cosmetic + correct specs for lqueue:peek{_r}2011-10-15T15:20:26+00:00Matthias Radestockmatthias@rabbitmq.com2011-10-15T15:20:26+00:00a97c0c40d238a1ac81d2f08c8ecd5bba1fb310b7
Revert lqueue so it does not cache ends of queue as we no longer need this to be O(1).2011-10-15T14:48:20+00:00Matthew Sackmanmatthew@rabbitmq.com2011-10-15T14:48:20+00:00ca5bdaaeef2bd3e4a3a9db820381de7d6aeac15a
cosmetic, and be somewhat smarter about deciding the minimum length of q3 (now must end on a segment boundary - minor optimisation)2011-10-12T13:40:12+00:00Matthew Sackmanmatthew@rabbitmq.com2011-10-12T13:40:12+00:00071488bfaada18897cb19ff3e568d837e0260bc4
Cache both ends of the queue to give O(1) peek and peek_r2011-10-12T13:15:22+00:00Matthew Sackmanmatthew@rabbitmq.com2011-10-12T13:15:22+00:00bf22ced93fa82abd9e1fd1833545bf55bd68c775
Add peek and peek_r to lqueue, and use them in needs_timeout. Whilst this is better, it's still O(N) per msg, so we shouldn't actually be doing even this...2011-10-11T17:02:04+00:00Matthew Sackmanmatthew@rabbitmq.com2011-10-11T17:02:04+00:00df36d772d5a6d6416cec61ade2e85453a2188f37
Make lqueue opaque2011-10-02T18:25:37+00:00Matthew Sackmanmatthew@rabbitmq.com2011-10-02T18:25:37+00:0091cc503da7eb78bdb308bb16a04fe74bd9b08886
Drop unneeded queue impls, tidy lqueue, add specs etc2011-10-02T14:13:35+00:00Matthew Sackmanmatthew@rabbitmq.com2011-10-02T14:13:35+00:0019cb6e7f8c890fd3c9b1fa563a54a80129843bd4
Avoid lots of combine_delta calls2011-10-02T13:53:31+00:00Matthew Sackmanmatthew@rabbitmq.com2011-10-02T13:53:31+00:00364b3898e14f9c801397639ce3df9d5f92588cb0
Count the quota properly on beta -> delta. Also, fprof shows that the queue:len calls in permitted_beta_count can be very expensive. So wrap the whole of queue in lqueue and associated refactoring.2011-09-29T16:09:09+00:00Matthew Sackmanmatthew@rabbitmq.com2011-09-29T16:09:09+00:006cca6fae7f00d19738f830684a6020f7975c284d