On Fri, Aug 14, 2009 at 04:44:42PM +0200, Corinna Vinschen wrote: >On Aug 14 10:15, Charles Wilson wrote: >> Corinna Vinschen wrote: >> >> > Oh, ok, sorry about that. I didn't realize there's a urxvt-X and a >> > urxvtc-X, and accidentally grabbed urxvtc-X when creating the shortcut. >> > >> > And yes, it's like you expected. When starting on W7 without run, a >> > command window briefly pops up and disappears again. The CPU usage is >> > low, afterwards. >> >> Thanks. >> >> OK, then here's how I'm going to "solve" the problem. I will modify run >> so that instead of using the "new" code on WinXP+ incl. W7, it will only >> use the "new" code on W7. That way, things will continue to work as >> they did in run-1.1.10 for existing OS's. This should also "solve" Ken >> Brown's problem with run -> batch file -> bash(.dotfiles) -> emacs.exe >> leading to poor performance -- at least until he upgrades to W7. >> >> For W7, "don't do that" -- don't launch urxvt-X (standalone) or urxvtd-X >> (daemon) using run, at all. Instead, rely on urxvt[d]-X's own code to >> hide the console window. The downside of this is a brief flicker. >> >> Now, you CAN use 'run' on W7 in conjunction with urxvtc-X (client), with >> no ill effects AFAICT. So, if you use the client/daemon urxvt, you get >> only ONE flicker when you first launch the daemon (perhaps thru >> .Xsession), and after that everything is fine. >> >> I'll also update the documentation to reflect this limitation. >> >> Later, when W7 in is more widespread use (e.g. when I have personal >> access to it), I'll see if I can improve the situation a bit. Maybe I'll >> add a switch "-f/--flicker-ok" to enable use of a mechanism similar to >> urxvt[d]-X's hide_console() [only, without the 'check to see if this >> console already existed before I was launched' intelligence] rather than >> cygwin's CREATE_NO_WINDOW, just for situations like urxvt[d]-X and Ken's >> emacs example. > >I just tested this further. If you omit the CREATE_NO_WINDOW flag >to CreateProcess, the console window still flickers when starting >urxvt, but the CPU usage does not stick to 100%. Since XWin and >xterm are apparently unaffected by this (no flickering cmd window) >it looks like that's the better interim solution.
Isn't just creating the process with SW_HIDE good enough? cgf -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple