Neil Horman
Neil Horman
@t8m hmm, I think I could see how that might work, but I'd be concerned about enforcing the behavior (If I understand it correctly). To elaborate, if we create a...
@t8m @paulidale @Sashan I think its something of a moot point. I thought @t8m 's idea was interesting, so I put together this branch last night: https://github.com/nhorman/openssl/tree/singlethreaded-contexts Which creates a...
so, we should prbably separate our thinking here. @t8m suggestion was a method to both improve performance and eliminate some amount of thread-local storage. The performance numbers seemed not to...
pushed to master, thank you!
@baentsch I think just changing a parameter name might not be considered abi breaking, but I'm not sure what our tests would say about it. Probably best to do in...
merged to master, thank you for the reviews
marking as inactive, will be closed at the completion of 3.4 dev without further input
Inactive means an issue has gone untouched in a sufficiently long time that I suspect no ones paying attention to it any more, so I'm threatening to close them to...
In reading over this again, a few things occured to me: 1) In lines 8 and 16 of the initial crash you encountered, we see calls to context_deinit, however they...
At exit problems have been a major issue for the 3.0 series, resulting in lots of odd bugs, like [this one](https://github.com/openssl/openssl/issues/22508), where we found that C++ libs used at exit...