Yes, I have tried lsof (its one of my favorites) and lsof is actually slower. Its even slower than "fuser -v", which does the same thing as "lsof". After suggestions and hints from others, my idea was premature. I understand that I'd have to set a lock on task struct, which in itself, might be more time consuming. Apparently, 2.4.x doesn't have a task list structure, rather it has a struct with pointers to the beginning and end. Even though my original idea is stalled, I am still very much open to ideas. Thanks, greg -----Original Message----- From: Ravi [mailto:kravi26@yahoo.com] Sent: Monday, June 09, 2003 7:25 AM To: Greg Nate; kernelnewbies@nl.linux.org Subject: Re: Help with writing a new "fuser" --- Greg Nate <Greg_Nate@adp.com> wrote: > Hello all, I hope this is the proper group. Here's where I need > direction. > > I want to write a new "fuser", which makes direct kernel table calls, > rather > than what the current "fuser" does, which seems to be a > file-tree-walk on > /proc. We've found that on a heavily loaded system, the current > "fuser" is > far too slow and can actually have a serious performance impact. Before you go ahead with this, have you tried using 'lsof'? I have found that many times 'lsof' is faster than fuser. Of course, I haven't looked into the source of either of them, so I don't know how different they really are. -Ravi. __________________________________ Do you Yahoo!? Yahoo! Calendar - Free online calendar with sync to Outlook(TM). http://calendar.yahoo.com -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
