On 2008-04-10, Dan Kegel <[EMAIL PROTECTED]> wrote: > On Wed, Apr 9, 2008 at 10:59 AM, David Flynn <[EMAIL PROTECTED]> wrote: >> NB, It is warned that: >> Do not use longjmp to transfer control out of an interrupt-handling routine >> >> No ill-effects have been observed by ignoring this. > > You're a braver man than I.
Heh, i've run it for 2**24 iterations and not spotted any resource leaks in the test program. I've run the test program 2e6 times (with 1 fault per program) and not noticed any OS leaks. This is on XP+sp2. I have looked up some more of the internal guts on SEH and have the magic to do it the 'proper' way. But until then, this gets you out of jail. (NB, the previous stuff just won't work on win32, so death soon happens if you have the wrong CPU). ..david ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone _______________________________________________ Schrodinger-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/schrodinger-devel
