1 Jan
2012
1 Jan
'12
12:42 a.m.
Yes. I should document it, but internally the library uses universal_time().
OK, that is a surprise ! You really need to document that all timed locks in interprocess use UTC and not local time. IMO saying "absolute time" as you do in the doc does not infer UTC, and it is more than natural to wrongly assume local time.
I'm not an expert on time interfaces, so maybe it was an incorrect choice. Hopefully porting it to C++11 will solve those problems. Let's see if we can find some time soon to do it. Thanks for the bug reports, Ion