Written by Garrett Wollman
This file is in the public domain.
.Dd August 10, 1994 .Dt PROCFS 5 .Os BSD 4.4 .Sh NAME .Nm procfs .Nd process file system .Sh SYNOPSIS d -literal proc /proc procfs rw 0 0 .Ed .Sh DESCRIPTION The process file system, or .Nm , implements a view of the system process table inside the file system. It is normally mounted on
a /proc , and is required for the complete operation of programs such as .Xr ps 1 and .Xr w 1 .
p The .Nm provides a two-level view of process space, unlike the previous .Fx 1.1 .Nm implementation. At the highest level, processes themselves are named, according to their process ids in decimal, with no leading zeros. There is also a special node called
a curproc which always refers to the process making the lookup request.
p Each node is a directory which contains the following entries: l -tag -width fpregsxx t Pa file the executable image (not currently implemented) t Pa status the process's current status t Pa ctl used to send control messages to the process, such as might be used by a debugger t Pa mem the complete virtual address space of the process t Pa note used to send signals to the process (not currently implemented) t Pa notepg used to send signals to the process group that the process belongs to (not currently implemented) t Pa regs the process register set .El
p Each node is owned by the process's user, and belongs to that user's primary group, except for the
a mem node, which belongs to the .Li kmem group instead. (This is not yet true, but will be eventually.)
p This manual page should eventually document what you can do with all these nodes. .Sh FILES l -tag -width /procxx t Pa /proc normal mount point for the .Nm procfs . .El .Sh AUTHOR This manual page written by Garrett Wollman, based on the description provided by Jan-Simon Pendry.