X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=drivers%2Fscsi%2Flibiscsi.c;h=703eb6a88790280c788b9fb88ce331d15ff282e1;hb=0c43ea544c1086fbbed5a6c99ea58eb64674ea8f;hp=d43f909a022c2916e7fe97185809c6e8f826abc9;hpb=6320377fd94316e58f75d0be6f6e7a644950a4ee;p=safe%2Fjmp%2Flinux-2.6 diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index d43f909..703eb6a 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -37,16 +38,50 @@ #include #include -static void fail_command(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, - int err); - -struct iscsi_session * -class_to_transport_session(struct iscsi_cls_session *cls_session) -{ - struct Scsi_Host *shost = iscsi_session_to_shost(cls_session); - return iscsi_hostdata(shost->hostdata); -} -EXPORT_SYMBOL_GPL(class_to_transport_session); +static int iscsi_dbg_lib_conn; +module_param_named(debug_libiscsi_conn, iscsi_dbg_lib_conn, int, + S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(debug_libiscsi_conn, + "Turn on debugging for connections in libiscsi module. " + "Set to 1 to turn on, and zero to turn off. Default is off."); + +static int iscsi_dbg_lib_session; +module_param_named(debug_libiscsi_session, iscsi_dbg_lib_session, int, + S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(debug_libiscsi_session, + "Turn on debugging for sessions in libiscsi module. " + "Set to 1 to turn on, and zero to turn off. Default is off."); + +static int iscsi_dbg_lib_eh; +module_param_named(debug_libiscsi_eh, iscsi_dbg_lib_eh, int, + S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(debug_libiscsi_eh, + "Turn on debugging for error handling in libiscsi module. " + "Set to 1 to turn on, and zero to turn off. Default is off."); + +#define ISCSI_DBG_CONN(_conn, dbg_fmt, arg...) \ + do { \ + if (iscsi_dbg_lib_conn) \ + iscsi_conn_printk(KERN_INFO, _conn, \ + "%s " dbg_fmt, \ + __func__, ##arg); \ + } while (0); + +#define ISCSI_DBG_SESSION(_session, dbg_fmt, arg...) \ + do { \ + if (iscsi_dbg_lib_session) \ + iscsi_session_printk(KERN_INFO, _session, \ + "%s " dbg_fmt, \ + __func__, ##arg); \ + } while (0); + +#define ISCSI_DBG_EH(_session, dbg_fmt, arg...) \ + do { \ + if (iscsi_dbg_lib_eh) \ + iscsi_session_printk(KERN_INFO, _session, \ + "%s " dbg_fmt, \ + __func__, ##arg); \ + } while (0); /* Serial Number Arithmetic, 32 bits, less than, RFC1982 */ #define SNA32_CHECK 2147483648UL @@ -64,12 +99,19 @@ static int iscsi_sna_lte(u32 n1, u32 n2) (n1 > n2 && (n2 - n1 < SNA32_CHECK))); } -void -iscsi_update_cmdsn(struct iscsi_session *session, struct iscsi_nopin *hdr) +inline void iscsi_conn_queue_work(struct iscsi_conn *conn) { - uint32_t max_cmdsn = be32_to_cpu(hdr->max_cmdsn); - uint32_t exp_cmdsn = be32_to_cpu(hdr->exp_cmdsn); + struct Scsi_Host *shost = conn->session->host; + struct iscsi_host *ihost = shost_priv(shost); + + if (ihost->workq) + queue_work(ihost->workq, &conn->xmitwork); +} +EXPORT_SYMBOL_GPL(iscsi_conn_queue_work); +static void __iscsi_update_cmdsn(struct iscsi_session *session, + uint32_t exp_cmdsn, uint32_t max_cmdsn) +{ /* * standard specifies this check for when to update expected and * max sequence numbers @@ -88,93 +130,290 @@ iscsi_update_cmdsn(struct iscsi_session *session, struct iscsi_nopin *hdr) * if the window closed with IO queued, then kick the * xmit thread */ - if (!list_empty(&session->leadconn->xmitqueue) || + if (!list_empty(&session->leadconn->cmdqueue) || !list_empty(&session->leadconn->mgmtqueue)) - scsi_queue_work(session->host, - &session->leadconn->xmitwork); + iscsi_conn_queue_work(session->leadconn); } } + +void iscsi_update_cmdsn(struct iscsi_session *session, struct iscsi_nopin *hdr) +{ + __iscsi_update_cmdsn(session, be32_to_cpu(hdr->exp_cmdsn), + be32_to_cpu(hdr->max_cmdsn)); +} EXPORT_SYMBOL_GPL(iscsi_update_cmdsn); -void iscsi_prep_unsolicit_data_pdu(struct iscsi_cmd_task *ctask, - struct iscsi_data *hdr) +/** + * iscsi_prep_data_out_pdu - initialize Data-Out + * @task: scsi command task + * @r2t: R2T info + * @hdr: iscsi data in pdu + * + * Notes: + * Initialize Data-Out within this R2T sequence and finds + * proper data_offset within this SCSI command. + * + * This function is called with connection lock taken. + **/ +void iscsi_prep_data_out_pdu(struct iscsi_task *task, struct iscsi_r2t_info *r2t, + struct iscsi_data *hdr) { - struct iscsi_conn *conn = ctask->conn; + struct iscsi_conn *conn = task->conn; + unsigned int left = r2t->data_length - r2t->sent; + + task->hdr_len = sizeof(struct iscsi_data); memset(hdr, 0, sizeof(struct iscsi_data)); - hdr->ttt = cpu_to_be32(ISCSI_RESERVED_TAG); - hdr->datasn = cpu_to_be32(ctask->unsol_datasn); - ctask->unsol_datasn++; + hdr->ttt = r2t->ttt; + hdr->datasn = cpu_to_be32(r2t->datasn); + r2t->datasn++; hdr->opcode = ISCSI_OP_SCSI_DATA_OUT; - memcpy(hdr->lun, ctask->hdr->lun, sizeof(hdr->lun)); - - hdr->itt = ctask->hdr->itt; - hdr->exp_statsn = cpu_to_be32(conn->exp_statsn); - hdr->offset = cpu_to_be32(ctask->unsol_offset); - - if (ctask->unsol_count > conn->max_xmit_dlength) { + memcpy(hdr->lun, task->lun, sizeof(hdr->lun)); + hdr->itt = task->hdr_itt; + hdr->exp_statsn = r2t->exp_statsn; + hdr->offset = cpu_to_be32(r2t->data_offset + r2t->sent); + if (left > conn->max_xmit_dlength) { hton24(hdr->dlength, conn->max_xmit_dlength); - ctask->data_count = conn->max_xmit_dlength; - ctask->unsol_offset += ctask->data_count; + r2t->data_count = conn->max_xmit_dlength; hdr->flags = 0; } else { - hton24(hdr->dlength, ctask->unsol_count); - ctask->data_count = ctask->unsol_count; + hton24(hdr->dlength, left); + r2t->data_count = left; hdr->flags = ISCSI_FLAG_CMD_FINAL; } + conn->dataout_pdus_cnt++; } -EXPORT_SYMBOL_GPL(iscsi_prep_unsolicit_data_pdu); +EXPORT_SYMBOL_GPL(iscsi_prep_data_out_pdu); -static int iscsi_add_hdr(struct iscsi_cmd_task *ctask, unsigned len) +static int iscsi_add_hdr(struct iscsi_task *task, unsigned len) { - unsigned exp_len = ctask->hdr_len + len; + unsigned exp_len = task->hdr_len + len; - if (exp_len > ctask->hdr_max) { + if (exp_len > task->hdr_max) { WARN_ON(1); return -EINVAL; } WARN_ON(len & (ISCSI_PAD_LEN - 1)); /* caller must pad the AHS */ - ctask->hdr_len = exp_len; + task->hdr_len = exp_len; + return 0; +} + +/* + * make an extended cdb AHS + */ +static int iscsi_prep_ecdb_ahs(struct iscsi_task *task) +{ + struct scsi_cmnd *cmd = task->sc; + unsigned rlen, pad_len; + unsigned short ahslength; + struct iscsi_ecdb_ahdr *ecdb_ahdr; + int rc; + + ecdb_ahdr = iscsi_next_hdr(task); + rlen = cmd->cmd_len - ISCSI_CDB_SIZE; + + BUG_ON(rlen > sizeof(ecdb_ahdr->ecdb)); + ahslength = rlen + sizeof(ecdb_ahdr->reserved); + + pad_len = iscsi_padding(rlen); + + rc = iscsi_add_hdr(task, sizeof(ecdb_ahdr->ahslength) + + sizeof(ecdb_ahdr->ahstype) + ahslength + pad_len); + if (rc) + return rc; + + if (pad_len) + memset(&ecdb_ahdr->ecdb[rlen], 0, pad_len); + + ecdb_ahdr->ahslength = cpu_to_be16(ahslength); + ecdb_ahdr->ahstype = ISCSI_AHSTYPE_CDB; + ecdb_ahdr->reserved = 0; + memcpy(ecdb_ahdr->ecdb, cmd->cmnd + ISCSI_CDB_SIZE, rlen); + + ISCSI_DBG_SESSION(task->conn->session, + "iscsi_prep_ecdb_ahs: varlen_cdb_len %d " + "rlen %d pad_len %d ahs_length %d iscsi_headers_size " + "%u\n", cmd->cmd_len, rlen, pad_len, ahslength, + task->hdr_len); + return 0; +} + +static int iscsi_prep_bidi_ahs(struct iscsi_task *task) +{ + struct scsi_cmnd *sc = task->sc; + struct iscsi_rlength_ahdr *rlen_ahdr; + int rc; + + rlen_ahdr = iscsi_next_hdr(task); + rc = iscsi_add_hdr(task, sizeof(*rlen_ahdr)); + if (rc) + return rc; + + rlen_ahdr->ahslength = + cpu_to_be16(sizeof(rlen_ahdr->read_length) + + sizeof(rlen_ahdr->reserved)); + rlen_ahdr->ahstype = ISCSI_AHSTYPE_RLENGTH; + rlen_ahdr->reserved = 0; + rlen_ahdr->read_length = cpu_to_be32(scsi_in(sc)->length); + + ISCSI_DBG_SESSION(task->conn->session, + "bidi-in rlen_ahdr->read_length(%d) " + "rlen_ahdr->ahslength(%d)\n", + be32_to_cpu(rlen_ahdr->read_length), + be16_to_cpu(rlen_ahdr->ahslength)); + return 0; +} + +/** + * iscsi_check_tmf_restrictions - check if a task is affected by TMF + * @task: iscsi task + * @opcode: opcode to check for + * + * During TMF a task has to be checked if it's affected. + * All unrelated I/O can be passed through, but I/O to the + * affected LUN should be restricted. + * If 'fast_abort' is set we won't be sending any I/O to the + * affected LUN. + * Otherwise the target is waiting for all TTTs to be completed, + * so we have to send all outstanding Data-Out PDUs to the target. + */ +static int iscsi_check_tmf_restrictions(struct iscsi_task *task, int opcode) +{ + struct iscsi_conn *conn = task->conn; + struct iscsi_tm *tmf = &conn->tmhdr; + unsigned int hdr_lun; + + if (conn->tmf_state == TMF_INITIAL) + return 0; + + if ((tmf->opcode & ISCSI_OPCODE_MASK) != ISCSI_OP_SCSI_TMFUNC) + return 0; + + switch (ISCSI_TM_FUNC_VALUE(tmf)) { + case ISCSI_TM_FUNC_LOGICAL_UNIT_RESET: + /* + * Allow PDUs for unrelated LUNs + */ + hdr_lun = scsilun_to_int((struct scsi_lun *)tmf->lun); + if (hdr_lun != task->sc->device->lun) + return 0; + /* fall through */ + case ISCSI_TM_FUNC_TARGET_WARM_RESET: + /* + * Fail all SCSI cmd PDUs + */ + if (opcode != ISCSI_OP_SCSI_DATA_OUT) { + iscsi_conn_printk(KERN_INFO, conn, + "task [op %x/%x itt " + "0x%x/0x%x] " + "rejected.\n", + task->hdr->opcode, opcode, + task->itt, task->hdr_itt); + return -EACCES; + } + /* + * And also all data-out PDUs in response to R2T + * if fast_abort is set. + */ + if (conn->session->fast_abort) { + iscsi_conn_printk(KERN_INFO, conn, + "task [op %x/%x itt " + "0x%x/0x%x] fast abort.\n", + task->hdr->opcode, opcode, + task->itt, task->hdr_itt); + return -EACCES; + } + break; + case ISCSI_TM_FUNC_ABORT_TASK: + /* + * the caller has already checked if the task + * they want to abort was in the pending queue so if + * we are here the cmd pdu has gone out already, and + * we will only hit this for data-outs + */ + if (opcode == ISCSI_OP_SCSI_DATA_OUT && + task->hdr_itt == tmf->rtt) { + ISCSI_DBG_SESSION(conn->session, + "Preventing task %x/%x from sending " + "data-out due to abort task in " + "progress\n", task->itt, + task->hdr_itt); + return -EACCES; + } + break; + } + return 0; } /** * iscsi_prep_scsi_cmd_pdu - prep iscsi scsi cmd pdu - * @ctask: iscsi cmd task + * @task: iscsi task * * Prep basic iSCSI PDU fields for a scsi cmd pdu. The LLD should set * fields like dlength or final based on how much data it sends */ -static int iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask) +static int iscsi_prep_scsi_cmd_pdu(struct iscsi_task *task) { - struct iscsi_conn *conn = ctask->conn; + struct iscsi_conn *conn = task->conn; struct iscsi_session *session = conn->session; - struct iscsi_cmd *hdr = ctask->hdr; - struct scsi_cmnd *sc = ctask->sc; - unsigned hdrlength; + struct scsi_cmnd *sc = task->sc; + struct iscsi_cmd *hdr; + unsigned hdrlength, cmd_len; + itt_t itt; int rc; - ctask->hdr_len = 0; - rc = iscsi_add_hdr(ctask, sizeof(*hdr)); + rc = iscsi_check_tmf_restrictions(task, ISCSI_OP_SCSI_CMD); + if (rc) + return rc; + + if (conn->session->tt->alloc_pdu) { + rc = conn->session->tt->alloc_pdu(task, ISCSI_OP_SCSI_CMD); + if (rc) + return rc; + } + hdr = (struct iscsi_cmd *) task->hdr; + itt = hdr->itt; + memset(hdr, 0, sizeof(*hdr)); + + if (session->tt->parse_pdu_itt) + hdr->itt = task->hdr_itt = itt; + else + hdr->itt = task->hdr_itt = build_itt(task->itt, + task->conn->session->age); + task->hdr_len = 0; + rc = iscsi_add_hdr(task, sizeof(*hdr)); if (rc) return rc; - hdr->opcode = ISCSI_OP_SCSI_CMD; - hdr->flags = ISCSI_ATTR_SIMPLE; - int_to_scsilun(sc->device->lun, (struct scsi_lun *)hdr->lun); - hdr->itt = build_itt(ctask->itt, conn->id, session->age); - hdr->data_length = cpu_to_be32(scsi_bufflen(sc)); - hdr->cmdsn = cpu_to_be32(session->cmdsn); - session->cmdsn++; - hdr->exp_statsn = cpu_to_be32(conn->exp_statsn); - memcpy(hdr->cdb, sc->cmnd, sc->cmd_len); - if (sc->cmd_len < MAX_COMMAND_SIZE) - memset(&hdr->cdb[sc->cmd_len], 0, - MAX_COMMAND_SIZE - sc->cmd_len); - - ctask->data_count = 0; - ctask->imm_count = 0; + hdr->opcode = ISCSI_OP_SCSI_CMD; + hdr->flags = ISCSI_ATTR_SIMPLE; + int_to_scsilun(sc->device->lun, (struct scsi_lun *)hdr->lun); + memcpy(task->lun, hdr->lun, sizeof(task->lun)); + hdr->exp_statsn = cpu_to_be32(conn->exp_statsn); + cmd_len = sc->cmd_len; + if (cmd_len < ISCSI_CDB_SIZE) + memset(&hdr->cdb[cmd_len], 0, ISCSI_CDB_SIZE - cmd_len); + else if (cmd_len > ISCSI_CDB_SIZE) { + rc = iscsi_prep_ecdb_ahs(task); + if (rc) + return rc; + cmd_len = ISCSI_CDB_SIZE; + } + memcpy(hdr->cdb, sc->cmnd, cmd_len); + + task->imm_count = 0; + if (scsi_bidi_cmnd(sc)) { + hdr->flags |= ISCSI_FLAG_CMD_READ; + rc = iscsi_prep_bidi_ahs(task); + if (rc) + return rc; + } if (sc->sc_data_direction == DMA_TO_DEVICE) { + unsigned out_len = scsi_out(sc)->length; + struct iscsi_r2t_info *r2t = &task->unsol_r2t; + + hdr->data_length = cpu_to_be32(out_len); hdr->flags |= ISCSI_FLAG_CMD_WRITE; /* * Write counters: @@ -186,44 +425,45 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask) * without R2T ack right after * immediate data * - * r2t_data_count bytes to be sent via R2T ack's + * r2t data_length bytes to be sent via R2T ack's * * pad_count bytes to be sent as zero-padding */ - ctask->unsol_count = 0; - ctask->unsol_offset = 0; - ctask->unsol_datasn = 0; + memset(r2t, 0, sizeof(*r2t)); if (session->imm_data_en) { - if (scsi_bufflen(sc) >= session->first_burst) - ctask->imm_count = min(session->first_burst, + if (out_len >= session->first_burst) + task->imm_count = min(session->first_burst, conn->max_xmit_dlength); else - ctask->imm_count = min(scsi_bufflen(sc), + task->imm_count = min(out_len, conn->max_xmit_dlength); - hton24(ctask->hdr->dlength, ctask->imm_count); + hton24(hdr->dlength, task->imm_count); } else - zero_data(ctask->hdr->dlength); + zero_data(hdr->dlength); if (!session->initial_r2t_en) { - ctask->unsol_count = min((session->first_burst), - (scsi_bufflen(sc))) - ctask->imm_count; - ctask->unsol_offset = ctask->imm_count; + r2t->data_length = min(session->first_burst, out_len) - + task->imm_count; + r2t->data_offset = task->imm_count; + r2t->ttt = cpu_to_be32(ISCSI_RESERVED_TAG); + r2t->exp_statsn = cpu_to_be32(conn->exp_statsn); } - if (!ctask->unsol_count) + if (!task->unsol_r2t.data_length) /* No unsolicit Data-Out's */ - ctask->hdr->flags |= ISCSI_FLAG_CMD_FINAL; + hdr->flags |= ISCSI_FLAG_CMD_FINAL; } else { hdr->flags |= ISCSI_FLAG_CMD_FINAL; zero_data(hdr->dlength); + hdr->data_length = cpu_to_be32(scsi_in(sc)->length); if (sc->sc_data_direction == DMA_FROM_DEVICE) hdr->flags |= ISCSI_FLAG_CMD_READ; } /* calculate size of additional header segments (AHSs) */ - hdrlength = ctask->hdr_len - sizeof(*hdr); + hdrlength = task->hdr_len - sizeof(*hdr); WARN_ON(hdrlength & (ISCSI_PAD_LEN-1)); hdrlength /= ISCSI_PAD_LEN; @@ -231,54 +471,351 @@ static int iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask) WARN_ON(hdrlength >= 256); hdr->hlength = hdrlength & 0xFF; - conn->scsicmd_pdus_cnt++; + if (session->tt->init_task && session->tt->init_task(task)) + return -EIO; - debug_scsi("iscsi prep [%s cid %d sc %p cdb 0x%x itt 0x%x len %d " - "cmdsn %d win %d]\n", - sc->sc_data_direction == DMA_TO_DEVICE ? "write" : "read", - conn->id, sc, sc->cmnd[0], ctask->itt, scsi_bufflen(sc), - session->cmdsn, session->max_cmdsn - session->exp_cmdsn + 1); + task->state = ISCSI_TASK_RUNNING; + hdr->cmdsn = task->cmdsn = cpu_to_be32(session->cmdsn); + session->cmdsn++; + + conn->scsicmd_pdus_cnt++; + ISCSI_DBG_SESSION(session, "iscsi prep [%s cid %d sc %p cdb 0x%x " + "itt 0x%x len %d bidi_len %d cmdsn %d win %d]\n", + scsi_bidi_cmnd(sc) ? "bidirectional" : + sc->sc_data_direction == DMA_TO_DEVICE ? + "write" : "read", conn->id, sc, sc->cmnd[0], + task->itt, scsi_bufflen(sc), + scsi_bidi_cmnd(sc) ? scsi_in(sc)->length : 0, + session->cmdsn, + session->max_cmdsn - session->exp_cmdsn + 1); return 0; } /** - * iscsi_complete_command - return command back to scsi-ml - * @ctask: iscsi cmd task + * iscsi_free_task - free a task + * @task: iscsi cmd task * * Must be called with session lock. - * This function returns the scsi command to scsi-ml and returns - * the cmd task to the pool of available cmd tasks. + * This function returns the scsi command to scsi-ml or cleans + * up mgmt tasks then returns the task to the pool. */ -static void iscsi_complete_command(struct iscsi_cmd_task *ctask) +static void iscsi_free_task(struct iscsi_task *task) { - struct iscsi_session *session = ctask->conn->session; - struct scsi_cmnd *sc = ctask->sc; + struct iscsi_conn *conn = task->conn; + struct iscsi_session *session = conn->session; + struct scsi_cmnd *sc = task->sc; - ctask->state = ISCSI_TASK_COMPLETED; - ctask->sc = NULL; - /* SCSI eh reuses commands to verify us */ - sc->SCp.ptr = NULL; - list_del_init(&ctask->running); - __kfifo_put(session->cmdpool.queue, (void*)&ctask, sizeof(void*)); - sc->scsi_done(sc); + ISCSI_DBG_SESSION(session, "freeing task itt 0x%x state %d sc %p\n", + task->itt, task->state, task->sc); + + session->tt->cleanup_task(task); + task->state = ISCSI_TASK_FREE; + task->sc = NULL; + /* + * login task is preallocated so do not free + */ + if (conn->login_task == task) + return; + + kfifo_in(&session->cmdpool.queue, (void*)&task, sizeof(void*)); + + if (sc) { + task->sc = NULL; + /* SCSI eh reuses commands to verify us */ + sc->SCp.ptr = NULL; + /* + * queue command may call this to free the task, but + * not have setup the sc callback + */ + if (sc->scsi_done) + sc->scsi_done(sc); + } +} + +void __iscsi_get_task(struct iscsi_task *task) +{ + atomic_inc(&task->refcount); +} +EXPORT_SYMBOL_GPL(__iscsi_get_task); + +static void __iscsi_put_task(struct iscsi_task *task) +{ + if (atomic_dec_and_test(&task->refcount)) + iscsi_free_task(task); +} + +void iscsi_put_task(struct iscsi_task *task) +{ + struct iscsi_session *session = task->conn->session; + + spin_lock_bh(&session->lock); + __iscsi_put_task(task); + spin_unlock_bh(&session->lock); +} +EXPORT_SYMBOL_GPL(iscsi_put_task); + +/** + * iscsi_complete_task - finish a task + * @task: iscsi cmd task + * @state: state to complete task with + * + * Must be called with session lock. + */ +static void iscsi_complete_task(struct iscsi_task *task, int state) +{ + struct iscsi_conn *conn = task->conn; + + ISCSI_DBG_SESSION(conn->session, + "complete task itt 0x%x state %d sc %p\n", + task->itt, task->state, task->sc); + if (task->state == ISCSI_TASK_COMPLETED || + task->state == ISCSI_TASK_ABRT_TMF || + task->state == ISCSI_TASK_ABRT_SESS_RECOV) + return; + WARN_ON_ONCE(task->state == ISCSI_TASK_FREE); + task->state = state; + + if (!list_empty(&task->running)) + list_del_init(&task->running); + + if (conn->task == task) + conn->task = NULL; + + if (conn->ping_task == task) + conn->ping_task = NULL; + + /* release get from queueing */ + __iscsi_put_task(task); +} + +/** + * iscsi_complete_scsi_task - finish scsi task normally + * @task: iscsi task for scsi cmd + * @exp_cmdsn: expected cmd sn in cpu format + * @max_cmdsn: max cmd sn in cpu format + * + * This is used when drivers do not need or cannot perform + * lower level pdu processing. + * + * Called with session lock + */ +void iscsi_complete_scsi_task(struct iscsi_task *task, + uint32_t exp_cmdsn, uint32_t max_cmdsn) +{ + struct iscsi_conn *conn = task->conn; + + ISCSI_DBG_SESSION(conn->session, "[itt 0x%x]\n", task->itt); + + conn->last_recv = jiffies; + __iscsi_update_cmdsn(conn->session, exp_cmdsn, max_cmdsn); + iscsi_complete_task(task, ISCSI_TASK_COMPLETED); +} +EXPORT_SYMBOL_GPL(iscsi_complete_scsi_task); + + +/* + * session lock must be held and if not called for a task that is + * still pending or from the xmit thread, then xmit thread must + * be suspended. + */ +static void fail_scsi_task(struct iscsi_task *task, int err) +{ + struct iscsi_conn *conn = task->conn; + struct scsi_cmnd *sc; + int state; + + /* + * if a command completes and we get a successful tmf response + * we will hit this because the scsi eh abort code does not take + * a ref to the task. + */ + sc = task->sc; + if (!sc) + return; + + if (task->state == ISCSI_TASK_PENDING) { + /* + * cmd never made it to the xmit thread, so we should not count + * the cmd in the sequencing + */ + conn->session->queued_cmdsn--; + /* it was never sent so just complete like normal */ + state = ISCSI_TASK_COMPLETED; + } else if (err == DID_TRANSPORT_DISRUPTED) + state = ISCSI_TASK_ABRT_SESS_RECOV; + else + state = ISCSI_TASK_ABRT_TMF; + + sc->result = err << 16; + if (!scsi_bidi_cmnd(sc)) + scsi_set_resid(sc, scsi_bufflen(sc)); + else { + scsi_out(sc)->resid = scsi_out(sc)->length; + scsi_in(sc)->resid = scsi_in(sc)->length; + } + + iscsi_complete_task(task, state); +} + +static int iscsi_prep_mgmt_task(struct iscsi_conn *conn, + struct iscsi_task *task) +{ + struct iscsi_session *session = conn->session; + struct iscsi_hdr *hdr = task->hdr; + struct iscsi_nopout *nop = (struct iscsi_nopout *)hdr; + uint8_t opcode = hdr->opcode & ISCSI_OPCODE_MASK; + + if (conn->session->state == ISCSI_STATE_LOGGING_OUT) + return -ENOTCONN; + + if (opcode != ISCSI_OP_LOGIN && opcode != ISCSI_OP_TEXT) + nop->exp_statsn = cpu_to_be32(conn->exp_statsn); + /* + * pre-format CmdSN for outgoing PDU. + */ + nop->cmdsn = cpu_to_be32(session->cmdsn); + if (hdr->itt != RESERVED_ITT) { + /* + * TODO: We always use immediate for normal session pdus. + * If we start to send tmfs or nops as non-immediate then + * we should start checking the cmdsn numbers for mgmt tasks. + * + * During discovery sessions iscsid sends TEXT as non immediate, + * but we always only send one PDU at a time. + */ + if (conn->c_stage == ISCSI_CONN_STARTED && + !(hdr->opcode & ISCSI_OP_IMMEDIATE)) { + session->queued_cmdsn++; + session->cmdsn++; + } + } + + if (session->tt->init_task && session->tt->init_task(task)) + return -EIO; + + if ((hdr->opcode & ISCSI_OPCODE_MASK) == ISCSI_OP_LOGOUT) + session->state = ISCSI_STATE_LOGGING_OUT; + + task->state = ISCSI_TASK_RUNNING; + ISCSI_DBG_SESSION(session, "mgmtpdu [op 0x%x hdr->itt 0x%x " + "datalen %d]\n", hdr->opcode & ISCSI_OPCODE_MASK, + hdr->itt, task->data_count); + return 0; } -static void __iscsi_get_ctask(struct iscsi_cmd_task *ctask) +static struct iscsi_task * +__iscsi_conn_send_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr, + char *data, uint32_t data_size) { - atomic_inc(&ctask->refcount); + struct iscsi_session *session = conn->session; + struct iscsi_host *ihost = shost_priv(session->host); + uint8_t opcode = hdr->opcode & ISCSI_OPCODE_MASK; + struct iscsi_task *task; + itt_t itt; + + if (session->state == ISCSI_STATE_TERMINATE) + return NULL; + + if (opcode == ISCSI_OP_LOGIN || opcode == ISCSI_OP_TEXT) { + /* + * Login and Text are sent serially, in + * request-followed-by-response sequence. + * Same task can be used. Same ITT must be used. + * Note that login_task is preallocated at conn_create(). + */ + if (conn->login_task->state != ISCSI_TASK_FREE) { + iscsi_conn_printk(KERN_ERR, conn, "Login/Text in " + "progress. Cannot start new task.\n"); + return NULL; + } + + task = conn->login_task; + } else { + if (session->state != ISCSI_STATE_LOGGED_IN) + return NULL; + + BUG_ON(conn->c_stage == ISCSI_CONN_INITIAL_STAGE); + BUG_ON(conn->c_stage == ISCSI_CONN_STOPPED); + + if (!kfifo_out(&session->cmdpool.queue, + (void*)&task, sizeof(void*))) + return NULL; + } + /* + * released in complete pdu for task we expect a response for, and + * released by the lld when it has transmitted the task for + * pdus we do not expect a response for. + */ + atomic_set(&task->refcount, 1); + task->conn = conn; + task->sc = NULL; + INIT_LIST_HEAD(&task->running); + task->state = ISCSI_TASK_PENDING; + + if (data_size) { + memcpy(task->data, data, data_size); + task->data_count = data_size; + } else + task->data_count = 0; + + if (conn->session->tt->alloc_pdu) { + if (conn->session->tt->alloc_pdu(task, hdr->opcode)) { + iscsi_conn_printk(KERN_ERR, conn, "Could not allocate " + "pdu for mgmt task.\n"); + goto free_task; + } + } + + itt = task->hdr->itt; + task->hdr_len = sizeof(struct iscsi_hdr); + memcpy(task->hdr, hdr, sizeof(struct iscsi_hdr)); + + if (hdr->itt != RESERVED_ITT) { + if (session->tt->parse_pdu_itt) + task->hdr->itt = itt; + else + task->hdr->itt = build_itt(task->itt, + task->conn->session->age); + } + + if (!ihost->workq) { + if (iscsi_prep_mgmt_task(conn, task)) + goto free_task; + + if (session->tt->xmit_task(task)) + goto free_task; + } else { + list_add_tail(&task->running, &conn->mgmtqueue); + iscsi_conn_queue_work(conn); + } + + return task; + +free_task: + __iscsi_put_task(task); + return NULL; } -static void __iscsi_put_ctask(struct iscsi_cmd_task *ctask) +int iscsi_conn_send_pdu(struct iscsi_cls_conn *cls_conn, struct iscsi_hdr *hdr, + char *data, uint32_t data_size) { - if (atomic_dec_and_test(&ctask->refcount)) - iscsi_complete_command(ctask); + struct iscsi_conn *conn = cls_conn->dd_data; + struct iscsi_session *session = conn->session; + int err = 0; + + spin_lock_bh(&session->lock); + if (!__iscsi_conn_send_pdu(conn, hdr, data, data_size)) + err = -EPERM; + spin_unlock_bh(&session->lock); + return err; } +EXPORT_SYMBOL_GPL(iscsi_conn_send_pdu); /** * iscsi_cmd_rsp - SCSI Command Response processing * @conn: iscsi connection * @hdr: iscsi header - * @ctask: scsi command task + * @task: scsi command task * @data: cmd data buffer * @datalen: len of buffer * @@ -286,12 +823,12 @@ static void __iscsi_put_ctask(struct iscsi_cmd_task *ctask) * then completes the command and task. **/ static void iscsi_scsi_cmd_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr, - struct iscsi_cmd_task *ctask, char *data, + struct iscsi_task *task, char *data, int datalen) { struct iscsi_cmd_rsp *rhdr = (struct iscsi_cmd_rsp *)hdr; struct iscsi_session *session = conn->session; - struct scsi_cmnd *sc = ctask->sc; + struct scsi_cmnd *sc = task->sc; iscsi_update_cmdsn(session, (struct iscsi_nopin*)rhdr); conn->exp_statsn = be32_to_cpu(rhdr->statsn) + 1; @@ -308,20 +845,34 @@ static void iscsi_scsi_cmd_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr, if (datalen < 2) { invalid_datalen: - printk(KERN_ERR "iscsi: Got CHECK_CONDITION but " - "invalid data buffer size of %d\n", datalen); + iscsi_conn_printk(KERN_ERR, conn, + "Got CHECK_CONDITION but invalid data " + "buffer size of %d\n", datalen); sc->result = DID_BAD_TARGET << 16; goto out; } - senselen = be16_to_cpu(get_unaligned((__be16 *) data)); + senselen = get_unaligned_be16(data); if (datalen < senselen) goto invalid_datalen; memcpy(sc->sense_buffer, data + 2, min_t(uint16_t, senselen, SCSI_SENSE_BUFFERSIZE)); - debug_scsi("copied %d bytes of sense\n", - min_t(uint16_t, senselen, SCSI_SENSE_BUFFERSIZE)); + ISCSI_DBG_SESSION(session, "copied %d bytes of sense\n", + min_t(uint16_t, senselen, + SCSI_SENSE_BUFFERSIZE)); + } + + if (rhdr->flags & (ISCSI_FLAG_CMD_BIDI_UNDERFLOW | + ISCSI_FLAG_CMD_BIDI_OVERFLOW)) { + int res_count = be32_to_cpu(rhdr->bi_residual_count); + + if (scsi_bidi_cmnd(sc) && res_count > 0 && + (rhdr->flags & ISCSI_FLAG_CMD_BIDI_OVERFLOW || + res_count <= scsi_in(sc)->length)) + scsi_in(sc)->resid = res_count; + else + sc->result = (DID_BAD_TARGET << 16) | rhdr->cmd_status; } if (rhdr->flags & (ISCSI_FLAG_CMD_UNDERFLOW | @@ -331,27 +882,62 @@ invalid_datalen: if (res_count > 0 && (rhdr->flags & ISCSI_FLAG_CMD_OVERFLOW || res_count <= scsi_bufflen(sc))) + /* write side for bidi or uni-io set_resid */ scsi_set_resid(sc, res_count); else sc->result = (DID_BAD_TARGET << 16) | rhdr->cmd_status; - } else if (rhdr->flags & (ISCSI_FLAG_CMD_BIDI_UNDERFLOW | - ISCSI_FLAG_CMD_BIDI_OVERFLOW)) - sc->result = (DID_BAD_TARGET << 16) | rhdr->cmd_status; - + } out: - debug_scsi("done [sc %lx res %d itt 0x%x]\n", - (long)sc, sc->result, ctask->itt); + ISCSI_DBG_SESSION(session, "cmd rsp done [sc %p res %d itt 0x%x]\n", + sc, sc->result, task->itt); conn->scsirsp_pdus_cnt++; - - __iscsi_put_ctask(ctask); + iscsi_complete_task(task, ISCSI_TASK_COMPLETED); } -static void iscsi_tmf_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr) +/** + * iscsi_data_in_rsp - SCSI Data-In Response processing + * @conn: iscsi connection + * @hdr: iscsi pdu + * @task: scsi command task + **/ +static void +iscsi_data_in_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr, + struct iscsi_task *task) { - struct iscsi_tm_rsp *tmf = (struct iscsi_tm_rsp *)hdr; + struct iscsi_data_rsp *rhdr = (struct iscsi_data_rsp *)hdr; + struct scsi_cmnd *sc = task->sc; - conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; - conn->tmfrsp_pdus_cnt++; + if (!(rhdr->flags & ISCSI_FLAG_DATA_STATUS)) + return; + + iscsi_update_cmdsn(conn->session, (struct iscsi_nopin *)hdr); + sc->result = (DID_OK << 16) | rhdr->cmd_status; + conn->exp_statsn = be32_to_cpu(rhdr->statsn) + 1; + if (rhdr->flags & (ISCSI_FLAG_DATA_UNDERFLOW | + ISCSI_FLAG_DATA_OVERFLOW)) { + int res_count = be32_to_cpu(rhdr->residual_count); + + if (res_count > 0 && + (rhdr->flags & ISCSI_FLAG_CMD_OVERFLOW || + res_count <= scsi_in(sc)->length)) + scsi_in(sc)->resid = res_count; + else + sc->result = (DID_BAD_TARGET << 16) | rhdr->cmd_status; + } + + ISCSI_DBG_SESSION(conn->session, "data in with status done " + "[sc %p res %d itt 0x%x]\n", + sc, sc->result, task->itt); + conn->scsirsp_pdus_cnt++; + iscsi_complete_task(task, ISCSI_TASK_COMPLETED); +} + +static void iscsi_tmf_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr) +{ + struct iscsi_tm_rsp *tmf = (struct iscsi_tm_rsp *)hdr; + + conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; + conn->tmfrsp_pdus_cnt++; if (conn->tmf_state != TMF_QUEUED) return; @@ -365,29 +951,161 @@ static void iscsi_tmf_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr) wake_up(&conn->ehwait); } +static void iscsi_send_nopout(struct iscsi_conn *conn, struct iscsi_nopin *rhdr) +{ + struct iscsi_nopout hdr; + struct iscsi_task *task; + + if (!rhdr && conn->ping_task) + return; + + memset(&hdr, 0, sizeof(struct iscsi_nopout)); + hdr.opcode = ISCSI_OP_NOOP_OUT | ISCSI_OP_IMMEDIATE; + hdr.flags = ISCSI_FLAG_CMD_FINAL; + + if (rhdr) { + memcpy(hdr.lun, rhdr->lun, 8); + hdr.ttt = rhdr->ttt; + hdr.itt = RESERVED_ITT; + } else + hdr.ttt = RESERVED_ITT; + + task = __iscsi_conn_send_pdu(conn, (struct iscsi_hdr *)&hdr, NULL, 0); + if (!task) + iscsi_conn_printk(KERN_ERR, conn, "Could not send nopout\n"); + else if (!rhdr) { + /* only track our nops */ + conn->ping_task = task; + conn->last_ping = jiffies; + } +} + +static int iscsi_nop_out_rsp(struct iscsi_task *task, + struct iscsi_nopin *nop, char *data, int datalen) +{ + struct iscsi_conn *conn = task->conn; + int rc = 0; + + if (conn->ping_task != task) { + /* + * If this is not in response to one of our + * nops then it must be from userspace. + */ + if (iscsi_recv_pdu(conn->cls_conn, (struct iscsi_hdr *)nop, + data, datalen)) + rc = ISCSI_ERR_CONN_FAILED; + } else + mod_timer(&conn->transport_timer, jiffies + conn->recv_timeout); + iscsi_complete_task(task, ISCSI_TASK_COMPLETED); + return rc; +} + static int iscsi_handle_reject(struct iscsi_conn *conn, struct iscsi_hdr *hdr, char *data, int datalen) { struct iscsi_reject *reject = (struct iscsi_reject *)hdr; struct iscsi_hdr rejected_pdu; - uint32_t itt; + int opcode, rc = 0; conn->exp_statsn = be32_to_cpu(reject->statsn) + 1; - if (reject->reason == ISCSI_REASON_DATA_DIGEST_ERROR) { - if (ntoh24(reject->dlength) > datalen) - return ISCSI_ERR_PROTO; + if (ntoh24(reject->dlength) > datalen || + ntoh24(reject->dlength) < sizeof(struct iscsi_hdr)) { + iscsi_conn_printk(KERN_ERR, conn, "Cannot handle rejected " + "pdu. Invalid data length (pdu dlength " + "%u, datalen %d\n", ntoh24(reject->dlength), + datalen); + return ISCSI_ERR_PROTO; + } + memcpy(&rejected_pdu, data, sizeof(struct iscsi_hdr)); + opcode = rejected_pdu.opcode & ISCSI_OPCODE_MASK; + + switch (reject->reason) { + case ISCSI_REASON_DATA_DIGEST_ERROR: + iscsi_conn_printk(KERN_ERR, conn, + "pdu (op 0x%x itt 0x%x) rejected " + "due to DataDigest error.\n", + rejected_pdu.itt, opcode); + break; + case ISCSI_REASON_IMM_CMD_REJECT: + iscsi_conn_printk(KERN_ERR, conn, + "pdu (op 0x%x itt 0x%x) rejected. Too many " + "immediate commands.\n", + rejected_pdu.itt, opcode); + /* + * We only send one TMF at a time so if the target could not + * handle it, then it should get fixed (RFC mandates that + * a target can handle one immediate TMF per conn). + * + * For nops-outs, we could have sent more than one if + * the target is sending us lots of nop-ins + */ + if (opcode != ISCSI_OP_NOOP_OUT) + return 0; - if (ntoh24(reject->dlength) >= sizeof(struct iscsi_hdr)) { - memcpy(&rejected_pdu, data, sizeof(struct iscsi_hdr)); - itt = get_itt(rejected_pdu.itt); - printk(KERN_ERR "itt 0x%x had pdu (op 0x%x) rejected " - "due to DataDigest error.\n", itt, - rejected_pdu.opcode); + if (rejected_pdu.itt == cpu_to_be32(ISCSI_RESERVED_TAG)) + /* + * nop-out in response to target's nop-out rejected. + * Just resend. + */ + iscsi_send_nopout(conn, + (struct iscsi_nopin*)&rejected_pdu); + else { + struct iscsi_task *task; + /* + * Our nop as ping got dropped. We know the target + * and transport are ok so just clean up + */ + task = iscsi_itt_to_task(conn, rejected_pdu.itt); + if (!task) { + iscsi_conn_printk(KERN_ERR, conn, + "Invalid pdu reject. Could " + "not lookup rejected task.\n"); + rc = ISCSI_ERR_BAD_ITT; + } else + rc = iscsi_nop_out_rsp(task, + (struct iscsi_nopin*)&rejected_pdu, + NULL, 0); } + break; + default: + iscsi_conn_printk(KERN_ERR, conn, + "pdu (op 0x%x itt 0x%x) rejected. Reason " + "code 0x%x\n", rejected_pdu.itt, + rejected_pdu.opcode, reject->reason); + break; } - return 0; + return rc; +} + +/** + * iscsi_itt_to_task - look up task by itt + * @conn: iscsi connection + * @itt: itt + * + * This should be used for mgmt tasks like login and nops, or if + * the LDD's itt space does not include the session age. + * + * The session lock must be held. + */ +struct iscsi_task *iscsi_itt_to_task(struct iscsi_conn *conn, itt_t itt) +{ + struct iscsi_session *session = conn->session; + int i; + + if (itt == RESERVED_ITT) + return NULL; + + if (session->tt->parse_pdu_itt) + session->tt->parse_pdu_itt(conn, itt, &i, NULL); + else + i = get_itt(itt); + if (i >= session->cmds_max) + return NULL; + + return session->cmds[i]; } +EXPORT_SYMBOL_GPL(iscsi_itt_to_task); /** * __iscsi_complete_pdu - complete pdu @@ -405,96 +1123,23 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr, { struct iscsi_session *session = conn->session; int opcode = hdr->opcode & ISCSI_OPCODE_MASK, rc = 0; - struct iscsi_cmd_task *ctask; - struct iscsi_mgmt_task *mtask; + struct iscsi_task *task; uint32_t itt; + conn->last_recv = jiffies; + rc = iscsi_verify_itt(conn, hdr->itt); + if (rc) + return rc; + if (hdr->itt != RESERVED_ITT) itt = get_itt(hdr->itt); else itt = ~0U; - if (itt < session->cmds_max) { - ctask = session->cmds[itt]; - - debug_scsi("cmdrsp [op 0x%x cid %d itt 0x%x len %d]\n", - opcode, conn->id, ctask->itt, datalen); - - switch(opcode) { - case ISCSI_OP_SCSI_CMD_RSP: - BUG_ON((void*)ctask != ctask->sc->SCp.ptr); - iscsi_scsi_cmd_rsp(conn, hdr, ctask, data, - datalen); - break; - case ISCSI_OP_SCSI_DATA_IN: - BUG_ON((void*)ctask != ctask->sc->SCp.ptr); - if (hdr->flags & ISCSI_FLAG_DATA_STATUS) { - conn->scsirsp_pdus_cnt++; - __iscsi_put_ctask(ctask); - } - break; - case ISCSI_OP_R2T: - /* LLD handles this for now */ - break; - default: - rc = ISCSI_ERR_BAD_OPCODE; - break; - } - } else if (itt >= ISCSI_MGMT_ITT_OFFSET && - itt < ISCSI_MGMT_ITT_OFFSET + session->mgmtpool_max) { - mtask = session->mgmt_cmds[itt - ISCSI_MGMT_ITT_OFFSET]; - - debug_scsi("immrsp [op 0x%x cid %d itt 0x%x len %d]\n", - opcode, conn->id, mtask->itt, datalen); - - iscsi_update_cmdsn(session, (struct iscsi_nopin*)hdr); - switch(opcode) { - case ISCSI_OP_LOGOUT_RSP: - if (datalen) { - rc = ISCSI_ERR_PROTO; - break; - } - conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; - /* fall through */ - case ISCSI_OP_LOGIN_RSP: - case ISCSI_OP_TEXT_RSP: - /* - * login related PDU's exp_statsn is handled in - * userspace - */ - if (iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen)) - rc = ISCSI_ERR_CONN_FAILED; - list_del_init(&mtask->running); - if (conn->login_mtask != mtask) - __kfifo_put(session->mgmtpool.queue, - (void*)&mtask, sizeof(void*)); - break; - case ISCSI_OP_SCSI_TMFUNC_RSP: - if (datalen) { - rc = ISCSI_ERR_PROTO; - break; - } - - iscsi_tmf_rsp(conn, hdr); - break; - case ISCSI_OP_NOOP_IN: - if (hdr->ttt != cpu_to_be32(ISCSI_RESERVED_TAG) || datalen) { - rc = ISCSI_ERR_PROTO; - break; - } - conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; + ISCSI_DBG_SESSION(session, "[op 0x%x cid %d itt 0x%x len %d]\n", + opcode, conn->id, itt, datalen); - if (iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen)) - rc = ISCSI_ERR_CONN_FAILED; - list_del_init(&mtask->running); - __kfifo_put(session->mgmtpool.queue, - (void*)&mtask, sizeof(void*)); - break; - default: - rc = ISCSI_ERR_BAD_OPCODE; - break; - } - } else if (itt == ~0U) { + if (itt == ~0U) { iscsi_update_cmdsn(session, (struct iscsi_nopin*)hdr); switch(opcode) { @@ -507,8 +1152,7 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr, if (hdr->ttt == cpu_to_be32(ISCSI_RESERVED_TAG)) break; - if (iscsi_recv_pdu(conn->cls_conn, hdr, NULL, 0)) - rc = ISCSI_ERR_CONN_FAILED; + iscsi_send_nopout(conn, (struct iscsi_nopin*)hdr); break; case ISCSI_OP_REJECT: rc = iscsi_handle_reject(conn, hdr, data, datalen); @@ -522,9 +1166,90 @@ int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr, rc = ISCSI_ERR_BAD_OPCODE; break; } - } else - rc = ISCSI_ERR_BAD_ITT; + goto out; + } + + switch(opcode) { + case ISCSI_OP_SCSI_CMD_RSP: + case ISCSI_OP_SCSI_DATA_IN: + task = iscsi_itt_to_ctask(conn, hdr->itt); + if (!task) + return ISCSI_ERR_BAD_ITT; + task->last_xfer = jiffies; + break; + case ISCSI_OP_R2T: + /* + * LLD handles R2Ts if they need to. + */ + return 0; + case ISCSI_OP_LOGOUT_RSP: + case ISCSI_OP_LOGIN_RSP: + case ISCSI_OP_TEXT_RSP: + case ISCSI_OP_SCSI_TMFUNC_RSP: + case ISCSI_OP_NOOP_IN: + task = iscsi_itt_to_task(conn, hdr->itt); + if (!task) + return ISCSI_ERR_BAD_ITT; + break; + default: + return ISCSI_ERR_BAD_OPCODE; + } + + switch(opcode) { + case ISCSI_OP_SCSI_CMD_RSP: + iscsi_scsi_cmd_rsp(conn, hdr, task, data, datalen); + break; + case ISCSI_OP_SCSI_DATA_IN: + iscsi_data_in_rsp(conn, hdr, task); + break; + case ISCSI_OP_LOGOUT_RSP: + iscsi_update_cmdsn(session, (struct iscsi_nopin*)hdr); + if (datalen) { + rc = ISCSI_ERR_PROTO; + break; + } + conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; + goto recv_pdu; + case ISCSI_OP_LOGIN_RSP: + case ISCSI_OP_TEXT_RSP: + iscsi_update_cmdsn(session, (struct iscsi_nopin*)hdr); + /* + * login related PDU's exp_statsn is handled in + * userspace + */ + goto recv_pdu; + case ISCSI_OP_SCSI_TMFUNC_RSP: + iscsi_update_cmdsn(session, (struct iscsi_nopin*)hdr); + if (datalen) { + rc = ISCSI_ERR_PROTO; + break; + } + + iscsi_tmf_rsp(conn, hdr); + iscsi_complete_task(task, ISCSI_TASK_COMPLETED); + break; + case ISCSI_OP_NOOP_IN: + iscsi_update_cmdsn(session, (struct iscsi_nopin*)hdr); + if (hdr->ttt != cpu_to_be32(ISCSI_RESERVED_TAG) || datalen) { + rc = ISCSI_ERR_PROTO; + break; + } + conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; + + rc = iscsi_nop_out_rsp(task, (struct iscsi_nopin*)hdr, + data, datalen); + break; + default: + rc = ISCSI_ERR_BAD_OPCODE; + break; + } +out: + return rc; +recv_pdu: + if (iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen)) + rc = ISCSI_ERR_CONN_FAILED; + iscsi_complete_task(task, ISCSI_TASK_COMPLETED); return rc; } EXPORT_SYMBOL_GPL(__iscsi_complete_pdu); @@ -541,137 +1266,120 @@ int iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr, } EXPORT_SYMBOL_GPL(iscsi_complete_pdu); -/* verify itt (itt encoding: age+cid+itt) */ -int iscsi_verify_itt(struct iscsi_conn *conn, struct iscsi_hdr *hdr, - uint32_t *ret_itt) +int iscsi_verify_itt(struct iscsi_conn *conn, itt_t itt) { struct iscsi_session *session = conn->session; - struct iscsi_cmd_task *ctask; - uint32_t itt; - - if (hdr->itt != RESERVED_ITT) { - if (((__force u32)hdr->itt & ISCSI_AGE_MASK) != - (session->age << ISCSI_AGE_SHIFT)) { - printk(KERN_ERR "iscsi: received itt %x expected " - "session age (%x)\n", (__force u32)hdr->itt, - session->age & ISCSI_AGE_MASK); - return ISCSI_ERR_BAD_ITT; - } - - if (((__force u32)hdr->itt & ISCSI_CID_MASK) != - (conn->id << ISCSI_CID_SHIFT)) { - printk(KERN_ERR "iscsi: received itt %x, expected " - "CID (%x)\n", (__force u32)hdr->itt, conn->id); - return ISCSI_ERR_BAD_ITT; - } - itt = get_itt(hdr->itt); - } else - itt = ~0U; + int age = 0, i = 0; - if (itt < session->cmds_max) { - ctask = session->cmds[itt]; + if (itt == RESERVED_ITT) + return 0; - if (!ctask->sc) { - printk(KERN_INFO "iscsi: dropping ctask with " - "itt 0x%x\n", ctask->itt); - /* force drop */ - return ISCSI_ERR_NO_SCSI_CMD; - } + if (session->tt->parse_pdu_itt) + session->tt->parse_pdu_itt(conn, itt, &i, &age); + else { + i = get_itt(itt); + age = ((__force u32)itt >> ISCSI_AGE_SHIFT) & ISCSI_AGE_MASK; + } - if (ctask->sc->SCp.phase != session->age) { - printk(KERN_ERR "iscsi: ctask's session age %d, " - "expected %d\n", ctask->sc->SCp.phase, - session->age); - return ISCSI_ERR_SESSION_FAILED; - } + if (age != session->age) { + iscsi_conn_printk(KERN_ERR, conn, + "received itt %x expected session age (%x)\n", + (__force u32)itt, session->age); + return ISCSI_ERR_BAD_ITT; } - *ret_itt = itt; + if (i >= session->cmds_max) { + iscsi_conn_printk(KERN_ERR, conn, + "received invalid itt index %u (max cmds " + "%u.\n", i, session->cmds_max); + return ISCSI_ERR_BAD_ITT; + } return 0; } EXPORT_SYMBOL_GPL(iscsi_verify_itt); -void iscsi_conn_failure(struct iscsi_conn *conn, enum iscsi_err err) +/** + * iscsi_itt_to_ctask - look up ctask by itt + * @conn: iscsi connection + * @itt: itt + * + * This should be used for cmd tasks. + * + * The session lock must be held. + */ +struct iscsi_task *iscsi_itt_to_ctask(struct iscsi_conn *conn, itt_t itt) { - struct iscsi_session *session = conn->session; + struct iscsi_task *task; + + if (iscsi_verify_itt(conn, itt)) + return NULL; + + task = iscsi_itt_to_task(conn, itt); + if (!task || !task->sc) + return NULL; + + if (task->sc->SCp.phase != conn->session->age) { + iscsi_session_printk(KERN_ERR, conn->session, + "task's session age %d, expected %d\n", + task->sc->SCp.phase, conn->session->age); + return NULL; + } + + return task; +} +EXPORT_SYMBOL_GPL(iscsi_itt_to_ctask); + +void iscsi_session_failure(struct iscsi_session *session, + enum iscsi_err err) +{ + struct iscsi_conn *conn; + struct device *dev; unsigned long flags; spin_lock_irqsave(&session->lock, flags); - if (session->state == ISCSI_STATE_FAILED) { + conn = session->leadconn; + if (session->state == ISCSI_STATE_TERMINATE || !conn) { spin_unlock_irqrestore(&session->lock, flags); return; } - if (conn->stop_stage == 0) - session->state = ISCSI_STATE_FAILED; + dev = get_device(&conn->cls_conn->dev); spin_unlock_irqrestore(&session->lock, flags); - set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); - set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_rx); - iscsi_conn_error(conn->cls_conn, err); -} -EXPORT_SYMBOL_GPL(iscsi_conn_failure); - -static void iscsi_prep_mtask(struct iscsi_conn *conn, - struct iscsi_mgmt_task *mtask) -{ - struct iscsi_session *session = conn->session; - struct iscsi_hdr *hdr = mtask->hdr; - struct iscsi_nopout *nop = (struct iscsi_nopout *)hdr; - - if (hdr->opcode != (ISCSI_OP_LOGIN | ISCSI_OP_IMMEDIATE) && - hdr->opcode != (ISCSI_OP_TEXT | ISCSI_OP_IMMEDIATE)) - nop->exp_statsn = cpu_to_be32(conn->exp_statsn); + if (!dev) + return; /* - * pre-format CmdSN for outgoing PDU. + * if the host is being removed bypass the connection + * recovery initialization because we are going to kill + * the session. */ - nop->cmdsn = cpu_to_be32(session->cmdsn); - if (hdr->itt != RESERVED_ITT) { - hdr->itt = build_itt(mtask->itt, conn->id, session->age); - /* - * TODO: We always use immediate, so we never hit this. - * If we start to send tmfs or nops as non-immediate then - * we should start checking the cmdsn numbers for mgmt tasks. - */ - if (conn->c_stage == ISCSI_CONN_STARTED && - !(hdr->opcode & ISCSI_OP_IMMEDIATE)) { - session->queued_cmdsn++; - session->cmdsn++; - } - } - - if (session->tt->init_mgmt_task) - session->tt->init_mgmt_task(conn, mtask); - - debug_scsi("mgmtpdu [op 0x%x hdr->itt 0x%x datalen %d]\n", - hdr->opcode & ISCSI_OPCODE_MASK, hdr->itt, - mtask->data_count); + if (err == ISCSI_ERR_INVALID_HOST) + iscsi_conn_error_event(conn->cls_conn, err); + else + iscsi_conn_failure(conn, err); + put_device(dev); } +EXPORT_SYMBOL_GPL(iscsi_session_failure); -static int iscsi_xmit_mtask(struct iscsi_conn *conn) +void iscsi_conn_failure(struct iscsi_conn *conn, enum iscsi_err err) { - struct iscsi_hdr *hdr = conn->mtask->hdr; - int rc, was_logout = 0; + struct iscsi_session *session = conn->session; + unsigned long flags; - spin_unlock_bh(&conn->session->lock); - if ((hdr->opcode & ISCSI_OPCODE_MASK) == ISCSI_OP_LOGOUT) { - conn->session->state = ISCSI_STATE_IN_RECOVERY; - iscsi_block_session(session_to_cls(conn->session)); - was_logout = 1; + spin_lock_irqsave(&session->lock, flags); + if (session->state == ISCSI_STATE_FAILED) { + spin_unlock_irqrestore(&session->lock, flags); + return; } - rc = conn->session->tt->xmit_mgmt_task(conn, conn->mtask); - spin_lock_bh(&conn->session->lock); - if (rc) - return rc; - /* done with this in-progress mtask */ - conn->mtask = NULL; + if (conn->stop_stage == 0) + session->state = ISCSI_STATE_FAILED; + spin_unlock_irqrestore(&session->lock, flags); - if (was_logout) { - set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); - return -ENODATA; - } - return 0; + set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); + set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_rx); + iscsi_conn_error_event(conn->cls_conn, err); } +EXPORT_SYMBOL_GPL(iscsi_conn_failure); static int iscsi_check_cmdsn_window_closed(struct iscsi_conn *conn) { @@ -681,46 +1389,57 @@ static int iscsi_check_cmdsn_window_closed(struct iscsi_conn *conn) * Check for iSCSI window and take care of CmdSN wrap-around */ if (!iscsi_sna_lte(session->queued_cmdsn, session->max_cmdsn)) { - debug_scsi("iSCSI CmdSN closed. ExpCmdSn %u MaxCmdSN %u " - "CmdSN %u/%u\n", session->exp_cmdsn, - session->max_cmdsn, session->cmdsn, - session->queued_cmdsn); + ISCSI_DBG_SESSION(session, "iSCSI CmdSN closed. ExpCmdSn " + "%u MaxCmdSN %u CmdSN %u/%u\n", + session->exp_cmdsn, session->max_cmdsn, + session->cmdsn, session->queued_cmdsn); return -ENOSPC; } return 0; } -static int iscsi_xmit_ctask(struct iscsi_conn *conn) +static int iscsi_xmit_task(struct iscsi_conn *conn) { - struct iscsi_cmd_task *ctask = conn->ctask; + struct iscsi_task *task = conn->task; int rc; - __iscsi_get_ctask(ctask); + if (test_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx)) + return -ENODATA; + + __iscsi_get_task(task); spin_unlock_bh(&conn->session->lock); - rc = conn->session->tt->xmit_cmd_task(conn, ctask); + rc = conn->session->tt->xmit_task(task); spin_lock_bh(&conn->session->lock); - __iscsi_put_ctask(ctask); - if (!rc) - /* done with this ctask */ - conn->ctask = NULL; + if (!rc) { + /* done with this task */ + task->last_xfer = jiffies; + conn->task = NULL; + } + __iscsi_put_task(task); return rc; } /** - * iscsi_requeue_ctask - requeue ctask to run from session workqueue - * @ctask: ctask to requeue + * iscsi_requeue_task - requeue task to run from session workqueue + * @task: task to requeue * - * LLDs that need to run a ctask from the session workqueue should call - * this. The session lock must be held. + * LLDs that need to run a task from the session workqueue should call + * this. The session lock must be held. This should only be called + * by software drivers. */ -void iscsi_requeue_ctask(struct iscsi_cmd_task *ctask) +void iscsi_requeue_task(struct iscsi_task *task) { - struct iscsi_conn *conn = ctask->conn; + struct iscsi_conn *conn = task->conn; - list_move_tail(&ctask->running, &conn->requeue); - scsi_queue_work(conn->session->host, &conn->xmitwork); + /* + * this may be on the requeue list already if the xmit_task callout + * is handling the r2ts while we are adding new ones + */ + if (list_empty(&task->running)) + list_add_tail(&task->running, &conn->requeue); + iscsi_conn_queue_work(conn); } -EXPORT_SYMBOL_GPL(iscsi_requeue_ctask); +EXPORT_SYMBOL_GPL(iscsi_requeue_task); /** * iscsi_data_xmit - xmit any command into the scheduled connection @@ -733,25 +1452,20 @@ EXPORT_SYMBOL_GPL(iscsi_requeue_ctask); **/ static int iscsi_data_xmit(struct iscsi_conn *conn) { + struct iscsi_task *task; int rc = 0; spin_lock_bh(&conn->session->lock); - if (unlikely(conn->suspend_tx)) { - debug_scsi("conn %d Tx suspended!\n", conn->id); + if (test_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx)) { + ISCSI_DBG_SESSION(conn->session, "Tx suspended!\n"); spin_unlock_bh(&conn->session->lock); return -ENODATA; } - if (conn->ctask) { - rc = iscsi_xmit_ctask(conn); - if (rc) - goto again; - } - - if (conn->mtask) { - rc = iscsi_xmit_mtask(conn); + if (conn->task) { + rc = iscsi_xmit_task(conn); if (rc) - goto again; + goto done; } /* @@ -761,34 +1475,44 @@ static int iscsi_data_xmit(struct iscsi_conn *conn) */ check_mgmt: while (!list_empty(&conn->mgmtqueue)) { - conn->mtask = list_entry(conn->mgmtqueue.next, - struct iscsi_mgmt_task, running); - iscsi_prep_mtask(conn, conn->mtask); - list_move_tail(conn->mgmtqueue.next, &conn->mgmt_run_list); - rc = iscsi_xmit_mtask(conn); + conn->task = list_entry(conn->mgmtqueue.next, + struct iscsi_task, running); + list_del_init(&conn->task->running); + if (iscsi_prep_mgmt_task(conn, conn->task)) { + __iscsi_put_task(conn->task); + conn->task = NULL; + continue; + } + rc = iscsi_xmit_task(conn); if (rc) - goto again; + goto done; } /* process pending command queue */ - while (!list_empty(&conn->xmitqueue)) { - if (conn->tmf_state == TMF_QUEUED) - break; - - conn->ctask = list_entry(conn->xmitqueue.next, - struct iscsi_cmd_task, running); - if (iscsi_prep_scsi_cmd_pdu(conn->ctask)) { - fail_command(conn, conn->ctask, DID_ABORT << 16); + while (!list_empty(&conn->cmdqueue)) { + conn->task = list_entry(conn->cmdqueue.next, struct iscsi_task, + running); + list_del_init(&conn->task->running); + if (conn->session->state == ISCSI_STATE_LOGGING_OUT) { + fail_scsi_task(conn->task, DID_IMM_RETRY); continue; } - conn->session->tt->init_cmd_task(conn->ctask); - conn->ctask->state = ISCSI_TASK_RUNNING; - list_move_tail(conn->xmitqueue.next, &conn->run_list); - rc = iscsi_xmit_ctask(conn); + rc = iscsi_prep_scsi_cmd_pdu(conn->task); + if (rc) { + if (rc == -ENOMEM || rc == -EACCES) { + list_add_tail(&conn->task->running, + &conn->cmdqueue); + conn->task = NULL; + goto done; + } else + fail_scsi_task(conn->task, DID_ABORT); + continue; + } + rc = iscsi_xmit_task(conn); if (rc) - goto again; + goto done; /* - * we could continuously get new ctask requests so + * we could continuously get new task requests so * we need to check the mgmt queue for nops that need to * be sent to aviod starvation */ @@ -797,25 +1521,30 @@ check_mgmt: } while (!list_empty(&conn->requeue)) { - if (conn->session->fast_abort && conn->tmf_state != TMF_INITIAL) + /* + * we always do fastlogout - conn stop code will clean up. + */ + if (conn->session->state == ISCSI_STATE_LOGGING_OUT) break; - conn->ctask = list_entry(conn->requeue.next, - struct iscsi_cmd_task, running); - conn->ctask->state = ISCSI_TASK_RUNNING; - list_move_tail(conn->requeue.next, &conn->run_list); - rc = iscsi_xmit_ctask(conn); + task = list_entry(conn->requeue.next, struct iscsi_task, + running); + if (iscsi_check_tmf_restrictions(task, ISCSI_OP_SCSI_DATA_OUT)) + break; + + conn->task = task; + list_del_init(&conn->task->running); + conn->task->state = ISCSI_TASK_RUNNING; + rc = iscsi_xmit_task(conn); if (rc) - goto again; + goto done; if (!list_empty(&conn->mgmtqueue)) goto check_mgmt; } spin_unlock_bh(&conn->session->lock); return -ENODATA; -again: - if (unlikely(conn->suspend_tx)) - rc = -ENODATA; +done: spin_unlock_bh(&conn->session->lock); return rc; } @@ -833,6 +1562,29 @@ static void iscsi_xmitworker(struct work_struct *work) } while (rc >= 0 || rc == -EAGAIN); } +static inline struct iscsi_task *iscsi_alloc_task(struct iscsi_conn *conn, + struct scsi_cmnd *sc) +{ + struct iscsi_task *task; + + if (!kfifo_out(&conn->session->cmdpool.queue, + (void *) &task, sizeof(void *))) + return NULL; + + sc->SCp.phase = conn->session->age; + sc->SCp.ptr = (char *) task; + + atomic_set(&task->refcount, 1); + task->state = ISCSI_TASK_PENDING; + task->conn = conn; + task->sc = sc; + task->have_checked_conn = false; + task->last_timeout = jiffies; + task->last_xfer = jiffies; + INIT_LIST_HEAD(&task->running); + return task; +} + enum { FAILURE_BAD_HOST = 1, FAILURE_SESSION_FAILED, @@ -842,55 +1594,80 @@ enum { FAILURE_SESSION_TERMINATE, FAILURE_SESSION_IN_RECOVERY, FAILURE_SESSION_RECOVERY_TIMEOUT, + FAILURE_SESSION_LOGGING_OUT, + FAILURE_SESSION_NOT_READY, }; int iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *)) { + struct iscsi_cls_session *cls_session; struct Scsi_Host *host; + struct iscsi_host *ihost; int reason = 0; struct iscsi_session *session; struct iscsi_conn *conn; - struct iscsi_cmd_task *ctask = NULL; + struct iscsi_task *task = NULL; sc->scsi_done = done; sc->result = 0; sc->SCp.ptr = NULL; host = sc->device->host; - session = iscsi_hostdata(host->hostdata); + ihost = shost_priv(host); + spin_unlock(host->host_lock); + cls_session = starget_to_session(scsi_target(sc->device)); + session = cls_session->dd_data; spin_lock(&session->lock); - /* - * ISCSI_STATE_FAILED is a temp. state. The recovery - * code will decide what is best to do with command queued - * during this time - */ - if (session->state != ISCSI_STATE_LOGGED_IN && - session->state != ISCSI_STATE_FAILED) { + reason = iscsi_session_chkready(cls_session); + if (reason) { + sc->result = reason; + goto fault; + } + + if (session->state != ISCSI_STATE_LOGGED_IN) { /* * to handle the race between when we set the recovery state * and block the session we requeue here (commands could * be entering our queuecommand while a block is starting * up because the block code is not locked) */ - if (session->state == ISCSI_STATE_IN_RECOVERY) { + switch (session->state) { + case ISCSI_STATE_FAILED: + case ISCSI_STATE_IN_RECOVERY: reason = FAILURE_SESSION_IN_RECOVERY; - goto reject; - } - - if (session->state == ISCSI_STATE_RECOVERY_FAILED) + sc->result = DID_IMM_RETRY << 16; + break; + case ISCSI_STATE_LOGGING_OUT: + reason = FAILURE_SESSION_LOGGING_OUT; + sc->result = DID_IMM_RETRY << 16; + break; + case ISCSI_STATE_RECOVERY_FAILED: reason = FAILURE_SESSION_RECOVERY_TIMEOUT; - else if (session->state == ISCSI_STATE_TERMINATE) + sc->result = DID_TRANSPORT_FAILFAST << 16; + break; + case ISCSI_STATE_TERMINATE: reason = FAILURE_SESSION_TERMINATE; - else + sc->result = DID_NO_CONNECT << 16; + break; + default: reason = FAILURE_SESSION_FREED; + sc->result = DID_NO_CONNECT << 16; + } goto fault; } conn = session->leadconn; if (!conn) { reason = FAILURE_SESSION_FREED; + sc->result = DID_NO_CONNECT << 16; + goto fault; + } + + if (test_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx)) { + reason = FAILURE_SESSION_IN_RECOVERY; + sc->result = DID_REQUEUE; goto fault; } @@ -899,164 +1676,95 @@ int iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *)) goto reject; } - if (!__kfifo_get(session->cmdpool.queue, (void*)&ctask, - sizeof(void*))) { + task = iscsi_alloc_task(conn, sc); + if (!task) { reason = FAILURE_OOM; goto reject; } - session->queued_cmdsn++; - sc->SCp.phase = session->age; - sc->SCp.ptr = (char *)ctask; - - atomic_set(&ctask->refcount, 1); - ctask->state = ISCSI_TASK_PENDING; - ctask->conn = conn; - ctask->sc = sc; - INIT_LIST_HEAD(&ctask->running); + if (!ihost->workq) { + reason = iscsi_prep_scsi_cmd_pdu(task); + if (reason) { + if (reason == -ENOMEM || reason == -EACCES) { + reason = FAILURE_OOM; + goto prepd_reject; + } else { + sc->result = DID_ABORT << 16; + goto prepd_fault; + } + } + if (session->tt->xmit_task(task)) { + session->cmdsn--; + reason = FAILURE_SESSION_NOT_READY; + goto prepd_reject; + } + } else { + list_add_tail(&task->running, &conn->cmdqueue); + iscsi_conn_queue_work(conn); + } - list_add_tail(&ctask->running, &conn->xmitqueue); + session->queued_cmdsn++; spin_unlock(&session->lock); - - scsi_queue_work(host, &conn->xmitwork); + spin_lock(host->host_lock); return 0; +prepd_reject: + sc->scsi_done = NULL; + iscsi_complete_task(task, ISCSI_TASK_COMPLETED); reject: spin_unlock(&session->lock); - debug_scsi("cmd 0x%x rejected (%d)\n", sc->cmnd[0], reason); - return SCSI_MLQUEUE_HOST_BUSY; - + ISCSI_DBG_SESSION(session, "cmd 0x%x rejected (%d)\n", + sc->cmnd[0], reason); + spin_lock(host->host_lock); + return SCSI_MLQUEUE_TARGET_BUSY; + +prepd_fault: + sc->scsi_done = NULL; + iscsi_complete_task(task, ISCSI_TASK_COMPLETED); fault: spin_unlock(&session->lock); - printk(KERN_ERR "iscsi: cmd 0x%x is not queued (%d)\n", - sc->cmnd[0], reason); - sc->result = (DID_NO_CONNECT << 16); - scsi_set_resid(sc, scsi_bufflen(sc)); - sc->scsi_done(sc); + ISCSI_DBG_SESSION(session, "iscsi: cmd 0x%x is not queued (%d)\n", + sc->cmnd[0], reason); + if (!scsi_bidi_cmnd(sc)) + scsi_set_resid(sc, scsi_bufflen(sc)); + else { + scsi_out(sc)->resid = scsi_out(sc)->length; + scsi_in(sc)->resid = scsi_in(sc)->length; + } + done(sc); + spin_lock(host->host_lock); return 0; } EXPORT_SYMBOL_GPL(iscsi_queuecommand); -int iscsi_change_queue_depth(struct scsi_device *sdev, int depth) +int iscsi_change_queue_depth(struct scsi_device *sdev, int depth, int reason) { - if (depth > ISCSI_MAX_CMD_PER_LUN) - depth = ISCSI_MAX_CMD_PER_LUN; - scsi_adjust_queue_depth(sdev, scsi_get_tag_type(sdev), depth); + switch (reason) { + case SCSI_QDEPTH_DEFAULT: + scsi_adjust_queue_depth(sdev, scsi_get_tag_type(sdev), depth); + break; + case SCSI_QDEPTH_QFULL: + scsi_track_queue_full(sdev, depth); + break; + case SCSI_QDEPTH_RAMP_UP: + scsi_adjust_queue_depth(sdev, scsi_get_tag_type(sdev), depth); + break; + default: + return -EOPNOTSUPP; + } return sdev->queue_depth; } EXPORT_SYMBOL_GPL(iscsi_change_queue_depth); -static struct iscsi_mgmt_task * -__iscsi_conn_send_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr, - char *data, uint32_t data_size) -{ - struct iscsi_session *session = conn->session; - struct iscsi_mgmt_task *mtask; - - if (session->state == ISCSI_STATE_TERMINATE) - return NULL; - - if (hdr->opcode == (ISCSI_OP_LOGIN | ISCSI_OP_IMMEDIATE) || - hdr->opcode == (ISCSI_OP_TEXT | ISCSI_OP_IMMEDIATE)) - /* - * Login and Text are sent serially, in - * request-followed-by-response sequence. - * Same mtask can be used. Same ITT must be used. - * Note that login_mtask is preallocated at conn_create(). - */ - mtask = conn->login_mtask; - else { - BUG_ON(conn->c_stage == ISCSI_CONN_INITIAL_STAGE); - BUG_ON(conn->c_stage == ISCSI_CONN_STOPPED); - - if (!__kfifo_get(session->mgmtpool.queue, - (void*)&mtask, sizeof(void*))) - return NULL; - } - - if (data_size) { - memcpy(mtask->data, data, data_size); - mtask->data_count = data_size; - } else - mtask->data_count = 0; - - memcpy(mtask->hdr, hdr, sizeof(struct iscsi_hdr)); - INIT_LIST_HEAD(&mtask->running); - list_add_tail(&mtask->running, &conn->mgmtqueue); - return mtask; -} - -int iscsi_conn_send_pdu(struct iscsi_cls_conn *cls_conn, struct iscsi_hdr *hdr, - char *data, uint32_t data_size) -{ - struct iscsi_conn *conn = cls_conn->dd_data; - struct iscsi_session *session = conn->session; - int err = 0; - - spin_lock_bh(&session->lock); - if (!__iscsi_conn_send_pdu(conn, hdr, data, data_size)) - err = -EPERM; - spin_unlock_bh(&session->lock); - scsi_queue_work(session->host, &conn->xmitwork); - return err; -} -EXPORT_SYMBOL_GPL(iscsi_conn_send_pdu); - -void iscsi_session_recovery_timedout(struct iscsi_cls_session *cls_session) -{ - struct iscsi_session *session = class_to_transport_session(cls_session); - - spin_lock_bh(&session->lock); - if (session->state != ISCSI_STATE_LOGGED_IN) { - session->state = ISCSI_STATE_RECOVERY_FAILED; - if (session->leadconn) - wake_up(&session->leadconn->ehwait); - } - spin_unlock_bh(&session->lock); -} -EXPORT_SYMBOL_GPL(iscsi_session_recovery_timedout); - -int iscsi_eh_host_reset(struct scsi_cmnd *sc) +int iscsi_target_alloc(struct scsi_target *starget) { - struct Scsi_Host *host = sc->device->host; - struct iscsi_session *session = iscsi_hostdata(host->hostdata); - struct iscsi_conn *conn = session->leadconn; - - spin_lock_bh(&session->lock); - if (session->state == ISCSI_STATE_TERMINATE) { -failed: - debug_scsi("failing host reset: session terminated " - "[CID %d age %d]\n", conn->id, session->age); - spin_unlock_bh(&session->lock); - return FAILED; - } + struct iscsi_cls_session *cls_session = starget_to_session(starget); + struct iscsi_session *session = cls_session->dd_data; - spin_unlock_bh(&session->lock); - - /* - * we drop the lock here but the leadconn cannot be destoyed while - * we are in the scsi eh - */ - iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); - - debug_scsi("iscsi_eh_host_reset wait for relogin\n"); - wait_event_interruptible(conn->ehwait, - session->state == ISCSI_STATE_TERMINATE || - session->state == ISCSI_STATE_LOGGED_IN || - session->state == ISCSI_STATE_RECOVERY_FAILED); - if (signal_pending(current)) - flush_signals(current); - - spin_lock_bh(&session->lock); - if (session->state == ISCSI_STATE_LOGGED_IN) - printk(KERN_INFO "iscsi: host reset succeeded\n"); - else - goto failed; - spin_unlock_bh(&session->lock); - - return SUCCESS; + starget->can_queue = session->scsi_cmds_max; + return 0; } -EXPORT_SYMBOL_GPL(iscsi_eh_host_reset); +EXPORT_SYMBOL_GPL(iscsi_target_alloc); static void iscsi_tmf_timedout(unsigned long data) { @@ -1066,7 +1774,7 @@ static void iscsi_tmf_timedout(unsigned long data) spin_lock(&session->lock); if (conn->tmf_state == TMF_QUEUED) { conn->tmf_state = TMF_TIMEDOUT; - debug_scsi("tmf timedout\n"); + ISCSI_DBG_EH(session, "tmf timedout\n"); /* unblock eh_abort() */ wake_up(&conn->ehwait); } @@ -1074,30 +1782,30 @@ static void iscsi_tmf_timedout(unsigned long data) } static int iscsi_exec_task_mgmt_fn(struct iscsi_conn *conn, - struct iscsi_tm *hdr, int age) + struct iscsi_tm *hdr, int age, + int timeout) { struct iscsi_session *session = conn->session; - struct iscsi_mgmt_task *mtask; + struct iscsi_task *task; - mtask = __iscsi_conn_send_pdu(conn, (struct iscsi_hdr *)hdr, + task = __iscsi_conn_send_pdu(conn, (struct iscsi_hdr *)hdr, NULL, 0); - if (!mtask) { + if (!task) { spin_unlock_bh(&session->lock); iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); spin_lock_bh(&session->lock); - debug_scsi("tmf exec failure\n"); + ISCSI_DBG_EH(session, "tmf exec failure\n"); return -EPERM; } conn->tmfcmd_pdus_cnt++; - conn->tmf_timer.expires = 30 * HZ + jiffies; + conn->tmf_timer.expires = timeout * HZ + jiffies; conn->tmf_timer.function = iscsi_tmf_timedout; conn->tmf_timer.data = (unsigned long)conn; add_timer(&conn->tmf_timer); - debug_scsi("tmf set timeout\n"); + ISCSI_DBG_EH(session, "tmf set timeout\n"); spin_unlock_bh(&session->lock); mutex_unlock(&session->eh_mutex); - scsi_queue_work(session->host, &conn->xmitwork); /* * block eh thread until: @@ -1116,119 +1824,300 @@ static int iscsi_exec_task_mgmt_fn(struct iscsi_conn *conn, mutex_lock(&session->eh_mutex); spin_lock_bh(&session->lock); - /* if the session drops it will clean up the mtask */ + /* if the session drops it will clean up the task */ if (age != session->age || session->state != ISCSI_STATE_LOGGED_IN) return -ENOTCONN; - - if (!list_empty(&mtask->running)) { - list_del_init(&mtask->running); - __kfifo_put(session->mgmtpool.queue, (void*)&mtask, - sizeof(void*)); - } return 0; } /* - * session lock must be held + * Fail commands. session lock held and recv side suspended and xmit + * thread flushed */ -static void fail_command(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, - int err) +static void fail_scsi_tasks(struct iscsi_conn *conn, unsigned lun, + int error) { - struct scsi_cmnd *sc; + struct iscsi_task *task; + int i; - sc = ctask->sc; - if (!sc) - return; + for (i = 0; i < conn->session->cmds_max; i++) { + task = conn->session->cmds[i]; + if (!task->sc || task->state == ISCSI_TASK_FREE) + continue; - if (ctask->state == ISCSI_TASK_PENDING) - /* - * cmd never made it to the xmit thread, so we should not count - * the cmd in the sequencing - */ - conn->session->queued_cmdsn--; - else - conn->session->tt->cleanup_cmd_task(conn, ctask); + if (lun != -1 && lun != task->sc->device->lun) + continue; + + ISCSI_DBG_SESSION(conn->session, + "failing sc %p itt 0x%x state %d\n", + task->sc, task->itt, task->state); + fail_scsi_task(task, error); + } +} + +/** + * iscsi_suspend_queue - suspend iscsi_queuecommand + * @conn: iscsi conn to stop queueing IO on + * + * This grabs the session lock to make sure no one is in + * xmit_task/queuecommand, and then sets suspend to prevent + * new commands from being queued. This only needs to be called + * by offload drivers that need to sync a path like ep disconnect + * with the iscsi_queuecommand/xmit_task. To start IO again libiscsi + * will call iscsi_start_tx and iscsi_unblock_session when in FFP. + */ +void iscsi_suspend_queue(struct iscsi_conn *conn) +{ + spin_lock_bh(&conn->session->lock); + set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); + spin_unlock_bh(&conn->session->lock); +} +EXPORT_SYMBOL_GPL(iscsi_suspend_queue); + +/** + * iscsi_suspend_tx - suspend iscsi_data_xmit + * @conn: iscsi conn tp stop processing IO on. + * + * This function sets the suspend bit to prevent iscsi_data_xmit + * from sending new IO, and if work is queued on the xmit thread + * it will wait for it to be completed. + */ +void iscsi_suspend_tx(struct iscsi_conn *conn) +{ + struct Scsi_Host *shost = conn->session->host; + struct iscsi_host *ihost = shost_priv(shost); + + set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); + if (ihost->workq) + flush_workqueue(ihost->workq); +} +EXPORT_SYMBOL_GPL(iscsi_suspend_tx); - sc->result = err; - scsi_set_resid(sc, scsi_bufflen(sc)); - if (conn->ctask == ctask) - conn->ctask = NULL; - /* release ref from queuecommand */ - __iscsi_put_ctask(ctask); +static void iscsi_start_tx(struct iscsi_conn *conn) +{ + clear_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); + iscsi_conn_queue_work(conn); } /* - * Fail commands. session lock held and recv side suspended and xmit - * thread flushed + * We want to make sure a ping is in flight. It has timed out. + * And we are not busy processing a pdu that is making + * progress but got started before the ping and is taking a while + * to complete so the ping is just stuck behind it in a queue. */ -static void fail_all_commands(struct iscsi_conn *conn, unsigned lun) +static int iscsi_has_ping_timed_out(struct iscsi_conn *conn) { - struct iscsi_cmd_task *ctask, *tmp; + if (conn->ping_task && + time_before_eq(conn->last_recv + (conn->recv_timeout * HZ) + + (conn->ping_timeout * HZ), jiffies)) + return 1; + else + return 0; +} - if (conn->ctask && (conn->ctask->sc->device->lun == lun || lun == -1)) - conn->ctask = NULL; +static enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc) +{ + enum blk_eh_timer_return rc = BLK_EH_NOT_HANDLED; + struct iscsi_task *task = NULL, *running_task; + struct iscsi_cls_session *cls_session; + struct iscsi_session *session; + struct iscsi_conn *conn; + int i; - /* flush pending */ - list_for_each_entry_safe(ctask, tmp, &conn->xmitqueue, running) { - if (lun == ctask->sc->device->lun || lun == -1) { - debug_scsi("failing pending sc %p itt 0x%x\n", - ctask->sc, ctask->itt); - fail_command(conn, ctask, DID_BUS_BUSY << 16); - } + cls_session = starget_to_session(scsi_target(sc->device)); + session = cls_session->dd_data; + + ISCSI_DBG_EH(session, "scsi cmd %p timedout\n", sc); + + spin_lock(&session->lock); + if (session->state != ISCSI_STATE_LOGGED_IN) { + /* + * We are probably in the middle of iscsi recovery so let + * that complete and handle the error. + */ + rc = BLK_EH_RESET_TIMER; + goto done; } - list_for_each_entry_safe(ctask, tmp, &conn->requeue, running) { - if (lun == ctask->sc->device->lun || lun == -1) { - debug_scsi("failing requeued sc %p itt 0x%x\n", - ctask->sc, ctask->itt); - fail_command(conn, ctask, DID_BUS_BUSY << 16); - } + conn = session->leadconn; + if (!conn) { + /* In the middle of shuting down */ + rc = BLK_EH_RESET_TIMER; + goto done; + } + + task = (struct iscsi_task *)sc->SCp.ptr; + if (!task) { + /* + * Raced with completion. Just reset timer, and let it + * complete normally + */ + rc = BLK_EH_RESET_TIMER; + goto done; + } + + /* + * If we have sent (at least queued to the network layer) a pdu or + * recvd one for the task since the last timeout ask for + * more time. If on the next timeout we have not made progress + * we can check if it is the task or connection when we send the + * nop as a ping. + */ + if (time_after(task->last_xfer, task->last_timeout)) { + ISCSI_DBG_EH(session, "Command making progress. Asking " + "scsi-ml for more time to complete. " + "Last data xfer at %lu. Last timeout was at " + "%lu\n.", task->last_xfer, task->last_timeout); + task->have_checked_conn = false; + rc = BLK_EH_RESET_TIMER; + goto done; } - /* fail all other running */ - list_for_each_entry_safe(ctask, tmp, &conn->run_list, running) { - if (lun == ctask->sc->device->lun || lun == -1) { - debug_scsi("failing in progress sc %p itt 0x%x\n", - ctask->sc, ctask->itt); - fail_command(conn, ctask, DID_BUS_BUSY << 16); + if (!conn->recv_timeout && !conn->ping_timeout) + goto done; + /* + * if the ping timedout then we are in the middle of cleaning up + * and can let the iscsi eh handle it + */ + if (iscsi_has_ping_timed_out(conn)) { + rc = BLK_EH_RESET_TIMER; + goto done; + } + + for (i = 0; i < conn->session->cmds_max; i++) { + running_task = conn->session->cmds[i]; + if (!running_task->sc || running_task == task || + running_task->state != ISCSI_TASK_RUNNING) + continue; + + /* + * Only check if cmds started before this one have made + * progress, or this could never fail + */ + if (time_after(running_task->sc->jiffies_at_alloc, + task->sc->jiffies_at_alloc)) + continue; + + if (time_after(running_task->last_xfer, task->last_timeout)) { + /* + * This task has not made progress, but a task + * started before us has transferred data since + * we started/last-checked. We could be queueing + * too many tasks or the LU is bad. + * + * If the device is bad the cmds ahead of us on + * other devs will complete, and this loop will + * eventually fail starting the scsi eh. + */ + ISCSI_DBG_EH(session, "Command has not made progress " + "but commands ahead of it have. " + "Asking scsi-ml for more time to " + "complete. Our last xfer vs running task " + "last xfer %lu/%lu. Last check %lu.\n", + task->last_xfer, running_task->last_xfer, + task->last_timeout); + rc = BLK_EH_RESET_TIMER; + goto done; } } -} -static void iscsi_suspend_tx(struct iscsi_conn *conn) -{ - set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); - scsi_flush_work(conn->session->host); + /* Assumes nop timeout is shorter than scsi cmd timeout */ + if (task->have_checked_conn) + goto done; + + /* + * Checking the transport already or nop from a cmd timeout still + * running + */ + if (conn->ping_task) { + task->have_checked_conn = true; + rc = BLK_EH_RESET_TIMER; + goto done; + } + + /* Make sure there is a transport check done */ + iscsi_send_nopout(conn, NULL); + task->have_checked_conn = true; + rc = BLK_EH_RESET_TIMER; + +done: + if (task) + task->last_timeout = jiffies; + spin_unlock(&session->lock); + ISCSI_DBG_EH(session, "return %s\n", rc == BLK_EH_RESET_TIMER ? + "timer reset" : "nh"); + return rc; } -static void iscsi_start_tx(struct iscsi_conn *conn) +static void iscsi_check_transport_timeouts(unsigned long data) { - clear_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); - scsi_queue_work(conn->session->host, &conn->xmitwork); + struct iscsi_conn *conn = (struct iscsi_conn *)data; + struct iscsi_session *session = conn->session; + unsigned long recv_timeout, next_timeout = 0, last_recv; + + spin_lock(&session->lock); + if (session->state != ISCSI_STATE_LOGGED_IN) + goto done; + + recv_timeout = conn->recv_timeout; + if (!recv_timeout) + goto done; + + recv_timeout *= HZ; + last_recv = conn->last_recv; + + if (iscsi_has_ping_timed_out(conn)) { + iscsi_conn_printk(KERN_ERR, conn, "ping timeout of %d secs " + "expired, recv timeout %d, last rx %lu, " + "last ping %lu, now %lu\n", + conn->ping_timeout, conn->recv_timeout, + last_recv, conn->last_ping, jiffies); + spin_unlock(&session->lock); + iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); + return; + } + + if (time_before_eq(last_recv + recv_timeout, jiffies)) { + /* send a ping to try to provoke some traffic */ + ISCSI_DBG_CONN(conn, "Sending nopout as ping\n"); + iscsi_send_nopout(conn, NULL); + next_timeout = conn->last_ping + (conn->ping_timeout * HZ); + } else + next_timeout = last_recv + recv_timeout; + + ISCSI_DBG_CONN(conn, "Setting next tmo %lu\n", next_timeout); + mod_timer(&conn->transport_timer, next_timeout); +done: + spin_unlock(&session->lock); } -static void iscsi_prep_abort_task_pdu(struct iscsi_cmd_task *ctask, +static void iscsi_prep_abort_task_pdu(struct iscsi_task *task, struct iscsi_tm *hdr) { memset(hdr, 0, sizeof(*hdr)); hdr->opcode = ISCSI_OP_SCSI_TMFUNC | ISCSI_OP_IMMEDIATE; hdr->flags = ISCSI_TM_FUNC_ABORT_TASK & ISCSI_FLAG_TM_FUNC_MASK; hdr->flags |= ISCSI_FLAG_CMD_FINAL; - memcpy(hdr->lun, ctask->hdr->lun, sizeof(hdr->lun)); - hdr->rtt = ctask->hdr->itt; - hdr->refcmdsn = ctask->hdr->cmdsn; + memcpy(hdr->lun, task->lun, sizeof(hdr->lun)); + hdr->rtt = task->hdr_itt; + hdr->refcmdsn = task->cmdsn; } int iscsi_eh_abort(struct scsi_cmnd *sc) { - struct Scsi_Host *host = sc->device->host; - struct iscsi_session *session = iscsi_hostdata(host->hostdata); + struct iscsi_cls_session *cls_session; + struct iscsi_session *session; struct iscsi_conn *conn; - struct iscsi_cmd_task *ctask; + struct iscsi_task *task; struct iscsi_tm *hdr; int rc, age; + cls_session = starget_to_session(scsi_target(sc->device)); + session = cls_session->dd_data; + + ISCSI_DBG_EH(session, "aborting sc %p\n", sc); + mutex_lock(&session->eh_mutex); spin_lock_bh(&session->lock); /* @@ -1236,7 +2125,8 @@ int iscsi_eh_abort(struct scsi_cmnd *sc) * got the command. */ if (!sc->SCp.ptr) { - debug_scsi("sc never reached iscsi layer or it completed.\n"); + ISCSI_DBG_EH(session, "sc never reached iscsi layer or " + "it completed.\n"); spin_unlock_bh(&session->lock); mutex_unlock(&session->eh_mutex); return SUCCESS; @@ -1250,6 +2140,8 @@ int iscsi_eh_abort(struct scsi_cmnd *sc) sc->SCp.phase != session->age) { spin_unlock_bh(&session->lock); mutex_unlock(&session->eh_mutex); + ISCSI_DBG_EH(session, "failing abort due to dropped " + "session.\n"); return FAILED; } @@ -1257,17 +2149,18 @@ int iscsi_eh_abort(struct scsi_cmnd *sc) conn->eh_abort_cnt++; age = session->age; - ctask = (struct iscsi_cmd_task *)sc->SCp.ptr; - debug_scsi("aborting [sc %p itt 0x%x]\n", sc, ctask->itt); + task = (struct iscsi_task *)sc->SCp.ptr; + ISCSI_DBG_EH(session, "aborting [sc %p itt 0x%x]\n", + sc, task->itt); - /* ctask completed before time out */ - if (!ctask->sc) { - debug_scsi("sc completed while abort in progress\n"); + /* task completed before time out */ + if (!task->sc) { + ISCSI_DBG_EH(session, "sc completed while abort in progress\n"); goto success; } - if (ctask->state == ISCSI_TASK_PENDING) { - fail_command(conn, ctask, DID_ABORT << 16); + if (task->state == ISCSI_TASK_PENDING) { + fail_scsi_task(task, DID_ABORT); goto success; } @@ -1276,82 +2169,256 @@ int iscsi_eh_abort(struct scsi_cmnd *sc) goto failed; conn->tmf_state = TMF_QUEUED; - hdr = &conn->tmhdr; - iscsi_prep_abort_task_pdu(ctask, hdr); + hdr = &conn->tmhdr; + iscsi_prep_abort_task_pdu(task, hdr); + + if (iscsi_exec_task_mgmt_fn(conn, hdr, age, session->abort_timeout)) { + rc = FAILED; + goto failed; + } + + switch (conn->tmf_state) { + case TMF_SUCCESS: + spin_unlock_bh(&session->lock); + /* + * stop tx side incase the target had sent a abort rsp but + * the initiator was still writing out data. + */ + iscsi_suspend_tx(conn); + /* + * we do not stop the recv side because targets have been + * good and have never sent us a successful tmf response + * then sent more data for the cmd. + */ + spin_lock_bh(&session->lock); + fail_scsi_task(task, DID_ABORT); + conn->tmf_state = TMF_INITIAL; + memset(hdr, 0, sizeof(*hdr)); + spin_unlock_bh(&session->lock); + iscsi_start_tx(conn); + goto success_unlocked; + case TMF_TIMEDOUT: + spin_unlock_bh(&session->lock); + iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); + goto failed_unlocked; + case TMF_NOT_FOUND: + if (!sc->SCp.ptr) { + conn->tmf_state = TMF_INITIAL; + memset(hdr, 0, sizeof(*hdr)); + /* task completed before tmf abort response */ + ISCSI_DBG_EH(session, "sc completed while abort in " + "progress\n"); + goto success; + } + /* fall through */ + default: + conn->tmf_state = TMF_INITIAL; + goto failed; + } + +success: + spin_unlock_bh(&session->lock); +success_unlocked: + ISCSI_DBG_EH(session, "abort success [sc %p itt 0x%x]\n", + sc, task->itt); + mutex_unlock(&session->eh_mutex); + return SUCCESS; + +failed: + spin_unlock_bh(&session->lock); +failed_unlocked: + ISCSI_DBG_EH(session, "abort failed [sc %p itt 0x%x]\n", sc, + task ? task->itt : 0); + mutex_unlock(&session->eh_mutex); + return FAILED; +} +EXPORT_SYMBOL_GPL(iscsi_eh_abort); + +static void iscsi_prep_lun_reset_pdu(struct scsi_cmnd *sc, struct iscsi_tm *hdr) +{ + memset(hdr, 0, sizeof(*hdr)); + hdr->opcode = ISCSI_OP_SCSI_TMFUNC | ISCSI_OP_IMMEDIATE; + hdr->flags = ISCSI_TM_FUNC_LOGICAL_UNIT_RESET & ISCSI_FLAG_TM_FUNC_MASK; + hdr->flags |= ISCSI_FLAG_CMD_FINAL; + int_to_scsilun(sc->device->lun, (struct scsi_lun *)hdr->lun); + hdr->rtt = RESERVED_ITT; +} + +int iscsi_eh_device_reset(struct scsi_cmnd *sc) +{ + struct iscsi_cls_session *cls_session; + struct iscsi_session *session; + struct iscsi_conn *conn; + struct iscsi_tm *hdr; + int rc = FAILED; + + cls_session = starget_to_session(scsi_target(sc->device)); + session = cls_session->dd_data; + + ISCSI_DBG_EH(session, "LU Reset [sc %p lun %u]\n", sc, sc->device->lun); + + mutex_lock(&session->eh_mutex); + spin_lock_bh(&session->lock); + /* + * Just check if we are not logged in. We cannot check for + * the phase because the reset could come from a ioctl. + */ + if (!session->leadconn || session->state != ISCSI_STATE_LOGGED_IN) + goto unlock; + conn = session->leadconn; + + /* only have one tmf outstanding at a time */ + if (conn->tmf_state != TMF_INITIAL) + goto unlock; + conn->tmf_state = TMF_QUEUED; + + hdr = &conn->tmhdr; + iscsi_prep_lun_reset_pdu(sc, hdr); + + if (iscsi_exec_task_mgmt_fn(conn, hdr, session->age, + session->lu_reset_timeout)) { + rc = FAILED; + goto unlock; + } + + switch (conn->tmf_state) { + case TMF_SUCCESS: + break; + case TMF_TIMEDOUT: + spin_unlock_bh(&session->lock); + iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); + goto done; + default: + conn->tmf_state = TMF_INITIAL; + goto unlock; + } + + rc = SUCCESS; + spin_unlock_bh(&session->lock); + + iscsi_suspend_tx(conn); + + spin_lock_bh(&session->lock); + memset(hdr, 0, sizeof(*hdr)); + fail_scsi_tasks(conn, sc->device->lun, DID_ERROR); + conn->tmf_state = TMF_INITIAL; + spin_unlock_bh(&session->lock); + + iscsi_start_tx(conn); + goto done; + +unlock: + spin_unlock_bh(&session->lock); +done: + ISCSI_DBG_EH(session, "dev reset result = %s\n", + rc == SUCCESS ? "SUCCESS" : "FAILED"); + mutex_unlock(&session->eh_mutex); + return rc; +} +EXPORT_SYMBOL_GPL(iscsi_eh_device_reset); + +void iscsi_session_recovery_timedout(struct iscsi_cls_session *cls_session) +{ + struct iscsi_session *session = cls_session->dd_data; - if (iscsi_exec_task_mgmt_fn(conn, hdr, age)) { - rc = FAILED; - goto failed; + spin_lock_bh(&session->lock); + if (session->state != ISCSI_STATE_LOGGED_IN) { + session->state = ISCSI_STATE_RECOVERY_FAILED; + if (session->leadconn) + wake_up(&session->leadconn->ehwait); } + spin_unlock_bh(&session->lock); +} +EXPORT_SYMBOL_GPL(iscsi_session_recovery_timedout); - switch (conn->tmf_state) { - case TMF_SUCCESS: - spin_unlock_bh(&session->lock); - iscsi_suspend_tx(conn); - /* - * clean up task if aborted. grab the recv lock as a writer - */ - write_lock_bh(conn->recv_lock); - spin_lock(&session->lock); - fail_command(conn, ctask, DID_ABORT << 16); - conn->tmf_state = TMF_INITIAL; - spin_unlock(&session->lock); - write_unlock_bh(conn->recv_lock); - iscsi_start_tx(conn); - goto success_unlocked; - case TMF_TIMEDOUT: +/** + * iscsi_eh_session_reset - drop session and attempt relogin + * @sc: scsi command + * + * This function will wait for a relogin, session termination from + * userspace, or a recovery/replacement timeout. + */ +static int iscsi_eh_session_reset(struct scsi_cmnd *sc) +{ + struct iscsi_cls_session *cls_session; + struct iscsi_session *session; + struct iscsi_conn *conn; + + cls_session = starget_to_session(scsi_target(sc->device)); + session = cls_session->dd_data; + conn = session->leadconn; + + mutex_lock(&session->eh_mutex); + spin_lock_bh(&session->lock); + if (session->state == ISCSI_STATE_TERMINATE) { +failed: + ISCSI_DBG_EH(session, + "failing session reset: Could not log back into " + "%s, %s [age %d]\n", session->targetname, + conn->persistent_address, session->age); spin_unlock_bh(&session->lock); - iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); - goto failed_unlocked; - case TMF_NOT_FOUND: - if (!sc->SCp.ptr) { - conn->tmf_state = TMF_INITIAL; - /* ctask completed before tmf abort response */ - debug_scsi("sc completed while abort in progress\n"); - goto success; - } - /* fall through */ - default: - conn->tmf_state = TMF_INITIAL; - goto failed; + mutex_unlock(&session->eh_mutex); + return FAILED; } -success: spin_unlock_bh(&session->lock); -success_unlocked: - debug_scsi("abort success [sc %lx itt 0x%x]\n", (long)sc, ctask->itt); mutex_unlock(&session->eh_mutex); - return SUCCESS; + /* + * we drop the lock here but the leadconn cannot be destoyed while + * we are in the scsi eh + */ + iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); -failed: + ISCSI_DBG_EH(session, "wait for relogin\n"); + wait_event_interruptible(conn->ehwait, + session->state == ISCSI_STATE_TERMINATE || + session->state == ISCSI_STATE_LOGGED_IN || + session->state == ISCSI_STATE_RECOVERY_FAILED); + if (signal_pending(current)) + flush_signals(current); + + mutex_lock(&session->eh_mutex); + spin_lock_bh(&session->lock); + if (session->state == ISCSI_STATE_LOGGED_IN) { + ISCSI_DBG_EH(session, + "session reset succeeded for %s,%s\n", + session->targetname, conn->persistent_address); + } else + goto failed; spin_unlock_bh(&session->lock); -failed_unlocked: - debug_scsi("abort failed [sc %p itt 0x%x]\n", sc, - ctask ? ctask->itt : 0); mutex_unlock(&session->eh_mutex); - return FAILED; + return SUCCESS; } -EXPORT_SYMBOL_GPL(iscsi_eh_abort); -static void iscsi_prep_lun_reset_pdu(struct scsi_cmnd *sc, struct iscsi_tm *hdr) +static void iscsi_prep_tgt_reset_pdu(struct scsi_cmnd *sc, struct iscsi_tm *hdr) { memset(hdr, 0, sizeof(*hdr)); hdr->opcode = ISCSI_OP_SCSI_TMFUNC | ISCSI_OP_IMMEDIATE; - hdr->flags = ISCSI_TM_FUNC_LOGICAL_UNIT_RESET & ISCSI_FLAG_TM_FUNC_MASK; + hdr->flags = ISCSI_TM_FUNC_TARGET_WARM_RESET & ISCSI_FLAG_TM_FUNC_MASK; hdr->flags |= ISCSI_FLAG_CMD_FINAL; - int_to_scsilun(sc->device->lun, (struct scsi_lun *)hdr->lun); - hdr->rtt = ISCSI_RESERVED_TAG; + hdr->rtt = RESERVED_ITT; } -int iscsi_eh_device_reset(struct scsi_cmnd *sc) +/** + * iscsi_eh_target_reset - reset target + * @sc: scsi command + * + * This will attempt to send a warm target reset. If that fails + * then we will drop the session and attempt ERL0 recovery. + */ +int iscsi_eh_target_reset(struct scsi_cmnd *sc) { - struct Scsi_Host *host = sc->device->host; - struct iscsi_session *session = iscsi_hostdata(host->hostdata); + struct iscsi_cls_session *cls_session; + struct iscsi_session *session; struct iscsi_conn *conn; struct iscsi_tm *hdr; int rc = FAILED; - debug_scsi("LU Reset [sc %p lun %u]\n", sc, sc->device->lun); + cls_session = starget_to_session(scsi_target(sc->device)); + session = cls_session->dd_data; + + ISCSI_DBG_EH(session, "tgt Reset [sc %p tgt %s]\n", sc, + session->targetname); mutex_lock(&session->eh_mutex); spin_lock_bh(&session->lock); @@ -1369,9 +2436,10 @@ int iscsi_eh_device_reset(struct scsi_cmnd *sc) conn->tmf_state = TMF_QUEUED; hdr = &conn->tmhdr; - iscsi_prep_lun_reset_pdu(sc, hdr); + iscsi_prep_tgt_reset_pdu(sc, hdr); - if (iscsi_exec_task_mgmt_fn(conn, hdr, session->age)) { + if (iscsi_exec_task_mgmt_fn(conn, hdr, session->age, + session->tgt_reset_timeout)) { rc = FAILED; goto unlock; } @@ -1392,13 +2460,12 @@ int iscsi_eh_device_reset(struct scsi_cmnd *sc) spin_unlock_bh(&session->lock); iscsi_suspend_tx(conn); - /* need to grab the recv lock then session lock */ - write_lock_bh(conn->recv_lock); - spin_lock(&session->lock); - fail_all_commands(conn, sc->device->lun); + + spin_lock_bh(&session->lock); + memset(hdr, 0, sizeof(*hdr)); + fail_scsi_tasks(conn, -1, DID_ERROR); conn->tmf_state = TMF_INITIAL; - spin_unlock(&session->lock); - write_unlock_bh(conn->recv_lock); + spin_unlock_bh(&session->lock); iscsi_start_tx(conn); goto done; @@ -1406,12 +2473,15 @@ int iscsi_eh_device_reset(struct scsi_cmnd *sc) unlock: spin_unlock_bh(&session->lock); done: - debug_scsi("iscsi_eh_device_reset %s\n", - rc == SUCCESS ? "SUCCESS" : "FAILED"); + ISCSI_DBG_EH(session, "tgt %s reset result = %s\n", session->targetname, + rc == SUCCESS ? "SUCCESS" : "FAILED"); mutex_unlock(&session->eh_mutex); + + if (rc == FAILED) + rc = iscsi_eh_session_reset(sc); return rc; } -EXPORT_SYMBOL_GPL(iscsi_eh_device_reset); +EXPORT_SYMBOL_GPL(iscsi_eh_target_reset); /* * Pre-allocate a pool of @max items of @item_size. By default, the pool @@ -1434,12 +2504,9 @@ iscsi_pool_init(struct iscsi_pool *q, int max, void ***items, int item_size) num_arrays++; q->pool = kzalloc(num_arrays * max * sizeof(void*), GFP_KERNEL); if (q->pool == NULL) - goto enomem; + return -ENOMEM; - q->queue = kfifo_init((void*)q->pool, max * sizeof(void*), - GFP_KERNEL, NULL); - if (q->queue == ERR_PTR(-ENOMEM)) - goto enomem; + kfifo_init(&q->queue, (void*)q->pool, max * sizeof(void*)); for (i = 0; i < max; i++) { q->pool[i] = kzalloc(item_size, GFP_KERNEL); @@ -1447,7 +2514,7 @@ iscsi_pool_init(struct iscsi_pool *q, int max, void ***items, int item_size) q->max = i; goto enomem; } - __kfifo_put(q->queue, (void*)&q->pool[i], sizeof(void*)); + kfifo_in(&q->queue, (void*)&q->pool[i], sizeof(void*)); } if (items) { @@ -1469,178 +2536,277 @@ void iscsi_pool_free(struct iscsi_pool *q) for (i = 0; i < q->max; i++) kfree(q->pool[i]); - if (q->pool) - kfree(q->pool); + kfree(q->pool); } EXPORT_SYMBOL_GPL(iscsi_pool_free); -/* - * iSCSI Session's hostdata organization: +/** + * iscsi_host_add - add host to system + * @shost: scsi host + * @pdev: parent device * - * *------------------* <== hostdata_session(host->hostdata) - * | ptr to class sess| - * |------------------| <== iscsi_hostdata(host->hostdata) - * | iscsi_session | - * *------------------* + * This should be called by partial offload and software iscsi drivers + * to add a host to the system. */ +int iscsi_host_add(struct Scsi_Host *shost, struct device *pdev) +{ + if (!shost->can_queue) + shost->can_queue = ISCSI_DEF_XMIT_CMDS_MAX; + + if (!shost->cmd_per_lun) + shost->cmd_per_lun = ISCSI_DEF_CMD_PER_LUN; + + if (!shost->transportt->eh_timed_out) + shost->transportt->eh_timed_out = iscsi_eh_cmd_timed_out; + return scsi_add_host(shost, pdev); +} +EXPORT_SYMBOL_GPL(iscsi_host_add); + +/** + * iscsi_host_alloc - allocate a host and driver data + * @sht: scsi host template + * @dd_data_size: driver host data size + * @xmit_can_sleep: bool indicating if LLD will queue IO from a work queue + * + * This should be called by partial offload and software iscsi drivers. + * To access the driver specific memory use the iscsi_host_priv() macro. + */ +struct Scsi_Host *iscsi_host_alloc(struct scsi_host_template *sht, + int dd_data_size, bool xmit_can_sleep) +{ + struct Scsi_Host *shost; + struct iscsi_host *ihost; + + shost = scsi_host_alloc(sht, sizeof(struct iscsi_host) + dd_data_size); + if (!shost) + return NULL; + ihost = shost_priv(shost); + + if (xmit_can_sleep) { + snprintf(ihost->workq_name, sizeof(ihost->workq_name), + "iscsi_q_%d", shost->host_no); + ihost->workq = create_singlethread_workqueue(ihost->workq_name); + if (!ihost->workq) + goto free_host; + } + + spin_lock_init(&ihost->lock); + ihost->state = ISCSI_HOST_SETUP; + ihost->num_sessions = 0; + init_waitqueue_head(&ihost->session_removal_wq); + return shost; + +free_host: + scsi_host_put(shost); + return NULL; +} +EXPORT_SYMBOL_GPL(iscsi_host_alloc); + +static void iscsi_notify_host_removed(struct iscsi_cls_session *cls_session) +{ + iscsi_session_failure(cls_session->dd_data, ISCSI_ERR_INVALID_HOST); +} + +/** + * iscsi_host_remove - remove host and sessions + * @shost: scsi host + * + * If there are any sessions left, this will initiate the removal and wait + * for the completion. + */ +void iscsi_host_remove(struct Scsi_Host *shost) +{ + struct iscsi_host *ihost = shost_priv(shost); + unsigned long flags; + + spin_lock_irqsave(&ihost->lock, flags); + ihost->state = ISCSI_HOST_REMOVED; + spin_unlock_irqrestore(&ihost->lock, flags); + + iscsi_host_for_each_session(shost, iscsi_notify_host_removed); + wait_event_interruptible(ihost->session_removal_wq, + ihost->num_sessions == 0); + if (signal_pending(current)) + flush_signals(current); -#define hostdata_privsize(_sz) (sizeof(unsigned long) + _sz + \ - _sz % sizeof(unsigned long)) + scsi_remove_host(shost); + if (ihost->workq) + destroy_workqueue(ihost->workq); +} +EXPORT_SYMBOL_GPL(iscsi_host_remove); + +void iscsi_host_free(struct Scsi_Host *shost) +{ + struct iscsi_host *ihost = shost_priv(shost); + + kfree(ihost->netdev); + kfree(ihost->hwaddress); + kfree(ihost->initiatorname); + scsi_host_put(shost); +} +EXPORT_SYMBOL_GPL(iscsi_host_free); + +static void iscsi_host_dec_session_cnt(struct Scsi_Host *shost) +{ + struct iscsi_host *ihost = shost_priv(shost); + unsigned long flags; + + shost = scsi_host_get(shost); + if (!shost) { + printk(KERN_ERR "Invalid state. Cannot notify host removal " + "of session teardown event because host already " + "removed.\n"); + return; + } -#define hostdata_session(_hostdata) (iscsi_ptr(*(unsigned long *)_hostdata)) + spin_lock_irqsave(&ihost->lock, flags); + ihost->num_sessions--; + if (ihost->num_sessions == 0) + wake_up(&ihost->session_removal_wq); + spin_unlock_irqrestore(&ihost->lock, flags); + scsi_host_put(shost); +} /** * iscsi_session_setup - create iscsi cls session and host and session - * @scsit: scsi transport template * @iscsit: iscsi transport template - * @cmds_max: scsi host can queue - * @qdepth: scsi host cmds per lun - * @cmd_task_size: LLD ctask private data size - * @mgmt_task_size: LLD mtask private data size + * @shost: scsi host + * @cmds_max: session can queue + * @cmd_task_size: LLD task private data size * @initial_cmdsn: initial CmdSN - * @hostno: host no allocated * * This can be used by software iscsi_transports that allocate * a session per scsi host. - **/ + * + * Callers should set cmds_max to the largest total numer (mgmt + scsi) of + * tasks they support. The iscsi layer reserves ISCSI_MGMT_CMDS_MAX tasks + * for nop handling and login/logout requests. + */ struct iscsi_cls_session * -iscsi_session_setup(struct iscsi_transport *iscsit, - struct scsi_transport_template *scsit, - uint16_t cmds_max, uint16_t qdepth, - int cmd_task_size, int mgmt_task_size, - uint32_t initial_cmdsn, uint32_t *hostno) +iscsi_session_setup(struct iscsi_transport *iscsit, struct Scsi_Host *shost, + uint16_t cmds_max, int dd_size, int cmd_task_size, + uint32_t initial_cmdsn, unsigned int id) { - struct Scsi_Host *shost; + struct iscsi_host *ihost = shost_priv(shost); struct iscsi_session *session; struct iscsi_cls_session *cls_session; - int cmd_i; + int cmd_i, scsi_cmds, total_cmds = cmds_max; + unsigned long flags; - if (qdepth > ISCSI_MAX_CMD_PER_LUN || qdepth < 1) { - if (qdepth != 0) - printk(KERN_ERR "iscsi: invalid queue depth of %d. " - "Queue depth must be between 1 and %d.\n", - qdepth, ISCSI_MAX_CMD_PER_LUN); - qdepth = ISCSI_DEF_CMD_PER_LUN; + spin_lock_irqsave(&ihost->lock, flags); + if (ihost->state == ISCSI_HOST_REMOVED) { + spin_unlock_irqrestore(&ihost->lock, flags); + return NULL; } + ihost->num_sessions++; + spin_unlock_irqrestore(&ihost->lock, flags); - if (cmds_max < 2 || (cmds_max & (cmds_max - 1)) || - cmds_max >= ISCSI_MGMT_ITT_OFFSET) { - if (cmds_max != 0) - printk(KERN_ERR "iscsi: invalid can_queue of %d. " - "can_queue must be a power of 2 and between " - "2 and %d - setting to %d.\n", cmds_max, - ISCSI_MGMT_ITT_OFFSET, ISCSI_DEF_XMIT_CMDS_MAX); - cmds_max = ISCSI_DEF_XMIT_CMDS_MAX; + if (!total_cmds) + total_cmds = ISCSI_DEF_XMIT_CMDS_MAX; + /* + * The iscsi layer needs some tasks for nop handling and tmfs, + * so the cmds_max must at least be greater than ISCSI_MGMT_CMDS_MAX + * + 1 command for scsi IO. + */ + if (total_cmds < ISCSI_TOTAL_CMDS_MIN) { + printk(KERN_ERR "iscsi: invalid can_queue of %d. can_queue " + "must be a power of two that is at least %d.\n", + total_cmds, ISCSI_TOTAL_CMDS_MIN); + goto dec_session_count; } - shost = scsi_host_alloc(iscsit->host_template, - hostdata_privsize(sizeof(*session))); - if (!shost) - return NULL; + if (total_cmds > ISCSI_TOTAL_CMDS_MAX) { + printk(KERN_ERR "iscsi: invalid can_queue of %d. can_queue " + "must be a power of 2 less than or equal to %d.\n", + cmds_max, ISCSI_TOTAL_CMDS_MAX); + total_cmds = ISCSI_TOTAL_CMDS_MAX; + } + + if (!is_power_of_2(total_cmds)) { + printk(KERN_ERR "iscsi: invalid can_queue of %d. can_queue " + "must be a power of 2.\n", total_cmds); + total_cmds = rounddown_pow_of_two(total_cmds); + if (total_cmds < ISCSI_TOTAL_CMDS_MIN) + return NULL; + printk(KERN_INFO "iscsi: Rounding can_queue to %d.\n", + total_cmds); + } + scsi_cmds = total_cmds - ISCSI_MGMT_CMDS_MAX; - /* the iscsi layer takes one task for reserve */ - shost->can_queue = cmds_max - 1; - shost->cmd_per_lun = qdepth; - shost->max_id = 1; - shost->max_channel = 0; - shost->max_lun = iscsit->max_lun; - shost->max_cmd_len = iscsit->max_cmd_len; - shost->transportt = scsit; - shost->transportt->create_work_queue = 1; - *hostno = shost->host_no; - - session = iscsi_hostdata(shost->hostdata); - memset(session, 0, sizeof(struct iscsi_session)); + cls_session = iscsi_alloc_session(shost, iscsit, + sizeof(struct iscsi_session) + + dd_size); + if (!cls_session) + goto dec_session_count; + session = cls_session->dd_data; + session->cls_session = cls_session; session->host = shost; session->state = ISCSI_STATE_FREE; - session->mgmtpool_max = ISCSI_MGMT_CMDS_MAX; - session->cmds_max = cmds_max; + session->fast_abort = 1; + session->tgt_reset_timeout = 30; + session->lu_reset_timeout = 15; + session->abort_timeout = 10; + session->scsi_cmds_max = scsi_cmds; + session->cmds_max = total_cmds; session->queued_cmdsn = session->cmdsn = initial_cmdsn; session->exp_cmdsn = initial_cmdsn + 1; session->max_cmdsn = initial_cmdsn + 1; session->max_r2t = 1; session->tt = iscsit; + session->dd_data = cls_session->dd_data + sizeof(*session); mutex_init(&session->eh_mutex); + spin_lock_init(&session->lock); /* initialize SCSI PDU commands pool */ if (iscsi_pool_init(&session->cmdpool, session->cmds_max, (void***)&session->cmds, - cmd_task_size + sizeof(struct iscsi_cmd_task))) + cmd_task_size + sizeof(struct iscsi_task))) goto cmdpool_alloc_fail; /* pre-format cmds pool with ITT */ for (cmd_i = 0; cmd_i < session->cmds_max; cmd_i++) { - struct iscsi_cmd_task *ctask = session->cmds[cmd_i]; + struct iscsi_task *task = session->cmds[cmd_i]; if (cmd_task_size) - ctask->dd_data = &ctask[1]; - ctask->itt = cmd_i; - INIT_LIST_HEAD(&ctask->running); - } - - spin_lock_init(&session->lock); - - /* initialize immediate command pool */ - if (iscsi_pool_init(&session->mgmtpool, session->mgmtpool_max, - (void***)&session->mgmt_cmds, - mgmt_task_size + sizeof(struct iscsi_mgmt_task))) - goto mgmtpool_alloc_fail; - - - /* pre-format immediate cmds pool with ITT */ - for (cmd_i = 0; cmd_i < session->mgmtpool_max; cmd_i++) { - struct iscsi_mgmt_task *mtask = session->mgmt_cmds[cmd_i]; - - if (mgmt_task_size) - mtask->dd_data = &mtask[1]; - mtask->itt = ISCSI_MGMT_ITT_OFFSET + cmd_i; - INIT_LIST_HEAD(&mtask->running); + task->dd_data = &task[1]; + task->itt = cmd_i; + task->state = ISCSI_TASK_FREE; + INIT_LIST_HEAD(&task->running); } - if (scsi_add_host(shost, NULL)) - goto add_host_fail; - if (!try_module_get(iscsit->owner)) - goto cls_session_fail; + goto module_get_fail; - cls_session = iscsi_create_session(shost, iscsit, 0); - if (!cls_session) - goto module_put; - *(unsigned long*)shost->hostdata = (unsigned long)cls_session; + if (iscsi_add_session(cls_session, id)) + goto cls_session_fail; return cls_session; -module_put: - module_put(iscsit->owner); cls_session_fail: - scsi_remove_host(shost); -add_host_fail: - iscsi_pool_free(&session->mgmtpool); -mgmtpool_alloc_fail: + module_put(iscsit->owner); +module_get_fail: iscsi_pool_free(&session->cmdpool); cmdpool_alloc_fail: - scsi_host_put(shost); + iscsi_free_session(cls_session); +dec_session_count: + iscsi_host_dec_session_cnt(shost); return NULL; } EXPORT_SYMBOL_GPL(iscsi_session_setup); /** * iscsi_session_teardown - destroy session, host, and cls_session - * shost: scsi host + * @cls_session: iscsi session * - * This can be used by software iscsi_transports that allocate - * a session per scsi host. - **/ + * The driver must have called iscsi_remove_session before + * calling this. + */ void iscsi_session_teardown(struct iscsi_cls_session *cls_session) { - struct Scsi_Host *shost = iscsi_session_to_shost(cls_session); - struct iscsi_session *session = iscsi_hostdata(shost->hostdata); + struct iscsi_session *session = cls_session->dd_data; struct module *owner = cls_session->transport->owner; + struct Scsi_Host *shost = session->host; - iscsi_unblock_session(cls_session); - scsi_remove_host(shost); - - iscsi_pool_free(&session->mgmtpool); iscsi_pool_free(&session->cmdpool); kfree(session->password); @@ -1648,12 +2814,11 @@ void iscsi_session_teardown(struct iscsi_cls_session *cls_session) kfree(session->username); kfree(session->username_in); kfree(session->targetname); - kfree(session->netdev); - kfree(session->hwaddress); kfree(session->initiatorname); + kfree(session->ifacename); iscsi_destroy_session(cls_session); - scsi_host_put(shost); + iscsi_host_dec_session_cnt(shost); module_put(owner); } EXPORT_SYMBOL_GPL(iscsi_session_teardown); @@ -1661,59 +2826,67 @@ EXPORT_SYMBOL_GPL(iscsi_session_teardown); /** * iscsi_conn_setup - create iscsi_cls_conn and iscsi_conn * @cls_session: iscsi_cls_session + * @dd_size: private driver data size * @conn_idx: cid - **/ + */ struct iscsi_cls_conn * -iscsi_conn_setup(struct iscsi_cls_session *cls_session, uint32_t conn_idx) +iscsi_conn_setup(struct iscsi_cls_session *cls_session, int dd_size, + uint32_t conn_idx) { - struct iscsi_session *session = class_to_transport_session(cls_session); + struct iscsi_session *session = cls_session->dd_data; struct iscsi_conn *conn; struct iscsi_cls_conn *cls_conn; char *data; - cls_conn = iscsi_create_conn(cls_session, conn_idx); + cls_conn = iscsi_create_conn(cls_session, sizeof(*conn) + dd_size, + conn_idx); if (!cls_conn) return NULL; conn = cls_conn->dd_data; - memset(conn, 0, sizeof(*conn)); + memset(conn, 0, sizeof(*conn) + dd_size); + conn->dd_data = cls_conn->dd_data + sizeof(*conn); conn->session = session; conn->cls_conn = cls_conn; conn->c_stage = ISCSI_CONN_INITIAL_STAGE; conn->id = conn_idx; conn->exp_statsn = 0; conn->tmf_state = TMF_INITIAL; - INIT_LIST_HEAD(&conn->run_list); - INIT_LIST_HEAD(&conn->mgmt_run_list); + + init_timer(&conn->transport_timer); + conn->transport_timer.data = (unsigned long)conn; + conn->transport_timer.function = iscsi_check_transport_timeouts; + INIT_LIST_HEAD(&conn->mgmtqueue); - INIT_LIST_HEAD(&conn->xmitqueue); + INIT_LIST_HEAD(&conn->cmdqueue); INIT_LIST_HEAD(&conn->requeue); INIT_WORK(&conn->xmitwork, iscsi_xmitworker); - /* allocate login_mtask used for the login/text sequences */ + /* allocate login_task used for the login/text sequences */ spin_lock_bh(&session->lock); - if (!__kfifo_get(session->mgmtpool.queue, - (void*)&conn->login_mtask, + if (!kfifo_out(&session->cmdpool.queue, + (void*)&conn->login_task, sizeof(void*))) { spin_unlock_bh(&session->lock); - goto login_mtask_alloc_fail; + goto login_task_alloc_fail; } spin_unlock_bh(&session->lock); - data = kmalloc(ISCSI_DEF_MAX_RECV_SEG_LEN, GFP_KERNEL); + data = (char *) __get_free_pages(GFP_KERNEL, + get_order(ISCSI_DEF_MAX_RECV_SEG_LEN)); if (!data) - goto login_mtask_data_alloc_fail; - conn->login_mtask->data = conn->data = data; + goto login_task_data_alloc_fail; + conn->login_task->data = conn->data = data; init_timer(&conn->tmf_timer); init_waitqueue_head(&conn->ehwait); return cls_conn; -login_mtask_data_alloc_fail: - __kfifo_put(session->mgmtpool.queue, (void*)&conn->login_mtask, +login_task_data_alloc_fail: + kfifo_in(&session->cmdpool.queue, (void*)&conn->login_task, sizeof(void*)); -login_mtask_alloc_fail: +login_task_alloc_fail: iscsi_destroy_conn(cls_conn); return NULL; } @@ -1732,6 +2905,8 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn) struct iscsi_session *session = conn->session; unsigned long flags; + del_timer_sync(&conn->transport_timer); + spin_lock_bh(&session->lock); conn->c_stage = ISCSI_CONN_CLEANUP_WAIT; if (session->leadconn == conn) { @@ -1755,9 +2930,10 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn) } spin_unlock_irqrestore(session->host->host_lock, flags); msleep_interruptible(500); - printk(KERN_INFO "iscsi: scsi conn_destroy(): host_busy %d " - "host_failed %d\n", session->host->host_busy, - session->host->host_failed); + iscsi_conn_printk(KERN_INFO, conn, "iscsi conn_destroy(): " + "host_busy %d host_failed %d\n", + session->host->host_busy, + session->host->host_failed); /* * force eh_abort() to unblock */ @@ -1768,9 +2944,10 @@ void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn) iscsi_suspend_tx(conn); spin_lock_bh(&session->lock); - kfree(conn->data); + free_pages((unsigned long) conn->data, + get_order(ISCSI_DEF_MAX_RECV_SEG_LEN)); kfree(conn->persistent_address); - __kfifo_put(session->mgmtpool.queue, (void*)&conn->login_mtask, + kfifo_in(&session->cmdpool.queue, (void*)&conn->login_task, sizeof(void*)); if (session->leadconn == conn) session->leadconn = NULL; @@ -1786,23 +2963,42 @@ int iscsi_conn_start(struct iscsi_cls_conn *cls_conn) struct iscsi_session *session = conn->session; if (!session) { - printk(KERN_ERR "iscsi: can't start unbound connection\n"); + iscsi_conn_printk(KERN_ERR, conn, + "can't start unbound connection\n"); return -EPERM; } if ((session->imm_data_en || !session->initial_r2t_en) && session->first_burst > session->max_burst) { - printk("iscsi: invalid burst lengths: " - "first_burst %d max_burst %d\n", - session->first_burst, session->max_burst); + iscsi_conn_printk(KERN_INFO, conn, "invalid burst lengths: " + "first_burst %d max_burst %d\n", + session->first_burst, session->max_burst); return -EINVAL; } + if (conn->ping_timeout && !conn->recv_timeout) { + iscsi_conn_printk(KERN_ERR, conn, "invalid recv timeout of " + "zero. Using 5 seconds\n."); + conn->recv_timeout = 5; + } + + if (conn->recv_timeout && !conn->ping_timeout) { + iscsi_conn_printk(KERN_ERR, conn, "invalid ping timeout of " + "zero. Using 5 seconds.\n"); + conn->ping_timeout = 5; + } + spin_lock_bh(&session->lock); conn->c_stage = ISCSI_CONN_STARTED; session->state = ISCSI_STATE_LOGGED_IN; session->queued_cmdsn = session->cmdsn; + conn->last_recv = jiffies; + conn->last_ping = jiffies; + if (conn->recv_timeout && conn->ping_timeout) + mod_timer(&conn->transport_timer, + jiffies + (conn->recv_timeout * HZ)); + switch(conn->stop_stage) { case STOP_CONN_RECOVER: /* @@ -1812,11 +3008,9 @@ int iscsi_conn_start(struct iscsi_cls_conn *cls_conn) conn->stop_stage = 0; conn->tmf_state = TMF_INITIAL; session->age++; - spin_unlock_bh(&session->lock); - - iscsi_unblock_session(session_to_cls(session)); - wake_up(&conn->ehwait); - return 0; + if (session->age == 16) + session->age = 0; + break; case STOP_CONN_TERM: conn->stop_stage = 0; break; @@ -1825,37 +3019,35 @@ int iscsi_conn_start(struct iscsi_cls_conn *cls_conn) } spin_unlock_bh(&session->lock); + iscsi_unblock_session(session->cls_session); + wake_up(&conn->ehwait); return 0; } EXPORT_SYMBOL_GPL(iscsi_conn_start); static void -flush_control_queues(struct iscsi_session *session, struct iscsi_conn *conn) +fail_mgmt_tasks(struct iscsi_session *session, struct iscsi_conn *conn) { - struct iscsi_mgmt_task *mtask, *tmp; + struct iscsi_task *task; + int i, state; - /* handle pending */ - list_for_each_entry_safe(mtask, tmp, &conn->mgmtqueue, running) { - debug_scsi("flushing pending mgmt task itt 0x%x\n", mtask->itt); - list_del_init(&mtask->running); - if (mtask == conn->login_mtask) + for (i = 0; i < conn->session->cmds_max; i++) { + task = conn->session->cmds[i]; + if (task->sc) continue; - __kfifo_put(session->mgmtpool.queue, (void*)&mtask, - sizeof(void*)); - } - /* handle running */ - list_for_each_entry_safe(mtask, tmp, &conn->mgmt_run_list, running) { - debug_scsi("flushing running mgmt task itt 0x%x\n", mtask->itt); - list_del_init(&mtask->running); - - if (mtask == conn->login_mtask) + if (task->state == ISCSI_TASK_FREE) continue; - __kfifo_put(session->mgmtpool.queue, (void*)&mtask, - sizeof(void*)); - } - conn->mtask = NULL; + ISCSI_DBG_SESSION(conn->session, + "failing mgmt itt 0x%x state %d\n", + task->itt, task->state); + state = ISCSI_TASK_ABRT_SESS_RECOV; + if (task->state == ISCSI_TASK_PENDING) + state = ISCSI_TASK_COMPLETED; + iscsi_complete_task(task, state); + + } } static void iscsi_start_session_recovery(struct iscsi_session *session, @@ -1872,17 +3064,6 @@ static void iscsi_start_session_recovery(struct iscsi_session *session, } /* - * The LLD either freed/unset the lock on us, or userspace called - * stop but did not create a proper connection (connection was never - * bound or it was unbound then stop was called). - */ - if (!conn->recv_lock) { - spin_unlock_bh(&session->lock); - mutex_unlock(&session->eh_mutex); - return; - } - - /* * When this is called for the in_login state, we only want to clean * up the login task and connection. We do not need to block and set * the recovery state again @@ -1891,18 +3072,17 @@ static void iscsi_start_session_recovery(struct iscsi_session *session, session->state = ISCSI_STATE_TERMINATE; else if (conn->stop_stage != STOP_CONN_RECOVER) session->state = ISCSI_STATE_IN_RECOVERY; + spin_unlock_bh(&session->lock); + + del_timer_sync(&conn->transport_timer); + iscsi_suspend_tx(conn); + spin_lock_bh(&session->lock); old_stop_stage = conn->stop_stage; conn->stop_stage = flag; conn->c_stage = ISCSI_CONN_STOPPED; spin_unlock_bh(&session->lock); - iscsi_suspend_tx(conn); - - write_lock_bh(conn->recv_lock); - set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_rx); - write_unlock_bh(conn->recv_lock); - /* * for connection level recovery we should not calculate * header digest. conn->hdr_size used for optimization @@ -1914,8 +3094,8 @@ static void iscsi_start_session_recovery(struct iscsi_session *session, conn->datadgst_en = 0; if (session->state == ISCSI_STATE_IN_RECOVERY && old_stop_stage != STOP_CONN_RECOVER) { - debug_scsi("blocking session\n"); - iscsi_block_session(session_to_cls(session)); + ISCSI_DBG_SESSION(session, "blocking session\n"); + iscsi_block_session(session->cls_session); } } @@ -1923,8 +3103,9 @@ static void iscsi_start_session_recovery(struct iscsi_session *session, * flush queues. */ spin_lock_bh(&session->lock); - fail_all_commands(conn, -1); - flush_control_queues(session, conn); + fail_scsi_tasks(conn, -1, DID_TRANSPORT_DISRUPTED); + fail_mgmt_tasks(session, conn); + memset(&conn->tmhdr, 0, sizeof(conn->tmhdr)); spin_unlock_bh(&session->lock); mutex_unlock(&session->eh_mutex); } @@ -1940,7 +3121,8 @@ void iscsi_conn_stop(struct iscsi_cls_conn *cls_conn, int flag) iscsi_start_session_recovery(session, conn, flag); break; default: - printk(KERN_ERR "iscsi: invalid stop flag %d\n", flag); + iscsi_conn_printk(KERN_ERR, conn, + "invalid stop flag %d\n", flag); } } EXPORT_SYMBOL_GPL(iscsi_conn_stop); @@ -1948,7 +3130,7 @@ EXPORT_SYMBOL_GPL(iscsi_conn_stop); int iscsi_conn_bind(struct iscsi_cls_session *cls_session, struct iscsi_cls_conn *cls_conn, int is_leading) { - struct iscsi_session *session = class_to_transport_session(cls_session); + struct iscsi_session *session = cls_session->dd_data; struct iscsi_conn *conn = cls_conn->dd_data; spin_lock_bh(&session->lock); @@ -1965,6 +3147,23 @@ int iscsi_conn_bind(struct iscsi_cls_session *cls_session, } EXPORT_SYMBOL_GPL(iscsi_conn_bind); +static int iscsi_switch_str_param(char **param, char *new_val_buf) +{ + char *new_val; + + if (*param) { + if (!strcmp(*param, new_val_buf)) + return 0; + } + + new_val = kstrdup(new_val_buf, GFP_NOIO); + if (!new_val) + return -ENOMEM; + + kfree(*param); + *param = new_val; + return 0; +} int iscsi_set_param(struct iscsi_cls_conn *cls_conn, enum iscsi_param param, char *buf, int buflen) @@ -1977,6 +3176,21 @@ int iscsi_set_param(struct iscsi_cls_conn *cls_conn, case ISCSI_PARAM_FAST_ABORT: sscanf(buf, "%d", &session->fast_abort); break; + case ISCSI_PARAM_ABORT_TMO: + sscanf(buf, "%d", &session->abort_timeout); + break; + case ISCSI_PARAM_LU_RESET_TMO: + sscanf(buf, "%d", &session->lu_reset_timeout); + break; + case ISCSI_PARAM_TGT_RESET_TMO: + sscanf(buf, "%d", &session->tgt_reset_timeout); + break; + case ISCSI_PARAM_PING_TMO: + sscanf(buf, "%d", &conn->ping_timeout); + break; + case ISCSI_PARAM_RECV_TMO: + sscanf(buf, "%d", &conn->recv_timeout); + break; case ISCSI_PARAM_MAX_RECV_DLENGTH: sscanf(buf, "%d", &conn->max_recv_dlength); break; @@ -2025,38 +3239,15 @@ int iscsi_set_param(struct iscsi_cls_conn *cls_conn, sscanf(buf, "%u", &conn->exp_statsn); break; case ISCSI_PARAM_USERNAME: - kfree(session->username); - session->username = kstrdup(buf, GFP_KERNEL); - if (!session->username) - return -ENOMEM; - break; + return iscsi_switch_str_param(&session->username, buf); case ISCSI_PARAM_USERNAME_IN: - kfree(session->username_in); - session->username_in = kstrdup(buf, GFP_KERNEL); - if (!session->username_in) - return -ENOMEM; - break; + return iscsi_switch_str_param(&session->username_in, buf); case ISCSI_PARAM_PASSWORD: - kfree(session->password); - session->password = kstrdup(buf, GFP_KERNEL); - if (!session->password) - return -ENOMEM; - break; + return iscsi_switch_str_param(&session->password, buf); case ISCSI_PARAM_PASSWORD_IN: - kfree(session->password_in); - session->password_in = kstrdup(buf, GFP_KERNEL); - if (!session->password_in) - return -ENOMEM; - break; + return iscsi_switch_str_param(&session->password_in, buf); case ISCSI_PARAM_TARGET_NAME: - /* this should not change between logins */ - if (session->targetname) - break; - - session->targetname = kstrdup(buf, GFP_KERNEL); - if (!session->targetname) - return -ENOMEM; - break; + return iscsi_switch_str_param(&session->targetname, buf); case ISCSI_PARAM_TPGT: sscanf(buf, "%d", &session->tpgt); break; @@ -2064,17 +3255,11 @@ int iscsi_set_param(struct iscsi_cls_conn *cls_conn, sscanf(buf, "%d", &conn->persistent_port); break; case ISCSI_PARAM_PERSISTENT_ADDRESS: - /* - * this is the address returned in discovery so it should - * not change between logins. - */ - if (conn->persistent_address) - break; - - conn->persistent_address = kstrdup(buf, GFP_KERNEL); - if (!conn->persistent_address) - return -ENOMEM; - break; + return iscsi_switch_str_param(&conn->persistent_address, buf); + case ISCSI_PARAM_IFACE_NAME: + return iscsi_switch_str_param(&session->ifacename, buf); + case ISCSI_PARAM_INITIATOR_NAME: + return iscsi_switch_str_param(&session->initiatorname, buf); default: return -ENOSYS; } @@ -2086,14 +3271,22 @@ EXPORT_SYMBOL_GPL(iscsi_set_param); int iscsi_session_get_param(struct iscsi_cls_session *cls_session, enum iscsi_param param, char *buf) { - struct Scsi_Host *shost = iscsi_session_to_shost(cls_session); - struct iscsi_session *session = iscsi_hostdata(shost->hostdata); + struct iscsi_session *session = cls_session->dd_data; int len; switch(param) { case ISCSI_PARAM_FAST_ABORT: len = sprintf(buf, "%d\n", session->fast_abort); break; + case ISCSI_PARAM_ABORT_TMO: + len = sprintf(buf, "%d\n", session->abort_timeout); + break; + case ISCSI_PARAM_LU_RESET_TMO: + len = sprintf(buf, "%d\n", session->lu_reset_timeout); + break; + case ISCSI_PARAM_TGT_RESET_TMO: + len = sprintf(buf, "%d\n", session->tgt_reset_timeout); + break; case ISCSI_PARAM_INITIAL_R2T_EN: len = sprintf(buf, "%d\n", session->initial_r2t_en); break; @@ -2136,6 +3329,12 @@ int iscsi_session_get_param(struct iscsi_cls_session *cls_session, case ISCSI_PARAM_PASSWORD_IN: len = sprintf(buf, "%s\n", session->password_in); break; + case ISCSI_PARAM_IFACE_NAME: + len = sprintf(buf, "%s\n", session->ifacename); + break; + case ISCSI_PARAM_INITIATOR_NAME: + len = sprintf(buf, "%s\n", session->initiatorname); + break; default: return -ENOSYS; } @@ -2151,6 +3350,12 @@ int iscsi_conn_get_param(struct iscsi_cls_conn *cls_conn, int len; switch(param) { + case ISCSI_PARAM_PING_TMO: + len = sprintf(buf, "%u\n", conn->ping_timeout); + break; + case ISCSI_PARAM_RECV_TMO: + len = sprintf(buf, "%u\n", conn->recv_timeout); + break; case ISCSI_PARAM_MAX_RECV_DLENGTH: len = sprintf(buf, "%u\n", conn->max_recv_dlength); break; @@ -2189,29 +3394,22 @@ EXPORT_SYMBOL_GPL(iscsi_conn_get_param); int iscsi_host_get_param(struct Scsi_Host *shost, enum iscsi_host_param param, char *buf) { - struct iscsi_session *session = iscsi_hostdata(shost->hostdata); + struct iscsi_host *ihost = shost_priv(shost); int len; switch (param) { case ISCSI_HOST_PARAM_NETDEV_NAME: - if (!session->netdev) - len = sprintf(buf, "%s\n", "default"); - else - len = sprintf(buf, "%s\n", session->netdev); + len = sprintf(buf, "%s\n", ihost->netdev); break; case ISCSI_HOST_PARAM_HWADDRESS: - if (!session->hwaddress) - len = sprintf(buf, "%s\n", "default"); - else - len = sprintf(buf, "%s\n", session->hwaddress); + len = sprintf(buf, "%s\n", ihost->hwaddress); break; case ISCSI_HOST_PARAM_INITIATOR_NAME: - if (!session->initiatorname) - len = sprintf(buf, "%s\n", "unknown"); - else - len = sprintf(buf, "%s\n", session->initiatorname); + len = sprintf(buf, "%s\n", ihost->initiatorname); + break; + case ISCSI_HOST_PARAM_IPADDRESS: + len = sprintf(buf, "%s\n", ihost->local_address); break; - default: return -ENOSYS; } @@ -2223,21 +3421,15 @@ EXPORT_SYMBOL_GPL(iscsi_host_get_param); int iscsi_host_set_param(struct Scsi_Host *shost, enum iscsi_host_param param, char *buf, int buflen) { - struct iscsi_session *session = iscsi_hostdata(shost->hostdata); + struct iscsi_host *ihost = shost_priv(shost); switch (param) { case ISCSI_HOST_PARAM_NETDEV_NAME: - if (!session->netdev) - session->netdev = kstrdup(buf, GFP_KERNEL); - break; + return iscsi_switch_str_param(&ihost->netdev, buf); case ISCSI_HOST_PARAM_HWADDRESS: - if (!session->hwaddress) - session->hwaddress = kstrdup(buf, GFP_KERNEL); - break; + return iscsi_switch_str_param(&ihost->hwaddress, buf); case ISCSI_HOST_PARAM_INITIATOR_NAME: - if (!session->initiatorname) - session->initiatorname = kstrdup(buf, GFP_KERNEL); - break; + return iscsi_switch_str_param(&ihost->initiatorname, buf); default: return -ENOSYS; }