Firstly, I want to say thank you to the wonderful cocos2d-x developers for such fine work, particularly for the awesomeness that is the end of the v3 branch.
I’ve been working on my game, grabbing large chunks of my spare time, for around 4 years now. At last, a few weeks ago, I reached a releasable point in the development, but all horror, it intermittedly crashed. It’s lua-based and the crashes were SEGVs, so I knew it was unlikely to be caused by my game code. I tried updating to v4, but I’m using some APIs that aren’t fully filled in for lua yet (I’ll do that sometime) I tried going back to v3.17.1, but still it crashed. Last I tried the end of the v3 branch, and it’s now rock solid. I’m so pleased.
Now I’m wondering if it would be a good time to cut a v3.17.3 release?
I’m not great with Lua problem solving, so I can’t help you there.
I don’t know if there will be a 3.17.3. I think probably not. v4 is our next release and I know the team is focusing on this version for the future. Perhaps if there is a compelling reason to release 3.17.3 the engineers may do so, but I don’t hear anything about a future 3.17.x version.
Thanks slackmoehrle for the reply. On this occasion, I’m not looking for help. I’m happy using end of v3 branch on git. It’s working perfectly for me.
I was suggesting a new v3 release just with the thought that it might help others. I’m sure the problems I found with v3.17.1 and v3.17.2 will hit others, and just thought, given how well the end of v3 branch is working, it might make a good v3.17.3. I quite understand that that might not fit with the dev teams release strategy.
The problems I was seeing were probably not Lua related. I mentioned Lua just as an explanation as to why I’d not suspected my own code as the cause of the crashes: a bug in lua code shouldn’t be able to cause a SEGV.
