The remove paths on send fail thing in Peer.cpp was not well thought out, and there is no point in mallocing the TCP write buffer.
This commit is contained in:
parent
c3b41c289d
commit
0e1fc06a6f
3 changed files with 20 additions and 49 deletions
|
@ -78,10 +78,8 @@ protected:
|
|||
Socket(t,s),
|
||||
_lastActivity(Utils::now()),
|
||||
_sm(sm),
|
||||
_outbuf((unsigned char *)0),
|
||||
_outptr(0),
|
||||
_outbufsize(0),
|
||||
_inptr(0),
|
||||
_outptr(0),
|
||||
_connecting(c),
|
||||
_remote(r)
|
||||
{
|
||||
|
@ -93,12 +91,11 @@ protected:
|
|||
|
||||
private:
|
||||
unsigned char _inbuf[ZT_SOCKET_MAX_MESSAGE_LEN];
|
||||
unsigned char _outbuf[ZT_SOCKET_MAX_MESSAGE_LEN * 4];
|
||||
uint64_t _lastActivity; // updated whenever data is received, checked directly by SocketManager for stale TCP cleanup
|
||||
SocketManager *_sm;
|
||||
unsigned char *_outbuf;
|
||||
unsigned int _outptr;
|
||||
unsigned int _outbufsize;
|
||||
unsigned int _inptr;
|
||||
unsigned int _outptr;
|
||||
bool _connecting; // manipulated directly by SocketManager, true if connect() is in progress
|
||||
InetAddress _remote;
|
||||
Mutex _writeLock;
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue