>> Also findutils lists S_IFMPB and S_IFMPC as "V7" but you
>> say "Coherent". I'm also collecting a list -- could you tell
>> me which is appropriate, or is it both: "Coherent & V7".
>
> From memory - V7 only had one type of special file which is
> non-standard by todays standards. These are multiplexor files
> which were being manipulated by the mpx(2) syscall.
SCO might document the mpx(2) call at http://doc.sco.com or
http://docs.sco.com in the OpenServer documentation. There
ought to be a section that covers the old Xenix calls.
> They never really did exist on the V7 filesystem because the
> V7 fsck considered multiplexor files as corroupted inodes and
> deleted them ... Wasn't the symbol used for them S_IFMPX?
I have these:
hex name ls octal description
3000 S_IFMPC 030000 multiplexed char device (Coherent & v7)
5000 S_IFNAM 050000 XENIX special named file
7000 S_IFMPB 070000 multiplexed block device (Coherent & v7)
The S_IFNAM type has subtypes S_INSEM and S_INSHD, using st_rdev
values of 1 and 2 to tell them apart. An S_INSEM file is a Xenix
semaphore, shown as type 's' in a file list. An S_INSHD file is a
Xenix shared data (memory) file, shown as type 'm' in a file list.
I believe SysV systems show both S_IFIFO and S_IFSOCK as type 'p',
eliminating the 's' conflict.
The S_IFMPC and S_IFMPB types might just show as normal character
and block devices.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/