Eli Zaretskii writes: >> Cc: agustin.martin@hispalinux.es, michael.cadilhac@lrde.org, >> monnier@iro.umontreal.ca, emacs-devel@gnu.org >> From: storm@cua.dk (Kim F. Storm) >> Date: Mon, 29 May 2006 10:14:22 +0200 >> >> But... if MS-Windows does not support SIGCHLD (or SIGCLD), how does >> emacs detect process termination on MS-Windows?? > > It uses a Windows specific API to probe for subprocess termination, > and when that tells us that a child process terminated, we call the > signal handler by hand. So, is there a case in which the deleted_pid list is not necessary, and in which we should not install the mechanism (to avoid leaks as well as unnecessary processing) ? It seems that the guard we put (#ifdef SIGCHLD) was not the right thing to do, isn't it ? -- | Michaël `Micha' Cadilhac | All your base are belong to us. | | Epita/LRDE Promo 2007 | You have no change to survive | | http://www.lrde.org/~cadilh_m | make your time, hahaha. | `-- - JID: micha@amessage.be --' -- Zero Wings - --'