SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


    [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

    Re: Bidirectional SCSI commands and iSCSI



    > Jim,
    > 
    > What is the intended use?  Are there already commands that use 
    > it?  Is there a major/minor distinction - like output data 
    > & input results or output parameters & input data - or is it 
    > intended to be completely symmetric?
    
    There can be different amounts of write and read data.  The
    data-out size and data-in size are set by the initiator.
    
    The target chooses:
    * whether to accept write data or provide read data first
    * when to switch direction
    * how often to switch direction
    
    The initiator is not allowed to change the write data after
    sending the CDB - read-modify-write (where the initiator modifies
    the data) is not supported.  Write-modify-read (where the
    target modifies the data) is supported.
    
    T10/00-315 describes a bidirectional command called XDWRITEREAD
    useful for RAID operations.  The object storage device (OSD) 
    command set plans to use bidirectional commands.  Other
    commands may appear in the future that use this feature.
    
    T10/00-314 describes how this is supported on Parallel SCSI.
    
    T10/00-377 describes how this is supported on Fibre Channel 
    (FCP-2).
    
    The 1394 community plans to revise SBP-2 to add bidirectional
    command support.
    
    (all T10 documents are on http://www.t10.org)
    
    > Julo
    > 
    > "Jim Hafner/Almaden/IBM" <hafner@almaden.ibm.com> on 20/09/2000 17:56:27
    > 
    > Please respond to "Jim Hafner/Almaden/IBM" <hafner@almaden.ibm.com>
    > 
    > To:   ips@ece.cmu.edu
    > Subject:  Bidirectional SCSI commands and iSCSI
    > 
    > Folks,
    > 
    > In case you missed this, T10 just approved changes to SAM-2 (rev14) which
    > remove the restriction that a single SCSI command can have a data flow in
    > only one direction (DataIn xor DataOut).   Transport specs (that want to
    > allow this) will need to change to allow for two sets of Data buffer fields
    > in the headers (length, over/underrun, etc.).
    > 
    > Is anybody looking to modify the iSCSI spec to get these concepts
    > supported?
    > 
    > Jim Hafner
    > 
    
    
    -- 
    Rob Elliott      UNIX mailto:relliott@unixmail.compaq.com    
    Houston, TX        PC mailto:Robert.Elliott@compaq.com
    


Home

Last updated: Tue Sep 04 01:07:00 2001
6315 messages in chronological order