D22339: Make sure solid backends are reentrant

Stefan BrĂ¼ns noreply at phabricator.kde.org
Tue Jul 16 15:13:58 BST 2019


bruns added a comment.


  I would have preferred this to only land when D22333 <https://phabricator.kde.org/D22333> had been addressed fully, I assumed this required no further notice.
  
  So from a pure technical view, this indeed seems to be correct, but from an architectural view this seems less than ideal (to put it mildly).
  
  One obvious issue is the multiple instantiation and initialization, i.e. when the UDisks2 backend is used from different threads each one has to introspect udisks itself, which seems wasteful for me.
  
  I will extend this thought in D22333 <https://phabricator.kde.org/D22333>, can we move any further discussion there?

REPOSITORY
  R245 Solid

REVISION DETAIL
  https://phabricator.kde.org/D22339

To: apol, #frameworks, davidedmundson, bruns
Cc: lbeltrame, bruns, kde-frameworks-devel, LeGast00n, sbergeron, michaelh, ngraham
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-frameworks-devel/attachments/20190716/20fe6f17/attachment.html>


More information about the Kde-frameworks-devel mailing list