failed to pass the threshold
ded.out: DSX:    1 out of  3500 tests failed to pass the threshold
dgd.out: DXV drivers:    200 out of   5000 tests failed to pass the
threshold
sgd.out: SXV drivers:     37 out of   5000 tests failed to pass the
threshold
ssep.out: SST:    1 out of  4662 tests failed to pass the threshold
ssep.out: SST:    2 out of  4662 tests failed to pass the threshold
ssep.out: SST drivers:      1 out of  14256 tests failed to pass the
threshold
ssvd.out: SBD:      1 out of   5510 tests failed to pass the threshold
zed.out: ZEV:    5 out of   929 tests failed to pass the threshold
zed.out: ZVX:   31 out of  5203 tests failed to pass the threshold
zgd.out: ZXV drivers:     24 out of   5000 tests failed to pass the
threshold

gcc 3.2
OPTS     = -g -O2 -funroll-loops
cgd.out: CGV drivers:     66 out of   1092 tests failed to pass the
threshold
csep.out: CST:    1 out of  4662 tests failed to pass the threshold
csep.out: CST drivers:      1 out of  11664 tests failed to pass the
threshold
csvd.out: CBD:      1 out of   4085 tests failed to pass the threshold
dgd.out: DXV drivers:    200 out of   5000 tests failed to pass the
threshold
sgd.out: SGV drivers:      6 out of   1092 tests failed to pass the
threshold
sgd.out: SXV drivers:     37 out of   5000 tests failed to pass the
threshold
ssep.out: SST drivers:      1 out of  14256 tests failed to pass the
threshold
zgd.out: ZGV drivers:     64 out of   1092 tests failed to pass the
threshold
zgd.out: ZXV drivers:     24 out of   5000 tests failed to pass the
threshold



> Arrghh, sorry, I forgot I specified the options in the description 
> for 
> the LAPACK runs.
> 
> But perhaps you could perform the test I can't anymore (because I 
> canned 
>  my ix86 laptop):
> 
> Compare your run with the options above with one where you add 
> -ffloat-store to those options.  This will show if the outcome of 
> the 
> LAPACK tests is sensitive to small differences in rounding.
> 
> Thanks in advance,
> 
> -- 
> Toon Moene - mailto:toon at moene dot indiv dot nluug dot nl - phoneto: +31 346 
> 214290Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
> Maintainer, GNU Fortran 77: 
> http://gcc.gnu.org/onlinedocs/g77_news.htmlGNU Fortran 95: 
> http://gcc-g95.sourceforge.net/ (under construction)
> 
> 

--- End Message ---

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]