[pooma-dev] Pooma Code Failures
Julian C. Cummings
cummings at linkline.com
Wed Oct 10 10:50:35 UTC 2001
I just wanted to note that I looked at the failure of
dynamic_array_test5. The cause of the failure is as
noted in the code comments; namely that patch-specific
destroy and copy requests using iterator pairs will
fail to compile if the iterators are not convertible
to const int*. Apparently, the standard iterators in
gcc 3.x are not convertible to raw pointers, so this
part of the test code fails. If we skip this one line
of dynamic_array_test5.cpp using a #if 0, as suggested
in the code comments, the remainder of the code runs
just fine under gcc 3.x.
I don't see any nice way of fixing this problem, unless
we make separate DestroyEventCodes for IPDomain<int*>
and IPDomain<Iter>, so that we can dynamically cast the
ObserverEvent to the appropriate type of DestroyEvent.
We would have to protect this extra code with a macro
guard, to avoid the extra code in cases where there is
no difference between the standard iterator and a raw
int* or const int*.
-- Julian C.
> -----Original Message-----
> From: Jeffrey Oldham [mailto:oldham at codesourcery.com]
> Sent: Monday, October 01, 2001 1:47 PM
> To: pooma-dev at pooma.codesourcery.com
> Subject: [pooma-dev] Pooma Code Failures
>
>
>
> One of this week's tasks is to eliminate all these Pooma regressions.
> I used sequential Linux and gcc3.1 so some of these might be gcc3.1
> failures.
>
> 1) Does anyone have a failure list when using KCC?
> 2) Some of the examples and benchmarks run for a long time. It would
> be nice to find these and reduce their running time.
>
> 34 ( 14%) tests FAIL
> examples.field.caramana.caramana
> : FAIL
> examples.field.laplace.laplace
> : FAIL
> examples.field.laplace2.laplace2
> : FAIL
> examples.field.scalaradvection.scalaradvection
> : FAIL
> examples.field.scalaradvection1d.scalaradvection1d
> : FAIL
> examples.field.statigraphicflow.statigraphicflow
> : FAIL
> examples.lattice.coordinate
> : FAIL
> examples.particles.bounce.bounce
> : FAIL
> examples.particles.oscillation.oscillation
> : FAIL
> examples.particles.pic2d.pic2d
> : FAIL
> examples.solvers.blockdecomposed.fivepoint.fivepointsolve
> : FAIL
> examples.solvers.blockdecomposed.ninepoint.ninepointsolve
> : FAIL
> regressions.array.scalaradvection
> : FAIL
> regressions.array.scalaradvectionxb
> : FAIL
> regressions.connect.lux.bytefield
> : FAIL
> regressions.connect.lux.lux_test1
> : FAIL
> regressions.connect.lux.lux_test2
> : FAIL
> regressions.databrowser.testdatabrowser
> : FAIL
> regressions.dynamicarray.dynamic_array_test5
> : FAIL
> regressions.field.crossbox
> : FAIL
> regressions.field.localpatch
> : FAIL
> regressions.field.nearestneighbors
> : FAIL
> regressions.field.positions
> : FAIL
> regressions.io.filesetreadertest1
> : FAIL
> regressions.io.filesetreadertest2
> : FAIL
> regressions.particles.interpolate
> : FAIL
> regressions.particles.particle_bench1
> : FAIL
> regressions.particles.particle_bench2
> : FAIL
> regressions.particles.particle_bench3
> : FAIL
> regressions.particles.particle_bench4
> : FAIL
> regressions.particles.particle_test1
> : FAIL
> regressions.particles.particle_test2
> : FAIL
> regressions.particles.particle_test3
> : FAIL
> regressions.particles.particle_test4
> : FAIL
>
> Thanks,
> Jeffrey D. Oldham
> oldham at codesourcery.com
>
More information about the pooma-dev
mailing list