On 6/27/16, 3:17 PM, "Michael Cyr" <target-devel-owner@xxxxxxxxxxxxxxx on behalf of mikecyr@xxxxxxxxxxxxxxxxxx> wrote:
<SNIP>
+ cmd->se_cmd.tag = be64_to_cpu(srp->tag);
+
+ spin_lock_bh(&vscsi->intr_lock);
+ list_add_tail(&cmd->list, &vscsi->active_q);
+ spin_unlock_bh(&vscsi->intr_lock);
+
+ srp->lun.scsi_lun[0] &= 0x3f;
+
+ pr_debug("calling submit_cmd, se_cmd %p, lun 0x%llx, cdb 0x%x, attr:%d\n",
+ &cmd->se_cmd, scsilun_to_int(&srp->lun), (int)srp->cdb[0],
+ attr);
+
+ rc = target_submit_cmd(&cmd->se_cmd, nexus->se_sess, srp->cdb,
+ cmd->sense_buf, scsilun_to_int(&srp->lun),
+ data_len, attr, dir, 0);
+ if (rc) {
<SNIP>
+ * ibmvscsis_parse_task() - Parse SRP Task Management Request
+ *
+ * Parse the srp task management request; if it is valid then submit it to tcm.
+ * Note: The return code does not reflect the status of the task management
+ * request.
+ *
+ * EXECUTION ENVIRONMENT:
+ * Processor level
+ */
+static void ibmvscsis_parse_task(struct scsi_info *vscsi,
+ struct ibmvscsis_cmd *cmd)
+{
<SNIP>
+ spin_lock_bh(&vscsi->intr_lock);
+ list_add_tail(&cmd->list, &vscsi->active_q);
+ spin_unlock_bh(&vscsi->intr_lock);
+
+ srp_tsk->lun.scsi_lun[0] &= 0x3f;
+
+ pr_debug("calling submit_tmr, func %d\n",
+ srp_tsk->tsk_mgmt_func);
+ rc = target_submit_tmr(&cmd->se_cmd, nexus->se_sess, NULL,
+ scsilun_to_int(&srp_tsk->lun), srp_tsk,
+ tcm_type, GFP_KERNEL, tag_to_abort, 0);
+ if (rc) {
<SNIP>
I was hoping someone in the community can help answer this:
If I were to remove the masking off of the lun addressing method prior to target_submit_cmd/target_submit_tmr then I get a hang within scsi probe
on bootup. (srp->lun.scsi_lun[0] &= 0x3f; and srp_tsk->lun.scsi_lun[0] &= 0x3f;)
[ 0.842648] ibmvscsi 3000000d: sent SRP login
[ 0.842667] ibmvscsi 3000000d: SRP_LOGIN succeeded
[ 0.842682] scsi 0:0:2:0: CD-ROM AIX VOPTA PQ: 0 ANSI: 4
[ OK ] Started Show Plymouth Boot Screen.
[ OK ] Reached target Paths.
[ OK ] Reached target Basic System.
[ 0.886179] sr 0:0:2:0: [sr0] scsi-1 drive
[ 0.886199] cdrom: Uniform CD-ROM driver Revision: 3.20
[ 0.888582] sd 0:0:1:0: [sda] 41943040 512-byte logical blocks: (21.4 GB/20.0 GiB)
[ 0.888657] sd 0:0:1:0: [sda] Write Protect is off
[ 0.888712] sd 0:0:1:0: [sda] Cache data unavailable
[ 0.888722] sd 0:0:1:0: [sda] Assuming drive cache: write through
[ 0.901443] sda: sda1 sda2 sda3
[ 0.901838] sd 0:0:1:0: [sda] Attached SCSI disk
[ 124.522778] dracut-initqueue[353]: Warning: dracut-initqueue timeout - starting timeout scripts
[ 125.151242] dracut-initqueue[353]: Warning: dracut-initqueue timeout - starting timeout scripts
[ 125.662808] dracut-initqueue[353]: Warning: dracut-initqueue timeout - starting timeout scripts
Does anyone know of the reasoning for having to mask off the addressing method prior
to submitting to target?
Thanks,
Bryant
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
