Clean up timer outputs

[efaulhaber]

1. The shifting callback was calling update_systems_and_nhs! outside an @timeit, which caused orphaned update timers like update density diffusion on the top level.
2. For save solution, interpolation and shifting, the system/nhs update is usually negligible (and if not, the percentages should still be the same as in kick!), so there is no point in having all the sub-timers. Especially with Split fluid and TLSPH time integration to allow for TLSPH substeps #794, the timer output otherwise becomes very cluttered when multiple callbacks are used.

Before:

──────────────────────────────────────────────────────────────────────────────────────────
           TrixiParticles.jl                     Time                    Allocations      
                                        ───────────────────────   ────────────────────────
           Tot / % measured:                 5.84s /  89.6%           85.6MiB /  76.8%    

Section                         ncalls     time    %tot     avg     alloc    %tot      avg
──────────────────────────────────────────────────────────────────────────────────────────
kick!                              212    4.58s   87.5%  21.6ms   2.03MiB    3.1%  9.81KiB
  update systems and nhs           212    2.73s   52.2%  12.9ms   1.14MiB    1.7%  5.52KiB
    update density diffusion       212    2.61s   50.0%  12.3ms    262KiB    0.4%  1.23KiB
    update nhs                     212   69.0ms    1.3%   325μs    398KiB    0.6%  1.88KiB
    compute boundary pressure      424   34.3ms    0.7%  80.8μs    341KiB    0.5%     824B
    ~update systems and nhs~       212   10.7ms    0.2%  50.3μs   44.6KiB    0.1%     215B
    inverse state equation         424   3.63ms    0.1%  8.56μs    126KiB    0.2%     304B
  system interaction               212    1.84s   35.2%  8.69ms    851KiB    1.3%  4.01KiB
    fluid1-fluid1                  212    1.57s   30.1%  7.42ms    129KiB    0.2%     624B
    fluid1-boundary2               212    160ms    3.1%   755μs    129KiB    0.2%     624B
    fluid1-boundary3               212    106ms    2.0%   501μs    129KiB    0.2%     624B
    ~system interaction~           212   2.15ms    0.0%  10.1μs    463KiB    0.7%  2.19KiB
    boundary3-boundary2            212   3.46μs    0.0%  16.3ns     0.00B    0.0%    0.00B
    boundary2-fluid1               212   2.87μs    0.0%  13.5ns     0.00B    0.0%    0.00B
    boundary3-boundary3            212   2.71μs    0.0%  12.8ns     0.00B    0.0%    0.00B
    boundary2-boundary2            212   2.63μs    0.0%  12.4ns     0.00B    0.0%    0.00B
    boundary3-fluid1               212   2.58μs    0.0%  12.1ns     0.00B    0.0%    0.00B
    boundary2-boundary3            212   2.29μs    0.0%  10.8ns     0.00B    0.0%    0.00B
  reset ∂v/∂t                      212   3.33ms    0.1%  15.7μs   6.62KiB    0.0%    32.0B
  source terms                     212   2.24ms    0.0%  10.5μs   49.7KiB    0.1%     240B
  ~kick!~                          212    137μs    0.0%   645ns   1.55KiB    0.0%    7.47B
update density diffusion            20    250ms    4.8%  12.5ms   24.7KiB    0.0%  1.23KiB
particle shifting                   20    160ms    3.1%  7.99ms   38.4KiB    0.1%  1.92KiB
save solution                        1    132ms    2.5%   132ms   39.0MiB   59.4%  39.0MiB
  write to vtk                       3   76.5ms    1.5%  25.5ms   33.0MiB   50.3%  11.0MiB
  ~save solution~                    1   42.0ms    0.8%  42.0ms   5.97MiB    9.1%  5.97MiB
  update systems                     1   13.4ms    0.3%  13.4ms   7.06KiB    0.0%  7.06KiB
    update density diffusion         1   12.5ms    0.2%  12.5ms   1.23KiB    0.0%  1.23KiB
    update nhs                       1    395μs    0.0%   395μs   1.88KiB    0.0%  1.88KiB
    compute boundary pressure        2    280μs    0.0%   140μs   1.61KiB    0.0%     824B
    ~update systems~                 1    157μs    0.0%   157μs   1.75KiB    0.0%  1.75KiB
    inverse state equation           2   52.0μs    0.0%  26.0μs      608B    0.0%     304B
apply postprocess cb                 1   95.1ms    1.8%  95.1ms   24.5MiB   37.2%  24.5MiB
  interpolate plane                  1   55.2ms    1.1%  55.2ms   23.4MiB   35.6%  23.4MiB
  write to vtk                       1   28.0ms    0.5%  28.0ms   1.08MiB    1.6%  1.08MiB
  update systems                     1   11.9ms    0.2%  11.9ms   7.06KiB    0.0%  7.06KiB
    update density diffusion         1   11.3ms    0.2%  11.3ms   1.23KiB    0.0%  1.23KiB
    update nhs                       1    358μs    0.0%   358μs   1.88KiB    0.0%  1.88KiB
    compute boundary pressure        2    155μs    0.0%  77.3μs   1.61KiB    0.0%     824B
    ~update systems~                 1   39.2μs    0.0%  39.2μs   1.75KiB    0.0%  1.75KiB
    inverse state equation           2   11.5μs    0.0%  5.77μs      608B    0.0%     304B
  ~apply postprocess cb~             1   56.1μs    0.0%  56.1μs   5.68KiB    0.0%  5.68KiB
update nhs                          20   5.42ms    0.1%   271μs   37.5KiB    0.1%  1.88KiB
drift!                             212   4.88ms    0.1%  23.0μs   54.0KiB    0.1%     261B
  reset ∂u/∂t                      212   3.15ms    0.1%  14.9μs   6.62KiB    0.0%    32.0B
  velocity                         212   1.68ms    0.0%  7.91μs   46.4KiB    0.1%     224B
  ~drift!~                         212   53.9μs    0.0%   254ns      976B    0.0%    4.60B
compute boundary pressure           40   3.58ms    0.1%  89.6μs   32.2KiB    0.0%     824B
inverse state equation              40    548μs    0.0%  13.7μs   11.9KiB    0.0%     304B
──────────────────────────────────────────────────────────────────────────────────────────

This PR:

───────────────────────────────────────────────────────────────────────────────────────────
            TrixiParticles.jl                     Time                    Allocations      
                                         ───────────────────────   ────────────────────────
            Tot / % measured:                 8.88s /  73.1%            129MiB /  59.9%    

Section                          ncalls     time    %tot     avg     alloc    %tot      avg
───────────────────────────────────────────────────────────────────────────────────────────
kick!                               212    4.80s   73.9%  22.6ms   2.03MiB    2.6%  9.81KiB
  update systems and nhs            212    2.90s   44.6%  13.7ms   1.14MiB    1.5%  5.52KiB
    update density diffusion        212    2.75s   42.3%  13.0ms    262KiB    0.3%  1.23KiB
    update nhs                      212   83.3ms    1.3%   393μs    398KiB    0.5%  1.88KiB
    compute boundary pressure       424   45.5ms    0.7%   107μs    341KiB    0.4%     824B
    ~update systems and nhs~        212   14.1ms    0.2%  66.6μs   44.6KiB    0.1%     215B
    inverse state equation          424   4.06ms    0.1%  9.57μs    126KiB    0.2%     304B
  system interaction                212    1.89s   29.1%  8.92ms    851KiB    1.1%  4.01KiB
    fluid1-fluid1                   212    1.61s   24.8%  7.61ms    129KiB    0.2%     624B
    fluid1-boundary2                212    165ms    2.5%   778μs    129KiB    0.2%     624B
    fluid1-boundary3                212    111ms    1.7%   524μs    129KiB    0.2%     624B
    ~system interaction~            212   3.47ms    0.1%  16.4μs    463KiB    0.6%  2.19KiB
    boundary2-fluid1                212   5.79μs    0.0%  27.3ns     0.00B    0.0%    0.00B
    boundary2-boundary2             212   5.21μs    0.0%  24.6ns     0.00B    0.0%    0.00B
    boundary3-boundary2             212   4.00μs    0.0%  18.9ns     0.00B    0.0%    0.00B
    boundary3-boundary3             212   3.79μs    0.0%  17.9ns     0.00B    0.0%    0.00B
    boundary2-boundary3             212   3.05μs    0.0%  14.4ns     0.00B    0.0%    0.00B
    boundary3-fluid1                212   2.83μs    0.0%  13.4ns     0.00B    0.0%    0.00B
  reset ∂v/∂t                       212   7.70ms    0.1%  36.3μs   6.62KiB    0.0%    32.0B
  source terms                      212   2.05ms    0.0%  9.68μs   49.7KiB    0.1%     240B
  ~kick!~                           212    165μs    0.0%   777ns   1.55KiB    0.0%    7.47B
save solution                         1    973ms   15.0%   973ms   48.6MiB   63.0%  48.6MiB
  ~save solution~                     1    811ms   12.5%   811ms   15.5MiB   20.1%  15.5MiB
  write to vtk                        3    146ms    2.2%  48.7ms   33.1MiB   42.9%  11.0MiB
  update systems                      1   15.7ms    0.2%  15.7ms   5.52KiB    0.0%  5.52KiB
particle shifting callback           20    454ms    7.0%  22.7ms    150KiB    0.2%  7.49KiB
  update systems and nhs             20    267ms    4.1%  13.4ms    110KiB    0.1%  5.52KiB
  particle shifting                  20    186ms    2.9%  9.32ms   38.4KiB    0.0%  1.92KiB
  ~particle shifting callback~       20   65.2μs    0.0%  3.26μs      976B    0.0%    48.8B
apply postprocess cb                  1    266ms    4.1%   266ms   26.3MiB   34.1%  26.3MiB
  interpolate plane                   1    173ms    2.7%   173ms   23.4MiB   30.3%  23.4MiB
  ~apply postprocess cb~              1   41.8ms    0.6%  41.8ms   1.84MiB    2.4%  1.84MiB
  write to vtk                        1   33.2ms    0.5%  33.2ms   1.08MiB    1.4%  1.08MiB
  update systems and nhs              1   18.8ms    0.3%  18.8ms   5.52KiB    0.0%  5.52KiB
drift!                              212   5.05ms    0.1%  23.8μs   54.0KiB    0.1%     261B
  reset ∂u/∂t                       212   2.94ms    0.0%  13.9μs   6.62KiB    0.0%    32.0B
  velocity                          212   2.00ms    0.0%  9.42μs   46.4KiB    0.1%     224B
  ~drift!~                          212    117μs    0.0%   554ns      976B    0.0%    4.60B
───────────────────────────────────────────────────────────────────────────────────────────

[Copilot]

Pull Request Overview

This PR aims to clean up timer outputs by grouping update calls into block scopes and using the @notimeit macro to prevent creating extraneous sub-timers. Key changes include:

• Wrapping update_systems_and_nhs calls in a begin…end block with @notimeit in I/O, post-process, and particle shifting callbacks.
• Updating timer label names to reflect the encompassed operations.
• Adjusting the TimerOutputs import to include the new @notimeit macro.

Reviewed Changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

File	Description
src/io/write_vtk.jl	Wrapped update_systems_and_nhs call in a block to suppress sub-timer output.
src/callbacks/post_process.jl	Similar block wrapping applied with an updated timer label for clarity.
src/callbacks/particle_shifting.jl	Nested timer blocks introduced to group update operations and prevent clutter.
src/TrixiParticles.jl	Updated TimerOutputs import to include the new @notimeit macro.

Comments suppressed due to low confidence (2)

src/callbacks/post_process.jl:241

• [nitpick] Consider standardizing the timer label names across modules. In src/io/write_vtk.jl the timer label remains 'update systems', whereas here it is 'update systems and nhs'. Choose one consistent label to avoid confusion.

@trixi_timeit timer() "update systems and nhs" begin

src/callbacks/particle_shifting.jl:40

• [nitpick] Ensure the timer label 'update systems and nhs' is used consistently with similar timer blocks elsewhere, or adjust it if a different naming convention is intended.

@trixi_timeit timer() "update systems and nhs" begin

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 11 lines in your changes missing coverage. Please review.

Project coverage is 70.70%. Comparing base (c4b008f) to head (d304e3c).

Files with missing lines	Patch %	Lines
src/callbacks/particle_shifting.jl	0.00%	7 Missing ⚠️
src/callbacks/post_process.jl	0.00%	2 Missing ⚠️
src/io/write_vtk.jl	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #819      +/-   ##
==========================================
- Coverage   70.74%   70.70%   -0.05%     
==========================================
  Files         106      106              
  Lines        6710     6714       +4     
==========================================
  Hits         4747     4747              
- Misses       1963     1967       +4     

Flag	Coverage Δ
unit	70.70% <0.00%> (-0.05%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.