SWIFTsim merge requestshttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests2024-01-23T14:11:57Zhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1837A different way of adding B fields to ICs, better choice for current unit too2024-01-23T14:11:57ZOrestis KarapiperisA different way of adding B fields to ICs, better choice for current unit tooMatthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1836Karapiperis/update maggntised cloud example2023-12-15T13:47:01ZOrestis KarapiperisKarapiperis/update maggntised cloud exampleMatthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1835Run python formater2023-12-14T14:08:04ZOrestis KarapiperisRun python formaterMatthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1834Modified Roberts_flow_acceleration forcing2023-12-20T09:20:51ZNikyta ShchutskyiModified Roberts_flow_acceleration forcingFede's effective viscosity divided by 2 * (3+2), where 3 - dimensionality of the run. Also modified RobertsFlow.yml by setting up constant artificial viscosity.Fede's effective viscosity divided by 2 * (3+2), where 3 - dimensionality of the run. Also modified RobertsFlow.yml by setting up constant artificial viscosity.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1833Small fix to make SWIFT compile afetr addition of new error metric code2023-12-10T12:08:24ZOrestis KarapiperisSmall fix to make SWIFT compile afetr addition of new error metric codeDon't need norm(B) in calculation of R2, could've caused compilers to complain if left there.Don't need norm(B) in calculation of R2, could've caused compilers to complain if left there.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1832Clean BW shock tube directory, runs easily out of the box2023-12-08T21:06:39ZOrestis KarapiperisClean BW shock tube directory, runs easily out of the boxMatthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1831Minor mpi fixes with MUST2023-12-02T18:36:42ZMladen IvkovicMinor mpi fixes with MUSTHad a go with the MUST tool on swift. It pointed out two minor issues (running SmallCosmoVolume_hydro)
- we pass `MPI_INFO_NULL` as an argument for `MPI_Info info`. Then we try to write into that using `MPI_Info_Set(&info)` . Apparently...Had a go with the MUST tool on swift. It pointed out two minor issues (running SmallCosmoVolume_hydro)
- we pass `MPI_INFO_NULL` as an argument for `MPI_Info info`. Then we try to write into that using `MPI_Info_Set(&info)` . Apparently that's illegal, because `MPI_INFO_NULL` is supposed to be `const` , I've been told, but I couldn't find that documented in the MPI standard (https://www.mpi-forum.org/docs/mpi-4.1/mpi41-report.pdf)
- the `MPI_Type lospart_mpi_type` is declared and freed, but never committed. I (and `grep`) don't see it used anywhere aside from the declaration and freeing. It's been a while since I wrote MPI stuff myself, but I'm pretty sure MPI would complain if you're trying to use a type that isn't committed. So I believe if the type were indeed used, we should have noticed. But I wouldn't bet money on that.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1830Check if the tools to support ipo are available. If not use the standard...2023-11-30T17:29:28ZPeter W. DraperCheck if the tools to support ipo are available. If not use the standard...Check if the tools to support ipo are available. If not use the standard versions as determined by libtool.
Fix for #876 as Apple do not have the `llvm-` command names.Check if the tools to support ipo are available. If not use the standard versions as determined by libtool.
Fix for #876 as Apple do not have the `llvm-` command names.Peter W. DraperPeter W. Draperhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1829Add new error monitoring ot the MHD code2023-12-08T21:07:20ZNikyta ShchutskyiAdd new error monitoring ot the MHD codeAdding error metrics calculation R0-R3Adding error metrics calculation R0-R3Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1828fix wrong parameter in advect_ions.yml2023-11-30T11:35:27ZMladen Ivkovicfix wrong parameter in advect_ions.ymlhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1827added check for valid thread number provided2023-11-29T12:57:42ZMladen Ivkovicadded check for valid thread number providedMatthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1826Draft: Magnetosonic speed in timescale used in updating the viscosity alphas2024-02-23T21:01:39ZOrestis KarapiperisDraft: Magnetosonic speed in timescale used in updating the viscosity alphasAlso changed hydro parameters to what works best for me in the shock tubes, and is consistent with the value suggested for hydro. Let me know what you think.Also changed hydro parameters to what works best for me in the shock tubes, and is consistent with the value suggested for hydro. Let me know what you think.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1825Chemistry API changes for metal fluxes2024-01-24T09:40:37ZYolan UyttenhoveChemistry API changes for metal fluxesThis merge request extends the chemistry API as discussed in #834 to correctly advect metals in hydro schemes with mass fluxes between particles.
The approach is very similar to @mivkov's approach to tracking the mass fractions of the i...This merge request extends the chemistry API as discussed in #834 to correctly advect metals in hydro schemes with mass fluxes between particles.
The approach is very similar to @mivkov's approach to tracking the mass fractions of the ionizing species in GEAR-RT.
## Current status
This is a working implementation for the EAGLE chemistry scheme.
Chemistry schemes other than `EAGLE` and `none` currently throw a compile time error trying to compile with a hydro scheme that does mass fluxes (in this branch, only `gizmo-mfv`).
## To-do
- [ ] Implement the new functions for other chemistry schemes
- [ ] How to treat diffusion in this case? (only necessary when implementing for GEAR-diffusion)
- [ ] Advect other quantities tracked by chemistry as well (e.g. `mass_from_SNIa`, `metal_mass_fraction_from_SNIa` and others)?
## Other thoughts
As discussed in #834, we would also need to disable metal smoothing when using this with e.g. moving mesh hydro.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1824Sink : bugs fixing2024-03-25T17:14:24ZDarwinSink : bugs fixingAll known bugs that lead to crashes are solved.All known bugs that lead to crashes are solved.https://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1823Support optimization for the AMD aocc compiler2023-11-30T15:24:19ZPeter W. DraperSupport optimization for the AMD aocc compilerMake sure we use the optimized maths library and interprocedural optimization. Both are needed to get
optimization that works as well as other clang based-compilers.
Note in this MR we enable interprocedural optimization by default, unl...Make sure we use the optimized maths library and interprocedural optimization. Both are needed to get
optimization that works as well as other clang based-compilers.
Note in this MR we enable interprocedural optimization by default, unless the `--enable-debug`
or `--disable-optimization`
flags are used.Peter W. DraperPeter W. Draperhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1822Fix radiated energy saved to statistics for EAGLE cooling.2023-11-23T22:43:39ZYolan UyttenhoveFix radiated energy saved to statistics for EAGLE cooling.(cherry picked from commit 340658c0a4ba9b4bfc37623002d9c7ef21199466)
I think the radiated energy stored in `xp.cooling_data` was incorrect for the EAGLE cooling scheme.
The variable `cooling_du_dt` actually stores the total `du_dt` (hy...(cherry picked from commit 340658c0a4ba9b4bfc37623002d9c7ef21199466)
I think the radiated energy stored in `xp.cooling_data` was incorrect for the EAGLE cooling scheme.
The variable `cooling_du_dt` actually stores the total `du_dt` (hydro + cooling), so the `hydro_du_dt` needs to be subtracted when computing the radiated energy.
Some examples (from CoolingSedovBlast3D):
| | Without fix | With fix |
|:----:|:----:|:----:|
| Sphenix: | ![energy_sphenix_before](/uploads/29b7bf22fa890bb7b09642f971b2c32a/energy_sphenix_before.png) | ![energy_sphenix_after](/uploads/bed18fab3b8606ed8ecbf3a3b173580c/energy_sphenix_after.png) |
| Gizmo-MFV: |![energy_gizmo-mfv_before](/uploads/df048e870074d2b795e6cab00a8d6922/energy_gizmo-mfv_before.png) |![energy_gizmo-mfv_after](/uploads/15b6424e22de51fc429cd41ad3e3ae22/energy_gizmo-mfv_after.png)Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1821Fstasys/fast rotor test multiple schemes2023-11-15T08:56:28ZFederico Andrés StasyszynFstasys/fast rotor test multiple schemesAdded scripts and plotting routines for the MHD Fast Rotor
added extra information also for the MHD Blast waveAdded scripts and plotting routines for the MHD Fast Rotor
added extra information also for the MHD Blast wavehttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1820Fstasys/blast wave additional files2023-11-14T14:45:38ZFederico Andrés StasyszynFstasys/blast wave additional filesmimimum work to comprare codes.... easy to check also MAGMA.
Still needed to add a pattern informationmimimum work to comprare codes.... easy to check also MAGMA.
Still needed to add a pattern informationhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1819Prevent overflows in time step calculations with RT2023-11-13T15:49:11ZMladen IvkovicPrevent overflows in time step calculations with RTIn certain cases, overflows may occur when computing the hydro time step sizes in the timestep tasks. Specifically, this may occur when a) the total physical simulation time to be covered is short; b) there are particles with long time s...In certain cases, overflows may occur when computing the hydro time step sizes in the timestep tasks. Specifically, this may occur when a) the total physical simulation time to be covered is short; b) there are particles with long time step sizes (e.g. like in voids); and c) a large number of sub-cycles is permitted.
I added a check and a fix for that.
I also noticed that I'm not properly initializing the RT time step data anywhere. It shouldn't be a problem, but I added it during first_init anyway. Better safe than sorry.Matthieu SchallerMatthieu Schallerhttps://gitlab.cosma.dur.ac.uk/swift/swiftsim/-/merge_requests/1818Fstasys/clean blast now with vp2023-11-10T21:23:50ZFederico Andrés StasyszynFstasys/clean blast now with vpNow the Blast wave with VP.
Note that I changed the IC a little bit following Seo & Ryu, 2023Now the Blast wave with VP.
Note that I changed the IC a little bit following Seo & Ryu, 2023Federico Andrés StasyszynFederico Andrés Stasyszyn