On Mon, 30 Jul 2007 22:56:27 CDT, Joseph Pingenot said: > From Al Viro on Tuesday, 31 July, 2007: > >On Mon, Jul 30, 2007 at 10:40:59PM -0500, Joseph Pingenot wrote: > >> I'm trying to implement pwait. It blocks until a specified PID exits, > >> and then it exits. > >er... ptrace(2)? > > Should work for most common usage scenarios, although will suspect that it > won't for for processes owned by another user (at least, I hope > it wouldn't). > > What is dangerous about inotify on a proc file? And if it's dangerous, should the inotify system call fail when trying to set the watch? Bailout with a -EBADIDEA or something?
Attachment:
pgpjs5uHSFDLM.pgp
Description: PGP signature
- References:
- inotify and /proc/<pid>
- From: Joseph Pingenot <[email protected]>
- Re: inotify and /proc/<pid>
- From: Al Viro <[email protected]>
- Re: inotify and /proc/<pid>
- From: Joseph Pingenot <[email protected]>
- Re: inotify and /proc/<pid>
- From: Joseph Pingenot <[email protected]>
- Re: inotify and /proc/<pid>
- From: Al Viro <[email protected]>
- Re: inotify and /proc/<pid>
- From: Joseph Pingenot <[email protected]>
- Re: inotify and /proc/<pid>
- From: Al Viro <[email protected]>
- Re: inotify and /proc/<pid>
- From: Joseph Pingenot <[email protected]>
- inotify and /proc/<pid>
- Prev by Date: [PATCH] Fix section mismatch warnings in sound/pci/hda/
- Next by Date: Re: [PATH 0/1] Kexec jump - v2 - the first step to kexec based hibernation
- Previous by thread: Re: inotify and /proc/<pid>
- Next by thread: Re: inotify and /proc/<pid>
- Index(es):
 |