Age | Commit message (Collapse) | Author | |
---|---|---|---|
2020-11-15 | 1.3.17 release prepv1.3.17 | Lee Salzman | |
2020-11-13 | avoid sending packets in unacknowledged window | Lee Salzman | |
2020-10-19 | fix for sending getting too far ahead of receiver | Lee Salzman | |
2020-09-08 | 1.3.16 release prepv1.3.16 | Lee Salzman | |
2020-09-08 | revert failed throttle changes | Lee Salzman | |
2020-09-05 | clamp RTT variance a bit more loosely for throttle | Lee Salzman | |
2020-09-03 | round RTT stats before comparing | Lee Salzman | |
2020-09-01 | clamp minimum highest RTT variance | Lee Salzman | |
2020-08-23 | more command queuing fixes | Lee Salzman | |
2020-08-23 | command queuing fix | Lee Salzman | |
2020-08-23 | make throttle more readily accelerate | Lee Salzman | |
2020-08-12 | make throttle even more tolerant of variance | Lee Salzman | |
2020-07-24 | clamp throttle variance from below based on RTT percentage | Lee Salzman | |
2020-07-23 | fix clearing of outgoing command queue | Lee Salzman | |
2020-07-23 | use unified outgoing command queue for reliable and unreliable commands | Lee Salzman | |
2020-04-20 | fix doc licensev1.3.15 | Lee Salzman | |
2020-04-19 | enet 1.3.15 release prep | Lee Salzman | |
2020-04-16 | update license dates | Lee Salzman | |
2020-04-16 | stabilize packet throttle when RTT variance is low | Lee Salzman | |
2020-04-13 | don't throttle on first RTT measurement | Lee Salzman | |
2020-04-07 | Merge pull request #120 from Vincenz099/master | Lee Salzman | |
Return 0 instead of -1 on enet_protocol_receive_incoming_commands | |||
2020-04-07 | Set DWORD sentLength = 0; | Maxim | |
Same case like receive, we should initialize this variable. | |||
2020-04-07 | Se trecvLength = 0; | Maxim | |
2020-04-07 | Make recvLength = 0; In the rare case of it having bogus data. | Maxim | |
I noticed from testing this that this might be the source of it, I once had a package returned with length total bogus. like 982349829 or something like that, it crashed my game, I suspect this is the source. | |||
2020-04-07 | Return 0 instead of -1 on enet_protocol_receive_incoming_commands | Maxim | |
Return 0 instead of -1 on enet_protocol_receive_incoming_commands when nothing received. This allows the Service loop to continue running and not return an error when there is nothing to do with the socket receive. From debugging I found sometimes the enet_protocol_receive_incoming_commands returns -1 because simply nothing happened in the 256 for loop. Other functions like enet_protocol_send_outgoing_commands return 0 when nothing happened. | |||
2020-04-04 | accumulate fractional RTT values | Lee Salzman | |
2020-04-03 | clamp roundTripTime | Lee Salzman | |
2020-03-03 | make RTT calculations more TCP-like (contributed by Vladimir 'virtul' Ivannikov) | Lee Salzman | |
2019-01-27 | enet 1.3.14 release prepv1.3.14 | Lee Salzman | |
2019-01-27 | delay handling of DISCONNECT_LATER until in-flight unreliable packets are sent | Lee Salzman | |
2018-09-11 | Merge pull request #96 from Subv/patch-2 | Lee Salzman | |
Add the <arpa/inet.h> include to unix.h | |||
2018-09-10 | Remove the now unneeded <arpa/inet.h> include in unix.c | Sebastian Valle | |
This is already included in unix.h | |||
2018-09-11 | Merge pull request #95 from Subv/patch-1 | Lee Salzman | |
Use <poll.h> instead of <sys/poll.h> in unix.c | |||
2018-09-10 | Add the <arpa/inet.h> include to unix.h. | Sebastian Valle | |
Some files were using ENET_HOST_TO_NET_32 and ENET_NET_TO_HOST_32 without having included the <arpa/inet.h> file beforehand, leading to compiler warnings about implicit declarations of ntohl and htonl. | |||
2018-09-10 | Use <poll.h> instead of <sys/poll.h> in unix.c | Sebastian Valle | |
<poll.h> is the correct POSIX header according to the POSIX standard. Some targets (like homebrew for the Nintendo Switch) do not provide a <sys/poll.h> so it makes compiling for them rather hard. | |||
2018-08-19 | remove bandwidth limits from tutorial | Lee Salzman | |
2017-10-20 | Merge pull request #83 from thelvyn/master | Lee Salzman | |
CMake source groups added for cleaner VS solutions. | |||
2017-10-15 | CMake source groups added for cleaner VS solutions. | Lukasz Fronc | |
2017-07-12 | Merge pull request #76 from jroweboy/mingw-fix | Lee Salzman | |
Fix mingw compilation when using CMake | |||
2017-07-12 | Fix mingw compilation | James Rowe | |
In order to compile enet on mingw, you need to link against winmm and ws2_32. This explicitly makes those libraries required on mingw | |||
2017-05-21 | Merge pull request #71 from cxong/patch-2 | Lee Salzman | |
Don't treat warnings as errors | |||
2017-05-22 | Don't treat warnings as errors | Cong | |
When enet is included as a child CMake project in a parent that has warnings treated as errors, enet fails to build. This ensures that the compiler doesn't treat warnings as errors for enet. | |||
2017-05-21 | Merge pull request #70 from cxong/patch-1 | Lee Salzman | |
reduce warnings on MSVC | |||
2017-05-21 | MSVC only warn up to level 3 | Cong | |
enet produces warnings at level 4, which can interrupt compiles if warnings are treated as errors. Turning down the warning level resolves the issue. | |||
2017-05-06 | add portable enet_address_set_host_ip that can properly parse broadcast address | Lee Salzman | |
2016-06-04 | typo fix | [email protected] | |
2016-02-19 | update copyright years | [email protected] | |
2016-02-19 | return -1 instead of 0 if getnameinfo returns an error | [email protected] | |
2015-12-30 | update packet flag documentation | Lee Salzman | |
2015-05-11 | use getaddrinfo and getnameinfo where available | Lee Salzman | |