freebsd-skq/sys/dev/twa/tw_cl_init.c

711 lines
22 KiB
C
Raw Normal View History

/*
* Copyright (c) 2004-07 Applied Micro Circuits Corporation.
* Copyright (c) 2004-05 Vinod Kashyap
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $FreeBSD$
*/
/*
* AMCC'S 3ware driver for 9000 series storage controllers.
*
* Author: Vinod Kashyap
* Modifications by: Adam Radford
* Modifications by: Manjunath Ranganathaiah
*/
/*
* Common Layer initialization functions.
*/
#include "tw_osl_share.h"
#include "tw_cl_share.h"
#include "tw_cl_fwif.h"
#include "tw_cl_ioctl.h"
#include "tw_cl.h"
#include "tw_cl_externs.h"
#include "tw_osl_ioctl.h"
/*
* Function name: tw_cl_ctlr_supported
* Description: Determines if a controller is supported.
*
* Input: vendor_id -- vendor id of the controller
* device_id -- device id of the controller
* Output: None
* Return value: TW_CL_TRUE-- controller supported
* TW_CL_FALSE-- controller not supported
*/
TW_INT32
tw_cl_ctlr_supported(TW_INT32 vendor_id, TW_INT32 device_id)
{
if ((vendor_id == TW_CL_VENDOR_ID) &&
((device_id == TW_CL_DEVICE_ID_9K) ||
(device_id == TW_CL_DEVICE_ID_9K_X) ||
(device_id == TW_CL_DEVICE_ID_9K_E) ||
(device_id == TW_CL_DEVICE_ID_9K_SA)))
return(TW_CL_TRUE);
return(TW_CL_FALSE);
}
/*
* Function name: tw_cl_get_pci_bar_info
* Description: Returns PCI BAR info.
*
* Input: device_id -- device id of the controller
* bar_type -- type of PCI BAR in question
* Output: bar_num -- PCI BAR number corresponding to bar_type
* bar0_offset -- byte offset from BAR 0 (0x10 in
* PCI config space)
* bar_size -- size, in bytes, of the BAR in question
* Return value: 0 -- success
* non-zero -- failure
*/
TW_INT32
tw_cl_get_pci_bar_info(TW_INT32 device_id, TW_INT32 bar_type,
TW_INT32 *bar_num, TW_INT32 *bar0_offset, TW_INT32 *bar_size)
{
TW_INT32 error = TW_OSL_ESUCCESS;
switch(device_id) {
case TW_CL_DEVICE_ID_9K:
switch(bar_type) {
case TW_CL_BAR_TYPE_IO:
*bar_num = 0;
*bar0_offset = 0;
*bar_size = 4;
break;
case TW_CL_BAR_TYPE_MEM:
*bar_num = 1;
*bar0_offset = 0x4;
*bar_size = 8;
break;
case TW_CL_BAR_TYPE_SBUF:
*bar_num = 2;
*bar0_offset = 0xC;
*bar_size = 8;
break;
}
break;
case TW_CL_DEVICE_ID_9K_X:
case TW_CL_DEVICE_ID_9K_E:
case TW_CL_DEVICE_ID_9K_SA:
switch(bar_type) {
case TW_CL_BAR_TYPE_IO:
*bar_num = 2;
*bar0_offset = 0x10;
*bar_size = 4;
break;
case TW_CL_BAR_TYPE_MEM:
*bar_num = 1;
*bar0_offset = 0x8;
*bar_size = 8;
break;
case TW_CL_BAR_TYPE_SBUF:
*bar_num = 0;
*bar0_offset = 0;
*bar_size = 8;
break;
}
break;
default:
error = TW_OSL_ENOTTY;
break;
}
return(error);
}
/*
* Function name: tw_cl_get_mem_requirements
* Description: Provides info about Common Layer requirements for a
* controller, given the controller type (in 'flags').
* Input: ctlr_handle -- controller handle
* flags -- more info passed by the OS Layer
* device_id -- device id of the controller
* max_simult_reqs -- maximum # of simultaneous
* requests that the OS Layer expects
* the Common Layer to support
* max_aens -- maximun # of AEN's needed to be supported
* Output: alignment -- alignment needed for all DMA'able
* buffers
* sg_size_factor -- every SG element should have a size
* that's a multiple of this number
* non_dma_mem_size -- # of bytes of memory needed for
* non-DMA purposes
* dma_mem_size -- # of bytes of DMA'able memory needed
* per_req_dma_mem_size -- # of bytes of DMA'able memory
* needed per request, if applicable
* per_req_non_dma_mem_size -- # of bytes of memory needed
* per request for non-DMA purposes,
* if applicable
* Output: None
* Return value: 0 -- success
* non-zero-- failure
*/
TW_INT32
tw_cl_get_mem_requirements(struct tw_cl_ctlr_handle *ctlr_handle,
TW_UINT32 flags, TW_INT32 device_id, TW_INT32 max_simult_reqs,
TW_INT32 max_aens, TW_UINT32 *alignment, TW_UINT32 *sg_size_factor,
TW_UINT32 *non_dma_mem_size, TW_UINT32 *dma_mem_size
)
{
if (device_id == 0)
device_id = TW_CL_DEVICE_ID_9K;
if (max_simult_reqs > TW_CL_MAX_SIMULTANEOUS_REQUESTS) {
tw_cl_create_event(ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1000, 0x1, TW_CL_SEVERITY_ERROR_STRING,
"Too many simultaneous requests to support!",
"requested = %d, supported = %d, error = %d\n",
max_simult_reqs, TW_CL_MAX_SIMULTANEOUS_REQUESTS,
TW_OSL_EBIG);
return(TW_OSL_EBIG);
}
*alignment = TWA_ALIGNMENT(device_id);
*sg_size_factor = TWA_SG_ELEMENT_SIZE_FACTOR(device_id);
/*
* Total non-DMA memory needed is the sum total of memory needed for
* the controller context, request packets (including the 1 needed for
* CL internal requests), and event packets.
*/
*non_dma_mem_size = sizeof(struct tw_cli_ctlr_context) +
Apply driver update from LSI. Many thanks to LSI for continuing to support FreeBSD. 1) Timeout ioctl command timeouts. Do not reset the controller if ioctl command completed successfully. 2) Remove G66_WORKAROUND code (this bug never shipped). 3) Remove unnecessary interrupt lock (intr_lock). 4) Timeout firmware handshake for PChip reset (don't wait forever). 5) Handle interrupts inline. 6) Unmask command interrupt ONLY when adding a command to the pending queue. 7) Mask command interrupt ONLY after removing the last command from the pending queue. 8) Remove TW_OSLI_DEFERRED_INTR_USED code. 9) Replace controller "state" with separate data fields to avoid races: TW_CLI_CTLR_STATE_ACTIVE ctlr->active TW_CLI_CTLR_STATE_INTR_ENABLED ctlr->interrupts_enabled TW_CLI_CTLR_STATE_INTERNAL_REQ_BUSY ctlr->internal_req_busy TW_CLI_CTLR_STATE_GET_MORE_AENS ctlr->get_more_aens TW_CLI_CTLR_STATE_RESET_IN_PROGRESS ctlr->reset_in_progress TW_CLI_CTLR_STATE_RESET_PHASE1_IN_PROGRESS ctlr->reset_phase1_in_progress 10) Fix "req" leak in twa_action() when simq is frozen and req is NOT null. 11) Replace softc "state" with separate data fields to avoid races: TW_OSLI_CTLR_STATE_OPEN sc->open TW_OSLI_CTLR_STATE_SIMQ_FROZEN sc->simq_frozen 12) Fix reference to TW_OSLI_REQ_FLAGS_IN_PROGRESS in tw_osl_complete_passthru() 13) Use correct CAM status values. Change CAM_REQ_CMP_ERR to CAM_REQ_INVALID. Remove use of CAM_RELEASE_SIMQ for physical data addresses. 14) Do not freeze/ release the simq with non I/O commands. When it is appropriate to temporarily freeze the simq with an I/O command use: xpt_freeze_simq(sim, 1); ccb->ccb_h.status |= CAM_RELEASE_SIMQ; otherwise use: xpt_freeze_simq(sim, 1); xpt_release_simq(sim, 1); Submitted by: Tom Couch <tom.couch lsi.com> PR: kern/147695 MFC after: 3 days
2010-06-09 21:40:38 +00:00
(sizeof(struct tw_cli_req_context) * max_simult_reqs) +
(sizeof(struct tw_cl_event_packet) * max_aens);
/*
* Total DMA'able memory needed is the sum total of memory needed for
* all command packets (including the 1 needed for CL internal
* requests), and memory needed to hold the payload for internal
* requests.
*/
*dma_mem_size = (sizeof(struct tw_cl_command_packet) *
Apply driver update from LSI. Many thanks to LSI for continuing to support FreeBSD. 1) Timeout ioctl command timeouts. Do not reset the controller if ioctl command completed successfully. 2) Remove G66_WORKAROUND code (this bug never shipped). 3) Remove unnecessary interrupt lock (intr_lock). 4) Timeout firmware handshake for PChip reset (don't wait forever). 5) Handle interrupts inline. 6) Unmask command interrupt ONLY when adding a command to the pending queue. 7) Mask command interrupt ONLY after removing the last command from the pending queue. 8) Remove TW_OSLI_DEFERRED_INTR_USED code. 9) Replace controller "state" with separate data fields to avoid races: TW_CLI_CTLR_STATE_ACTIVE ctlr->active TW_CLI_CTLR_STATE_INTR_ENABLED ctlr->interrupts_enabled TW_CLI_CTLR_STATE_INTERNAL_REQ_BUSY ctlr->internal_req_busy TW_CLI_CTLR_STATE_GET_MORE_AENS ctlr->get_more_aens TW_CLI_CTLR_STATE_RESET_IN_PROGRESS ctlr->reset_in_progress TW_CLI_CTLR_STATE_RESET_PHASE1_IN_PROGRESS ctlr->reset_phase1_in_progress 10) Fix "req" leak in twa_action() when simq is frozen and req is NOT null. 11) Replace softc "state" with separate data fields to avoid races: TW_OSLI_CTLR_STATE_OPEN sc->open TW_OSLI_CTLR_STATE_SIMQ_FROZEN sc->simq_frozen 12) Fix reference to TW_OSLI_REQ_FLAGS_IN_PROGRESS in tw_osl_complete_passthru() 13) Use correct CAM status values. Change CAM_REQ_CMP_ERR to CAM_REQ_INVALID. Remove use of CAM_RELEASE_SIMQ for physical data addresses. 14) Do not freeze/ release the simq with non I/O commands. When it is appropriate to temporarily freeze the simq with an I/O command use: xpt_freeze_simq(sim, 1); ccb->ccb_h.status |= CAM_RELEASE_SIMQ; otherwise use: xpt_freeze_simq(sim, 1); xpt_release_simq(sim, 1); Submitted by: Tom Couch <tom.couch lsi.com> PR: kern/147695 MFC after: 3 days
2010-06-09 21:40:38 +00:00
(max_simult_reqs)) + (TW_CLI_SECTOR_SIZE);
return(0);
}
/*
* Function name: tw_cl_init_ctlr
* Description: Initializes driver data structures for the controller.
*
* Input: ctlr_handle -- controller handle
* flags -- more info passed by the OS Layer
* device_id -- device id of the controller
* max_simult_reqs -- maximum # of simultaneous requests
* that the OS Layer expects the Common
* Layer to support
* max_aens -- maximun # of AEN's needed to be supported
* non_dma_mem -- ptr to allocated non-DMA memory
* dma_mem -- ptr to allocated DMA'able memory
* dma_mem_phys -- physical address of dma_mem
* Output: None
* Return value: 0 -- success
* non-zero-- failure
*/
TW_INT32
tw_cl_init_ctlr(struct tw_cl_ctlr_handle *ctlr_handle, TW_UINT32 flags,
TW_INT32 device_id, TW_INT32 max_simult_reqs, TW_INT32 max_aens,
TW_VOID *non_dma_mem, TW_VOID *dma_mem, TW_UINT64 dma_mem_phys
)
{
struct tw_cli_ctlr_context *ctlr;
struct tw_cli_req_context *req;
TW_UINT8 *free_non_dma_mem;
TW_INT32 error = TW_OSL_ESUCCESS;
TW_INT32 i;
tw_cli_dbg_printf(3, ctlr_handle, tw_osl_cur_func(), "entered");
if (flags & TW_CL_START_CTLR_ONLY) {
ctlr = (struct tw_cli_ctlr_context *)
(ctlr_handle->cl_ctlr_ctxt);
goto start_ctlr;
}
if (max_simult_reqs > TW_CL_MAX_SIMULTANEOUS_REQUESTS) {
tw_cl_create_event(ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1000, 0x1, TW_CL_SEVERITY_ERROR_STRING,
"Too many simultaneous requests to support!",
"requested = %d, supported = %d, error = %d\n",
max_simult_reqs, TW_CL_MAX_SIMULTANEOUS_REQUESTS,
TW_OSL_EBIG);
return(TW_OSL_EBIG);
}
if ((non_dma_mem == TW_CL_NULL) || (dma_mem == TW_CL_NULL)
) {
tw_cl_create_event(ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1001, 0x1, TW_CL_SEVERITY_ERROR_STRING,
"Insufficient memory for Common Layer's internal usage",
"error = %d\n", TW_OSL_ENOMEM);
return(TW_OSL_ENOMEM);
}
tw_osl_memzero(non_dma_mem, sizeof(struct tw_cli_ctlr_context) +
Apply driver update from LSI. Many thanks to LSI for continuing to support FreeBSD. 1) Timeout ioctl command timeouts. Do not reset the controller if ioctl command completed successfully. 2) Remove G66_WORKAROUND code (this bug never shipped). 3) Remove unnecessary interrupt lock (intr_lock). 4) Timeout firmware handshake for PChip reset (don't wait forever). 5) Handle interrupts inline. 6) Unmask command interrupt ONLY when adding a command to the pending queue. 7) Mask command interrupt ONLY after removing the last command from the pending queue. 8) Remove TW_OSLI_DEFERRED_INTR_USED code. 9) Replace controller "state" with separate data fields to avoid races: TW_CLI_CTLR_STATE_ACTIVE ctlr->active TW_CLI_CTLR_STATE_INTR_ENABLED ctlr->interrupts_enabled TW_CLI_CTLR_STATE_INTERNAL_REQ_BUSY ctlr->internal_req_busy TW_CLI_CTLR_STATE_GET_MORE_AENS ctlr->get_more_aens TW_CLI_CTLR_STATE_RESET_IN_PROGRESS ctlr->reset_in_progress TW_CLI_CTLR_STATE_RESET_PHASE1_IN_PROGRESS ctlr->reset_phase1_in_progress 10) Fix "req" leak in twa_action() when simq is frozen and req is NOT null. 11) Replace softc "state" with separate data fields to avoid races: TW_OSLI_CTLR_STATE_OPEN sc->open TW_OSLI_CTLR_STATE_SIMQ_FROZEN sc->simq_frozen 12) Fix reference to TW_OSLI_REQ_FLAGS_IN_PROGRESS in tw_osl_complete_passthru() 13) Use correct CAM status values. Change CAM_REQ_CMP_ERR to CAM_REQ_INVALID. Remove use of CAM_RELEASE_SIMQ for physical data addresses. 14) Do not freeze/ release the simq with non I/O commands. When it is appropriate to temporarily freeze the simq with an I/O command use: xpt_freeze_simq(sim, 1); ccb->ccb_h.status |= CAM_RELEASE_SIMQ; otherwise use: xpt_freeze_simq(sim, 1); xpt_release_simq(sim, 1); Submitted by: Tom Couch <tom.couch lsi.com> PR: kern/147695 MFC after: 3 days
2010-06-09 21:40:38 +00:00
(sizeof(struct tw_cli_req_context) * max_simult_reqs) +
(sizeof(struct tw_cl_event_packet) * max_aens));
tw_osl_memzero(dma_mem,
(sizeof(struct tw_cl_command_packet) *
Apply driver update from LSI. Many thanks to LSI for continuing to support FreeBSD. 1) Timeout ioctl command timeouts. Do not reset the controller if ioctl command completed successfully. 2) Remove G66_WORKAROUND code (this bug never shipped). 3) Remove unnecessary interrupt lock (intr_lock). 4) Timeout firmware handshake for PChip reset (don't wait forever). 5) Handle interrupts inline. 6) Unmask command interrupt ONLY when adding a command to the pending queue. 7) Mask command interrupt ONLY after removing the last command from the pending queue. 8) Remove TW_OSLI_DEFERRED_INTR_USED code. 9) Replace controller "state" with separate data fields to avoid races: TW_CLI_CTLR_STATE_ACTIVE ctlr->active TW_CLI_CTLR_STATE_INTR_ENABLED ctlr->interrupts_enabled TW_CLI_CTLR_STATE_INTERNAL_REQ_BUSY ctlr->internal_req_busy TW_CLI_CTLR_STATE_GET_MORE_AENS ctlr->get_more_aens TW_CLI_CTLR_STATE_RESET_IN_PROGRESS ctlr->reset_in_progress TW_CLI_CTLR_STATE_RESET_PHASE1_IN_PROGRESS ctlr->reset_phase1_in_progress 10) Fix "req" leak in twa_action() when simq is frozen and req is NOT null. 11) Replace softc "state" with separate data fields to avoid races: TW_OSLI_CTLR_STATE_OPEN sc->open TW_OSLI_CTLR_STATE_SIMQ_FROZEN sc->simq_frozen 12) Fix reference to TW_OSLI_REQ_FLAGS_IN_PROGRESS in tw_osl_complete_passthru() 13) Use correct CAM status values. Change CAM_REQ_CMP_ERR to CAM_REQ_INVALID. Remove use of CAM_RELEASE_SIMQ for physical data addresses. 14) Do not freeze/ release the simq with non I/O commands. When it is appropriate to temporarily freeze the simq with an I/O command use: xpt_freeze_simq(sim, 1); ccb->ccb_h.status |= CAM_RELEASE_SIMQ; otherwise use: xpt_freeze_simq(sim, 1); xpt_release_simq(sim, 1); Submitted by: Tom Couch <tom.couch lsi.com> PR: kern/147695 MFC after: 3 days
2010-06-09 21:40:38 +00:00
max_simult_reqs) +
TW_CLI_SECTOR_SIZE);
free_non_dma_mem = (TW_UINT8 *)non_dma_mem;
ctlr = (struct tw_cli_ctlr_context *)free_non_dma_mem;
free_non_dma_mem += sizeof(struct tw_cli_ctlr_context);
ctlr_handle->cl_ctlr_ctxt = ctlr;
ctlr->ctlr_handle = ctlr_handle;
ctlr->device_id = (TW_UINT32)device_id;
ctlr->arch_id = TWA_ARCH_ID(device_id);
ctlr->flags = flags;
ctlr->sg_size_factor = TWA_SG_ELEMENT_SIZE_FACTOR(device_id);
Apply driver update from LSI. Many thanks to LSI for continuing to support FreeBSD. 1) Timeout ioctl command timeouts. Do not reset the controller if ioctl command completed successfully. 2) Remove G66_WORKAROUND code (this bug never shipped). 3) Remove unnecessary interrupt lock (intr_lock). 4) Timeout firmware handshake for PChip reset (don't wait forever). 5) Handle interrupts inline. 6) Unmask command interrupt ONLY when adding a command to the pending queue. 7) Mask command interrupt ONLY after removing the last command from the pending queue. 8) Remove TW_OSLI_DEFERRED_INTR_USED code. 9) Replace controller "state" with separate data fields to avoid races: TW_CLI_CTLR_STATE_ACTIVE ctlr->active TW_CLI_CTLR_STATE_INTR_ENABLED ctlr->interrupts_enabled TW_CLI_CTLR_STATE_INTERNAL_REQ_BUSY ctlr->internal_req_busy TW_CLI_CTLR_STATE_GET_MORE_AENS ctlr->get_more_aens TW_CLI_CTLR_STATE_RESET_IN_PROGRESS ctlr->reset_in_progress TW_CLI_CTLR_STATE_RESET_PHASE1_IN_PROGRESS ctlr->reset_phase1_in_progress 10) Fix "req" leak in twa_action() when simq is frozen and req is NOT null. 11) Replace softc "state" with separate data fields to avoid races: TW_OSLI_CTLR_STATE_OPEN sc->open TW_OSLI_CTLR_STATE_SIMQ_FROZEN sc->simq_frozen 12) Fix reference to TW_OSLI_REQ_FLAGS_IN_PROGRESS in tw_osl_complete_passthru() 13) Use correct CAM status values. Change CAM_REQ_CMP_ERR to CAM_REQ_INVALID. Remove use of CAM_RELEASE_SIMQ for physical data addresses. 14) Do not freeze/ release the simq with non I/O commands. When it is appropriate to temporarily freeze the simq with an I/O command use: xpt_freeze_simq(sim, 1); ccb->ccb_h.status |= CAM_RELEASE_SIMQ; otherwise use: xpt_freeze_simq(sim, 1); xpt_release_simq(sim, 1); Submitted by: Tom Couch <tom.couch lsi.com> PR: kern/147695 MFC after: 3 days
2010-06-09 21:40:38 +00:00
ctlr->max_simult_reqs = max_simult_reqs;
ctlr->max_aens_supported = max_aens;
/* Initialize queues of CL internal request context packets. */
tw_cli_req_q_init(ctlr, TW_CLI_FREE_Q);
tw_cli_req_q_init(ctlr, TW_CLI_BUSY_Q);
tw_cli_req_q_init(ctlr, TW_CLI_PENDING_Q);
tw_cli_req_q_init(ctlr, TW_CLI_COMPLETE_Q);
/* Initialize all locks used by CL. */
ctlr->gen_lock = &(ctlr->gen_lock_handle);
tw_osl_init_lock(ctlr_handle, "tw_cl_gen_lock", ctlr->gen_lock);
ctlr->io_lock = &(ctlr->io_lock_handle);
tw_osl_init_lock(ctlr_handle, "tw_cl_io_lock", ctlr->io_lock);
/* Initialize CL internal request context packets. */
ctlr->req_ctxt_buf = (struct tw_cli_req_context *)free_non_dma_mem;
free_non_dma_mem += (sizeof(struct tw_cli_req_context) *
Apply driver update from LSI. Many thanks to LSI for continuing to support FreeBSD. 1) Timeout ioctl command timeouts. Do not reset the controller if ioctl command completed successfully. 2) Remove G66_WORKAROUND code (this bug never shipped). 3) Remove unnecessary interrupt lock (intr_lock). 4) Timeout firmware handshake for PChip reset (don't wait forever). 5) Handle interrupts inline. 6) Unmask command interrupt ONLY when adding a command to the pending queue. 7) Mask command interrupt ONLY after removing the last command from the pending queue. 8) Remove TW_OSLI_DEFERRED_INTR_USED code. 9) Replace controller "state" with separate data fields to avoid races: TW_CLI_CTLR_STATE_ACTIVE ctlr->active TW_CLI_CTLR_STATE_INTR_ENABLED ctlr->interrupts_enabled TW_CLI_CTLR_STATE_INTERNAL_REQ_BUSY ctlr->internal_req_busy TW_CLI_CTLR_STATE_GET_MORE_AENS ctlr->get_more_aens TW_CLI_CTLR_STATE_RESET_IN_PROGRESS ctlr->reset_in_progress TW_CLI_CTLR_STATE_RESET_PHASE1_IN_PROGRESS ctlr->reset_phase1_in_progress 10) Fix "req" leak in twa_action() when simq is frozen and req is NOT null. 11) Replace softc "state" with separate data fields to avoid races: TW_OSLI_CTLR_STATE_OPEN sc->open TW_OSLI_CTLR_STATE_SIMQ_FROZEN sc->simq_frozen 12) Fix reference to TW_OSLI_REQ_FLAGS_IN_PROGRESS in tw_osl_complete_passthru() 13) Use correct CAM status values. Change CAM_REQ_CMP_ERR to CAM_REQ_INVALID. Remove use of CAM_RELEASE_SIMQ for physical data addresses. 14) Do not freeze/ release the simq with non I/O commands. When it is appropriate to temporarily freeze the simq with an I/O command use: xpt_freeze_simq(sim, 1); ccb->ccb_h.status |= CAM_RELEASE_SIMQ; otherwise use: xpt_freeze_simq(sim, 1); xpt_release_simq(sim, 1); Submitted by: Tom Couch <tom.couch lsi.com> PR: kern/147695 MFC after: 3 days
2010-06-09 21:40:38 +00:00
max_simult_reqs);
ctlr->cmd_pkt_buf = (struct tw_cl_command_packet *)dma_mem;
ctlr->cmd_pkt_phys = dma_mem_phys;
ctlr->internal_req_data = (TW_UINT8 *)
(ctlr->cmd_pkt_buf +
Apply driver update from LSI. Many thanks to LSI for continuing to support FreeBSD. 1) Timeout ioctl command timeouts. Do not reset the controller if ioctl command completed successfully. 2) Remove G66_WORKAROUND code (this bug never shipped). 3) Remove unnecessary interrupt lock (intr_lock). 4) Timeout firmware handshake for PChip reset (don't wait forever). 5) Handle interrupts inline. 6) Unmask command interrupt ONLY when adding a command to the pending queue. 7) Mask command interrupt ONLY after removing the last command from the pending queue. 8) Remove TW_OSLI_DEFERRED_INTR_USED code. 9) Replace controller "state" with separate data fields to avoid races: TW_CLI_CTLR_STATE_ACTIVE ctlr->active TW_CLI_CTLR_STATE_INTR_ENABLED ctlr->interrupts_enabled TW_CLI_CTLR_STATE_INTERNAL_REQ_BUSY ctlr->internal_req_busy TW_CLI_CTLR_STATE_GET_MORE_AENS ctlr->get_more_aens TW_CLI_CTLR_STATE_RESET_IN_PROGRESS ctlr->reset_in_progress TW_CLI_CTLR_STATE_RESET_PHASE1_IN_PROGRESS ctlr->reset_phase1_in_progress 10) Fix "req" leak in twa_action() when simq is frozen and req is NOT null. 11) Replace softc "state" with separate data fields to avoid races: TW_OSLI_CTLR_STATE_OPEN sc->open TW_OSLI_CTLR_STATE_SIMQ_FROZEN sc->simq_frozen 12) Fix reference to TW_OSLI_REQ_FLAGS_IN_PROGRESS in tw_osl_complete_passthru() 13) Use correct CAM status values. Change CAM_REQ_CMP_ERR to CAM_REQ_INVALID. Remove use of CAM_RELEASE_SIMQ for physical data addresses. 14) Do not freeze/ release the simq with non I/O commands. When it is appropriate to temporarily freeze the simq with an I/O command use: xpt_freeze_simq(sim, 1); ccb->ccb_h.status |= CAM_RELEASE_SIMQ; otherwise use: xpt_freeze_simq(sim, 1); xpt_release_simq(sim, 1); Submitted by: Tom Couch <tom.couch lsi.com> PR: kern/147695 MFC after: 3 days
2010-06-09 21:40:38 +00:00
max_simult_reqs);
ctlr->internal_req_data_phys = ctlr->cmd_pkt_phys +
(sizeof(struct tw_cl_command_packet) *
Apply driver update from LSI. Many thanks to LSI for continuing to support FreeBSD. 1) Timeout ioctl command timeouts. Do not reset the controller if ioctl command completed successfully. 2) Remove G66_WORKAROUND code (this bug never shipped). 3) Remove unnecessary interrupt lock (intr_lock). 4) Timeout firmware handshake for PChip reset (don't wait forever). 5) Handle interrupts inline. 6) Unmask command interrupt ONLY when adding a command to the pending queue. 7) Mask command interrupt ONLY after removing the last command from the pending queue. 8) Remove TW_OSLI_DEFERRED_INTR_USED code. 9) Replace controller "state" with separate data fields to avoid races: TW_CLI_CTLR_STATE_ACTIVE ctlr->active TW_CLI_CTLR_STATE_INTR_ENABLED ctlr->interrupts_enabled TW_CLI_CTLR_STATE_INTERNAL_REQ_BUSY ctlr->internal_req_busy TW_CLI_CTLR_STATE_GET_MORE_AENS ctlr->get_more_aens TW_CLI_CTLR_STATE_RESET_IN_PROGRESS ctlr->reset_in_progress TW_CLI_CTLR_STATE_RESET_PHASE1_IN_PROGRESS ctlr->reset_phase1_in_progress 10) Fix "req" leak in twa_action() when simq is frozen and req is NOT null. 11) Replace softc "state" with separate data fields to avoid races: TW_OSLI_CTLR_STATE_OPEN sc->open TW_OSLI_CTLR_STATE_SIMQ_FROZEN sc->simq_frozen 12) Fix reference to TW_OSLI_REQ_FLAGS_IN_PROGRESS in tw_osl_complete_passthru() 13) Use correct CAM status values. Change CAM_REQ_CMP_ERR to CAM_REQ_INVALID. Remove use of CAM_RELEASE_SIMQ for physical data addresses. 14) Do not freeze/ release the simq with non I/O commands. When it is appropriate to temporarily freeze the simq with an I/O command use: xpt_freeze_simq(sim, 1); ccb->ccb_h.status |= CAM_RELEASE_SIMQ; otherwise use: xpt_freeze_simq(sim, 1); xpt_release_simq(sim, 1); Submitted by: Tom Couch <tom.couch lsi.com> PR: kern/147695 MFC after: 3 days
2010-06-09 21:40:38 +00:00
max_simult_reqs);
for (i = 0; i < max_simult_reqs; i++) {
req = &(ctlr->req_ctxt_buf[i]);
req->cmd_pkt = &(ctlr->cmd_pkt_buf[i]);
req->cmd_pkt_phys = ctlr->cmd_pkt_phys +
(i * sizeof(struct tw_cl_command_packet));
req->request_id = i;
req->ctlr = ctlr;
/* Insert request into the free queue. */
tw_cli_req_q_insert_tail(req, TW_CLI_FREE_Q);
}
/* Initialize the AEN queue. */
ctlr->aen_queue = (struct tw_cl_event_packet *)free_non_dma_mem;
start_ctlr:
/*
* Disable interrupts. Interrupts will be enabled in tw_cli_start_ctlr
* (only) if initialization succeeded.
*/
tw_cli_disable_interrupts(ctlr);
/* Initialize the controller. */
if ((error = tw_cli_start_ctlr(ctlr))) {
/* Soft reset the controller, and try one more time. */
tw_cl_create_event(ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1002, 0x1, TW_CL_SEVERITY_ERROR_STRING,
"Controller initialization failed. Retrying...",
"error = %d\n", error);
if ((error = tw_cli_soft_reset(ctlr))) {
tw_cl_create_event(ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1003, 0x1, TW_CL_SEVERITY_ERROR_STRING,
"Controller soft reset failed",
"error = %d\n", error);
return(error);
} else if ((error = tw_cli_start_ctlr(ctlr))) {
tw_cl_create_event(ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1004, 0x1, TW_CL_SEVERITY_ERROR_STRING,
"Controller initialization retry failed",
"error = %d\n", error);
return(error);
}
}
/* Notify some info about the controller to the OSL. */
tw_cli_notify_ctlr_info(ctlr);
Apply driver update from LSI. Many thanks to LSI for continuing to support FreeBSD. 1) Timeout ioctl command timeouts. Do not reset the controller if ioctl command completed successfully. 2) Remove G66_WORKAROUND code (this bug never shipped). 3) Remove unnecessary interrupt lock (intr_lock). 4) Timeout firmware handshake for PChip reset (don't wait forever). 5) Handle interrupts inline. 6) Unmask command interrupt ONLY when adding a command to the pending queue. 7) Mask command interrupt ONLY after removing the last command from the pending queue. 8) Remove TW_OSLI_DEFERRED_INTR_USED code. 9) Replace controller "state" with separate data fields to avoid races: TW_CLI_CTLR_STATE_ACTIVE ctlr->active TW_CLI_CTLR_STATE_INTR_ENABLED ctlr->interrupts_enabled TW_CLI_CTLR_STATE_INTERNAL_REQ_BUSY ctlr->internal_req_busy TW_CLI_CTLR_STATE_GET_MORE_AENS ctlr->get_more_aens TW_CLI_CTLR_STATE_RESET_IN_PROGRESS ctlr->reset_in_progress TW_CLI_CTLR_STATE_RESET_PHASE1_IN_PROGRESS ctlr->reset_phase1_in_progress 10) Fix "req" leak in twa_action() when simq is frozen and req is NOT null. 11) Replace softc "state" with separate data fields to avoid races: TW_OSLI_CTLR_STATE_OPEN sc->open TW_OSLI_CTLR_STATE_SIMQ_FROZEN sc->simq_frozen 12) Fix reference to TW_OSLI_REQ_FLAGS_IN_PROGRESS in tw_osl_complete_passthru() 13) Use correct CAM status values. Change CAM_REQ_CMP_ERR to CAM_REQ_INVALID. Remove use of CAM_RELEASE_SIMQ for physical data addresses. 14) Do not freeze/ release the simq with non I/O commands. When it is appropriate to temporarily freeze the simq with an I/O command use: xpt_freeze_simq(sim, 1); ccb->ccb_h.status |= CAM_RELEASE_SIMQ; otherwise use: xpt_freeze_simq(sim, 1); xpt_release_simq(sim, 1); Submitted by: Tom Couch <tom.couch lsi.com> PR: kern/147695 MFC after: 3 days
2010-06-09 21:40:38 +00:00
/* Mark the controller active. */
ctlr->active = TW_CL_TRUE;
return(error);
}
/*
* Function name: tw_cli_start_ctlr
* Description: Establishes a logical connection with the controller.
* Determines whether or not the driver is compatible
* with the firmware on the controller, before proceeding
* to work with it.
*
* Input: ctlr -- ptr to per ctlr structure
* Output: None
* Return value: 0 -- success
* non-zero-- failure
*/
TW_INT32
tw_cli_start_ctlr(struct tw_cli_ctlr_context *ctlr)
{
TW_UINT16 fw_on_ctlr_srl = 0;
TW_UINT16 fw_on_ctlr_arch_id = 0;
TW_UINT16 fw_on_ctlr_branch = 0;
TW_UINT16 fw_on_ctlr_build = 0;
TW_UINT32 init_connect_result = 0;
TW_INT32 error = TW_OSL_ESUCCESS;
tw_cli_dbg_printf(3, ctlr->ctlr_handle, tw_osl_cur_func(), "entered");
/* Wait for the controller to become ready. */
if ((error = tw_cli_poll_status(ctlr,
TWA_STATUS_MICROCONTROLLER_READY,
TW_CLI_REQUEST_TIMEOUT_PERIOD))) {
tw_cl_create_event(ctlr->ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1009, 0x1, TW_CL_SEVERITY_ERROR_STRING,
"Microcontroller not ready",
"error = %d", error);
return(error);
}
/* Drain the response queue. */
if ((error = tw_cli_drain_response_queue(ctlr))) {
tw_cl_create_event(ctlr->ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x100A, 0x1, TW_CL_SEVERITY_ERROR_STRING,
"Can't drain response queue",
"error = %d", error);
return(error);
}
/* Establish a logical connection with the controller. */
if ((error = tw_cli_init_connection(ctlr,
(TW_UINT16)(ctlr->max_simult_reqs),
TWA_EXTENDED_INIT_CONNECT, TWA_CURRENT_FW_SRL,
(TW_UINT16)(ctlr->arch_id),
TWA_CURRENT_FW_BRANCH(ctlr->arch_id),
TWA_CURRENT_FW_BUILD(ctlr->arch_id),
&fw_on_ctlr_srl, &fw_on_ctlr_arch_id,
&fw_on_ctlr_branch, &fw_on_ctlr_build,
&init_connect_result))) {
tw_cl_create_event(ctlr->ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x100B, 0x2, TW_CL_SEVERITY_WARNING_STRING,
"Can't initialize connection in current mode",
"error = %d", error);
return(error);
}
{
/* See if we can at least work with the firmware on the
* controller in the current mode.
*/
if (init_connect_result & TWA_CTLR_FW_COMPATIBLE) {
/* Yes, we can. Make note of the operating mode. */
if (init_connect_result & TWA_CTLR_FW_SAME_OR_NEWER) {
ctlr->working_srl = TWA_CURRENT_FW_SRL;
ctlr->working_branch =
TWA_CURRENT_FW_BRANCH(ctlr->arch_id);
ctlr->working_build =
TWA_CURRENT_FW_BUILD(ctlr->arch_id);
} else {
ctlr->working_srl = fw_on_ctlr_srl;
ctlr->working_branch = fw_on_ctlr_branch;
ctlr->working_build = fw_on_ctlr_build;
}
} else {
/*
* No, we can't. See if we can at least work with
* it in the base mode.
*/
tw_cl_create_event(ctlr->ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1010, 0x2, TW_CL_SEVERITY_WARNING_STRING,
"Driver/Firmware mismatch. "
"Negotiating for base level...",
" ");
if ((error = tw_cli_init_connection(ctlr,
(TW_UINT16)(ctlr->max_simult_reqs),
TWA_EXTENDED_INIT_CONNECT,
TWA_BASE_FW_SRL,
(TW_UINT16)(ctlr->arch_id),
TWA_BASE_FW_BRANCH, TWA_BASE_FW_BUILD,
&fw_on_ctlr_srl, &fw_on_ctlr_arch_id,
&fw_on_ctlr_branch, &fw_on_ctlr_build,
&init_connect_result))) {
tw_cl_create_event(ctlr->ctlr_handle,
TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1011, 0x1,
TW_CL_SEVERITY_ERROR_STRING,
"Can't initialize connection in "
"base mode",
" ");
return(error);
}
if (!(init_connect_result & TWA_CTLR_FW_COMPATIBLE)) {
/*
* The firmware on the controller is not even
* compatible with our base mode. We cannot
* work with it. Bail...
*/
return(1);
}
/*
* We can work with this firmware, but only in
* base mode.
*/
ctlr->working_srl = TWA_BASE_FW_SRL;
ctlr->working_branch = TWA_BASE_FW_BRANCH;
ctlr->working_build = TWA_BASE_FW_BUILD;
ctlr->operating_mode = TWA_BASE_MODE;
}
ctlr->fw_on_ctlr_srl = fw_on_ctlr_srl;
ctlr->fw_on_ctlr_branch = fw_on_ctlr_branch;
ctlr->fw_on_ctlr_build = fw_on_ctlr_build;
}
/* Drain the AEN queue */
if ((error = tw_cli_drain_aen_queue(ctlr)))
/*
* We will just print that we couldn't drain the AEN queue.
* There's no need to bail out.
*/
tw_cl_create_event(ctlr->ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1014, 0x2, TW_CL_SEVERITY_WARNING_STRING,
"Can't drain AEN queue",
"error = %d", error);
/* Enable interrupts. */
tw_cli_enable_interrupts(ctlr);
return(TW_OSL_ESUCCESS);
}
/*
* Function name: tw_cl_shutdown_ctlr
* Description: Closes logical connection with the controller.
*
* Input: ctlr -- ptr to per ctlr structure
* flags -- more info passed by the OS Layer
* Output: None
* Return value: 0 -- success
* non-zero-- failure
*/
TW_INT32
tw_cl_shutdown_ctlr(struct tw_cl_ctlr_handle *ctlr_handle, TW_UINT32 flags)
{
struct tw_cli_ctlr_context *ctlr =
(struct tw_cli_ctlr_context *)(ctlr_handle->cl_ctlr_ctxt);
TW_INT32 error;
tw_cli_dbg_printf(3, ctlr_handle, tw_osl_cur_func(), "entered");
/*
* Mark the controller as inactive, disable any further interrupts,
* and notify the controller that we are going down.
*/
Apply driver update from LSI. Many thanks to LSI for continuing to support FreeBSD. 1) Timeout ioctl command timeouts. Do not reset the controller if ioctl command completed successfully. 2) Remove G66_WORKAROUND code (this bug never shipped). 3) Remove unnecessary interrupt lock (intr_lock). 4) Timeout firmware handshake for PChip reset (don't wait forever). 5) Handle interrupts inline. 6) Unmask command interrupt ONLY when adding a command to the pending queue. 7) Mask command interrupt ONLY after removing the last command from the pending queue. 8) Remove TW_OSLI_DEFERRED_INTR_USED code. 9) Replace controller "state" with separate data fields to avoid races: TW_CLI_CTLR_STATE_ACTIVE ctlr->active TW_CLI_CTLR_STATE_INTR_ENABLED ctlr->interrupts_enabled TW_CLI_CTLR_STATE_INTERNAL_REQ_BUSY ctlr->internal_req_busy TW_CLI_CTLR_STATE_GET_MORE_AENS ctlr->get_more_aens TW_CLI_CTLR_STATE_RESET_IN_PROGRESS ctlr->reset_in_progress TW_CLI_CTLR_STATE_RESET_PHASE1_IN_PROGRESS ctlr->reset_phase1_in_progress 10) Fix "req" leak in twa_action() when simq is frozen and req is NOT null. 11) Replace softc "state" with separate data fields to avoid races: TW_OSLI_CTLR_STATE_OPEN sc->open TW_OSLI_CTLR_STATE_SIMQ_FROZEN sc->simq_frozen 12) Fix reference to TW_OSLI_REQ_FLAGS_IN_PROGRESS in tw_osl_complete_passthru() 13) Use correct CAM status values. Change CAM_REQ_CMP_ERR to CAM_REQ_INVALID. Remove use of CAM_RELEASE_SIMQ for physical data addresses. 14) Do not freeze/ release the simq with non I/O commands. When it is appropriate to temporarily freeze the simq with an I/O command use: xpt_freeze_simq(sim, 1); ccb->ccb_h.status |= CAM_RELEASE_SIMQ; otherwise use: xpt_freeze_simq(sim, 1); xpt_release_simq(sim, 1); Submitted by: Tom Couch <tom.couch lsi.com> PR: kern/147695 MFC after: 3 days
2010-06-09 21:40:38 +00:00
ctlr->active = TW_CL_FALSE;
tw_cli_disable_interrupts(ctlr);
/* Let the controller know that we are going down. */
if ((error = tw_cli_init_connection(ctlr, TWA_SHUTDOWN_MESSAGE_CREDITS,
0, 0, 0, 0, 0, TW_CL_NULL, TW_CL_NULL, TW_CL_NULL,
TW_CL_NULL, TW_CL_NULL)))
tw_cl_create_event(ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1015, 0x1, TW_CL_SEVERITY_ERROR_STRING,
"Can't close connection with controller",
"error = %d", error);
if (flags & TW_CL_STOP_CTLR_ONLY)
goto ret;
/* Destroy all locks used by CL. */
tw_osl_destroy_lock(ctlr_handle, ctlr->gen_lock);
tw_osl_destroy_lock(ctlr_handle, ctlr->io_lock);
ret:
return(error);
}
/*
* Function name: tw_cli_init_connection
* Description: Sends init_connection cmd to firmware
*
* Input: ctlr -- ptr to per ctlr structure
* message_credits -- max # of requests that we might send
* down simultaneously. This will be
* typically set to 256 at init-time or
* after a reset, and to 1 at shutdown-time
* set_features -- indicates if we intend to use 64-bit
* sg, also indicates if we want to do a
* basic or an extended init_connection;
*
* Note: The following input/output parameters are valid, only in case of an
* extended init_connection:
*
* current_fw_srl -- srl of fw we are bundled
* with, if any; 0 otherwise
* current_fw_arch_id -- arch_id of fw we are bundled
* with, if any; 0 otherwise
* current_fw_branch -- branch # of fw we are bundled
* with, if any; 0 otherwise
* current_fw_build -- build # of fw we are bundled
* with, if any; 0 otherwise
* Output: fw_on_ctlr_srl -- srl of fw on ctlr
* fw_on_ctlr_arch_id -- arch_id of fw on ctlr
* fw_on_ctlr_branch -- branch # of fw on ctlr
* fw_on_ctlr_build -- build # of fw on ctlr
* init_connect_result -- result bitmap of fw response
* Return value: 0 -- success
* non-zero-- failure
*/
TW_INT32
tw_cli_init_connection(struct tw_cli_ctlr_context *ctlr,
TW_UINT16 message_credits, TW_UINT32 set_features,
TW_UINT16 current_fw_srl, TW_UINT16 current_fw_arch_id,
TW_UINT16 current_fw_branch, TW_UINT16 current_fw_build,
TW_UINT16 *fw_on_ctlr_srl, TW_UINT16 *fw_on_ctlr_arch_id,
TW_UINT16 *fw_on_ctlr_branch, TW_UINT16 *fw_on_ctlr_build,
TW_UINT32 *init_connect_result)
{
struct tw_cli_req_context *req;
struct tw_cl_command_init_connect *init_connect;
TW_INT32 error = TW_OSL_EBUSY;
tw_cli_dbg_printf(3, ctlr->ctlr_handle, tw_osl_cur_func(), "entered");
/* Get a request packet. */
if ((req = tw_cli_get_request(ctlr
)) == TW_CL_NULL)
goto out;
req->flags |= TW_CLI_REQ_FLAGS_INTERNAL;
/* Build the cmd pkt. */
init_connect = &(req->cmd_pkt->command.cmd_pkt_7k.init_connect);
req->cmd_pkt->cmd_hdr.header_desc.size_header = 128;
init_connect->res1__opcode =
BUILD_RES__OPCODE(0, TWA_FW_CMD_INIT_CONNECTION);
init_connect->request_id =
(TW_UINT8)(TW_CL_SWAP16(req->request_id));
init_connect->message_credits = TW_CL_SWAP16(message_credits);
init_connect->features = TW_CL_SWAP32(set_features);
if (ctlr->flags & TW_CL_64BIT_ADDRESSES)
init_connect->features |= TW_CL_SWAP32(TWA_64BIT_SG_ADDRESSES);
if (set_features & TWA_EXTENDED_INIT_CONNECT) {
/*
* Fill in the extra fields needed for an extended
* init_connect.
*/
init_connect->size = 6;
init_connect->fw_srl = TW_CL_SWAP16(current_fw_srl);
init_connect->fw_arch_id = TW_CL_SWAP16(current_fw_arch_id);
init_connect->fw_branch = TW_CL_SWAP16(current_fw_branch);
init_connect->fw_build = TW_CL_SWAP16(current_fw_build);
} else
init_connect->size = 3;
/* Submit the command, and wait for it to complete. */
error = tw_cli_submit_and_poll_request(req,
TW_CLI_REQUEST_TIMEOUT_PERIOD);
if (error == TW_OSL_ETIMEDOUT)
/* Clean-up done by tw_cli_submit_and_poll_request. */
return(error);
if (error)
goto out;
if ((error = init_connect->status)) {
tw_cli_create_ctlr_event(ctlr,
TW_CL_MESSAGE_SOURCE_CONTROLLER_ERROR,
&(req->cmd_pkt->cmd_hdr));
goto out;
}
if (set_features & TWA_EXTENDED_INIT_CONNECT) {
*fw_on_ctlr_srl = TW_CL_SWAP16(init_connect->fw_srl);
*fw_on_ctlr_arch_id = TW_CL_SWAP16(init_connect->fw_arch_id);
*fw_on_ctlr_branch = TW_CL_SWAP16(init_connect->fw_branch);
*fw_on_ctlr_build = TW_CL_SWAP16(init_connect->fw_build);
*init_connect_result = TW_CL_SWAP32(init_connect->result);
}
tw_cli_req_q_insert_tail(req, TW_CLI_FREE_Q);
return(error);
out:
tw_cl_create_event(ctlr->ctlr_handle, TW_CL_FALSE,
TW_CL_MESSAGE_SOURCE_COMMON_LAYER_ERROR,
0x1016, 0x1, TW_CL_SEVERITY_ERROR_STRING,
"init_connection failed",
"error = %d", error);
if (req)
tw_cli_req_q_insert_tail(req, TW_CLI_FREE_Q);
return(error);
}