Re: init's children list is long and slows reaping children.

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
From: Jeff Garzik
Date: Monday, April 9, 2007 - 11:53 pm

Eric W. Biederman wrote:


I disagree there is only a little bloat:  the current mechanism in place 
does not scale as NR_CPUS increases, as this thread demonstrates.

Beyond a certain point, on an 8-CPU box, it gets silly.  You certainly 
don't need eight kblockd threads or eight ata threads.

	Jeff


-
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Thu Apr 5, 7:15 pm)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Fri Apr 6, 8:38 am)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Fri Apr 6, 11:02 am)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Fri Apr 6, 11:04 am)
Re: init's children list is long and slows reaping children., Christoph Hellwig, (Fri Apr 6, 11:05 am)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Fri Apr 6, 11:30 am)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Fri Apr 6, 11:56 am)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Fri Apr 6, 12:39 pm)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Mon Apr 9, 1:00 pm)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Mon Apr 9, 5:48 pm)
Re: init's children list is long and slows reaping children., Alexey Dobriyan, (Mon Apr 9, 10:07 pm)
Re: init's children list is long and slows reaping children., Jeff Garzik, (Mon Apr 9, 11:53 pm)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Tue Apr 10, 7:51 am)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Tue Apr 10, 8:00 am)
[PATCH] Only send pdeath_signal when getppid changes., Eric W. Biederman, (Tue Apr 10, 8:16 am)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Tue Apr 10, 8:22 am)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Tue Apr 10, 9:17 am)
Re: [PATCH] Only send pdeath_signal when getppid changes., Oleg Nesterov, (Tue Apr 10, 9:37 am)
Re: [PATCH] Only send pdeath_signal when getppid changes., Eric W. Biederman, (Tue Apr 10, 10:41 am)
Re: [PATCH] Only send pdeath_signal when getppid changes., Roland McGrath, (Tue Apr 10, 10:48 am)
Re: init's children list is long and slows reaping children., Davide Libenzi, (Tue Apr 10, 12:17 pm)
Re: [PATCH] Only send pdeath_signal when getppid changes., Albert Cahalan, (Tue Apr 10, 8:17 pm)
[patch] uninline remove/add_parent() APIs, Ingo Molnar, (Tue Apr 10, 11:20 pm)
Re: [patch] uninline remove/add_parent() APIs, Eric W. Biederman, (Wed Apr 11, 12:00 am)
Re: init's children list is long and slows reaping children., Eric W. Biederman, (Wed Apr 11, 1:17 pm)
Re: [patch] uninline remove/add_parent() APIs, Andrew Morton, (Wed Apr 11, 3:06 pm)
Re: [patch] uninline remove/add_parent() APIs, Eric W. Biederman, (Thu Apr 12, 3:45 am)
Re: [patch] uninline remove/add_parent() APIs, Roland McGrath, (Thu Apr 12, 3:50 pm)