Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | tweak loggingbug24407 | Matthias Radestock | 2013-01-04 | 1 | -2/+1 |
| | |||||
* | cosmetic(ish) | Matthias Radestock | 2013-01-04 | 1 | -1/+1 |
| | |||||
* | cosmetic | Matthias Radestock | 2013-01-03 | 1 | -3/+2 |
| | |||||
* | simplify syncer | Matthias Radestock | 2013-01-03 | 1 | -33/+23 |
| | | | | we don't need to track monitors | ||||
* | make credit waiting less brittle | Matthias Radestock | 2013-01-03 | 2 | -12/+11 |
| | | | | | | | | | | | | We were relying on running out of credit for all slaves *simultaneously*, which requires in-depth knowledge of the credit flow logic and that no other credit-requiring messages are sent to a slave prior to this. Fortunately, since we are running in a fresh separate process we can simply handle *any* credit bumping message and *any* DOWN message. As a bonus we can revert to making the type of the bump msg opaque. | ||||
* | fix bug found by dialyzer | Matthias Radestock | 2013-01-03 | 1 | -2/+2 |
| | |||||
* | bring up to date with 'default' | Matthias Radestock | 2013-01-03 | 1 | -1/+1 |
| | |||||
* | merge default into bug24407 | Matthias Radestock | 2013-01-03 | 25 | -615/+677 |
|\ | |||||
| * | Merge bug25371 | Simon MacMullen | 2013-01-03 | 1 | -22/+21 |
| |\ | |||||
| | * | change rabbit_exchange:route/2's traversal orderbug25371 | Matthias Radestock | 2012-12-29 | 1 | -22/+21 |
| | | | | | | | | | | | | | | | | | | | | | | | | from breadth-first to depth-first, thus eliminating the queue data structure and hence simplifying the code and improving performance. Plus a little bit of refactoring in the area. | ||||
| * | | Merge bug25369 | Simon MacMullen | 2013-01-03 | 9 | -344/+324 |
| |\ \ | |||||
| | * \ | Merge bug25370 | Simon MacMullen | 2013-01-03 | 1 | -71/+64 |
| | |\ \ | |||||
| | | * | | move tx related state fields into sub-recordbug25370 | Matthias Radestock | 2012-12-29 | 1 | -71/+64 |
| | | |/ | |||||
| | * | | Merge bug25368 | Simon MacMullen | 2013-01-03 | 2 | -79/+81 |
| | |\ \ | |||||
| | | * | | move throttle related reader state into sub-statebug25368 | Matthias Radestock | 2012-12-28 | 1 | -26/+34 |
| | | | | | |||||
| | | * | | move some more reader state into #connection | Matthias Radestock | 2012-12-28 | 1 | -20/+22 |
| | | | | | | | | | | | | | | | | | | | | | | | | | and also clear auth_state once auth has been completed, so we don't hang on to some potentially large piece of state. | ||||
| | | * | | move a bunch of reader state fields into 'connection' | Matthias Radestock | 2012-12-28 | 2 | -39/+31 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | These fields are constant, seldom read, and associated with the connection. Also, move #connection definition from rabbit.hrl to rabbit_reader - nothing else is using this and it is really private to the reader. | ||||
| | * | | | Merge bug25366 | Simon MacMullen | 2013-01-03 | 1 | -25/+11 |
| | |\ \ \ | |||||
| | | * | | | get rid of rabbit_channel MASKED_CALL macrobug25366 | Matthias Radestock | 2012-12-28 | 1 | -25/+11 |
| | | |/ / | | | | | | | | | | | | | | | | which simplifies the code and eliminates a performance hotspot | ||||
| | * | | | Typo | Simon MacMullen | 2013-01-03 | 1 | -1/+1 |
| | | | | | |||||
| | * | | | Merge bug25364. | Simon MacMullen | 2013-01-03 | 9 | -151/+135 |
| | |\ \ \ | |||||
| | | * \ \ | Merge bug25367 | Simon MacMullen | 2013-01-03 | 9 | -131/+120 |
| | | |\ \ \ | |||||
| | | | * \ \ | Merge bug25365 | Simon MacMullen | 2013-01-03 | 1 | -19/+19 |
| | | | |\ \ \ | |||||
| | | | | * | | | oops; put exception handling back inbug25365 | Matthias Radestock | 2013-01-03 | 1 | -1/+1 |
| | | | | | | | | |||||
| | | | | * | | | small refactor | Matthias Radestock | 2012-12-29 | 1 | -4/+5 |
| | | | | | | | | |||||
| | | | * | | | | Merge heads | Emile Joubert | 2013-01-03 | 1 | -27/+24 |
| | | | |\ \ \ \ | |||||
| | | | | * \ \ \ | Merged bug25373 | Emile Joubert | 2013-01-03 | 1 | -27/+24 |
| | | | | |\ \ \ \ | |||||
| | | | | | * | | | | only retain ram msgs when inserting into pending_ackbug25373 | Matthias Radestock | 2013-01-02 | 1 | -27/+24 |
| | | | | |/ / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | which keeps memory use constant during fetch operations | ||||
| | | | * | | | | | Merge bug25363. | Simon MacMullen | 2013-01-03 | 8 | -108/+99 |
| | | | |\ \ \ \ \ | | | | | |/ / / / | |||||
| | | | | * | | | | refactor: return less from vq:remove | Matthias Radestock | 2013-01-02 | 1 | -14/+14 |
| | | | | | | | | | |||||
| | | | | * | | | | Merged bug25372 | Emile Joubert | 2013-01-02 | 5 | -45/+42 |
| | | | | |\ \ \ \ | |||||
| | | | | | * | | | | drop IsDelivered from bq:{fetchwhile,ackfold}bug25372 | Matthias Radestock | 2013-01-01 | 4 | -22/+18 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | since we don't need it | ||||
| | | | | | * | | | | replace bq:foreach_ack with a more versatile ackfold | Matthias Radestock | 2013-01-01 | 5 | -34/+35 |
| | | | | | | |_|/ | | | | | | |/| | | |||||
| | | | | * | | | | refactor: rename vq:intern_fetch to 'remove' | Matthias Radestock | 2013-01-02 | 1 | -18/+18 |
| | | | | | | | | | |||||
| | | | | * | | | | cosmetic | Matthias Radestock | 2013-01-02 | 1 | -1/+1 |
| | | | | | | | | | |||||
| | | | | * | | | | cosmetic | Matthias Radestock | 2013-01-02 | 1 | -6/+5 |
| | | | | | | | | | |||||
| | | | | * | | | | cosmetic | Matthias Radestock | 2013-01-01 | 1 | -2/+2 |
| | | | | | | | | | |||||
| | | | | * | | | | refactor: rename drop_expired_messages to drop_expired_msgs | Matthias Radestock | 2013-01-01 | 1 | -7/+7 |
| | | | | |/ / / | |||||
| | | | | * | | | cosmetic | Matthias Radestock | 2012-12-29 | 1 | -6/+3 |
| | | | | | | | | |||||
| | | * | | | | | make reader react to flow control soonerbug25367 | Matthias Radestock | 2012-12-28 | 1 | -7/+4 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ...by blocking when encountering content headers/body frames, rather than method frames of content-bearing commands. | ||||
| | | * | | | | | merge bug25365 into bug25367 | Matthias Radestock | 2012-12-28 | 0 | -0/+0 |
| | | |\ \ \ \ \ | | | | |_|_|/ / | | | |/| | | / | | | | | |_|/ | | | | |/| | | |||||
| | | * | | | | only invoke control_throttle when necesseary | Matthias Radestock | 2012-12-27 | 1 | -18/+17 |
| | | | | | | | |||||
| | | * | | | | merge bug25363 into bug25365 | Matthias Radestock | 2012-12-27 | 1 | -2/+1 |
| | | |\ \ \ \ | | | | | |/ / | | | | |/| | | |||||
| | | | * | | | get rid of bump_reductionsbug25363 | Matthias Radestock | 2012-12-27 | 1 | -2/+1 |
| | | | |/ / | | | | | | | | | | | | | | | | | | | it hurts performance in some cases and credit_flow does a better job | ||||
| | | * | | | create branch | Matthias Radestock | 2012-12-27 | 0 | -0/+0 |
| | | |/ / | |||||
| | * | | | Explainbug25364 | Simon MacMullen | 2013-01-03 | 1 | -0/+3 |
| | | | | | |||||
| | * | | | replace credit_flow:update/3 with a macro | Matthias Radestock | 2012-12-27 | 1 | -31/+41 |
| | |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | un order to eliminate closure creation and thus improve performance. Also... - get rid of a 'get/2' call by processing result of 'erase' instead - inline remaining get/2 calls for performance | ||||
| * | | | optimisation: macro-fy rabbit_channel:incr_stats/3bug25369 | Matthias Radestock | 2012-12-29 | 1 | -21/+23 |
| | | | | | | | | | | | | | | | | | | | | so we can avoid constructing complex terms for stats when stats emission is disabled | ||||
| * | | | optimisation: ditch guards on rabbit_misc:r/{2,3} | Matthias Radestock | 2012-12-28 | 1 | -4/+3 |
| | | | | |||||
| * | | | optimisation: inline rabbit_guid:blocks_to_binary/1 | Matthias Radestock | 2012-12-28 | 1 | -10/+9 |
| | | | |