Re: qla2xxx: Add LLD target-mode infrastructure for >= 24xx series

["Jörn Engel" <joern@logfs.org>]

Hello Nick!

On Thu, 15 March 2012 16:10:15 -0700, Nicholas A. Bellinger wrote:
> 
> iff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c
> index e415d1e..2af075b 100644
> --- a/drivers/scsi/qla2xxx/qla_target.c
> +++ b/drivers/scsi/qla2xxx/qla_target.c
> @@ -3169,11 +3169,11 @@ static void qla_tgt_reject_free_srr_imm(struct scsi_qla_host *vha, struct qla_tg
>  static void qla_tgt_handle_srr_work(struct work_struct *work)
>  {
>         struct qla_tgt *tgt = container_of(work, struct qla_tgt, srr_work);
> -       struct scsi_qla_host *vha = NULL;
> +       struct scsi_qla_host *vha = tgt->vha;
>         struct qla_tgt_srr_ctio *sctio;
>         unsigned long flags;
>  
> -       ql_dbg(ql_dbg_tgt_mgt, tgt->vha, 0xe12e, "Entering SRR work (tgt %p)\n", tgt);
> +       ql_dbg(ql_dbg_tgt_mgt, vha, 0xe12e, "Entering SRR work (tgt %p)\n", tgt);
>  
>  restart:
>         spin_lock_irqsave(&tgt->srr_lock, flags);
> @@ -3192,17 +3192,17 @@ restart:
>                                           "be only one IMM SRR per CTIO SRR "
>                                           "(IMM SRR %p, id %d, CTIO %p\n",
>                                           vha->vp_idx, i, i->srr_id, sctio);
> -                                       qla_tgt_reject_free_srr_imm(vha, i, 0);
> +                                       qla_tgt_reject_free_srr_imm(tgt->vha, i, 0);

Everywhere else you move from tgt->vha to vha.  Is there a reason for
doing the reverse here?

Jörn