Bitset improvements and more bitset usage

[szhorvat]

No description provided.

[codecov]

Codecov Report

Attention: Patch coverage is 97.82609% with 4 lines in your changes are missing coverage. Please review.

Project coverage is 84.30%. Comparing base (2f5179c) to head (87f264c).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2607      +/-   ##
==========================================
- Coverage   84.31%   84.30%   -0.01%     
==========================================
  Files         379      379              
  Lines       61951    61932      -19     
  Branches    12137    12136       -1     
==========================================
- Hits        52232    52213      -19     
  Misses       9719     9719              

Files	Coverage Δ
src/community/voronoi.c	86.47% <100.00%> (ø)
src/connectivity/components.c	95.34% <100.00%> (-0.06%)	⬇️
src/core/bitset.c	82.78% <100.00%> (+0.58%)	⬆️
src/flow/st-cuts.c	97.32% <100.00%> (-0.02%)	⬇️
src/graph/adjlist.c	91.93% <100.00%> (ø)
src/misc/spanning_trees.c	97.33% <100.00%> (-0.14%)	⬇️
src/paths/bellman_ford.c	90.09% <100.00%> (ø)
src/paths/eulerian.c	96.78% <100.00%> (ø)
src/paths/sparsifier.c	98.05% <100.00%> (ø)
src/properties/trees.c	94.71% <100.00%> (ø)
... and 2 more

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2f5179c...87f264c. Read the comment docs.

[szhorvat]

@Tagl Can you review this please?

[szhorvat]

The failure on Travis is real. Something might be wrong with bitsets on s390x. This needs to be checked before merging.

[szhorvat]

It's fixed now, it was a silly mistake on my part.

[Tagl]

Looks good as far as I can tell. Have you benchmarked anything yet?

[szhorvat]

Looks good as far as I can tell. Have you benchmarked anything yet?

I benchmarked several functions, and none of them show a noticeable change in performance. I'll consider that a win, as memory usage has been reduced, sometimes considerably.

[Tagl]

Looks good as far as I can tell. Have you benchmarked anything yet?

I benchmarked several functions, and none of them show a noticeable change in performance. I'll consider that a win, as memory usage has been reduced, sometimes considerably.

That makes sense and absolutely it is a win. Some of functions changed can probably be improved upon furhter by utilizing bitmasks in some way and then a speed increase will be noticeable. But this is already good!

[ntamas]

Thanks!