added test cases for mp, p1, and p2added DMMGSetFromOptions() for FortranHg-commit: 7252c216c1b03955f761e13b5001e1ebf5960e77
added DMMGSetFromOptions() for use with DMMGSetSNES()Hg-commit: 3ea7cf5f0b7da0cd72ac5580040339eec4823885
commit after mergeHg-commit: 78daef0359b6275ad4044a71b43f9e7286eb5096
changed DA_COOKIE to DM_COOKIE changed classname of DA, DMComposite and Sliced to DM, also gave them all type_namesHg-commit: 9299c78699d24417f4d7ca39368733ce3a989ace
adic fixes from barryHg-commit: d9adc325e929498574353c6e81ef98746ea4f26e
made new ex32.c (ex19.c) that ONLY does the multiphysics, much simplier easier to understandHg-commit: 1f7bb1d1679b17450331aab3611975049b0ecf5f
fixed some code in tfs to handle long doublesadded crude support for using Intel's icc _quad type instead of long double (cause long double does not match real*16Hg-commit: f61b45ea0e8f9447c84a535
fixed some code in tfs to handle long doublesadded crude support for using Intel's icc _quad type instead of long double (cause long double does not match real*16Hg-commit: f61b45ea0e8f9447c84a5352d72cec225a371329
show more ...
mv zpetsc.h (a meaningless name) to fortranimpl.h (a slightly more meaningful name)Hg-commit: d2687f12f92bff49ceff434100d018f6898829c3
added PetscObjectAllocateFortranPointers() replaced direct use of malloc for theseHg-commit: 2c4d80a83ccb656c1ab09c5dcec9c685c121655f
started support for in-block shift for ILUHg-commit: eb0884a63911c1ee92e1267fbd4eb2e94fcb5b3d
Small change in DAMG for MeshHg-commit: 16aa9d5dc774e2480a1175dcfe202dc2148bd8ca
Compile fixesHg-commit: ffc891097b62c4692d63af2797d41875fca28a23
petsc-maint 16799 ==6657== Conditional jump or move depends on uninitialised value(s) ==6657== at 0x8735AD1: petscoptionsgetstring_ (in /home/go
petsc-maint 16799 ==6657== Conditional jump or move depends on uninitialised value(s) ==6657== at 0x8735AD1: petscoptionsgetstring_ (in /home/gormo/fluidity/trunk/bin/petsc_readnsolve) ==6657== by 0x80F426D: petsc_readnsolve_options.878 (petsc_readnsolve.F90:264) ==6657== by 0x80F197C: petsc_readnsolve_ (petsc_readnsolve.F90:73) ==6657== by 0x80F18AD: main (petsc_readnsolve_main.cpp:28)moved zpetsc.h to private/zpetsc.hchanged Fortran interface function PetscOptionsGetXXX functions to allow passing PETSC_NULL_TRUTH as flag argumentHg-commit: a08dc4d2e15e97c0d3849b1cb02e6c1b0214b0c8
FAS seems to work for bratu- Finally got constant section correct for coarse solve- Set SNES iterates for FAS- Check for convergence after postsmooth as wellHg-commit: 0864d81f6384e042be22280180
FAS seems to work for bratu- Finally got constant section correct for coarse solve- Set SNES iterates for FAS- Check for convergence after postsmooth as wellHg-commit: 0864d81f6384e042be22280180c0bb982b6d318b
Fixed compilationHg-commit: 54cf8a6dd5222c716f6a9bb589bc104a62c0d252
Work on adding constant to unstructured nonlinear solver - This should be redone - Added axpy() to SectionHg-commit: 385fbd72c0396d598f1d9a76848e80afaa7a6f7d
Multilevel FAS seems to work- Need to mark boundary cells in coarse meshes- Added boundary cell marking with boundary edges marked- Can now give DMMG a negative number of levels, which causes it t
Multilevel FAS seems to work- Need to mark boundary cells in coarse meshes- Added boundary cell marking with boundary edges marked- Can now give DMMG a negative number of levels, which causes it to ignore -dmmg_nlevelsHg-commit: ba8dc9b9d32e3ec38df6496519b82d6f87f0a3c4
Small fixesHg-commit: 408410e2ad7394760aab783ea5b8470fbfa84fab
MergeHg-commit: 390efcf688260cab5f228bb6fe3a24edd2cc463f
Work on the null space for Stokes in FASHg-commit: aa76afe07d1811ca30eda765fb26d4f4635f9c2b
mergeHg-commit: 7414b34d106e13dde45172d49c8fc215d176acfe
MergeHg-commit: 0ea643e5e5a4778b32bccc4e39377f439d272331
More work on FAS for StokesHg-commit: ebc04b9ead4415aaca3e404738053ec18e7f4d77
replace // with /* */ for comments as complained by the nightly testHg-commit: 268be99ed1c601ff04ee54da613d44b43f3ad6a1
mergeHg-commit: 3fb77c8f0fffe636795e142bd1be924d13e15296
1...<<31323334353637383940>>...43