SORT BY:

LIST ORDER
THREAD
AUTHOR
SUBJECT


SEARCH

IPS HOME


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

    Re: iSCSI: Use of the A bit



    
    I am not sure that was the problem.  The point was if they have such a
    device that has such a shortage of memory that it can never complete the
    I/O without running out of buffer.  It wants to free up the buffer so that
    it can send some more data so that it can fulfill the Read request.
    
    If the Initiator supports the A-bit SNACK (which is optional), then some
    buffer control would be possible.  But as was pointed out by Matthew, with
    ErrorRecoveryLevel=0, there is no need to even use the A-bit, since there
    is no error recovery within command, you can just send the data, then
    refill the buffer.  If there is an error, the session has to be torn down.
    
    However, if you have that same type of device but support
    ErrorRecoveryLevel=1, then SNACK with A-bit support seems like an strong
    preference for a corresponding initiator.  If the device detects that the
    A-bit is not supported at the initiator (because A-bit was set, but no
    response was received at the approprate after a period of time), then it
    probably should "promote" the recovery to level=0, and start refilling the
    buffers as mentioned above.
    
    
    .
    .
    .
    John L. Hufferd
    Senior Technical Staff Member (STSM)
    IBM/SSG San Jose Ca
    Main Office (408) 256-0403, Tie: 276-0403,  eFax: (408) 904-4688
    Home Office (408) 997-6136, Cell: (408) 499-9702
    Internet address: hufferd@us.ibm.com
    
    
    Santosh Rao <santoshr@hpcuhe.cup.hp.com>@ece.cmu.edu on 03/13/2002 07:25:21
    PM
    
    Sent by:    owner-ips@ece.cmu.edu
    
    
    To:    ips@ece.cmu.edu (ips)
    cc:
    Subject:    Re: iSCSI: Use of the A bit
    
    
    
    >
    > Actually it does not actually say that.  It says "This is limited to
    > sessions that support error recovery and is implemented through the A bit
    > ..."
    >
    > I know what I am about to say, is nit picking, but ....
    > ErrorRecoveryLevel=0 is an error recovery technique.
    >
    > Having said that, we probably should just change that statement to "This
    is
    > implemented through the A bit ..."
    
    John,
    
    Initiators that use "ErrorRecoveryLevel=0" must not be required to respond
    to
    an A bit setting with a DataACK. "ErrorRecoveryLevel=0" implies the
    initiator
    need not support/use SNACK pdu.
    
    A target can recycle its data buffers once it has completed sending the
    data,
    since it knows that the initiator will not issue Data SNACK while operating
    as
    "ErrorRecoveryLevel=0".
    
    - Santosh
    
    --
    ##################################
    Santosh Rao
    Software Design Engineer,
    HP-UX iSCSI Driver Team,
    Hewlett Packard, Cupertino.
    email : santoshr@cup.hp.com
    Phone : 408-447-3751
    ##################################
    
    
    
    


Home

Last updated: Thu Mar 14 08:21:40 2002
9111 messages in chronological order