- Feb 11, 2025
-
-
Markus Blatt authored
-
Markus Blatt authored
-
- Feb 09, 2025
-
-
Markus Blatt authored
-
Markus Blatt authored
-
- Feb 07, 2025
-
-
Markus Blatt authored
-
- Nov 17, 2024
-
-
Markus Blatt authored
-
Markus Blatt authored
This is in line with DUNE since 2015 and fixes problems when trying to use opm-grid as a DUNE module. In this case the CMake find module of DUNE 2.10 would be used and the check in OPM for the components would fail because of the different case used there.
-
Markus Blatt authored
Only values 0 or 1 were handled correctly. A value defined to TRUE, ON, OFF would have made the value appear in the config.h as text e.g. ``` #define VAR TRUE ``` This caused a lot of trouble with refactored DUNE build system.
-
- Nov 15, 2024
-
-
Markus Blatt authored
-
Markus Blatt authored
Without this we might pickup a wrong version that is missing macros below debian needed for unstable.
-
Markus Blatt authored
-
Markus Blatt authored
-
- Nov 07, 2024
-
-
Markus Blatt authored
-
Markus Blatt authored
We do this by backporting the commits in upstream branch for this: - 0011-Removed-HAVE_UMFPACK-alias.patch - 0012-Workaround-possibly-set-HAVE_DUNE_MODULE-DUNE-2.10.patch
-
Markus Blatt authored
-
Markus Blatt authored
Update to upstream version '2024.10+ds' with Debian dir f21a3c9db03319a02e9215b02984cf0fd57166d7
-
Markus Blatt authored
-
Markus Blatt authored
-
- Nov 06, 2024
-
-
Markus Blatt authored
-
Bård Skaflestad authored
-
Markus Blatt authored
-
Markus Blatt authored
Update to upstream version '2024.10~rc4+ds' with Debian dir ac2bd00f399688ce25638c64cfa460514e8dfe56
-
Markus Blatt authored
-
- Nov 05, 2024
-
-
Bård Skaflestad authored
Update manual pages for release
-
Markus Blatt authored
-
Markus Blatt authored
It does not have to be the full path.
-
Markus Blatt authored
-
- Oct 31, 2024
-
-
Bård Skaflestad authored
-
Bård Skaflestad authored
fixed: out-of-bounds indexing in FieldProps
-
Arne Morten Kvarving authored
this happens when grid processing removed cells (pinch, minpv, ..). in particular the actnum in the grid_ptr member (ie EclipseState::getInputGrid()) does not reflect the additional inactive cells, and thus the active_index in the Box are not the same as those in the FieldData (which has been modified in the call to FieldProps::reset_actnum). this causes oob indexing in handle_schedule_keywords / multiply_deck. this is a minimal fix to fix the oob indexing. there might be more dragons hidden in the output where the inconsistent actnum from the input grid is used.
-
- Oct 30, 2024
-
-
Bård Skaflestad authored
Record ACTIONX Wells Affected by WELPI Separately
-
Bård Skaflestad authored
This way, the simulator layer has more information by which to decide if applying a CTF update is safe/sufficient or if it needs to fully rebuild its internal representation of the simulation wells.
-
Bård Skaflestad authored
While here, also use proper Doxygen-style comments.
-
- Oct 28, 2024
-
-
Bård Skaflestad authored
Fix serialization of shared pointers
-
Arne Morten Kvarving authored
-
Arne Morten Kvarving authored
commit 067d6d44 aimed to keep the shared nature of shared pointers when deserializing. the problem is that assigning multiple shared ptrs to the same raw pointer does not enable any sharing, it only leads to multiple deletions. fix this by properly holding a shared pointer in the map. convert the adresses that is serialized to keep track of what pointers were shared on the sender side to integers (uintptr_t) rather than serializing them as void pointers. this is clearer and in principle more correct as it is rather unclear how a serializer should handle a void pointer, and it only worked by "accident" since std::is_pod_v happens to return true for a void*
-
Bård Skaflestad authored
Fix some parallel object serialization issues
-
Arne Morten Kvarving authored
-
Arne Morten Kvarving authored
-
Arne Morten Kvarving authored
-