On Wed, Jun 21, 2006 at 07:44:15AM +0100, Kaveh Goudarzi wrote: >cgf wrote: >>The problem may be due to the fact that the injected thread doesn't have >>a cygtls area. If the thread is injected prior to the completion of >>cygwin's initialization or if it somehow bypasses the DLL_THREAD_ATTACH >>phase of the initialization (which was one problem that I had to deal >>with in the dreaded "cygwin loops forever" problem) then there will be >>no cygtls area and using cygwin functions will be problematic. > >[snip] > >As for the cygtls area ... is this something necessary for proper >functioning of cygwin calls? (I looked at the source ... but none the >wise I'm afraid :-( ) ...
"Using cygwin functions will be problematic": Some simple functions will work some other functions will crash, some may hang. >Empirically I see getenv working now "all" the time but the call to >cyg_internal seems to fail with regularity. Is this expected? Sorry. Don't know. I've given you the basic facts but researching this further is not something that I'm interested in doing. Your use is very much a corner case and if it doesn't work then I'm not terribly concerned. However, if you think you have a fix for the behavior you can certainly submit it to cygwin-patches. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
