Matthieu Schaller authored Feb 02, 2017 and Peter W. Draper committed Feb 03, 2017

Added the missing SSE function in vector.h. Compiles but accuracy is not identical to AVX when running test27.

Update the task plot scripts to include new tasks and subtasks.

Also fixes an issue where gaps between successive tasks of the same type
were not seen as these were joined (for plotting efficiency I guess).
A gap is now seen.

Also change the colour handling so that the task/subtask types are
easier to differentiate.

Fixes issue #260.

See merge request !311

Fixes an issue where gaps between successive tasks of the same type
were not seen as these were joined (for plotting efficiency I guess).
A gap is now seen.

Also change the colour handling so that the task/subtask types are
easier to differentiate.

Fix compilation error when compiling without affinity turned on.

Does this seem like a reasonable fix to #261, @pdraper @matthieu?

See merge request !309

MPI send/recv fixes for inactive cells

This is open so that we have a live diff of the changes to solve the MPI crisis. Also helpful to keep track of the changes.

Contains all of @nnrw56 @pdraper @matthieu's changes to this stage.

Also see #256.

See merge request !303

Call runner_doself1_density_vec in DOSUB_SELF1 if vectorisation is enabled.

Changed all `WITH_VECTORIZATION` macros to WITH_OLD_VECTORIZATION in `runner_doiact.h` so Pedro's old vectorised functions are not called.

See merge request !308

Fix cooling bug

Gas particles no longer cool below the energy floor if the timestep increases.

Changed CoolingHaloWithSpin example so that gas particle masses are more sensible.

See merge request !307

Call runner_doself1_density_vec in DOSUB_SELF1 if vectorisation is enabled. Changed all WITH_VECTORIZATION macros to WITH_OLD_VECTORIZATION.

Add another communication sub-type for the gparts. Send the gparts when these tasks get enqueued.

This follows suit with the addition of task sub-types for the communications. I have added one to ship the gparts between nodes. 

This also allows to send/recv the gparts instead of the parts at enqueuing time. Note that the gpart communications are currently unused since I never pushed the support for gravity over MPI to the master. 

It also allows to only loop over the relevant bits we received and not over the whole particle zoo to construct the time information on the receiving side. 

This merges into the `new_timeline_mpi` branch and not in `master` as it is yet another MPI bug/feature fix.

See merge request !305

Revert "Force a sort of the particles at every time-step for active cells and at reception of data."

This reverts commit 078d450f

Added test that benchmarks the interaction functions against their vectorised counterparts.



See merge request !306

Merge branch 'benchmark_interactions' of gitlab.cosma.dur.ac.uk:swift/swiftsim into benchmark_interactions