st ioctl documentation error (MT_ST_SYSV)

[Aaron Traas <atraas@syncsort.com>]

Hello. I'm trying to fix some bugs in a port of my company's proprietary 
tape backup software. Our software works flawlessly on most commercial 
UNIX platforms. When handling tape reads, we use one of two different 
subroutines for handling the positioning of the read-head after a read. 
On BSD-flavored systems (such as *BSD), we do more-or-less nothing, as 
after a read, the tape is spaced forward to the next filemark. On SysV 
flavored systems (like Solaris 8), we advance the filemark manually.

We have a problem on Linux, as it states in the st(4) manpage:

MT_ST_SYSV (Default: false)
    When  this  option  is  enabled,  the  tape
    devices use the SystemV  semantics.  Other­
    wise  the  BSD semantics are used. The most
    important difference between the  semantics
    is  what  happens  when  a  device used for
    reading is closed: in  SYSV  semantics  the
    tape  is spaced forward past the next file­
    mark if this has not happened  while  using
    the device. In BSD semantics the tape posi­
    tion is not changed.

Which switches what we have defined as BSD and SYSV semantics. It states 
that Linux uses BSD semantics by default, and then says SYSV semantics 
are what we know are BSD semantics, and that BSD semantics are what we 
know as SYSV semantics. How do we know? BSD and Solaris' man pages on 
the subject contradict Linux's, and our software properly works on those 
  systems.

So, going by the description in the page on what the semantics are, we 
have defined Linux in our build system as using non-BSD style tape read 
semantics. This results in fsf errors.


-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html