NAME

bind, mount, unmount − change file name space

SYNOPSIS

include "sys.m";
sys := load Sys Sys->PATH;
 bind:     fn(name, old: string, flag: int): int;
mount:    fn(fd: ref FD; old: string, flag: int, aname: string):
             int;
unmount:  fn(name, old: string): int;

DESCRIPTION

­Bind and ­mount modify the file name space of the current process and its name space group.  For both calls, ­old is the name of an existing file or directory in the current name space where the modification is to be made.  The name ­old is evaluated as described in sys-intro(2) except that no translation of the final path element is done.

For bind, ­name is the name of another (or possibly the same) existing file or directory in the current name space.  After a successful ­bind call, the file name ­old is an alias for the object originally named by name; if the modification does not hide the original, ­name will also still refer to its original file.  The evaluation of ­name happens at the time of the bind, not when the binding is later used. 

The ­fd argument to ­mount is a file descriptor of an open pipe or network connection to a file server ready to receive Styx messages.  The ­old file must be a directory.  After a successful mount, the file tree ­served (see below) by ­fd will be visible with its root directory having name old.

The ­flag controls details of the modification made to the name space.  In the following, ­new refers to the file as defined by ­name or the root directory served by fd. Either both ­old and new files must be directories, or both must not be directories.  ­Flag can be one of:

­Sys->MREPL Replace the ­old file by the new one.  Henceforth, an evaluation of ­old will be translated to the new file.  If they are directories (for mount, this condition is true by definition), ­old becomes a ­union directory consisting of one directory (the new file). 

­Sys->MBEFORE Both the ­old and new files must be directories.  Add the constituent files of the new directory to the union directory at ­old so its contents appear first in the union.  After a ­Sys->MBEFORE ­bind or mount, the new directory will be searched first when evaluating file names in the union directory. 

­Sys->MAFTER Like ­Sys->MBEFORE but the new directory goes at the end of the union. 

In addition, there is a ­Sys->MCREATE flag that can be OR’d with any of the above.  When a ­create call (see sys-open(2)) attempts to create in a union directory, and the file does not exist, the elements of the union are searched in order until one is found with ­Sys->MCREATE set.  The file is created in that directory; if that attempt fails, the ­create fails. 

With mount, the file descriptor ­fd must be open for reading and writing and connected to a file server.  After the mount, the file tree starting at ­old is served by a kernel mnt(3) device. That device will turn operations in the tree into messages to the server on fd. ­Aname selects among different file trees on the server; the empty (or nil) string chooses the default tree. 

The effects of ­bind and ­mount can be undone by unmount.  If ­name is nil, everything bound to or mounted upon ­old is unbound or unmounted.  If ­name is not nil, it is evaluated as described above for bind, and the effect of binding or mounting that particular result on ­old is undone. 

SEE ALSO

sys-intro(2)

DIAGNOSTICS

The return value is a positive integer (a unique sequence number) for success, −1 for failure. 

BUGS

­Mount will not return until it has successfully attached to the file server, so the thread doing a ­mount cannot be the one serving. 

If the target directory is not already a union, ­MBEFORE and ­MAFTER are equivalent to MREPL; bind a directory to itself to create a union if necessary. 

Plan 9  —  June 29, 2000