OpenCloudOS-Kernel/drivers/scsi/lpfc/lpfc_vport.c

843 lines
24 KiB
C
Raw Normal View History

/*******************************************************************
* This file is part of the Emulex Linux Device Driver for *
* Fibre Channel Host Bus Adapters. *
* Copyright (C) 2017-2020 Broadcom. All Rights Reserved. The term *
* Broadcom refers to Broadcom Inc. and/or its subsidiaries. *
* Copyright (C) 2004-2016 Emulex. All rights reserved. *
* EMULEX and SLI are trademarks of Emulex. *
* www.broadcom.com *
* Portions Copyright (C) 2004-2005 Christoph Hellwig *
* *
* This program is free software; you can redistribute it and/or *
* modify it under the terms of version 2 of the GNU General *
* Public License as published by the Free Software Foundation. *
* This program is distributed in the hope that it will be useful. *
* ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND *
* WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, *
* FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT, ARE *
* DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD *
* TO BE LEGALLY INVALID. See the GNU General Public License for *
* more details, a copy of which can be found in the file COPYING *
* included with this package. *
*******************************************************************/
#include <linux/blkdev.h>
#include <linux/delay.h>
#include <linux/dma-mapping.h>
#include <linux/idr.h>
#include <linux/interrupt.h>
#include <linux/kthread.h>
#include <linux/pci.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/sched/signal.h>
#include <scsi/scsi.h>
#include <scsi/scsi_device.h>
#include <scsi/scsi_host.h>
#include <scsi/scsi_transport_fc.h>
#include "lpfc_hw4.h"
#include "lpfc_hw.h"
#include "lpfc_sli.h"
#include "lpfc_sli4.h"
#include "lpfc_nl.h"
#include "lpfc_disc.h"
#include "lpfc_scsi.h"
#include "lpfc.h"
#include "lpfc_logmsg.h"
#include "lpfc_crtn.h"
#include "lpfc_version.h"
#include "lpfc_vport.h"
inline void lpfc_vport_set_state(struct lpfc_vport *vport,
enum fc_vport_state new_state)
{
struct fc_vport *fc_vport = vport->fc_vport;
if (fc_vport) {
/*
* When the transport defines fc_vport_set state we will replace
* this code with the following line
*/
/* fc_vport_set_state(fc_vport, new_state); */
if (new_state != FC_VPORT_INITIALIZING)
fc_vport->vport_last_state = fc_vport->vport_state;
fc_vport->vport_state = new_state;
}
/* for all the error states we will set the invternal state to FAILED */
switch (new_state) {
case FC_VPORT_NO_FABRIC_SUPP:
case FC_VPORT_NO_FABRIC_RSCS:
case FC_VPORT_FABRIC_LOGOUT:
case FC_VPORT_FABRIC_REJ_WWN:
case FC_VPORT_FAILED:
vport->port_state = LPFC_VPORT_FAILED;
break;
case FC_VPORT_LINKDOWN:
vport->port_state = LPFC_VPORT_UNKNOWN;
break;
default:
/* do nothing */
break;
}
}
int
lpfc_alloc_vpi(struct lpfc_hba *phba)
{
unsigned long vpi;
spin_lock_irq(&phba->hbalock);
/* Start at bit 1 because vpi zero is reserved for the physical port */
vpi = find_next_zero_bit(phba->vpi_bmask, (phba->max_vpi + 1), 1);
if (vpi > phba->max_vpi)
vpi = 0;
else
set_bit(vpi, phba->vpi_bmask);
if (phba->sli_rev == LPFC_SLI_REV4)
phba->sli4_hba.max_cfg_param.vpi_used++;
spin_unlock_irq(&phba->hbalock);
return vpi;
}
static void
lpfc_free_vpi(struct lpfc_hba *phba, int vpi)
{
if (vpi == 0)
return;
spin_lock_irq(&phba->hbalock);
clear_bit(vpi, phba->vpi_bmask);
if (phba->sli_rev == LPFC_SLI_REV4)
phba->sli4_hba.max_cfg_param.vpi_used--;
spin_unlock_irq(&phba->hbalock);
}
static int
lpfc_vport_sparm(struct lpfc_hba *phba, struct lpfc_vport *vport)
{
LPFC_MBOXQ_t *pmb;
MAILBOX_t *mb;
struct lpfc_dmabuf *mp;
int rc;
pmb = mempool_alloc(phba->mbox_mem_pool, GFP_KERNEL);
if (!pmb) {
return -ENOMEM;
}
mb = &pmb->u.mb;
rc = lpfc_read_sparam(phba, pmb, vport->vpi);
if (rc) {
mempool_free(pmb, phba->mbox_mem_pool);
return -ENOMEM;
}
/*
* Grab buffer pointer and clear context1 so we can use
* lpfc_sli_issue_box_wait
*/
mp = (struct lpfc_dmabuf *)pmb->ctx_buf;
pmb->ctx_buf = NULL;
pmb->vport = vport;
rc = lpfc_sli_issue_mbox_wait(phba, pmb, phba->fc_ratov * 2);
if (rc != MBX_SUCCESS) {
if (signal_pending(current)) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1830 Signal aborted mbxCmd x%x\n",
mb->mbxCommand);
lpfc_mbuf_free(phba, mp->virt, mp->phys);
kfree(mp);
if (rc != MBX_TIMEOUT)
mempool_free(pmb, phba->mbox_mem_pool);
return -EINTR;
} else {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1818 VPort failed init, mbxCmd x%x "
"READ_SPARM mbxStatus x%x, rc = x%x\n",
mb->mbxCommand, mb->mbxStatus, rc);
lpfc_mbuf_free(phba, mp->virt, mp->phys);
kfree(mp);
if (rc != MBX_TIMEOUT)
mempool_free(pmb, phba->mbox_mem_pool);
return -EIO;
}
}
memcpy(&vport->fc_sparam, mp->virt, sizeof (struct serv_parm));
memcpy(&vport->fc_nodename, &vport->fc_sparam.nodeName,
sizeof (struct lpfc_name));
memcpy(&vport->fc_portname, &vport->fc_sparam.portName,
sizeof (struct lpfc_name));
lpfc_mbuf_free(phba, mp->virt, mp->phys);
kfree(mp);
mempool_free(pmb, phba->mbox_mem_pool);
return 0;
}
static int
lpfc_valid_wwn_format(struct lpfc_hba *phba, struct lpfc_name *wwn,
const char *name_type)
{
/* ensure that IEEE format 1 addresses
* contain zeros in bits 59-48
*/
if (!((wwn->u.wwn[0] >> 4) == 1 &&
((wwn->u.wwn[0] & 0xf) != 0 || (wwn->u.wwn[1] & 0xf) != 0)))
return 1;
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_log(phba, KERN_ERR, LOG_TRACE_EVENT,
"1822 Invalid %s: %02x:%02x:%02x:%02x:"
"%02x:%02x:%02x:%02x\n",
name_type,
wwn->u.wwn[0], wwn->u.wwn[1],
wwn->u.wwn[2], wwn->u.wwn[3],
wwn->u.wwn[4], wwn->u.wwn[5],
wwn->u.wwn[6], wwn->u.wwn[7]);
return 0;
}
static int
lpfc_unique_wwpn(struct lpfc_hba *phba, struct lpfc_vport *new_vport)
{
struct lpfc_vport *vport;
unsigned long flags;
spin_lock_irqsave(&phba->port_list_lock, flags);
list_for_each_entry(vport, &phba->port_list, listentry) {
if (vport == new_vport)
continue;
/* If they match, return not unique */
if (memcmp(&vport->fc_sparam.portName,
&new_vport->fc_sparam.portName,
sizeof(struct lpfc_name)) == 0) {
spin_unlock_irqrestore(&phba->port_list_lock, flags);
return 0;
}
}
spin_unlock_irqrestore(&phba->port_list_lock, flags);
return 1;
}
/**
* lpfc_discovery_wait - Wait for driver discovery to quiesce
* @vport: The virtual port for which this call is being executed.
*
* This driver calls this routine specifically from lpfc_vport_delete
* to enforce a synchronous execution of vport
* delete relative to discovery activities. The
* lpfc_vport_delete routine should not return until it
* can reasonably guarantee that discovery has quiesced.
* Post FDISC LOGO, the driver must wait until its SAN teardown is
* complete and all resources recovered before allowing
* cleanup.
*
* This routine does not require any locks held.
**/
static void lpfc_discovery_wait(struct lpfc_vport *vport)
{
struct lpfc_hba *phba = vport->phba;
uint32_t wait_flags = 0;
unsigned long wait_time_max;
unsigned long start_time;
wait_flags = FC_RSCN_MODE | FC_RSCN_DISCOVERY | FC_NLP_MORE |
FC_RSCN_DEFERRED | FC_NDISC_ACTIVE | FC_DISC_TMO;
/*
* The time constraint on this loop is a balance between the
* fabric RA_TOV value and dev_loss tmo. The driver's
* devloss_tmo is 10 giving this loop a 3x multiplier minimally.
*/
wait_time_max = msecs_to_jiffies(((phba->fc_ratov * 3) + 3) * 1000);
wait_time_max += jiffies;
start_time = jiffies;
while (time_before(jiffies, wait_time_max)) {
if ((vport->num_disc_nodes > 0) ||
(vport->fc_flag & wait_flags) ||
((vport->port_state > LPFC_VPORT_FAILED) &&
(vport->port_state < LPFC_VPORT_READY))) {
lpfc_printf_vlog(vport, KERN_INFO, LOG_VPORT,
"1833 Vport discovery quiesce Wait:"
" state x%x fc_flags x%x"
" num_nodes x%x, waiting 1000 msecs"
" total wait msecs x%x\n",
vport->port_state, vport->fc_flag,
vport->num_disc_nodes,
jiffies_to_msecs(jiffies - start_time));
msleep(1000);
} else {
/* Base case. Wait variants satisfied. Break out */
lpfc_printf_vlog(vport, KERN_INFO, LOG_VPORT,
"1834 Vport discovery quiesced:"
" state x%x fc_flags x%x"
" wait msecs x%x\n",
vport->port_state, vport->fc_flag,
jiffies_to_msecs(jiffies
- start_time));
break;
}
}
if (time_after(jiffies, wait_time_max))
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1835 Vport discovery quiesce failed:"
" state x%x fc_flags x%x wait msecs x%x\n",
vport->port_state, vport->fc_flag,
jiffies_to_msecs(jiffies - start_time));
}
int
lpfc_vport_create(struct fc_vport *fc_vport, bool disable)
{
struct lpfc_nodelist *ndlp;
struct Scsi_Host *shost = fc_vport->shost;
struct lpfc_vport *pport = (struct lpfc_vport *) shost->hostdata;
struct lpfc_hba *phba = pport->phba;
struct lpfc_vport *vport = NULL;
int instance;
int vpi;
int rc = VPORT_ERROR;
int status;
if ((phba->sli_rev < 3) || !(phba->cfg_enable_npiv)) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_log(phba, KERN_ERR, LOG_TRACE_EVENT,
"1808 Create VPORT failed: "
"NPIV is not enabled: SLImode:%d\n",
phba->sli_rev);
rc = VPORT_INVAL;
goto error_out;
}
/* NPIV is not supported if HBA has NVME Target enabled */
if (phba->nvmet_support) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_log(phba, KERN_ERR, LOG_TRACE_EVENT,
"3189 Create VPORT failed: "
"NPIV is not supported on NVME Target\n");
rc = VPORT_INVAL;
goto error_out;
}
vpi = lpfc_alloc_vpi(phba);
if (vpi == 0) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_log(phba, KERN_ERR, LOG_TRACE_EVENT,
"1809 Create VPORT failed: "
"Max VPORTs (%d) exceeded\n",
phba->max_vpi);
rc = VPORT_NORESOURCES;
goto error_out;
}
/* Assign an unused board number */
if ((instance = lpfc_get_instance()) < 0) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_log(phba, KERN_ERR, LOG_TRACE_EVENT,
"1810 Create VPORT failed: Cannot get "
"instance number\n");
lpfc_free_vpi(phba, vpi);
rc = VPORT_NORESOURCES;
goto error_out;
}
vport = lpfc_create_port(phba, instance, &fc_vport->dev);
if (!vport) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_log(phba, KERN_ERR, LOG_TRACE_EVENT,
"1811 Create VPORT failed: vpi x%x\n", vpi);
lpfc_free_vpi(phba, vpi);
rc = VPORT_NORESOURCES;
goto error_out;
}
vport->vpi = vpi;
lpfc_debugfs_initialize(vport);
if ((status = lpfc_vport_sparm(phba, vport))) {
if (status == -EINTR) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1831 Create VPORT Interrupted.\n");
rc = VPORT_ERROR;
} else {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1813 Create VPORT failed. "
"Cannot get sparam\n");
rc = VPORT_NORESOURCES;
}
lpfc_free_vpi(phba, vpi);
destroy_port(vport);
goto error_out;
}
u64_to_wwn(fc_vport->node_name, vport->fc_nodename.u.wwn);
u64_to_wwn(fc_vport->port_name, vport->fc_portname.u.wwn);
memcpy(&vport->fc_sparam.portName, vport->fc_portname.u.wwn, 8);
memcpy(&vport->fc_sparam.nodeName, vport->fc_nodename.u.wwn, 8);
if (!lpfc_valid_wwn_format(phba, &vport->fc_sparam.nodeName, "WWNN") ||
!lpfc_valid_wwn_format(phba, &vport->fc_sparam.portName, "WWPN")) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1821 Create VPORT failed. "
"Invalid WWN format\n");
lpfc_free_vpi(phba, vpi);
destroy_port(vport);
rc = VPORT_INVAL;
goto error_out;
}
if (!lpfc_unique_wwpn(phba, vport)) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1823 Create VPORT failed. "
"Duplicate WWN on HBA\n");
lpfc_free_vpi(phba, vpi);
destroy_port(vport);
rc = VPORT_INVAL;
goto error_out;
}
/* Create binary sysfs attribute for vport */
lpfc_alloc_sysfs_attr(vport);
/* Set the DFT_LUN_Q_DEPTH accordingly */
vport->cfg_lun_queue_depth = phba->pport->cfg_lun_queue_depth;
/* Only the physical port can support NVME for now */
vport->cfg_enable_fc4_type = LPFC_ENABLE_FCP;
*(struct lpfc_vport **)fc_vport->dd_data = vport;
vport->fc_vport = fc_vport;
/* At this point we are fully registered with SCSI Layer. */
vport->load_flag |= FC_ALLOW_FDMI;
if (phba->cfg_enable_SmartSAN ||
(phba->cfg_fdmi_on == LPFC_FDMI_SUPPORT)) {
/* Setup appropriate attribute masks */
vport->fdmi_hba_mask = phba->pport->fdmi_hba_mask;
vport->fdmi_port_mask = phba->pport->fdmi_port_mask;
}
/*
* In SLI4, the vpi must be activated before it can be used
* by the port.
*/
if ((phba->sli_rev == LPFC_SLI_REV4) &&
2010-11-21 12:11:48 +08:00
(pport->fc_flag & FC_VFI_REGISTERED)) {
rc = lpfc_sli4_init_vpi(vport);
if (rc) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_log(phba, KERN_ERR, LOG_TRACE_EVENT,
"1838 Failed to INIT_VPI on vpi %d "
"status %d\n", vpi, rc);
rc = VPORT_NORESOURCES;
lpfc_free_vpi(phba, vpi);
goto error_out;
}
} else if (phba->sli_rev == LPFC_SLI_REV4) {
/*
* Driver cannot INIT_VPI now. Set the flags to
* init_vpi when reg_vfi complete.
*/
vport->fc_flag |= FC_VPORT_NEEDS_INIT_VPI;
lpfc_vport_set_state(vport, FC_VPORT_LINKDOWN);
rc = VPORT_OK;
goto out;
}
if ((phba->link_state < LPFC_LINK_UP) ||
(pport->port_state < LPFC_FABRIC_CFG_LINK) ||
2010-11-21 12:11:48 +08:00
(phba->fc_topology == LPFC_TOPOLOGY_LOOP)) {
lpfc_vport_set_state(vport, FC_VPORT_LINKDOWN);
rc = VPORT_OK;
goto out;
}
if (disable) {
lpfc_vport_set_state(vport, FC_VPORT_DISABLED);
rc = VPORT_OK;
goto out;
}
/* Use the Physical nodes Fabric NDLP to determine if the link is
* up and ready to FDISC.
*/
ndlp = lpfc_findnode_did(phba->pport, Fabric_DID);
scsi: lpfc: Rework remote port ref counting and node freeing When a remote port is disconnected and disappears, its node structure (ndlp) stays allocated and on a vport node list. While on the list it can be matched, thus requires validation checks on state to be added in numerous code paths. If the node comes back, its possible for there to be multiple node structures for the same device on the vport node list. There is no reason to keep the node structure around after it is no longer in existence, and the current implementation creates problems for itself (multiple nodes) and lots of unnecessary code for state validation. Additionally, the reference taking on the node structure didn't follow the normal model used by the kernel kref api. It included lots of odd logic to match state with reference count. The combination of this odd logic plus the way it was implicitly used in the discovery engine made its reference taking implementation suspect and extremely hard to follow. Change the driver such that the reference taking routines are now normal ref increments/decrements and callout on refcount=0. With this in place, the rework can be done such that the node structure is fully removed and deallocated when the remote port no longer exists and all references are removed. This removal logic, and the basic ref counting are intrically tied, thus in a single patch. Link: https://lore.kernel.org/r/20201115192646.12977-2-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:30 +08:00
if (ndlp &&
ndlp->nlp_state == NLP_STE_UNMAPPED_NODE) {
if (phba->link_flag & LS_NPIV_FAB_SUPPORTED) {
lpfc_set_disctmo(vport);
lpfc_initial_fdisc(vport);
} else {
lpfc_vport_set_state(vport, FC_VPORT_NO_FABRIC_SUPP);
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"0262 No NPIV Fabric support\n");
}
} else {
lpfc_vport_set_state(vport, FC_VPORT_FAILED);
}
rc = VPORT_OK;
out:
lpfc_printf_vlog(vport, KERN_ERR, LOG_VPORT,
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
"1825 Vport Created.\n");
lpfc_host_attrib_init(lpfc_shost_from_vport(vport));
error_out:
return rc;
}
static int
disable_vport(struct fc_vport *fc_vport)
{
struct lpfc_vport *vport = *(struct lpfc_vport **)fc_vport->dd_data;
struct lpfc_hba *phba = vport->phba;
struct lpfc_nodelist *ndlp = NULL, *next_ndlp = NULL;
long timeout;
struct Scsi_Host *shost = lpfc_shost_from_vport(vport);
ndlp = lpfc_findnode_did(vport, Fabric_DID);
scsi: lpfc: Rework remote port ref counting and node freeing When a remote port is disconnected and disappears, its node structure (ndlp) stays allocated and on a vport node list. While on the list it can be matched, thus requires validation checks on state to be added in numerous code paths. If the node comes back, its possible for there to be multiple node structures for the same device on the vport node list. There is no reason to keep the node structure around after it is no longer in existence, and the current implementation creates problems for itself (multiple nodes) and lots of unnecessary code for state validation. Additionally, the reference taking on the node structure didn't follow the normal model used by the kernel kref api. It included lots of odd logic to match state with reference count. The combination of this odd logic plus the way it was implicitly used in the discovery engine made its reference taking implementation suspect and extremely hard to follow. Change the driver such that the reference taking routines are now normal ref increments/decrements and callout on refcount=0. With this in place, the rework can be done such that the node structure is fully removed and deallocated when the remote port no longer exists and all references are removed. This removal logic, and the basic ref counting are intrically tied, thus in a single patch. Link: https://lore.kernel.org/r/20201115192646.12977-2-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:30 +08:00
if (ndlp && phba->link_state >= LPFC_LINK_UP) {
vport->unreg_vpi_cmpl = VPORT_INVAL;
timeout = msecs_to_jiffies(phba->fc_ratov * 2000);
if (!lpfc_issue_els_npiv_logo(vport, ndlp))
while (vport->unreg_vpi_cmpl == VPORT_INVAL && timeout)
timeout = schedule_timeout(timeout);
}
lpfc_sli_host_down(vport);
/* Mark all nodes for discovery so we can remove them by
* calling lpfc_cleanup_rpis(vport, 1)
*/
list_for_each_entry_safe(ndlp, next_ndlp, &vport->fc_nodes, nlp_listp) {
if (ndlp->nlp_state == NLP_STE_UNUSED_NODE)
continue;
lpfc_disc_state_machine(vport, ndlp, NULL,
NLP_EVT_DEVICE_RECOVERY);
}
lpfc_cleanup_rpis(vport, 1);
lpfc_stop_vport_timers(vport);
lpfc_unreg_all_rpis(vport);
lpfc_unreg_default_rpis(vport);
/*
* Completion of unreg_vpi (lpfc_mbx_cmpl_unreg_vpi) does the
* scsi_host_put() to release the vport.
*/
lpfc_mbx_unreg_vpi(vport);
if (phba->sli_rev == LPFC_SLI_REV4) {
spin_lock_irq(shost->host_lock);
vport->fc_flag |= FC_VPORT_NEEDS_INIT_VPI;
spin_unlock_irq(shost->host_lock);
}
lpfc_vport_set_state(vport, FC_VPORT_DISABLED);
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1826 Vport Disabled.\n");
return VPORT_OK;
}
static int
enable_vport(struct fc_vport *fc_vport)
{
struct lpfc_vport *vport = *(struct lpfc_vport **)fc_vport->dd_data;
struct lpfc_hba *phba = vport->phba;
struct lpfc_nodelist *ndlp = NULL;
struct Scsi_Host *shost = lpfc_shost_from_vport(vport);
if ((phba->link_state < LPFC_LINK_UP) ||
2010-11-21 12:11:48 +08:00
(phba->fc_topology == LPFC_TOPOLOGY_LOOP)) {
lpfc_vport_set_state(vport, FC_VPORT_LINKDOWN);
return VPORT_OK;
}
spin_lock_irq(shost->host_lock);
vport->load_flag |= FC_LOADING;
if (vport->fc_flag & FC_VPORT_NEEDS_INIT_VPI) {
spin_unlock_irq(shost->host_lock);
lpfc_issue_init_vpi(vport);
goto out;
}
vport->fc_flag |= FC_VPORT_NEEDS_REG_VPI;
spin_unlock_irq(shost->host_lock);
/* Use the Physical nodes Fabric NDLP to determine if the link is
* up and ready to FDISC.
*/
ndlp = lpfc_findnode_did(phba->pport, Fabric_DID);
scsi: lpfc: Rework remote port ref counting and node freeing When a remote port is disconnected and disappears, its node structure (ndlp) stays allocated and on a vport node list. While on the list it can be matched, thus requires validation checks on state to be added in numerous code paths. If the node comes back, its possible for there to be multiple node structures for the same device on the vport node list. There is no reason to keep the node structure around after it is no longer in existence, and the current implementation creates problems for itself (multiple nodes) and lots of unnecessary code for state validation. Additionally, the reference taking on the node structure didn't follow the normal model used by the kernel kref api. It included lots of odd logic to match state with reference count. The combination of this odd logic plus the way it was implicitly used in the discovery engine made its reference taking implementation suspect and extremely hard to follow. Change the driver such that the reference taking routines are now normal ref increments/decrements and callout on refcount=0. With this in place, the rework can be done such that the node structure is fully removed and deallocated when the remote port no longer exists and all references are removed. This removal logic, and the basic ref counting are intrically tied, thus in a single patch. Link: https://lore.kernel.org/r/20201115192646.12977-2-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:30 +08:00
if (ndlp && ndlp->nlp_state == NLP_STE_UNMAPPED_NODE) {
if (phba->link_flag & LS_NPIV_FAB_SUPPORTED) {
lpfc_set_disctmo(vport);
lpfc_initial_fdisc(vport);
} else {
lpfc_vport_set_state(vport, FC_VPORT_NO_FABRIC_SUPP);
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"0264 No NPIV Fabric support\n");
}
} else {
lpfc_vport_set_state(vport, FC_VPORT_FAILED);
}
out:
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1827 Vport Enabled.\n");
return VPORT_OK;
}
int
lpfc_vport_disable(struct fc_vport *fc_vport, bool disable)
{
if (disable)
return disable_vport(fc_vport);
else
return enable_vport(fc_vport);
}
int
lpfc_vport_delete(struct fc_vport *fc_vport)
{
struct lpfc_nodelist *ndlp = NULL;
struct lpfc_vport *vport = *(struct lpfc_vport **)fc_vport->dd_data;
struct Scsi_Host *shost = lpfc_shost_from_vport(vport);
scsi: lpfc: Fix refcounting around SCSI and NVMe transport APIs Due to bug history and code review, the node reference counting approach in the driver isn't implemented consistently with how the scsi and nvme transport perform registrations and unregistrations and their callbacks. This resulted in many bad/stale node pointers. Reword the driver so that reference handling is performed as follows: - The initial node reference is taken on structure allocation - Take a reference on any add/register call to the transport - Remove a reference on any delete/unregister call to the transport - After the node has fully removed from both the SCSI and NVMEe transports (dev_loss_callbacks have called back) call the discovery engine DEVICE_RM event which will remove the final reference and release the node structure. - Alter dev_loss handling when a vport or base port is unloading. - Remove the put_node handling - no longer needed. - Rewrite the vport_delete handling on reference counts. Part of this effort was driven from the FDISC not registering with the transport and disrupting the model for node reference counting. - Deleted lpfc_nlp_remove. Pushed it's remaining ops into lpfc_nlp_release. - Several other small code cleanups. Link: https://lore.kernel.org/r/20201115192646.12977-5-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:33 +08:00
struct lpfc_hba *phba = vport->phba;
long timeout;
if (vport->port_type == LPFC_PHYSICAL_PORT) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1812 vport_delete failed: Cannot delete "
"physical host\n");
return VPORT_ERROR;
}
/* If the vport is a static vport fail the deletion. */
if ((vport->vport_flag & STATIC_VPORT) &&
!(phba->pport->load_flag & FC_UNLOADING)) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1837 vport_delete failed: Cannot delete "
"static vport.\n");
return VPORT_ERROR;
}
scsi: lpfc: Fix refcounting around SCSI and NVMe transport APIs Due to bug history and code review, the node reference counting approach in the driver isn't implemented consistently with how the scsi and nvme transport perform registrations and unregistrations and their callbacks. This resulted in many bad/stale node pointers. Reword the driver so that reference handling is performed as follows: - The initial node reference is taken on structure allocation - Take a reference on any add/register call to the transport - Remove a reference on any delete/unregister call to the transport - After the node has fully removed from both the SCSI and NVMEe transports (dev_loss_callbacks have called back) call the discovery engine DEVICE_RM event which will remove the final reference and release the node structure. - Alter dev_loss handling when a vport or base port is unloading. - Remove the put_node handling - no longer needed. - Rewrite the vport_delete handling on reference counts. Part of this effort was driven from the FDISC not registering with the transport and disrupting the model for node reference counting. - Deleted lpfc_nlp_remove. Pushed it's remaining ops into lpfc_nlp_release. - Several other small code cleanups. Link: https://lore.kernel.org/r/20201115192646.12977-5-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:33 +08:00
spin_lock_irq(&phba->hbalock);
vport->load_flag |= FC_UNLOADING;
spin_unlock_irq(&phba->hbalock);
scsi: lpfc: Fix refcounting around SCSI and NVMe transport APIs Due to bug history and code review, the node reference counting approach in the driver isn't implemented consistently with how the scsi and nvme transport perform registrations and unregistrations and their callbacks. This resulted in many bad/stale node pointers. Reword the driver so that reference handling is performed as follows: - The initial node reference is taken on structure allocation - Take a reference on any add/register call to the transport - Remove a reference on any delete/unregister call to the transport - After the node has fully removed from both the SCSI and NVMEe transports (dev_loss_callbacks have called back) call the discovery engine DEVICE_RM event which will remove the final reference and release the node structure. - Alter dev_loss handling when a vport or base port is unloading. - Remove the put_node handling - no longer needed. - Rewrite the vport_delete handling on reference counts. Part of this effort was driven from the FDISC not registering with the transport and disrupting the model for node reference counting. - Deleted lpfc_nlp_remove. Pushed it's remaining ops into lpfc_nlp_release. - Several other small code cleanups. Link: https://lore.kernel.org/r/20201115192646.12977-5-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:33 +08:00
/*
* If we are not unloading the driver then prevent the vport_delete
* from happening until after this vport's discovery is finished.
*/
if (!(phba->pport->load_flag & FC_UNLOADING)) {
int check_count = 0;
while (check_count < ((phba->fc_ratov * 3) + 3) &&
vport->port_state > LPFC_VPORT_FAILED &&
vport->port_state < LPFC_VPORT_READY) {
check_count++;
msleep(1000);
}
if (vport->port_state > LPFC_VPORT_FAILED &&
vport->port_state < LPFC_VPORT_READY)
return -EAGAIN;
}
/*
* Take early refcount for outstanding I/O requests we schedule during
* delete processing for unreg_vpi. Always keep this before
* scsi_remove_host() as we can no longer obtain a reference through
* scsi_host_get() after scsi_host_remove as shost is set to SHOST_DEL.
*/
if (!scsi_host_get(shost))
return VPORT_INVAL;
lpfc_free_sysfs_attr(vport);
lpfc_debugfs_terminate(vport);
/* Remove FC host to break driver binding. */
fc_remove_host(shost);
scsi: lpfc: Fix refcounting around SCSI and NVMe transport APIs Due to bug history and code review, the node reference counting approach in the driver isn't implemented consistently with how the scsi and nvme transport perform registrations and unregistrations and their callbacks. This resulted in many bad/stale node pointers. Reword the driver so that reference handling is performed as follows: - The initial node reference is taken on structure allocation - Take a reference on any add/register call to the transport - Remove a reference on any delete/unregister call to the transport - After the node has fully removed from both the SCSI and NVMEe transports (dev_loss_callbacks have called back) call the discovery engine DEVICE_RM event which will remove the final reference and release the node structure. - Alter dev_loss handling when a vport or base port is unloading. - Remove the put_node handling - no longer needed. - Rewrite the vport_delete handling on reference counts. Part of this effort was driven from the FDISC not registering with the transport and disrupting the model for node reference counting. - Deleted lpfc_nlp_remove. Pushed it's remaining ops into lpfc_nlp_release. - Several other small code cleanups. Link: https://lore.kernel.org/r/20201115192646.12977-5-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:33 +08:00
scsi_remove_host(shost);
scsi: lpfc: Fix refcounting around SCSI and NVMe transport APIs Due to bug history and code review, the node reference counting approach in the driver isn't implemented consistently with how the scsi and nvme transport perform registrations and unregistrations and their callbacks. This resulted in many bad/stale node pointers. Reword the driver so that reference handling is performed as follows: - The initial node reference is taken on structure allocation - Take a reference on any add/register call to the transport - Remove a reference on any delete/unregister call to the transport - After the node has fully removed from both the SCSI and NVMEe transports (dev_loss_callbacks have called back) call the discovery engine DEVICE_RM event which will remove the final reference and release the node structure. - Alter dev_loss handling when a vport or base port is unloading. - Remove the put_node handling - no longer needed. - Rewrite the vport_delete handling on reference counts. Part of this effort was driven from the FDISC not registering with the transport and disrupting the model for node reference counting. - Deleted lpfc_nlp_remove. Pushed it's remaining ops into lpfc_nlp_release. - Several other small code cleanups. Link: https://lore.kernel.org/r/20201115192646.12977-5-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:33 +08:00
/* Send the DA_ID and Fabric LOGO to cleanup Nameserver entries. */
ndlp = lpfc_findnode_did(vport, Fabric_DID);
if (!ndlp)
goto skip_logo;
scsi: lpfc: Rework remote port ref counting and node freeing When a remote port is disconnected and disappears, its node structure (ndlp) stays allocated and on a vport node list. While on the list it can be matched, thus requires validation checks on state to be added in numerous code paths. If the node comes back, its possible for there to be multiple node structures for the same device on the vport node list. There is no reason to keep the node structure around after it is no longer in existence, and the current implementation creates problems for itself (multiple nodes) and lots of unnecessary code for state validation. Additionally, the reference taking on the node structure didn't follow the normal model used by the kernel kref api. It included lots of odd logic to match state with reference count. The combination of this odd logic plus the way it was implicitly used in the discovery engine made its reference taking implementation suspect and extremely hard to follow. Change the driver such that the reference taking routines are now normal ref increments/decrements and callout on refcount=0. With this in place, the rework can be done such that the node structure is fully removed and deallocated when the remote port no longer exists and all references are removed. This removal logic, and the basic ref counting are intrically tied, thus in a single patch. Link: https://lore.kernel.org/r/20201115192646.12977-2-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:30 +08:00
if (ndlp && ndlp->nlp_state == NLP_STE_UNMAPPED_NODE &&
[SCSI] lpfc 8.2.6 : Multiple discovery fixes Multiple Discovery Fixes: - Fix race on discovery due to link events coinciding with vport_delete. - Use NLP_FABRIC state to filter out switch-based pseudo initiators that reuse the same WWNs. - Correct erroneous setting of DID=0 in lpfc_matchdid() - Correct extra reference count that was in the lookup path for the remoteid from an unsolicited ELS. - Correct double-free bug in els abort path. - Correct FDMI server discovery logic for switch that return a WWN of 0. - Fix bugs in ndlp mgmt when a node changes address - Correct bug that did not delete RSCNs for vports upon link transitions - Fix "0216 Link event during NS query" error which pops up when vports are swapped to different switch ports. - Add sanity checks on ndlp structures - Fix devloss log message to dump WWN correctly - Hold off mgmt commands that were interferring with discovery mailbox cmds - Remove unnecessary FC_ESTABLISH_LINK logic. - Correct some race conditions in the worker thread, resulting in devloss: - Clear the work_port_events field before handling the work port events - Clear the deferred ring event before handling a deferred ring event - Hold the hba lock when waking up the work thread - Send an acc for the rscn even when we aren't going to handle it - Fix locking behavior that was not properly protecting the ACTIVE flag, thus allowing mailbox command order to shift. Signed-off-by: James Smart <james.smart@emulex.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-04-07 22:15:56 +08:00
phba->link_state >= LPFC_LINK_UP &&
2010-11-21 12:11:48 +08:00
phba->fc_topology != LPFC_TOPOLOGY_LOOP) {
if (vport->cfg_enable_da_id) {
scsi: lpfc: Fix refcounting around SCSI and NVMe transport APIs Due to bug history and code review, the node reference counting approach in the driver isn't implemented consistently with how the scsi and nvme transport perform registrations and unregistrations and their callbacks. This resulted in many bad/stale node pointers. Reword the driver so that reference handling is performed as follows: - The initial node reference is taken on structure allocation - Take a reference on any add/register call to the transport - Remove a reference on any delete/unregister call to the transport - After the node has fully removed from both the SCSI and NVMEe transports (dev_loss_callbacks have called back) call the discovery engine DEVICE_RM event which will remove the final reference and release the node structure. - Alter dev_loss handling when a vport or base port is unloading. - Remove the put_node handling - no longer needed. - Rewrite the vport_delete handling on reference counts. Part of this effort was driven from the FDISC not registering with the transport and disrupting the model for node reference counting. - Deleted lpfc_nlp_remove. Pushed it's remaining ops into lpfc_nlp_release. - Several other small code cleanups. Link: https://lore.kernel.org/r/20201115192646.12977-5-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:33 +08:00
/* Send DA_ID and wait for a completion. */
timeout = msecs_to_jiffies(phba->fc_ratov * 2000);
if (!lpfc_ns_cmd(vport, SLI_CTNS_DA_ID, 0, 0))
while (vport->ct_flags && timeout)
timeout = schedule_timeout(timeout);
else
lpfc_printf_log(vport->phba, KERN_WARNING,
LOG_VPORT,
"1829 CT command failed to "
"delete objects on fabric\n");
}
2011-10-11 09:32:10 +08:00
/*
* If the vpi is not registered, then a valid FDISC doesn't
* exist and there is no need for a ELS LOGO. Just cleanup
* the ndlp.
*/
scsi: lpfc: Fix refcounting around SCSI and NVMe transport APIs Due to bug history and code review, the node reference counting approach in the driver isn't implemented consistently with how the scsi and nvme transport perform registrations and unregistrations and their callbacks. This resulted in many bad/stale node pointers. Reword the driver so that reference handling is performed as follows: - The initial node reference is taken on structure allocation - Take a reference on any add/register call to the transport - Remove a reference on any delete/unregister call to the transport - After the node has fully removed from both the SCSI and NVMEe transports (dev_loss_callbacks have called back) call the discovery engine DEVICE_RM event which will remove the final reference and release the node structure. - Alter dev_loss handling when a vport or base port is unloading. - Remove the put_node handling - no longer needed. - Rewrite the vport_delete handling on reference counts. Part of this effort was driven from the FDISC not registering with the transport and disrupting the model for node reference counting. - Deleted lpfc_nlp_remove. Pushed it's remaining ops into lpfc_nlp_release. - Several other small code cleanups. Link: https://lore.kernel.org/r/20201115192646.12977-5-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:33 +08:00
if (!(vport->vpi_state & LPFC_VPI_REGISTERED))
goto skip_logo;
2011-10-11 09:32:10 +08:00
scsi: lpfc: Fix refcounting around SCSI and NVMe transport APIs Due to bug history and code review, the node reference counting approach in the driver isn't implemented consistently with how the scsi and nvme transport perform registrations and unregistrations and their callbacks. This resulted in many bad/stale node pointers. Reword the driver so that reference handling is performed as follows: - The initial node reference is taken on structure allocation - Take a reference on any add/register call to the transport - Remove a reference on any delete/unregister call to the transport - After the node has fully removed from both the SCSI and NVMEe transports (dev_loss_callbacks have called back) call the discovery engine DEVICE_RM event which will remove the final reference and release the node structure. - Alter dev_loss handling when a vport or base port is unloading. - Remove the put_node handling - no longer needed. - Rewrite the vport_delete handling on reference counts. Part of this effort was driven from the FDISC not registering with the transport and disrupting the model for node reference counting. - Deleted lpfc_nlp_remove. Pushed it's remaining ops into lpfc_nlp_release. - Several other small code cleanups. Link: https://lore.kernel.org/r/20201115192646.12977-5-james.smart@broadcom.com Co-developed-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <james.smart@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-11-16 03:26:33 +08:00
/* Issue a Fabric LOGO to cleanup fabric resources. */
ndlp = lpfc_findnode_did(vport, Fabric_DID);
if (!ndlp)
goto skip_logo;
vport->unreg_vpi_cmpl = VPORT_INVAL;
timeout = msecs_to_jiffies(phba->fc_ratov * 2000);
if (!lpfc_issue_els_npiv_logo(vport, ndlp))
while (vport->unreg_vpi_cmpl == VPORT_INVAL && timeout)
timeout = schedule_timeout(timeout);
}
if (!(phba->pport->load_flag & FC_UNLOADING))
lpfc_discovery_wait(vport);
skip_logo:
lpfc_cleanup(vport);
/* Remove scsi host now. The nodes are cleaned up. */
lpfc_sli_host_down(vport);
lpfc_stop_vport_timers(vport);
if (!(phba->pport->load_flag & FC_UNLOADING)) {
lpfc_unreg_all_rpis(vport);
lpfc_unreg_default_rpis(vport);
/*
* Completion of unreg_vpi (lpfc_mbx_cmpl_unreg_vpi)
* does the scsi_host_put() to release the vport.
*/
if (!(vport->vpi_state & LPFC_VPI_REGISTERED) ||
lpfc_mbx_unreg_vpi(vport))
scsi_host_put(shost);
} else {
scsi_host_put(shost);
}
lpfc_free_vpi(phba, vport->vpi);
vport->work_port_events = 0;
spin_lock_irq(&phba->port_list_lock);
list_del_init(&vport->listentry);
spin_unlock_irq(&phba->port_list_lock);
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR, LOG_TRACE_EVENT,
"1828 Vport Deleted.\n");
scsi_host_put(shost);
return VPORT_OK;
}
struct lpfc_vport **
lpfc_create_vport_work_array(struct lpfc_hba *phba)
{
struct lpfc_vport *port_iterator;
struct lpfc_vport **vports;
int index = 0;
treewide: kzalloc() -> kcalloc() The kzalloc() function has a 2-factor argument form, kcalloc(). This patch replaces cases of: kzalloc(a * b, gfp) with: kcalloc(a * b, gfp) as well as handling cases of: kzalloc(a * b * c, gfp) with: kzalloc(array3_size(a, b, c), gfp) as it's slightly less ugly than: kzalloc_array(array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: kzalloc(4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( kzalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | kzalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( kzalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | kzalloc( - sizeof(u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(__u8) * COUNT + COUNT , ...) | kzalloc( - sizeof(char) * COUNT + COUNT , ...) | kzalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ - kzalloc + kcalloc ( - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( kzalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kzalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( kzalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | kzalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( kzalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kzalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( kzalloc(C1 * C2 * C3, ...) | kzalloc( - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | kzalloc( - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | kzalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( kzalloc(sizeof(THING) * C2, ...) | kzalloc(sizeof(TYPE) * C2, ...) | kzalloc(C1 * C2 * C3, ...) | kzalloc(C1 * C2, ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - kzalloc + kcalloc ( - (E1) * E2 + E1, E2 , ...) | - kzalloc + kcalloc ( - (E1) * (E2) + E1, E2 , ...) | - kzalloc + kcalloc ( - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-13 05:03:40 +08:00
vports = kcalloc(phba->max_vports + 1, sizeof(struct lpfc_vport *),
GFP_KERNEL);
if (vports == NULL)
return NULL;
spin_lock_irq(&phba->port_list_lock);
list_for_each_entry(port_iterator, &phba->port_list, listentry) {
if (port_iterator->load_flag & FC_UNLOADING)
continue;
if (!scsi_host_get(lpfc_shost_from_vport(port_iterator))) {
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(port_iterator, KERN_ERR,
LOG_TRACE_EVENT,
"1801 Create vport work array FAILED: "
"cannot do scsi_host_get\n");
continue;
}
vports[index++] = port_iterator;
}
spin_unlock_irq(&phba->port_list_lock);
return vports;
}
void
[SCSI] lpfc 8.2.4 : Miscellaneous Fixes Miscellaneous Fixes: - Fix a couple of sparse complaints - Reset the FCP recovery flag when the node is not a FCP2 device. - Speed up offline prep delays - Fixed a memory leak in lpfc_mem_alloc failure path - Fixed external loopback test. - Fixed error code returned from the driver when HBA is over heated. - Correct Max NPIV vport to limits read from adapter - Add missing locks around fc_flag and FC_NEEDS_REG_VPI - Add missing hba ids for device identification - Added support for SET_VARIABLE and MBX_WRITE_WWN mailbox commands - Changed all temperature event messages from warning to error - Fix reporting of link speed when link is down - Added support for MBX_WRITE_WWN mailbox command - Change del_timer_sync() in ISR to del_timer() in interrupt handler - Correct instances of beXX_to_cpu() that should be cpu_to_beXX() - Perform target flush before releasing node references on module unload - Avoid bogus devloss_tmo messages when driver unloads - Fix panic when HBA generates ERATT interupt - Fix mbox race condition and a workaround on back-to-back mailbox commands - Force NPIV off for pt2pt mode between 2 NPorts - Stop worker thread before removing fc_host. - Fix up discovery timeout error case due to missing clear_la - Tighten mailbox polling code to speed up detection of fast completions - Only allow DUMP_MEMORY if adapter offline due to overtemp errors - Added extended error information to the log messages in chip init. Signed-off-by: James Smart <James.Smart@emulex.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-01-11 14:52:54 +08:00
lpfc_destroy_vport_work_array(struct lpfc_hba *phba, struct lpfc_vport **vports)
{
int i;
if (vports == NULL)
return;
for (i = 0; i <= phba->max_vports && vports[i] != NULL; i++)
scsi_host_put(lpfc_shost_from_vport(vports[i]));
kfree(vports);
}
/**
* lpfc_vport_reset_stat_data - Reset the statistical data for the vport
* @vport: Pointer to vport object.
*
* This function resets the statistical data for the vport. This function
* is called with the host_lock held
**/
void
lpfc_vport_reset_stat_data(struct lpfc_vport *vport)
{
struct lpfc_nodelist *ndlp = NULL, *next_ndlp = NULL;
list_for_each_entry_safe(ndlp, next_ndlp, &vport->fc_nodes, nlp_listp) {
if (ndlp->lat_data)
memset(ndlp->lat_data, 0, LPFC_MAX_BUCKET_COUNT *
sizeof(struct lpfc_scsicmd_bkt));
}
}
/**
* lpfc_alloc_bucket - Allocate data buffer required for statistical data
* @vport: Pointer to vport object.
*
* This function allocates data buffer required for all the FC
* nodes of the vport to collect statistical data.
**/
void
lpfc_alloc_bucket(struct lpfc_vport *vport)
{
struct lpfc_nodelist *ndlp = NULL, *next_ndlp = NULL;
list_for_each_entry_safe(ndlp, next_ndlp, &vport->fc_nodes, nlp_listp) {
kfree(ndlp->lat_data);
ndlp->lat_data = NULL;
if (ndlp->nlp_state == NLP_STE_MAPPED_NODE) {
ndlp->lat_data = kcalloc(LPFC_MAX_BUCKET_COUNT,
sizeof(struct lpfc_scsicmd_bkt),
GFP_ATOMIC);
if (!ndlp->lat_data)
scsi: lpfc: Add an internal trace log buffer The current logging methods typically end up requesting a reproduction with a different logging level set to figure out what happened. This was mainly by design to not clutter the kernel log messages with things that were typically not interesting and the messages themselves could cause other issues. When looking to make a better system, it was seen that in many cases when more data was wanted was when another message, usually at KERN_ERR level, was logged. And in most cases, what the additional logging that was then enabled was typically. Most of these areas fell into the discovery machine. Based on this summary, the following design has been put in place: The driver will maintain an internal log (256 elements of 256 bytes). The "additional logging" messages that are usually enabled in a reproduction will be changed to now log all the time to the internal log. A new logging level is defined - LOG_TRACE_EVENT. When this level is set (it is not by default) and a message marked as KERN_ERR is logged, all the messages in the internal log will be dumped to the kernel log before the KERN_ERR message is logged. There is a timestamp on each message added to the internal log. However, this timestamp is not converted to wall time when logged. The value of the timestamp is solely to give a crude time reference for the messages. Link: https://lore.kernel.org/r/20200630215001.70793-14-jsmart2021@gmail.com Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com> Signed-off-by: James Smart <jsmart2021@gmail.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-07-01 05:50:00 +08:00
lpfc_printf_vlog(vport, KERN_ERR,
LOG_TRACE_EVENT,
"0287 lpfc_alloc_bucket failed to "
"allocate statistical data buffer DID "
"0x%x\n", ndlp->nlp_DID);
}
}
}
/**
* lpfc_free_bucket - Free data buffer required for statistical data
* @vport: Pointer to vport object.
*
* Th function frees statistical data buffer of all the FC
* nodes of the vport.
**/
void
lpfc_free_bucket(struct lpfc_vport *vport)
{
struct lpfc_nodelist *ndlp = NULL, *next_ndlp = NULL;
list_for_each_entry_safe(ndlp, next_ndlp, &vport->fc_nodes, nlp_listp) {
kfree(ndlp->lat_data);
ndlp->lat_data = NULL;
}
}