SWIFTsim merge requestshttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests2024-02-02T08:41:25Zhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1853Chemistry: Metal fluxes for GEAR and AGORA (and QLA)2024-02-02T08:41:25ZYolan UyttenhoveChemistry: Metal fluxes for GEAR and AGORA (and QLA)I also implemented the metal advection for the `GEAR` and `AGORA` schemes (`AGORA` seems to just be a specific variant of GEAR?). See also #834 and !1825.
The `QLA` chemistry scheme looks at first glance like it is just a rename of the ...I also implemented the metal advection for the `GEAR` and `AGORA` schemes (`AGORA` seems to just be a specific variant of GEAR?). See also #834 and !1825.
The `QLA` chemistry scheme looks at first glance like it is just a rename of the `none` chemistry scheme? So, I added the same no-ops for the metal advection, so that we can also compile with `gizmo-mfv` and `QLA` chemistry.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1839update Grackle cooling2023-12-30T13:47:43ZYves Revazupdate Grackle coolingAdd Grackle parameters to form H2 on dust grainsAdd Grackle parameters to form H2 on dust grainsMatthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1811Small changes in the Grackle cooling2023-11-08T15:15:09ZYves RevazSmall changes in the Grackle coolingSome imporovements of the Grackle coolingSome imporovements of the Grackle coolingMatthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1741Grackle Cooling updates2023-09-22T07:47:08ZYves RevazGrackle Cooling updatesAdd new parameters that allow now to reproduce the Iliev06 test0.Add new parameters that allow now to reproduce the Iliev06 test0.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1742GEARRT: update grackle2023-07-25T15:11:55ZMladen IvkovicGEARRT: update grackle- update the grackle API for newer version of grackle
- includes minor other fixes
- I spent waaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaay to much time trying to get autoconf to somehow check for the correct grackle version being installed by expli...- update the grackle API for newer version of grackle
- includes minor other fixes
- I spent waaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaay to much time trying to get autoconf to somehow check for the correct grackle version being installed by explicitly checking the grackle version. Can't get it to work. So now I'm checking for 2 specific new functions in the grackle API, and hoping that will catch most issues with deprecated grackle versions.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1703Fix GEAR pressure floor2023-03-28T08:01:08ZMatthieu SchallerFix GEAR pressure floorBig change to the pressure floor API and corresponding changes to the hydro flavours as we need to pass the engine-carried floor
throughout. This also has consequences on the cooling API and on the feedback.
At the same time, cleaned up...Big change to the pressure floor API and corresponding changes to the hydro flavours as we need to pass the engine-carried floor
throughout. This also has consequences on the cooling API and on the feedback.
At the same time, cleaned up a bit the grackle cooling header.
Also added a few checks to prevent certain feedback modes and hydro models to run with the pressure floor where it is not implemented.
Cleaned up the Agora galaxy example.
Todo:
- [x] Add doxygen strings.
- [x] Apply the same API change to the other hydro schemes.
- [x] Add the higher-res. ICs to the repo. Update run script.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1488add a new paprameter density_threashold in the GEAR star formation scheme.2022-02-04T10:50:58ZYves Revazadd a new paprameter density_threashold in the GEAR star formation scheme.Add a new paprameter density_threashold in the GEAR star formation scheme.
The value is given in g/cm3.Add a new paprameter density_threashold in the GEAR star formation scheme.
The value is given in g/cm3.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1341Use only min mass gas in time step2021-10-15T10:58:12ZLoic HausammannUse only min mass gas in time stepFix #758
My student finished its simulations and everything seems alright. Do you wish to keep the current behavior for your simulations or are you fine if merge it like this?Fix #758
My student finished its simulations and everything seems alright. Do you wish to keep the current behavior for your simulations or are you fine if merge it like this?Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1307Sink star formation2021-03-24T18:05:24ZLoic HausammannSink star formationImplement the star formation from the sink particles. The idea is to reuse the "normal" star formation task but to loop over the sink particles and not the hydro ones. I am also slightly changing the dependencies in order to move it afte...Implement the star formation from the sink particles. The idea is to reuse the "normal" star formation task but to loop over the sink particles and not the hydro ones. I am also slightly changing the dependencies in order to move it after the sink business.
I still need to review the code a fix a small issue.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1210Sink merger2021-02-17T18:10:45ZLoic HausammannSink mergerI have implemented the merger of two sink particles:
![dependency_graph](/uploads/f9eb976777d0db687f534d0c84307247/dependency_graph.png)
I realized that my logic for the activation of sink's tasks and my interaction loop for the comput...I have implemented the merger of two sink particles:
![dependency_graph](/uploads/f9eb976777d0db687f534d0c84307247/dependency_graph.png)
I realized that my logic for the activation of sink's tasks and my interaction loop for the computation of the sink formation were not correct.
As I need the dependencies between the `compute_formation` and `sink_merger`, I am using the `sink_formation` task as a ghost. Therefore it is activated even if there is no sink formation (in order to have the dependencies for the sink mergers).
So my logic is now: If active (hydro or sink), I activate everything for the self. For the pair, I drift the sink of both cells (for the merger) and the part only in the active cell (for the `compute_formation`). For the `sink_formation` task, I activate it for both cells.
The logic for the interaction:
* Compute formation: interact all the active hydro particles with the (active and non active) sink particles.
* Sink merger: interact all the active sink particles with the (active and non active) sink particles.
My next step is to implement the accretion function and then I will be done with adding new tasks (still a lot of work, but the most important one will be done).Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1300Add missing parameter2021-02-15T15:05:42ZLoic HausammannAdd missing parameterI forgot a parameter in the yaml file.I forgot a parameter in the yaml file.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1270Fix feedback in GEAR2021-01-18T08:29:18ZLoic HausammannFix feedback in GEARDepending on the time step, the criterion of `1e-6` was not enough. Now I am only checking that the time at the end of the step is larger than 0 and I am modifying the time at the beginning of the step in order to ensure a positive age.Depending on the time step, the criterion of `1e-6` was not enough. Now I am only checking that the time at the end of the step is larger than 0 and I am modifying the time at the beginning of the step in order to ensure a positive age.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1271Fix particle separation2021-01-15T16:48:06ZLoic HausammannFix particle separationConclusion of the merge request: I hate my past self.
Tiny mistake due to a bad habit of copy/paste and while debugging I found that the displacements where a bit too large at my taste. I reduced them by a factor of 2.Conclusion of the merge request: I hate my past self.
Tiny mistake due to a bad habit of copy/paste and while debugging I found that the displacements where a bit too large at my taste. I reduced them by a factor of 2.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1268GEAR: Move return condition before checks2021-01-13T20:06:16ZLoic HausammannGEAR: Move return condition before checksCurrently, GEAR is not working with the feedback on and debugging checks. Here I am simply moving a return condition before the non working check.Currently, GEAR is not working with the feedback on and debugging checks. Here I am simply moving a return condition before the non working check.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1225Implement gear's stars skipping2020-11-23T09:28:14ZLoic HausammannImplement gear's stars skippingAs the stars are spending most of their time without any feedback, I am skipping them when no supernovae are produced.
@matthieu Do you accept the modifications done to `runner_time_integration.c` or should I use some `#ifdef` and kee...As the stars are spending most of their time without any feedback, I am skipping them when no supernovae are produced.
@matthieu Do you accept the modifications done to `runner_time_integration.c` or should I use some `#ifdef` and keep only a single evolution function?Loic HausammannLoic Hausammannhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1232Update zoom in example and add comment to error message2020-11-20T12:29:48ZLoic HausammannUpdate zoom in example and add comment to error messageThe zoom in example was not up to date with my last simulations. I have changed a bit the parameter file and the list of snapshots was missing.
I have also updated the comment for the error `Reduction in top-level cells number not handl...The zoom in example was not up to date with my last simulations. I have changed a bit the parameter file and the list of snapshots was missing.
I have also updated the comment for the error `Reduction in top-level cells number not handled.`.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1201Gear metal diffusion2020-11-13T12:48:42ZLoic HausammannGear metal diffusionIn this merge request, I am implementing the metal diffusion from Shen et al. 2010. We still need to test it in a cosmological context, conservation of mass and find the best coefficient / time step (e.g. work on the ideas from Greif et ...In this merge request, I am implementing the metal diffusion from Shen et al. 2010. We still need to test it in a cosmological context, conservation of mass and find the best coefficient / time step (e.g. work on the ideas from Greif et al. 2009), but this will be for a future PhD student.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1186Sink formation check2020-10-20T12:49:44ZLoic HausammannSink formation checkIn this merge request, I am implementing the neighbors loop for the sink formation. I am planning to form a sink particle only if it is not a neighbor of an existing sink particle.
![dependency_graph](/uploads/21dcc38bef4f26bcd44f80c5c7...In this merge request, I am implementing the neighbors loop for the sink formation. I am planning to form a sink particle only if it is not a neighbor of an existing sink particle.
![dependency_graph](/uploads/21dcc38bef4f26bcd44f80c5c7caf2b4/dependency_graph.png)
I have tested the code for EAGLE 6 with MPI and full physics, everything seems fine.
In order to check my neighbors loop, I have removed the `engine_drift_all` before a snapshot and run the isolated galaxy example with some sink formation and a single time step for all the particles (I need to ensure that the neighbors loop is done with the position written in the snapshot). Once enough particle are formed, I stopped the simulation and compared the number of neighbors found in swift and with a simple python script on the snapshot. In the snapshot, 1942 sink particles are present. 1941 had the correct number of neighbors and 1 was formed in the last time step (thus the number of neighbors was set to 0).Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1181Fix first stars22020-10-05T15:32:14ZLoic HausammannFix first stars2Fix #712
@mivkov can you try it please?
I have simplified a bit the error logic. I am not sure why the code was working before the last merge request.Fix #712
@mivkov can you try it please?
I have simplified a bit the error logic. I am not sure why the code was working before the last merge request.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1177Skip supernovae producing negative mass2020-10-01T13:47:30ZLoic HausammannSkip supernovae producing negative massAs the supernovae are random events, you can end up with a bit more supernovae than expected. In the case of the first stars, it means that you can end up ejecting too much mass and have negative masses. To avoid this, in GEAR, we skip a...As the supernovae are random events, you can end up with a bit more supernovae than expected. In the case of the first stars, it means that you can end up ejecting too much mass and have negative masses. To avoid this, in GEAR, we skip all the cases where this happens.Matthieu SchallerMatthieu Schaller