On Monday 04 January 2010, Øyvind Harboe wrote: > On Mon, Jan 4, 2010 at 8:59 PM, David Brownell <davi...@pacbell.net> wrote: > > On Monday 04 January 2010, Øyvind Harboe wrote: > >> > >> This is a hack of course, but I'm wondering if it is the best > >> hack for 0.4... > > > > I'd like a solution that gets rid of *all* the errors on > > execution of the "shutdown" command. > > > > Is there a reason why shutdown doesn't just cause exit()? > > It's less "clean",
But it's "correct" ... where the current code isn't. > e.g. for embedded hosts there is no resource > tracking that happens then + it's nice to be able to have > something like valgrind print out a list of all leaked objects, > which a call to exit() would prevent. I don't seem to use valgrind. There's a reason that it doesn't call atexit()? Or is your isssue that you don't have even a stubbed version of exit() available, which calls the atexit() handlers and then loops forever? Depending on the embedding model, I'd think that the policy would be just to turn the "shutdown" operation into a NOP. That is, if running the OpenOCD server is the entire purpose of the box, then it must not support shutting down. - Dave _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
