Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1030359AbVIITiN (ORCPT ); Fri, 9 Sep 2005 15:38:13 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1030362AbVIIThn (ORCPT ); Fri, 9 Sep 2005 15:37:43 -0400 Received: from magic.adaptec.com ([216.52.22.17]:60613 "EHLO magic.adaptec.com") by vger.kernel.org with ESMTP id S1030359AbVIITh3 (ORCPT ); Fri, 9 Sep 2005 15:37:29 -0400 Message-ID: <4321E473.2020002@adaptec.com> Date: Fri, 09 Sep 2005 15:37:23 -0400 From: Luben Tuikov User-Agent: Mozilla Thunderbird 1.0.6 (X11/20050716) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Linux Kernel Mailing List , SCSI Mailing List Subject: [PATCH 2.6.13 20/20] aic94xx: aic94xx_tmf.c Implements the Task Management Funcions defined by SAS Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 09 Sep 2005 19:37:28.0049 (UTC) FILETIME=[E9568610:01C5B575] Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 19705 Lines: 647 Signed-off-by: Luben Tuikov diff -X linux-2.6.13/Documentation/dontdiff -Naur linux-2.6.13-orig/drivers/scsi/aic94xx/aic94xx_tmf.c linux-2.6.13/drivers/scsi/aic94xx/aic94xx_tmf.c --- linux-2.6.13-orig/drivers/scsi/aic94xx/aic94xx_tmf.c 1969-12-31 19:00:00.000000000 -0500 +++ linux-2.6.13/drivers/scsi/aic94xx/aic94xx_tmf.c 2005-09-09 11:21:23.000000000 -0400 @@ -0,0 +1,634 @@ +/* + * Aic94xx Task Management Functions + * + * Copyright (C) 2005 Adaptec, Inc. All rights reserved. + * Copyright (C) 2005 Luben Tuikov + * + * This file is licensed under GPLv2. + * + * This file is part of the aic94xx driver. + * + * The aic94xx driver is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation; version 2 of the + * License. + * + * The aic94xx driver is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with the aic94xx driver; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + * + * $Id: //depot/aic94xx/aic94xx_tmf.c#34 $ + */ + +#include +#include +#include "aic94xx.h" +#include "aic94xx_sas.h" +#include "aic94xx_hwi.h" + +/* ---------- Internal enqueue ---------- */ + +static int asd_enqueue_internal(struct asd_ascb *ascb, + void (*tasklet_complete)(struct asd_ascb *, + struct done_list_struct *), + void (*timed_out)(unsigned long)) +{ + int res; + + ascb->tasklet_complete = tasklet_complete; + ascb->uldd_timer = 1; + + ascb->timer.data = (unsigned long) ascb; + ascb->timer.function = timed_out; + ascb->timer.expires = jiffies + AIC94XX_SCB_TIMEOUT; + + add_timer(&ascb->timer); + + res = asd_post_ascb_list(ascb->ha, ascb, 1); + if (unlikely(res)) + del_timer(&ascb->timer); + return res; +} + +static inline void asd_timedout_common(unsigned long data) +{ + struct asd_ascb *ascb = (void *) data; + struct asd_seq_data *seq = &ascb->ha->seq; + unsigned long flags; + + spin_lock_irqsave(&seq->pend_q_lock, flags); + seq->pending--; + list_del_init(&ascb->list); + spin_unlock_irqrestore(&seq->pend_q_lock, flags); +} + +/* ---------- CLEAR NEXUS ---------- */ + +static void asd_clear_nexus_tasklet_complete(struct asd_ascb *ascb, + struct done_list_struct *dl) +{ + ASD_DPRINTK("%s: here\n", __FUNCTION__); + if (!del_timer(&ascb->timer)) { + ASD_DPRINTK("%s: couldn't delete timer\n", __FUNCTION__); + return; + } + ASD_DPRINTK("%s: opcode: 0x%x\n", __FUNCTION__, dl->opcode); + ascb->uldd_task = (void *) (unsigned long) dl->opcode; + complete(&ascb->completion); +} + +static void asd_clear_nexus_timedout(unsigned long data) +{ + struct asd_ascb *ascb = (void *) data; + + ASD_DPRINTK("%s: here\n", __FUNCTION__); + asd_timedout_common(data); + ascb->uldd_task = (void *) TMF_RESP_FUNC_FAILED; + complete(&ascb->completion); +} + +#define CLEAR_NEXUS_PRE \ + ASD_DPRINTK("%s: PRE\n", __FUNCTION__); \ + res = 1; \ + ascb = asd_ascb_alloc_list(asd_ha, &res, GFP_KERNEL); \ + if (!ascb) \ + return -ENOMEM; \ + \ + scb = ascb->scb; \ + scb->header.opcode = CLEAR_NEXUS + +#define CLEAR_NEXUS_POST \ + ASD_DPRINTK("%s: POST\n", __FUNCTION__); \ + res = asd_enqueue_internal(ascb, asd_clear_nexus_tasklet_complete, \ + asd_clear_nexus_timedout); \ + if (res) \ + goto out_err; \ + ASD_DPRINTK("%s: clear nexus posted, waiting...\n", __FUNCTION__); \ + wait_for_completion(&ascb->completion); \ + res = (int) (unsigned long) ascb->uldd_task; \ + if (res == TC_NO_ERROR) \ + res = TMF_RESP_FUNC_COMPLETE; \ +out_err: \ + asd_ascb_free(ascb); \ + return res + +int asd_clear_nexus_ha(struct sas_ha_struct *sas_ha) +{ + struct asd_ha_struct *asd_ha = sas_ha->lldd_ha; + struct asd_ascb *ascb; + struct scb *scb; + int res; + + CLEAR_NEXUS_PRE; + scb->clear_nexus.nexus = NEXUS_ADAPTER; + CLEAR_NEXUS_POST; +} + +int asd_clear_nexus_port(struct sas_port *port) +{ + struct asd_ha_struct *asd_ha = port->ha->lldd_ha; + struct asd_ascb *ascb; + struct scb *scb; + int res; + + CLEAR_NEXUS_PRE; + scb->clear_nexus.nexus = NEXUS_PORT; + scb->clear_nexus.conn_mask = port->phy_mask; + CLEAR_NEXUS_POST; +} + +#if 0 +static int asd_clear_nexus_I_T(struct domain_device *dev) +{ + struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha; + struct asd_ascb *ascb; + struct scb *scb; + int res; + + CLEAR_NEXUS_PRE; + scb->clear_nexus.nexus = NEXUS_I_T; + scb->clear_nexus.flags = SEND_Q | EXEC_Q | NOTINQ; + if (dev->tproto) + scb->clear_nexus.flags |= SUSPEND_TX; + scb->clear_nexus.conn_handle = cpu_to_le16((u16)(unsigned long) + dev->lldd_dev); + CLEAR_NEXUS_POST; +} +#endif + +static int asd_clear_nexus_I_T_L(struct domain_device *dev, u8 *lun) +{ + struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha; + struct asd_ascb *ascb; + struct scb *scb; + int res; + + CLEAR_NEXUS_PRE; + scb->clear_nexus.nexus = NEXUS_I_T_L; + scb->clear_nexus.flags = SEND_Q | EXEC_Q | NOTINQ; + if (dev->tproto) + scb->clear_nexus.flags |= SUSPEND_TX; + memcpy(scb->clear_nexus.ssp_task.lun, lun, 8); + scb->clear_nexus.conn_handle = cpu_to_le16((u16)(unsigned long) + dev->lldd_dev); + CLEAR_NEXUS_POST; +} + +static int asd_clear_nexus_tag(struct sas_task *task) +{ + struct asd_ha_struct *asd_ha = task->dev->port->ha->lldd_ha; + struct asd_ascb *tascb = task->lldd_task; + struct asd_ascb *ascb; + struct scb *scb; + int res; + + CLEAR_NEXUS_PRE; + scb->clear_nexus.nexus = NEXUS_TAG; + memcpy(scb->clear_nexus.ssp_task.lun, task->ssp_task.LUN, 8); + scb->clear_nexus.ssp_task.tag = tascb->tag; + if (task->dev->tproto) + scb->clear_nexus.conn_handle = cpu_to_le16((u16)(unsigned long) + task->dev->lldd_dev); + CLEAR_NEXUS_POST; +} + +static int asd_clear_nexus_index(struct sas_task *task) +{ + struct asd_ha_struct *asd_ha = task->dev->port->ha->lldd_ha; + struct asd_ascb *tascb = task->lldd_task; + struct asd_ascb *ascb; + struct scb *scb; + int res; + + CLEAR_NEXUS_PRE; + scb->clear_nexus.nexus = NEXUS_TRANS_CX; + if (task->dev->tproto) + scb->clear_nexus.conn_handle = cpu_to_le16((u16)(unsigned long) + task->dev->lldd_dev); + scb->clear_nexus.index = cpu_to_le16(tascb->tc_index); + CLEAR_NEXUS_POST; +} + +/* ---------- TMFs ---------- */ + +static void asd_tmf_timedout(unsigned long data) +{ + struct asd_ascb *ascb = (void *) data; + + ASD_DPRINTK("tmf timed out\n"); + asd_timedout_common(data); + ascb->uldd_task = (void *) TMF_RESP_FUNC_FAILED; + complete(&ascb->completion); +} + +static int asd_get_tmf_resp_tasklet(struct asd_ascb *ascb, + struct done_list_struct *dl) +{ + struct asd_ha_struct *asd_ha = ascb->ha; + unsigned long flags; + struct tc_resp_sb_struct { + __le16 index_escb; + u8 len_lsb; + u8 flags; + } __attribute__ ((packed)) *resp_sb = (void *) dl->status_block; + + int edb_id = ((resp_sb->flags & 0x70) >> 4)-1; + struct asd_ascb *escb; + struct asd_dma_tok *edb; + struct ssp_frame_hdr *fh; + struct ssp_response_iu *ru; + int res = TMF_RESP_FUNC_FAILED; + + ASD_DPRINTK("tmf resp tasklet\n"); + + spin_lock_irqsave(&asd_ha->seq.tc_index_lock, flags); + escb = asd_tc_index_find(&asd_ha->seq, + (int)le16_to_cpu(resp_sb->index_escb)); + spin_unlock_irqrestore(&asd_ha->seq.tc_index_lock, flags); + + if (!escb) { + ASD_DPRINTK("Uh-oh! No escb for this dl?!\n"); + return res; + } + + edb = asd_ha->seq.edb_arr[edb_id + escb->edb_index]; + ascb->tag = *(__be16 *)(edb->vaddr+4); + fh = edb->vaddr + 16; + ru = edb->vaddr + 16 + sizeof(*fh); + res = ru->status; + if (ru->datapres == 1) /* Response data present */ + res = ru->resp_data[3]; +#if 0 + ascb->tag = fh->tag; +#endif + ascb->tag_valid = 1; + + asd_invalidate_edb(escb, edb_id); + return res; +} + +static void asd_tmf_tasklet_complete(struct asd_ascb *ascb, + struct done_list_struct *dl) +{ + if (!del_timer(&ascb->timer)) + return; + + ASD_DPRINTK("tmf tasklet complete\n"); + + if (dl->opcode == TC_SSP_RESP) + ascb->uldd_task = (void *) (unsigned long) + asd_get_tmf_resp_tasklet(ascb, dl); + else + ascb->uldd_task = (void *) 0xFF00 + (unsigned long) dl->opcode; + + complete(&ascb->completion); +} + +static inline int asd_clear_nexus(struct sas_task *task) +{ + int res = TMF_RESP_FUNC_FAILED; + struct asd_ascb *tascb = task->lldd_task; + unsigned long flags; + + ASD_DPRINTK("task not done, clearing nexus\n"); + if (tascb->tag_valid) + res = asd_clear_nexus_tag(task); + else + res = asd_clear_nexus_index(task); + wait_for_completion_timeout(&tascb->completion, + AIC94XX_SCB_TIMEOUT); + ASD_DPRINTK("came back from clear nexus\n"); + spin_lock_irqsave(&task->task_state_lock, flags); + if (task->task_state_flags & SAS_TASK_STATE_DONE) + res = TMF_RESP_FUNC_COMPLETE; + spin_unlock_irqrestore(&task->task_state_lock, flags); + + return res; +} + +/** + * asd_abort_task -- ABORT TASK TMF + * @task: the task to be aborted + * + * Before calling ABORT TASK the task state flags should be ORed with + * SAS_TASK_STATE_ABORTED (unless SAS_TASK_STATE_DONE is set) under + * the task_state_lock IRQ spinlock, then ABORT TASK *must* be called. + * + * Implements the ABORT TASK TMF, I_T_L_Q nexus. + * Returns: SAS TMF responses (see sas_task.h), + * -ENOMEM, + * -SAS_QUEUE_FULL. + * + * When ABORT TASK returns, the caller of ABORT TASK checks first the + * task->task_state_flags, and then the return value of ABORT TASK. + * + * If the task has task state bit SAS_TASK_STATE_DONE set, then the + * task was completed successfully prior to it being aborted. The + * caller of ABORT TASK has responsibility to call task->task_done() + * xor free the task, depending on their framework. The return code + * is TMF_RESP_FUNC_FAILED in this case. + * + * Else the SAS_TASK_STATE_DONE bit is not set, + * If the return code is TMF_RESP_FUNC_COMPLETE, then + * the task was aborted successfully. The caller of + * ABORT TASK has responsibility to call task->task_done() + * to finish the task, xor free the task depending on their + * framework. + * else + * the ABORT TASK returned some kind of error. The task + * was _not_ cancelled. Nothing can be assumed. + * The caller of ABORT TASK may wish to retry. + */ +int asd_abort_task(struct sas_task *task) +{ + struct asd_ascb *tascb = task->lldd_task; + struct asd_ha_struct *asd_ha = tascb->ha; + int res = 1; + unsigned long flags; + struct asd_ascb *ascb = NULL; + struct scb *scb; + + spin_lock_irqsave(&task->task_state_lock, flags); + if (task->task_state_flags & SAS_TASK_STATE_DONE) { + spin_unlock_irqrestore(&task->task_state_lock, flags); + res = TMF_RESP_FUNC_COMPLETE; + ASD_DPRINTK("%s: task 0x%p done\n", __FUNCTION__, task); + goto out_done; + } + spin_unlock_irqrestore(&task->task_state_lock, flags); + + ascb = asd_ascb_alloc_list(asd_ha, &res, GFP_KERNEL); + if (!ascb) + return -ENOMEM; + scb = ascb->scb; + + scb->header.opcode = ABORT_TASK; + + switch (task->task_proto) { + case SATA_PROTO: + case SAS_PROTO_STP: + scb->abort_task.proto_conn_rate = (1 << 5); /* STP */ + break; + case SAS_PROTO_SSP: + scb->abort_task.proto_conn_rate = (1 << 4); /* SSP */ + scb->abort_task.proto_conn_rate |= task->dev->linkrate; + break; + case SAS_PROTO_SMP: + break; + default: + break; + } + + if (task->task_proto == SAS_PROTO_SSP) { + scb->abort_task.ssp_frame.frame_type = SSP_TASK; + memcpy(scb->abort_task.ssp_frame.hashed_dest_addr, + task->dev->hashed_sas_addr, HASHED_SAS_ADDR_SIZE); + memcpy(scb->abort_task.ssp_frame.hashed_src_addr, + task->dev->port->ha->hashed_sas_addr, + HASHED_SAS_ADDR_SIZE); + scb->abort_task.ssp_frame.tptt = cpu_to_be16(0xFFFF); + + memcpy(scb->abort_task.ssp_task.lun, task->ssp_task.LUN, 8); + scb->abort_task.ssp_task.tmf = TMF_ABORT_TASK; + scb->abort_task.ssp_task.tag = cpu_to_be16(0xFFFF); + } + + scb->abort_task.sister_scb = cpu_to_le16(0xFFFF); + scb->abort_task.conn_handle = cpu_to_le16( + (u16)(unsigned long)task->dev->lldd_dev); + scb->abort_task.retry_count = 1; + scb->abort_task.index = cpu_to_le16((u16)tascb->tc_index); + scb->abort_task.itnl_to = cpu_to_le16(ITNL_TIMEOUT_CONST); + + res = asd_enqueue_internal(ascb, asd_tmf_tasklet_complete, + asd_tmf_timedout); + if (res) + goto out; + wait_for_completion(&ascb->completion); + ASD_DPRINTK("tmf came back\n"); + + res = (int) (unsigned long) ascb->uldd_task; + tascb->tag = ascb->tag; + tascb->tag_valid = ascb->tag_valid; + + spin_lock_irqsave(&task->task_state_lock, flags); + if (task->task_state_flags & SAS_TASK_STATE_DONE) { + spin_unlock_irqrestore(&task->task_state_lock, flags); + res = TMF_RESP_FUNC_COMPLETE; + ASD_DPRINTK("%s: task 0x%p done\n", __FUNCTION__, task); + goto out_done; + } + spin_unlock_irqrestore(&task->task_state_lock, flags); + + switch (res) { + /* The task to be aborted has been sent to the device. + * We got a Response IU for the ABORT TASK TMF. */ + case TC_NO_ERROR + 0xFF00: + case TMF_RESP_FUNC_COMPLETE: + case TMF_RESP_FUNC_FAILED: + res = asd_clear_nexus(task); + break; + case TMF_RESP_INVALID_FRAME: + case TMF_RESP_OVERLAPPED_TAG: + case TMF_RESP_FUNC_ESUPP: + case TMF_RESP_NO_LUN: + goto out_done; break; + } + /* In the following we assume that the managing layer + * will _never_ make a mistake, when issuing ABORT TASK. + */ + switch (res) { + default: + res = asd_clear_nexus(task); + /* fallthrough */ + case TC_NO_ERROR + 0xFF00: + case TMF_RESP_FUNC_COMPLETE: + break; + /* The task hasn't been sent to the device xor we never got + * a (sane) Response IU for the ABORT TASK TMF. + */ + case TF_NAK_RECV + 0xFF00: + res = TMF_RESP_INVALID_FRAME; + break; + case TF_TMF_TASK_DONE + 0xFF00: /* done but not reported yet */ + res = TMF_RESP_FUNC_FAILED; + wait_for_completion_timeout(&tascb->completion, + AIC94XX_SCB_TIMEOUT); + spin_lock_irqsave(&task->task_state_lock, flags); + if (task->task_state_flags & SAS_TASK_STATE_DONE) + res = TMF_RESP_FUNC_COMPLETE; + spin_unlock_irqrestore(&task->task_state_lock, flags); + goto out_done; break; + case TF_TMF_NO_TAG + 0xFF00: + case TF_TMF_TAG_FREE + 0xFF00: /* the tag is in the free list */ + case TF_TMF_NO_CTX + 0xFF00: /* not in seq, or proto != SSP */ + case TF_TMF_NO_CONN_HANDLE + 0xFF00: /* no such device */ + res = TMF_RESP_FUNC_COMPLETE; + goto out_done; break; + } +out_done: + if (res == TMF_RESP_FUNC_COMPLETE) { + task->lldd_task = NULL; + mb(); + asd_ascb_free(tascb); + } +out: + asd_ascb_free(ascb); + ASD_DPRINTK("task 0x%p aborted, res: 0x%x\n", task, res); + return res; +} + +/** + * asd_initiate_ssp_tmf -- send a TMF to an I_T_L or I_T_L_Q nexus + * @dev: pointer to struct domain_device of interest + * @lun: pointer to u8[8] which is the LUN + * @tmf: the TMF to be performed (see sas_task.h or the SAS spec) + * @index: the transaction context of the task to be queried if QT TMF + * + * This function is used to send ABORT TASK SET, CLEAR ACA, + * CLEAR TASK SET, LU RESET and QUERY TASK TMFs. + * + * No SCBs should be queued to the I_T_L nexus when this SCB is + * pending. + * + * Returns: TMF response code (see sas_task.h or the SAS spec) + */ +static int asd_initiate_ssp_tmf(struct domain_device *dev, u8 *lun, + int tmf, int index) +{ + struct asd_ha_struct *asd_ha = dev->port->ha->lldd_ha; + struct asd_ascb *ascb; + int res = 1; + struct scb *scb; + + if (!(dev->tproto & SAS_PROTO_SSP)) + return TMF_RESP_FUNC_ESUPP; + + ascb = asd_ascb_alloc_list(asd_ha, &res, GFP_KERNEL); + if (!ascb) + return -ENOMEM; + scb = ascb->scb; + + if (tmf == TMF_QUERY_TASK) + scb->header.opcode = QUERY_SSP_TASK; + else + scb->header.opcode = INITIATE_SSP_TMF; + + scb->ssp_tmf.proto_conn_rate = (1 << 4); /* SSP */ + scb->ssp_tmf.proto_conn_rate |= dev->linkrate; + /* SSP frame header */ + scb->ssp_tmf.ssp_frame.frame_type = SSP_TASK; + memcpy(scb->ssp_tmf.ssp_frame.hashed_dest_addr, + dev->hashed_sas_addr, HASHED_SAS_ADDR_SIZE); + memcpy(scb->ssp_tmf.ssp_frame.hashed_src_addr, + dev->port->ha->hashed_sas_addr, HASHED_SAS_ADDR_SIZE); + scb->ssp_tmf.ssp_frame.tptt = cpu_to_be16(0xFFFF); + /* SSP Task IU */ + memcpy(scb->ssp_tmf.ssp_task.lun, lun, 8); + scb->ssp_tmf.ssp_task.tmf = tmf; + + scb->ssp_tmf.sister_scb = cpu_to_le16(0xFFFF); + scb->ssp_tmf.conn_handle= cpu_to_le16((u16)(unsigned long) + dev->lldd_dev); + scb->ssp_tmf.retry_count = 1; + scb->ssp_tmf.itnl_to = cpu_to_le16(ITNL_TIMEOUT_CONST); + if (tmf == TMF_QUERY_TASK) + scb->ssp_tmf.index = cpu_to_le16(index); + + res = asd_enqueue_internal(ascb, asd_tmf_tasklet_complete, + asd_tmf_timedout); + if (res) + goto out_err; + wait_for_completion(&ascb->completion); + res = (int) (unsigned long) ascb->uldd_task; + + switch (res) { + case TC_NO_ERROR + 0xFF00: + res = TMF_RESP_FUNC_COMPLETE; + break; + case TF_NAK_RECV + 0xFF00: + res = TMF_RESP_INVALID_FRAME; + break; + case TF_TMF_TASK_DONE + 0xFF00: + res = TMF_RESP_FUNC_FAILED; + break; + case TF_TMF_NO_TAG + 0xFF00: + case TF_TMF_TAG_FREE + 0xFF00: /* the tag is in the free list */ + case TF_TMF_NO_CTX + 0xFF00: /* not in seq, or proto != SSP */ + case TF_TMF_NO_CONN_HANDLE + 0xFF00: /* no such device */ + res = TMF_RESP_FUNC_COMPLETE; + break; + default: + ASD_DPRINTK("%s: converting result 0x%x to TMF_RESP_FUNC_FAILED\n", + __FUNCTION__, res); + res = TMF_RESP_FUNC_FAILED; + break; + } +out_err: + asd_ascb_free(ascb); + return res; +} + +int asd_abort_task_set(struct domain_device *dev, u8 *lun) +{ + int res = asd_initiate_ssp_tmf(dev, lun, TMF_ABORT_TASK_SET, 0); + + if (res == TMF_RESP_FUNC_COMPLETE) + asd_clear_nexus_I_T_L(dev, lun); + return res; +} + +int asd_clear_aca(struct domain_device *dev, u8 *lun) +{ + int res = asd_initiate_ssp_tmf(dev, lun, TMF_CLEAR_ACA, 0); + + if (res == TMF_RESP_FUNC_COMPLETE) + asd_clear_nexus_I_T_L(dev, lun); + return res; +} + +int asd_clear_task_set(struct domain_device *dev, u8 *lun) +{ + int res = asd_initiate_ssp_tmf(dev, lun, TMF_CLEAR_TASK_SET, 0); + + if (res == TMF_RESP_FUNC_COMPLETE) + asd_clear_nexus_I_T_L(dev, lun); + return res; +} + +int asd_lu_reset(struct domain_device *dev, u8 *lun) +{ + int res = asd_initiate_ssp_tmf(dev, lun, TMF_LU_RESET, 0); + + if (res == TMF_RESP_FUNC_COMPLETE) + asd_clear_nexus_I_T_L(dev, lun); + return res; +} + +/** + * asd_query_task -- send a QUERY TASK TMF to an I_T_L_Q nexus + * task: pointer to sas_task struct of interest + * + * Returns: TMF_RESP_FUNC_COMPLETE if the task is not in the task set, + * or TMF_RESP_FUNC_SUCC if the task is in the task set. + * + * Normally the management layer sets the task to aborted state, + * and then calls query task and then abort task. + */ +int asd_query_task(struct sas_task *task) +{ + struct asd_ascb *ascb = task->lldd_task; + int index; + + if (ascb) { + index = ascb->tc_index; + return asd_initiate_ssp_tmf(task->dev, task->ssp_task.LUN, + TMF_QUERY_TASK, index); + } + return TMF_RESP_FUNC_COMPLETE; +} - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/