> >> in the "select" and the socket's "write" there is sometimes a > >> non-conclusion happening. Anybody want to offer me a suggestion? > >>
Well actually it is a crapy bug in select() on Windows since years. We discovered it at Cygwin/Xfree86 project, last week. Select was doing bizzare things on Windows ME, someone from IEEE.ORG pointed out to the bugs in Winsock, which according to him M$ developers do intentionally as an inginious way to get things working. Nothing much can be done about it, except to code DX in a way to work around M$ bugs. Suhaib > -----Original Message----- > From: David Thompson [mailto:[EMAIL PROTECTED] > Sent: Wednesday, April 18, 2001 12:31 PM > To: [email protected] > Subject: RE: [opendx-dev] Windows--winsock > > > Suhaib, > > Your code exhibits the same behavior. I ran your version and it > appears to be something with the way windows is working with the > queuing aspect of DXLink. It is not specific to the Windows sockets. > > David > > >I do not see this problem with my binaries. I suspect it is > something in > >code. Have you tried VC++ 6.0 latest Service Pack 5? > > > >Suhaib > > > > > >> -----Original Message----- > >> From: David Thompson [mailto:[EMAIL PROTECTED] > >> Sent: Monday, April 16, 2001 2:12 PM > >> To: [email protected] > >> Subject: [opendx-dev] Windows--winsock > >> > >> > >> Wondering if any of the Windows gurus have seen problems > with the DX > >> link code on windows. I'm noticing something strange in > terms dx link > >> working within the startup window. If I start up DX run one sample > >> everything is fine, click on Samples again (not closing > the previous > >> one), sometimes it will start the new one, sometimes it will just > >> change execution modes to "End Execution", other times do nothing, > >> and yet others--dump. I've tried tracing the code and it > seems that > >> in the "select" and the socket's "write" there is sometimes a > >> non-conclusion happening. Anybody want to offer me a suggestion? > >> > >> I'm using the Feb '01 psdk and VC++ 6. Is there an easy way > >> to trace sockets? > >> > >> David > >> -- > >> .............................................................. > >> ............... > >> David L. Thompson The University > of Montana > >> mailto:[EMAIL PROTECTED] Computer > Science Department > >> http://www.cs.umt.edu/u/dthompsn Missoula, MT 59812 > >> Work Phone : > (406)257-8530 > >> > > -- > .............................................................. > ............... > David L. Thompson The University of Montana > mailto:[EMAIL PROTECTED] Computer Science Department > http://www.cs.umt.edu/u/dthompsn Missoula, MT 59812 > Work Phone : (406)257-8530 >
