yes. we need a flag to achieve that goal. but there are 2 flags.
[mingz@amd64 iscsitarget-0.4.11]$ grep -rn ISCSI_FLG_STATUS *
kernel/iscsi_hdr.h:242:#define ISCSI_FLG_STATUS 0x01
kernel/iscsi.c:271: rsp->flags = ISCSI_FLG_FINAL |
ISCSI_FLG_STATUS;
[mingz@amd64 iscsitarget-0.4.11]$ grep -rn ISCSI_FLG_FINAL *
kernel/iscsi_hdr.h:351:#define ISCSI_FLG_FINAL 0x80
kernel/iscsi.c:271: rsp->flags = ISCSI_FLG_FINAL |
ISCSI_FLG_STATUS;
kernel/iscsi.c:310: rsp_hdr->flags = ISCSI_FLG_FINAL;
kernel/iscsi.c:349: rsp_hdr->flags = ISCSI_FLG_FINAL;
kernel/iscsi.c:767: rsp_hdr->flags = ISCSI_FLG_FINAL;
kernel/iscsi.c:1029: if (scsi_cmnd->pdu.bhs.flags &
ISCSI_FLG_FINAL) {
kernel/iscsi.c:1070: if (req->flags & ISCSI_FLG_FINAL) {
kernel/iscsi.c:1077: if (!(req->flags & ISCSI_FLG_FINAL) &&
scsi_cmnd->r2t_length == 0)
kernel/iscsi.c:1080: if (!(req->flags & ISCSI_FLG_FINAL))
kernel/iscsi.c:1173: rsp_hdr->flags = ISCSI_FLG_FINAL;
kernel/iscsi.c:1236: rsp_hdr->flags = ISCSI_FLG_FINAL;
kernel/iscsi.c:1267: rsp_hdr->flags = ISCSI_FLG_FINAL;
kernel/iscsi.c:1423: cmnd_set_sn(cmnd, (rsp->flags &
ISCSI_FLG_FINAL) ? 1 : 0);
usr/iscsi_hdr.h:135:#define ISCSI_FLG_FINAL 0x80
usr/iscsid.c:605: if (req->flags & ISCSI_FLG_FINAL)
usr/iscsid.c:606: rsp->flags = ISCSI_FLG_FINAL;
usr/iscsid.c:620: rsp->flags = ISCSI_FLG_FINAL;
see ISCSI_FLG_STATUS, only set while never be evaluated.
see ISCSI_FLG_FINAL, set and evaluated here and there.
ming
On Tue, 2005-07-19 at 07:35 +0200, Arne Redlich wrote:
> Am Montag, den 18.07.2005, 17:11 -0400 schrieb Ming Zhang:
> > Hi,
> >
> > in do_send_data_rsp(), there is a
> > rsp->flags = ISCSI_FLG_FINAL | ISCSI_FLG_STATUS;
> >
> > but
> > [mingz@amd64 iscsitarget-0.4.11]$ grep -rn ISCSI_FLG_STATUS *
> > kernel/iscsi_hdr.h:242:#define ISCSI_FLG_STATUS 0x01
> > kernel/iscsi.c:271: rsp->flags = ISCSI_FLG_FINAL |
> > ISCSI_FLG_STATUS;
> >
> > i guess we shall get it removed?
>
> No! The status flag is set in the last Data-In PDU of a sequence (hence
> the F-bit is set, too) and indicates that the data contains SCSI status
> information and that the residual count flags and field of the PDU are
> to be evaluated - see RFC3720, 10.7.
>
> Arne
>
>
> -------------------------------------------------------
> SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
> from IBM. Find simple to follow Roadmaps, straightforward articles,
> informative Webcasts and more! Get everything you need to get up to
> speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
> _______________________________________________
> Iscsitarget-devel mailing list
> Iscsitarget-devel@xxxxxxxxxxxxxxxxxxxxx
> https://lists.sourceforge.net/lists/listinfo/iscsitarget-devel
-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
|
