On Mon, 24 Mar 2003, Ian Holsman wrote: > Peter Stamfest wrote: > > On Mon, 24 Mar 2003, Jake Brutlag wrote:
> >>>ps.. can you get the CVS-HEAD to build where you are on win32? > >> > >>My current CVS refresh is from 1/29/03, apparently before the > >>thread-safety code was added. This compiled on Win32 with VC++ 6.0 once > >>I added rrd_nan_inf.c to the project and a #include <math.h> in > >>rrd_man_inf.c. But I haven't done any testing with this version on > >>Win32. I'll give the tip of tree version a try if I have time later this > >>week. > > > > In case my thread safety stuff breaks on certain architectures (it is > > streamlined for POSIX threads and I was very aware of the possibility of > > problems in this area) the suggested solution is to write an > > "instantiation" of the rrd_thread_safe.c file for the used thread package. > > [snip] > > the problem isn't with the rrd_thread_safe code, per se, it is with the > addition of > the re-entrant functions localtime_r ctime_r gmtime_r & strtok_r through the > code, which > are not implemented on win32. I submitted a rrd_thread_safe_nt.c which > contains implementations BTW: Are the non *_r functions thread-safe on win32? Are there thread-safe version? I'm just curious. > of the missig re-entrant versions, as well as a win32 specfic implementaion > of a thread safe > get_context call. OK, I have to admit I didn't check your changes. However, I wanted to make this particular point clear in public as it is important for other thread-packages as well. I think I only wrote to Tobi in a private thread about it previously. But it seems you did the Right Thing anyway. Great! I guess this should also find its way into doc/rrdthreads.pod. What do you think about the overall design of the thread-safety stuff? Did it provide for all that was needed to get you going on win32? What could be improved? peter -- Unsubscribe mailto:[EMAIL PROTECTED] Help mailto:[EMAIL PROTECTED] Archive http://www.ee.ethz.ch/~slist/rrd-developers WebAdmin http://www.ee.ethz.ch/~slist/lsg2.cgi
