Cory Fields
4d712e366c
net: add a new message queue for the message processor
...
This separates the storage of messages from the net and queued messages for
processing, allowing the locks to be split.
2017-01-12 23:05:25 -05:00
Cory Fields
c5a8b1b946
net: rework the way that the messagehandler sleeps
...
In order to sleep accurately, the message handler needs to know if _any_ node
has more processing that it should do before the entire thread sleeps.
Rather than returning a value that represents whether ProcessMessages
encountered a message that should trigger a disconnnect, interpret the return
value as whether or not that node has more work to do.
Also, use a global fProcessWake value that can be set by other threads,
which takes precedence (for one cycle) over the messagehandler's decision.
Note that the previous behavior was to only process one message per loop
(except in the case of a bad checksum or invalid header). That was changed in
PR #3180 .
The only change here in that regard is that the current node now falls to the
back of the processing queue for the bad checksum/invalid header cases.
2017-01-12 23:05:24 -05:00
Cory Fields
c72cc88ed3
net: remove useless comments
2017-01-12 23:05:09 -05:00
Cory Fields
56212e20ac
net: set message deserialization version when it's actually time to deserialize
...
We'll soon no longer have access to vRecvMsg, and this is more intuitive anyway.
2017-01-04 10:10:35 -05:00
Cory Fields
0e973d970a
net: remove redundant max sendbuffer size check
...
This is left-over from before there was proper accounting. Hitting 2x the
sendbuffer size should not be possible.
2017-01-04 10:10:35 -05:00
Cory Fields
e5bcd9c84f
net: make vRecvMsg a list so that we can use splice()
2017-01-04 09:29:14 -05:00
Wladimir J. van der Laan
869781c51c
Merge #9283 : A few more CTransactionRef optimizations
...
91335ba6713f0f62607d7c44e4c4
2017-01-04 12:23:10 +01:00
Cory Fields
5cb0fcee81
net: remove thread_interrupted catch
...
This is now a std::thread, so there's no hope of catching a boost interruption
point.
2017-01-03 17:56:21 -05:00
Cory Fields
d3d7056d2a
net: make net processing interruptible
2017-01-03 17:56:20 -05:00
Pieter Wuille
ce5c1f4aca
Merge #9252 : Release cs_main before calling ProcessNewBlock, or processing headers (cmpctblock handling)
...
bd02bdd680b0c0
2017-01-03 13:52:16 -08:00
Suhas Daftuar
bd02bddb93
Release cs_main before processing cmpctblock as header
2016-12-26 05:34:54 -05:00
Suhas Daftuar
680b0c0138
Release cs_main before calling ProcessNewBlock (cmpctblock handling)
2016-12-26 05:30:26 -05:00
Matt Corallo
4cd373aea8
Un-expose mapArgs from utils.h
2016-12-24 11:29:33 -05:00
Matt Corallo
0cf86a6678
Introduce (and use) an IsArgSet accessor method
2016-12-23 21:30:16 -05:00
Pieter Wuille
62607d796c
Convert COrphanTx to keep a CTransactionRef
2016-12-21 18:18:28 -08:00
Pieter Wuille
c44e4c467c
Make AcceptToMemoryPool take CTransactionRef
2016-12-21 18:18:23 -08:00
Wladimir J. van der Laan
e8cfe1ee2d
Merge #8589 : Inline CTxInWitness inside CTxIn
...
f6fb7ac
2016-12-21 14:04:06 +01:00
Wladimir J. van der Laan
a7f76512d9
Merge #9352 : Attempt reconstruction from all compact block announcements
...
813ede97017298
2016-12-19 08:52:42 +01:00
MarcoFalke
fa16b8fdba
If we don't allow free txs, always send a fee filter (take 2)
2016-12-16 14:27:55 +01:00
Wladimir J. van der Laan
756374c522
Merge #9313 : If we don't allow free txs, always send a fee filter
...
01fea7a
2016-12-15 19:45:54 +01:00
Suhas Daftuar
7017298eb2
Allow compactblock reconstruction when block is in flight
2016-12-15 11:47:04 -05:00
Wladimir J. van der Laan
b68685a16a
Merge #9273 : Remove unused CDiskBlockPos* argument from ProcessNewBlock
...
a13fa4c
2016-12-14 10:56:28 +01:00
Gregory Maxwell
da9cdd2c9c
Do not run functions with necessary side-effects in assert()
2016-12-14 01:50:00 +00:00
Alex Morcos
01fea7a048
If we don't allow free txs, always send a fee filter
2016-12-09 15:25:46 -05:00
Pieter Wuille
09c4fd157c
Merge #9268 : Fix rounding privacy leak introduced in #9260
...
9b9324e#9260 (Matt Corallo)
2016-12-06 13:33:01 -08:00
Wladimir J. van der Laan
43e8150ef6
Merge #9269 : Align struct COrphan definition
...
2efc438
2016-12-05 11:09:26 +01:00
Pieter Wuille
2efc43874c
Align struct COrphan definition
2016-12-05 00:35:07 -08:00
Pieter Wuille
f6fb7acda4
Move CTxInWitness inside CTxIn
2016-12-04 23:47:12 -08:00
Wladimir J. van der Laan
d04aebaec7
Merge #9014 : Fix block-connection performance regression
...
dd0df812d6e5612736c44ae4db44fd9d8906fdd43b
2016-12-05 08:07:45 +01:00
Matt Corallo
a13fa4c80f
Remove unused CDiskBlockPos* argument from ProcessNewBlock
2016-12-04 00:23:17 -08:00
Matt Corallo
2d6e5619af
Switch pblock in ProcessNewBlock to a shared_ptr
...
This (finally) fixes a performance regression in
b3b3c2a562
2016-12-04 00:17:30 -08:00
Matt Corallo
9b9324ee49
Fix rounding privacy leak introduced in #9260
2016-12-02 21:30:36 -08:00
Pieter Wuille
81e3228fcb
Make CTransaction actually immutable
2016-12-02 18:37:43 -08:00
Matt Corallo
76faa3cdfe
Rename the remaining main.{h,cpp} to validation.{h,cpp}
2016-12-02 09:42:51 -08:00
Matt Corallo
e736772c56
Move network-msg-processing code out of main to its own file
2016-12-02 09:42:51 -08:00
