qla2xxx: Updates for the driver
Subsystems (2)
Nilesh Javali
maintainercover qla2xxx: Updates for the driver
Hi Martin,
Please apply the qla2xxx driver bug fixes to the scsi tree at your earliest convenience.
Thanks,
Nilesh
Arun Easi (2):
qla2xxx: Fix MPI failure AEN (8200) handling.
qla2xxx: Fix hang when issuing nvme disconnect-all in NPIV.
Quinn Tran (1):
qla2xxx: delete all sessions before unregister local nvme port
drivers/scsi/qla2xxx/qla_attr.c | 32 +++++++-- drivers/scsi/qla2xxx/qla_def.h | 13 +++- drivers/scsi/qla2xxx/qla_gbl.h | 3 + drivers/scsi/qla2xxx/qla_init.c | 2 + drivers/scsi/qla2xxx/qla_isr.c | 54 ++++++++++----- drivers/scsi/qla2xxx/qla_mbx.c | 2 +- drivers/scsi/qla2xxx/qla_os.c | 6 ++ drivers/scsi/qla2xxx/qla_tmpl.c | 118 ++++++++++++++++++++++++++------ 8 files changed, 187 insertions(+), 43 deletions(-)
Signature · 1 line
2.19.0.rc0
Nilesh Javali
maintainerPATCH 1/3 qla2xxx: Fix MPI failure AEN (8200) handling.
From: Arun Easi <[email protected]>
Today, upon an MPI failure AEN, on top of collecting an MPI dump, a regular firmware dump is also taken and the chip reset. This is disruptive to IOs and not required. So, make the firmware dump collection, followed by chip reset optional (not done by default).
Firmware dump buffer and MPI dump buffer are independent of each other with this change and each can have dump that was taken at two different times for two different issues. The MPI dump is saved in a separate buffer and is retrieved differently from firmware dump.
To collect full dump on MPI failure AEN, a module parameter is introduced:
ql2xfulldump_on_mpifail (default: 0)
--- drivers/scsi/qla2xxx/qla_attr.c | 30 +++++++- drivers/scsi/qla2xxx/qla_def.h | 13 +++- drivers/scsi/qla2xxx/qla_gbl.h | 3 + drivers/scsi/qla2xxx/qla_init.c | 2 + drivers/scsi/qla2xxx/qla_isr.c | 54 ++++++++++----- drivers/scsi/qla2xxx/qla_os.c | 6 ++ drivers/scsi/qla2xxx/qla_tmpl.c | 118 ++++++++++++++++++++++++++------ 7 files changed, 185 insertions(+), 41 deletions(-)
drivers/scsi/qla2xxx/qla_attr.c +27 -3
@@ -26,7 +26,8 @@ qla2x00_sysfs_read_fw_dump(struct file *filp, struct kobject *kobj,struct qla_hw_data *ha = vha->hw;int rval =0 ; -if (!(ha->fw_dump_reading || ha->mctp_dump_reading)) +if (!(ha->fw_dump_reading || ha->mctp_dump_reading || + ha->mpi_fw_dump_reading))return 0 ;mutex_lock (&ha->optrom_mutex);
@@ -42,6 +43,10 @@ qla2x00_sysfs_read_fw_dump(struct file *filp, struct kobject *kobj, }else if (ha->mctp_dumped && ha->mctp_dump_reading) { rval =memory_read_from_buffer (buf, count, &off, ha->mctp_dump,MCTP_DUMP_SIZE ); + }else if (ha->mpi_fw_dumped && ha->mpi_fw_dump_reading) { + rval =memory_read_from_buffer (buf, count, &off, + ha->mpi_fw_dump, + ha->mpi_fw_dump_len); }else if (ha->fw_dump_reading) { rval =memory_read_from_buffer (buf, count, &off, ha->fw_dump, ha->fw_dump_len);
@@ -103,7 +108,6 @@ qla2x00_sysfs_write_fw_dump(struct file *filp, struct kobject *kobj,qla82xx_set_reset_owner (vha);qla8044_idc_unlock (ha); }else { - ha->fw_dump_mpi =1 ;qla2x00_system_error (vha); }break ;
@@ -137,6 +141,22 @@ qla2x00_sysfs_write_fw_dump(struct file *filp, struct kobject *kobj, vha->host_no); }break ; +case 8 : +if (!ha->mpi_fw_dump_reading) +break ; +ql_log (ql_log_info, vha,0x70e7 , +"MPI firmware dump cleared on (%ld).\n" , vha->host_no); + ha->mpi_fw_dump_reading =0 ; + ha->mpi_fw_dumped =0 ; +break ; +case 9 : +if (ha->mpi_fw_dumped && !ha->mpi_fw_dump_reading) { + ha->mpi_fw_dump_reading =1 ; +ql_log (ql_log_info, vha,0x70e8 , +"Raw MPI firmware dump ready for read on (%ld).\n" , + vha->host_no); + } +break ; }return count; }
@@ -706,7 +726,8 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj,scsi_unblock_requests (vha->host);break ;case 0x2025d : -if (!IS_QLA81XX (ha) && !IS_QLA83XX (ha)) +if (!IS_QLA81XX (ha) && !IS_QLA83XX (ha) && + !IS_QLA27XX (ha) && !IS_QLA28XX (ha))return -EPERM ; ql_log(ql_log_info, vha, 0x706f,
@@ -724,6 +745,8 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj,qla83xx_idc_audit (vha,IDC_AUDIT_TIMESTAMP );qla83xx_idc_unlock (vha,0 );break ; + }else if (IS_QLA27XX (ha) ||IS_QLA28XX (ha)) { +qla27xx_reset_mpi (vha); }else {/* Make sure FC side is not in reset */ WARN_ON_ONCE(qla2x00_wait_for_hba_online(vha) !=
@@ -737,6 +760,7 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj,scsi_unblock_requests (vha->host);break ; } +break ;case 0x2025e :if (!IS_P3P_TYPE (ha) || vha != base_vha) { ql_log(ql_log_info, vha, 0x7071,
drivers/scsi/qla2xxx/qla_def.h +12 -1
@@ -3223,6 +3223,7 @@ struct isp_operations { uint32_t);void (*fw_dump) (struct scsi_qla_host *,int ); +void (*mpi_fw_dump)(struct scsi_qla_host *,int );int (*beacon_on) (struct scsi_qla_host *);int (*beacon_off) (struct scsi_qla_host *);
@@ -3748,6 +3749,11 @@ struct qlt_hw_data {#define LEAK_EXCHG_THRESH_HOLD_PERCENT 75/* 75 percent */ +struct qla_hw_data_stat { +u32 num_fw_dump; +u32 num_mpi_reset; +}; +/* * Qlogic host adapter specific data structure. */
@@ -4230,7 +4236,6 @@ struct qla_hw_data {uint32_t fw_dump_len;u32 fw_dump_alloc_len;bool fw_dumped; -bool fw_dump_mpi;unsigned long fw_dump_cap_flags;#define RISC_PAUSE_CMPL 0#define DMA_SHUTDOWN_CMPL 1
@@ -4241,6 +4246,10 @@ struct qla_hw_data {#define ISP_MBX_RDY 6#define ISP_SOFT_RESET_CMPL 7int fw_dump_reading; +void *mpi_fw_dump; +u32 mpi_fw_dump_len; + int mpi_fw_dump_reading:1; + int mpi_fw_dumped:1;int prev_minidump_failed;dma_addr_t eft_dma;void *eft;
@@ -4454,6 +4463,8 @@ struct qla_hw_data {uint16_t last_zio_threshold;#define DEFAULT_ZIO_THRESHOLD 5 + +struct qla_hw_data_stat stat; };struct active_regions {
drivers/scsi/qla2xxx/qla_gbl.h +3 -0
@@ -173,6 +173,7 @@ extern int ql2xenablemsix;extern int qla2xuseresexchforels;extern int ql2xexlogins;extern int ql2xdifbundlinginternalbuffers; +extern int ql2xfulldump_on_mpifail;extern int qla2x00_loop_reset (scsi_qla_host_t *);extern void qla2x00_abort_all_cmds (scsi_qla_host_t *,int );
@@ -645,6 +646,7 @@ extern void qla82xx_fw_dump(scsi_qla_host_t *, int);extern void qla8044_fw_dump (scsi_qla_host_t *,int );extern void qla27xx_fwdump (scsi_qla_host_t *,int ); +extern void qla27xx_mpi_fwdump (scsi_qla_host_t *,int );extern ulong qla27xx_fwdt_calculate_dump_size (struct scsi_qla_host *,void *);extern int qla27xx_fwdt_template_valid (void *);extern ulong qla27xx_fwdt_template_size (void *);
@@ -933,5 +935,6 @@ extern void qla24xx_process_purex_list(struct purex_list *);/* nvme.c */ void qla_nvme_unregister_remote_port (struct fc_port *fcport); +void qla27xx_reset_mpi (scsi_qla_host_t *vha);void qla_handle_els_plogi_done (scsi_qla_host_t *vha,struct event_arg *ea);#endif /* _QLA_GBL_H */
drivers/scsi/qla2xxx/qla_init.c +2 -0
@@ -3339,6 +3339,8 @@ qla2x00_alloc_fw_dump(scsi_qla_host_t *vha) dump_size / 1024);if (IS_QLA27XX (ha) ||IS_QLA28XX (ha)) { + ha->mpi_fw_dump = (char *)fw_dump + + ha->fwdt[1 ].dump_size;mutex_unlock (&ha->optrom_mutex);return ; }
drivers/scsi/qla2xxx/qla_isr.c +37 -17
72 lines not on this page · raw message
drivers/scsi/qla2xxx/qla_os.c +6 -0
@@ -35,6 +35,11 @@ static int apidev_major;*/ struct kmem_cache *srb_cachep; +int ql2xfulldump_on_mpifail; +module_param (ql2xfulldump_on_mpifail, int,S_IRUGO |S_IWUSR ); +MODULE_PARM_DESC (ql2xfulldump_on_mpifail, +"Set this to take full dump on MPI hang." ); +/* * CT6 CTX allocation cache */
@@ -2518,6 +2523,7 @@ static struct isp_operations qla27xx_isp_ops = { .read_nvram = NULL, .write_nvram = NULL, .fw_dump = qla27xx_fwdump, + .mpi_fw_dump = qla27xx_mpi_fwdump, .beacon_on = qla24xx_beacon_on, .beacon_off = qla24xx_beacon_off, .beacon_blink = qla83xx_beacon_blink,
drivers/scsi/qla2xxx/qla_tmpl.c +98 -20
147 lines not on this page · raw message
Signature · 1 line
2.19.0.rc0
Re: [PATCH 1/3] qla2xxx: Fix MPI failure AEN (8200) handling.
Quoted 456 lines · Nilesh Javali
From: Arun Easi <[email protected]>
Today, upon an MPI failure AEN, on top of collecting an MPI dump, a regular firmware dump is also taken and the chip reset. This is disruptive to IOs and not required. So, make the firmware dump collection, followed by chip reset optional (not done by default).
Firmware dump buffer and MPI dump buffer are independent of each other with this change and each can have dump that was taken at two different times for two different issues. The MPI dump is saved in a separate buffer and is retrieved differently from firmware dump.
To collect full dump on MPI failure AEN, a module parameter is introduced:
ql2xfulldump_on_mpifail(default: 0)Signed-off-by Arun Easi <[email protected]>Signed-off-by Himanshu Madhani <[email protected]>Signed-off-by Nilesh Javali <[email protected]>--- drivers/scsi/qla2xxx/qla_attr.c | 30 +++++++- drivers/scsi/qla2xxx/qla_def.h | 13 +++- drivers/scsi/qla2xxx/qla_gbl.h | 3 + drivers/scsi/qla2xxx/qla_init.c | 2 + drivers/scsi/qla2xxx/qla_isr.c | 54 ++++++++++----- drivers/scsi/qla2xxx/qla_os.c | 6 ++ drivers/scsi/qla2xxx/qla_tmpl.c | 118 ++++++++++++++++++++++++++------ 7 files changed, 185 insertions(+), 41 deletions(-)drivers/scsi/qla2xxx/qla_attr.c +27
-3@@ -26,7 +26,8 @@ qla2x00_sysfs_read_fw_dump(struct file *filp, struct kobject *kobj,struct qla_hw_data *ha = vha->hw;int rval =0 ; -if (!(ha->fw_dump_reading || ha->mctp_dump_reading)) +if (!(ha->fw_dump_reading || ha->mctp_dump_reading || + ha->mpi_fw_dump_reading))return 0 ;mutex_lock (&ha->optrom_mutex);@@ -42,6 +43,10 @@ qla2x00_sysfs_read_fw_dump(struct file *filp, struct kobject *kobj, }else if (ha->mctp_dumped && ha->mctp_dump_reading) { rval =memory_read_from_buffer (buf, count, &off, ha->mctp_dump,MCTP_DUMP_SIZE ); + }else if (ha->mpi_fw_dumped && ha->mpi_fw_dump_reading) { + rval =memory_read_from_buffer (buf, count, &off, + ha->mpi_fw_dump, + ha->mpi_fw_dump_len); }else if (ha->fw_dump_reading) { rval =memory_read_from_buffer (buf, count, &off, ha->fw_dump, ha->fw_dump_len);@@ -103,7 +108,6 @@ qla2x00_sysfs_write_fw_dump(struct file *filp, struct kobject *kobj,qla82xx_set_reset_owner (vha);qla8044_idc_unlock (ha); }else { - ha->fw_dump_mpi =1 ;qla2x00_system_error (vha); }break ;@@ -137,6 +141,22 @@ qla2x00_sysfs_write_fw_dump(struct file *filp, struct kobject *kobj, vha->host_no); }break ; +case 8 : +if (!ha->mpi_fw_dump_reading) +break ; +ql_log (ql_log_info, vha,0x70e7 , +"MPI firmware dump cleared on (%ld).\n" , vha->host_no); + ha->mpi_fw_dump_reading =0 ; + ha->mpi_fw_dumped =0 ; +break ; +case 9 : +if (ha->mpi_fw_dumped && !ha->mpi_fw_dump_reading) { + ha->mpi_fw_dump_reading =1 ; +ql_log (ql_log_info, vha,0x70e8 , +"Raw MPI firmware dump ready for read on (%ld).\n" , + vha->host_no); + } +break ; }return count; }@@ -706,7 +726,8 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj,scsi_unblock_requests (vha->host);break ;case 0x2025d : -if (!IS_QLA81XX (ha) && !IS_QLA83XX (ha)) +if (!IS_QLA81XX (ha) && !IS_QLA83XX (ha) && + !IS_QLA27XX (ha) && !IS_QLA28XX (ha))return -EPERM ; ql_log(ql_log_info, vha, 0x706f,@@ -724,6 +745,8 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj,qla83xx_idc_audit (vha,IDC_AUDIT_TIMESTAMP );qla83xx_idc_unlock (vha,0 );break ; + }else if (IS_QLA27XX (ha) ||IS_QLA28XX (ha)) { +qla27xx_reset_mpi (vha); }else {/* Make sure FC side is not in reset */ WARN_ON_ONCE(qla2x00_wait_for_hba_online(vha) !=@@ -737,6 +760,7 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj,scsi_unblock_requests (vha->host);break ; } +break ;case 0x2025e :if (!IS_P3P_TYPE (ha) || vha != base_vha) { ql_log(ql_log_info, vha, 0x7071,drivers/scsi/qla2xxx/qla_def.h +12
-1@@ -3223,6 +3223,7 @@ struct isp_operations { uint32_t);void (*fw_dump) (struct scsi_qla_host *,int ); +void (*mpi_fw_dump)(struct scsi_qla_host *,int );int (*beacon_on) (struct scsi_qla_host *);int (*beacon_off) (struct scsi_qla_host *);@@ -3748,6 +3749,11 @@ struct qlt_hw_data {#define LEAK_EXCHG_THRESH_HOLD_PERCENT 75/* 75 percent */ +struct qla_hw_data_stat { +u32 num_fw_dump; +u32 num_mpi_reset; +}; +/* * Qlogic host adapter specific data structure. */ @@ -4230,7 +4236,6 @@ struct qla_hw_data {uint32_t fw_dump_len;u32 fw_dump_alloc_len;bool fw_dumped; -bool fw_dump_mpi;unsigned long fw_dump_cap_flags;#define RISC_PAUSE_CMPL 0#define DMA_SHUTDOWN_CMPL 1@@ -4241,6 +4246,10 @@ struct qla_hw_data {#define ISP_MBX_RDY 6#define ISP_SOFT_RESET_CMPL 7int fw_dump_reading; +void *mpi_fw_dump; +u32 mpi_fw_dump_len; + int mpi_fw_dump_reading:1; + int mpi_fw_dumped:1;int prev_minidump_failed;dma_addr_t eft_dma;void *eft;@@ -4454,6 +4463,8 @@ struct qla_hw_data {uint16_t last_zio_threshold;#define DEFAULT_ZIO_THRESHOLD 5 + +struct qla_hw_data_stat stat; };struct active_regions {drivers/scsi/qla2xxx/qla_gbl.h +3
-0@@ -173,6 +173,7 @@ extern int ql2xenablemsix;extern int qla2xuseresexchforels;extern int ql2xexlogins;extern int ql2xdifbundlinginternalbuffers; +extern int ql2xfulldump_on_mpifail;extern int qla2x00_loop_reset (scsi_qla_host_t *);extern void qla2x00_abort_all_cmds (scsi_qla_host_t *,int );@@ -645,6 +646,7 @@ extern void qla82xx_fw_dump(scsi_qla_host_t *, int);extern void qla8044_fw_dump (scsi_qla_host_t *,int );extern void qla27xx_fwdump (scsi_qla_host_t *,int ); +extern void qla27xx_mpi_fwdump (scsi_qla_host_t *,int );extern ulong qla27xx_fwdt_calculate_dump_size (struct scsi_qla_host *,void *);extern int qla27xx_fwdt_template_valid (void *);extern ulong qla27xx_fwdt_template_size (void *);@@ -933,5 +935,6 @@ extern void qla24xx_process_purex_list(struct purex_list *);/* nvme.c */ void qla_nvme_unregister_remote_port (struct fc_port *fcport); +void qla27xx_reset_mpi (scsi_qla_host_t *vha);void qla_handle_els_plogi_done (scsi_qla_host_t *vha,struct event_arg *ea);#endif /* _QLA_GBL_H */ drivers/scsi/qla2xxx/qla_init.c +2
-0@@ -3339,6 +3339,8 @@ qla2x00_alloc_fw_dump(scsi_qla_host_t *vha) dump_size / 1024);if (IS_QLA27XX (ha) ||IS_QLA28XX (ha)) { + ha->mpi_fw_dump = (char *)fw_dump + + ha->fwdt[1 ].dump_size;mutex_unlock (&ha->optrom_mutex);return ; }drivers/scsi/qla2xxx/qla_isr.c +37
-17@@ -757,6 +757,39 @@ qla2x00_find_fcport_by_nportid(scsi_qla_host_t *vha, port_id_t *id,return NULL ; } +/* Shall be called only on supported adapters. */ +static void +qla27xx_handle_8200_aen (scsi_qla_host_t *vha,uint16_t *mb) +{ +struct qla_hw_data *ha = vha->hw; +bool reset_isp_needed =0 ; + +ql_log (ql_log_warn, vha,0x02f0 , +"MPI Heartbeat stop. MPI reset is%s needed. " +"MB0[%xh] MB1[%xh] MB2[%xh] MB3[%xh]\n" , + mb[0 ] &BIT_8 ?"" :" not" , + mb[0 ], mb[1 ], mb[2 ], mb[3 ]); + +if ((mb[1 ] &BIT_8 ) ==0 ) +return ; + +ql_log (ql_log_warn, vha,0x02f1 , +"MPI Heartbeat stop. FW dump needed\n" ); + +if (ql2xfulldump_on_mpifail) { + ha->isp_ops->fw_dump (vha,1 ); + reset_isp_needed =1 ; + } + + ha->isp_ops->mpi_fw_dump (vha,1 ); + +if (reset_isp_needed) { + vha->hw->flags.fw_init_done =0 ; +set_bit (ISP_ABORT_NEEDED , &vha->dpc_flags); +qla2xxx_wake_dpc (vha); + } +} +/** * qla2x00_async_event() - Process aynchronous events. * @vha: SCSI driver HA context @@ -872,9 +905,9 @@ qla2x00_async_event(scsi_qla_host_t *vha, struct rsp_que *rsp, uint16_t *mb)"ISP System Error - mbx1=%xh mbx2=%xh mbx3=%xh.\n " , mb[1], mb[2], mb[3]); - ha->fw_dump_mpi = - (IS_QLA27XX (ha) ||IS_QLA28XX (ha)) && -RD_REG_WORD (®24->mailbox7) &BIT_8 ; +if ((IS_QLA27XX (ha) ||IS_QLA28XX (ha)) && +RD_REG_WORD (®24->mailbox7) &BIT_8 ) + ha->isp_ops->mpi_fw_dump (vha,1 ); ha->isp_ops->fw_dump (vha,1 ); ha->flags.fw_init_done =0 ;QLA_FW_STOPPED (ha);@@ -1375,20 +1408,7 @@ qla2x00_async_event(scsi_qla_host_t *vha, struct rsp_que *rsp, uint16_t *mb)case MBA_IDC_AEN :if (IS_QLA27XX (ha) ||IS_QLA28XX (ha)) { - ha->flags.fw_init_done =0 ; -ql_log (ql_log_warn, vha,0xffff , -"MPI Heartbeat stop. Chip reset needed. MB0[%xh] MB1[%xh] MB2[%xh] MB3[%xh]\n" , - mb[0 ], mb[1 ], mb[2 ], mb[3 ]); - -if ((mb[1 ] &BIT_8 ) || - (mb[2 ] &BIT_8 )) { -ql_log (ql_log_warn, vha,0xd013 , -"MPI Heartbeat stop. FW dump needed\n" ); - ha->fw_dump_mpi =1 ; - ha->isp_ops->fw_dump (vha,1 ); - } -set_bit (ISP_ABORT_NEEDED , &vha->dpc_flags); -qla2xxx_wake_dpc (vha); +qla27xx_handle_8200_aen (vha, mb); }else if (IS_QLA83XX (ha)) { mb[4 ] =RD_REG_WORD (®24->mailbox4); mb[5 ] =RD_REG_WORD (®24->mailbox5);drivers/scsi/qla2xxx/qla_os.c +6
-0@@ -35,6 +35,11 @@ static int apidev_major;*/ struct kmem_cache *srb_cachep; +int ql2xfulldump_on_mpifail; +module_param (ql2xfulldump_on_mpifail, int,S_IRUGO |S_IWUSR ); +MODULE_PARM_DESC (ql2xfulldump_on_mpifail, +"Set this to take full dump on MPI hang." ); +/* * CT6 CTX allocation cache */ @@ -2518,6 +2523,7 @@ static struct isp_operations qla27xx_isp_ops = { .read_nvram = NULL, .write_nvram = NULL, .fw_dump = qla27xx_fwdump, + .mpi_fw_dump = qla27xx_mpi_fwdump, .beacon_on = qla24xx_beacon_on, .beacon_off = qla24xx_beacon_off, .beacon_blink = qla83xx_beacon_blink,drivers/scsi/qla2xxx/qla_tmpl.c +98
-19@@ -12,6 +12,33 @@#define IOBASE (vha) IOBAR(ISPREG(vha))#define INVALID_ENTRY ((struct qla27xx_fwdt_entry *)0xffffffffffffffffUL) +/* hardware_lock assumed held. */ +static void +qla27xx_write_remote_reg (struct scsi_qla_host *vha, +u32 addr,u32 data) +{ +char *reg = (char *)ISPREG (vha); + +ql_dbg (ql_dbg_misc, vha,0xd300 , +"%s: addr/data = %xh/%xh\n" , __func__, addr, data); + +WRT_REG_DWORD (reg +IOBASE (vha),0x40 ); +WRT_REG_DWORD (reg +0xc4 , data); +WRT_REG_DWORD (reg +0xc0 , addr); +} + +void +qla27xx_reset_mpi (scsi_qla_host_t *vha) +{ +ql_dbg (ql_dbg_misc + ql_dbg_verbose, vha,0xd301 , +"Entered %s.\n" , __func__); + +qla27xx_write_remote_reg (vha,0x104050 ,0x40004 ); +qla27xx_write_remote_reg (vha,0x10405c ,0x4 ); + + vha->hw->stat.num_mpi_reset++; +} +static inline void qla27xx_insert16 (uint16_t value,void *buf,ulong *len) {@@ -997,6 +1024,62 @@ qla27xx_fwdt_template_valid(void *p)return true; } +void +qla27xx_mpi_fwdump (scsi_qla_host_t *vha,int hardware_locked) +{ +ulong flags =0 ; +bool need_mpi_reset =1 ; + +#ifndef __CHECKER__ +if (!hardware_locked) +spin_lock_irqsave (&vha->hw->hardware_lock, flags); +#endif +if (!vha->hw->mpi_fw_dump) { +ql_log (ql_log_warn, vha,0x02f3 ,"-> mpi_fwdump no buffer\n" ); + }else if (vha->hw->mpi_fw_dumped) { +ql_log (ql_log_warn, vha,0x02f4 , +"-> MPI firmware already dumped (%p) -- ignoring request\n" , + vha->hw->mpi_fw_dump); + }else { +struct fwdt *fwdt = &vha->hw->fwdt[1 ]; +ulong len; +void *buf = vha->hw->mpi_fw_dump; + +ql_log (ql_log_warn, vha,0x02f5 ,"-> fwdt1 running...\n" ); +if (!fwdt->template) { +ql_log (ql_log_warn, vha,0x02f6 , +"-> fwdt1 no template\n" ); + goto bailout; + } + len =qla27xx_execute_fwdt_template (vha, fwdt->template, buf); +if (len ==0 ) { + goto bailout; + }else if (len != fwdt->dump_size) { +ql_log (ql_log_warn, vha,0x02f7 , +"-> fwdt1 fwdump residual=%+ld\n" , + fwdt->dump_size - len); + }else { + need_mpi_reset =0 ; + } + + vha->hw->mpi_fw_dump_len = len; + vha->hw->mpi_fw_dumped =1 ; + +ql_log (ql_log_warn, vha,0x02f8 , +"-> MPI firmware dump saved to buffer (%lu/%p)\n" , + vha->host_no, vha->hw->mpi_fw_dump); +qla2x00_post_uevent_work (vha,QLA_UEVENT_CODE_FW_DUMP ); + } + +bailout: +if (need_mpi_reset) +qla27xx_reset_mpi (vha); +#ifndef __CHECKER__ +if (!hardware_locked) +spin_unlock_irqrestore (&vha->hw->hardware_lock, flags); +#endif +} +void qla27xx_fwdump (scsi_qla_host_t *vha,int hardware_locked) {@@ -1015,30 +1098,26 @@ qla27xx_fwdump(scsi_qla_host_t *vha, int hardware_locked) vha->hw->fw_dump); }else {struct fwdt *fwdt = vha->hw->fwdt; -uint j;ulong len;void *buf = vha->hw->fw_dump;uint count = vha->hw->fw_dump_mpi ?2 :1 ; -for (j =0 ; j < count; j++, fwdt++, buf += len) { -ql_log (ql_log_warn, vha,0xd011 , -"-> fwdt%u running...\n" , j); -if (!fwdt->template) { -ql_log (ql_log_warn, vha,0xd012 , -"-> fwdt%u no template\n" , j); -break ; - } - len =qla27xx_execute_fwdt_template (vha, - fwdt->template, buf); -if (len ==0 ) { - goto bailout; - }else if (len != fwdt->dump_size) { -ql_log (ql_log_warn, vha,0xd013 , -"-> fwdt%u fwdump residual=%+ld\n" , - j, fwdt->dump_size - len); - } +ql_log (ql_log_warn, vha,0xd011 ,"-> fwdt0 running...\n" ); +if (!fwdt->template) { +ql_log (ql_log_warn, vha,0xd012 , +"-> fwdt0 no template\n" ); + goto bailout; } - vha->hw->fw_dump_len = buf - (void *)vha->hw->fw_dump; + len =qla27xx_execute_fwdt_template (vha, fwdt->template, buf); +if (len ==0 ) { + goto bailout; + }else if (len != fwdt->dump_size) { +ql_log (ql_log_warn, vha,0xd013 , +"-> fwdt0 fwdump residual=%+ld\n" , + fwdt->dump_size - len); + } + + vha->hw->fw_dump_len = len; vha->hw->fw_dumped =1 ; ql_log(ql_log_warn, vha, 0xd015,
drivers/scsi/qla2xxx/qla_tmpl.c qla27xx_fwdump(scsi_qla_host_t *vha, int hardware_locked) · line 1132
Quoted 7 lines
drivers/scsi/qla2xxx/qla_tmpl.c +0
-1@@ -1048,7 +1127,6 @@ qla27xx_fwdump(scsi_qla_host_t *vha, int hardware_locked) } bailout: - vha->hw->fw_dump_mpi =0 ;#ifndef __CHECKER__if (!hardware_locked)spin_unlock_irqrestore (&vha->hw->hardware_lock, flags);
Re: [PATCH 1/3] qla2xxx: Fix MPI failure AEN (8200) handling.
Quoted 451 lines · Himanshu Madhani
Quoted 450 lines · Nilesh Javali
From: Arun Easi <[email protected]>
Today, upon an MPI failure AEN, on top of collecting an MPI dump, a regular firmware dump is also taken and the chip reset. This is disruptive to IOs and not required. So, make the firmware dump collection, followed by chip reset optional (not done by default).
Firmware dump buffer and MPI dump buffer are independent of each other with this change and each can have dump that was taken at two different times for two different issues. The MPI dump is saved in a separate buffer and is retrieved differently from firmware dump.
To collect full dump on MPI failure AEN, a module parameter is introduced:
ql2xfulldump_on_mpifail(default: 0)Signed-off-by Arun Easi <[email protected]>Signed-off-by Himanshu Madhani <[email protected]>Signed-off-by Nilesh Javali <[email protected]>---
drivers/scsi/qla2xxx/qla_attr.c| 30 +++++++- drivers/scsi/qla2xxx/qla_def.h | 13 +++- drivers/scsi/qla2xxx/qla_gbl.h | 3 +drivers/scsi/qla2xxx/qla_init.c| 2 + drivers/scsi/qla2xxx/qla_isr.c | 54 ++++++++++----- drivers/scsi/qla2xxx/qla_os.c | 6 ++drivers/scsi/qla2xxx/qla_tmpl.c| 118 ++++++++++++++++++++++++++------7 files changed, 185 insertions(+), 41 deletions(-)drivers/scsi/qla2xxx/qla_attr.c +6
-1@@ -26,7 +26,8 @@ qla2x00_sysfs_read_fw_dump(struct file *filp, struct kobject *kobj, struct qla_hw_data *ha = vha->hw; int rval = 0; - if (!(ha->fw_dump_reading || ha->mctp_dump_reading)) + if (!(ha->fw_dump_reading || ha->mctp_dump_reading || + ha->mpi_fw_dump_reading)) return 0; mutex_lock(&ha->optrom_mutex);@@ -42,6 +43,10 @@ qla2x00_sysfs_read_fw_dump(struct file *filp,struct kobject *kobj, } else if (ha->mctp_dumped && ha->mctp_dump_reading) { rval = memory_read_from_buffer(buf, count, &off, ha->mctp_dump, MCTP_DUMP_SIZE); + } else if (ha->mpi_fw_dumped && ha->mpi_fw_dump_reading) { + rval = memory_read_from_buffer(buf, count, &off, + ha->mpi_fw_dump, + ha->mpi_fw_dump_len); } else if (ha->fw_dump_reading) { rval = memory_read_from_buffer(buf, count, &off, ha->fw_dump,
ha->fw_dump_len);drivers/scsi/qla2xxx/qla_attr.c +0
-1@@ -103,7 +108,6 @@ qla2x00_sysfs_write_fw_dump(struct file *filp,struct kobject *kobj, qla82xx_set_reset_owner(vha); qla8044_idc_unlock(ha); } else { - ha->fw_dump_mpi = 1; qla2x00_system_error(vha); }break;
drivers/scsi/qla2xxx/qla_attr.c +16
-0@@ -137,6 +141,22 @@ qla2x00_sysfs_write_fw_dump(struct file *filp,struct kobject *kobj, vha->host_no); } break; + case 8: + if (!ha->mpi_fw_dump_reading) + break; + ql_log(ql_log_info, vha, 0x70e7, + "MPI firmware dump cleared on (%ld).\n", vha->host_no); + ha->mpi_fw_dump_reading = 0; + ha->mpi_fw_dumped = 0; + break; + case 9: + if (ha->mpi_fw_dumped && !ha->mpi_fw_dump_reading) { + ha->mpi_fw_dump_reading = 1; + ql_log(ql_log_info, vha, 0x70e8, + "Raw MPI firmware dump ready for read on (%ld).\n", + vha->host_no); + } + break; } return count;}
drivers/scsi/qla2xxx/qla_attr.c +0
-0@@ -706,7 +726,8 @@ qla2x00_sysfs_write_reset(struct file *filp,struct kobject *kobj, scsi_unblock_requests(vha->host); break;case
0x2025d:
- if (!IS_QLA81XX(ha) && !IS_QLA83XX(ha)) + if (!IS_QLA81XX(ha) && !IS_QLA83XX(ha) && + !IS_QLA27XX(ha) && !IS_QLA28XX(ha)) return-EPERM; ql_log(ql_log_info, vha,0x706f,drivers/scsi/qla2xxx/qla_attr.c +2
-0@@ -724,6 +745,8 @@ qla2x00_sysfs_write_reset(struct file *filp,struct kobject *kobj, qla83xx_idc_audit(vha, IDC_AUDIT_TIMESTAMP); qla83xx_idc_unlock(vha, 0); break; + } else if (IS_QLA27XX(ha) || IS_QLA28XX(ha)) { + qla27xx_reset_mpi(vha); } else { /* Make sure FC side is not in reset */WARN_ON_ONCE(qla2x00_wait_for_hba_online(vha) !=
drivers/scsi/qla2xxx/qla_attr.c +1
-0@@ -737,6 +760,7 @@ qla2x00_sysfs_write_reset(struct file *filp,struct kobject *kobj, scsi_unblock_requests(vha->host); break; } + break;case
0x2025e:
if (!IS_P3P_TYPE(ha) || vha !=base_vha) { ql_log(ql_log_info, vha,0x7071,drivers/scsi/qla2xxx/qla_def.h +12
-1@@ -3223,6 +3223,7 @@ struct isp_operations { uint32_t); void (*fw_dump) (struct scsi_qla_host *, int); + void (*mpi_fw_dump)(struct scsi_qla_host *, int); int (*beacon_on) (struct scsi_qla_host *); int (*beacon_off) (struct scsi_qla_host *);@@ -3748,6 +3749,11 @@ struct qlt_hw_data { #define LEAK_EXCHG_THRESH_HOLD_PERCENT 75 /* 75 percent */ +struct qla_hw_data_stat { + u32 num_fw_dump; + u32 num_mpi_reset; +}; + /** Qlogic host adapter specific data structure. */ @@ -4230,7 +4236,6 @@ struct qla_hw_data { uint32_t fw_dump_len; u32 fw_dump_alloc_len; bool fw_dumped; - bool fw_dump_mpi; unsigned long fw_dump_cap_flags; #define RISC_PAUSE_CMPL 0 #define DMA_SHUTDOWN_CMPL 1@@ -4241,6 +4246,10 @@ struct qla_hw_data { #define ISP_MBX_RDY 6 #define ISP_SOFT_RESET_CMPL 7 int fw_dump_reading; + void *mpi_fw_dump; + u32 mpi_fw_dump_len; + int mpi_fw_dump_reading:1; + int mpi_fw_dumped:1; int prev_minidump_failed; dma_addr_t eft_dma; void *eft;@@ -4454,6 +4463,8 @@ struct qla_hw_data { uint16_t last_zio_threshold; #define DEFAULT_ZIO_THRESHOLD 5 + + struct qla_hw_data_stat stat; }; struct active_regions {drivers/scsi/qla2xxx/qla_gbl.h +3
-0@@ -173,6 +173,7 @@ extern int ql2xenablemsix; extern int qla2xuseresexchforels; extern int ql2xexlogins; extern int ql2xdifbundlinginternalbuffers; +extern int ql2xfulldump_on_mpifail; extern int qla2x00_loop_reset(scsi_qla_host_t *); extern void qla2x00_abort_all_cmds(scsi_qla_host_t *, int);@@ -645,6 +646,7 @@ extern void qla82xx_fw_dump(scsi_qla_host_t *, int); extern void qla8044_fw_dump(scsi_qla_host_t *, int); extern void qla27xx_fwdump(scsi_qla_host_t *, int); +extern void qla27xx_mpi_fwdump (scsi_qla_host_t *,int ); extern ulong qla27xx_fwdt_calculate_dump_size(struct scsi_qla_host *, void *); extern int qla27xx_fwdt_template_valid(void *); extern ulong qla27xx_fwdt_template_size(void *);@@ -933,5 +935,6 @@ extern void qla24xx_process_purex_list(struct purex_list *); /* nvme.c */ void qla_nvme_unregister_remote_port(struct fc_port *fcport); +void qla27xx_reset_mpi (scsi_qla_host_t *vha); void qla_handle_els_plogi_done(scsi_qla_host_t *vha, struct event_arg *ea); #endif /* _QLA_GBL_H */drivers/scsi/qla2xxx/qla_init.c +2
-0@@ -3339,6 +3339,8 @@ qla2x00_alloc_fw_dump(scsi_qla_host_t *vha) dump_size / 1024); if (IS_QLA27XX(ha) || IS_QLA28XX(ha)) { + ha->mpi_fw_dump = (char *)fw_dump + + ha->fwdt[1].dump_size; mutex_unlock(&ha->optrom_mutex); return; }drivers/scsi/qla2xxx/qla_isr.c +36
-3@@ -757,6 +757,39 @@ qla2x00_find_fcport_by_nportid(scsi_qla_host_t *vha, port_id_t *id, return NULL; } +/* Shall be called only on supported adapters. */ +static void +qla27xx_handle_8200_aen (scsi_qla_host_t *vha,uint16_t *mb) +{ + struct qla_hw_data *ha = vha->hw; + bool reset_isp_needed = 0; + + ql_log(ql_log_warn, vha, 0x02f0, + "MPI Heartbeat stop. MPI reset is%s needed. " + "MB0[%xh] MB1[%xh] MB2[%xh] MB3[%xh]\n", + mb[0] & BIT_8 ? "" : " not", + mb[0], mb[1], mb[2], mb[3]); + + if ((mb[1] & BIT_8) == 0) + return; + + ql_log(ql_log_warn, vha, 0x02f1, + "MPI Heartbeat stop. FW dump needed\n"); + + if (ql2xfulldump_on_mpifail) { + ha->isp_ops->fw_dump(vha, 1); + reset_isp_needed = 1; + } + + ha->isp_ops->mpi_fw_dump(vha, 1); + + if (reset_isp_needed) { + vha->hw->flags.fw_init_done = 0; + set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags); + qla2xxx_wake_dpc(vha); + } +} + /*** qla2x00_async_event() - Process aynchronous events. * @vha: SCSI driver HA context @@ -872,9 +905,9 @@ qla2x00_async_event(scsi_qla_host_t *vha, struct rsp_que *rsp, uint16_t *mb) "ISP System Error - mbx1=%xh mbx2=%xh mbx3=%xh.\n ", mb[1], mb[2], mb[3]); - ha->fw_dump_mpi = - (IS_QLA27XX(ha) || IS_QLA28XX(ha)) && - RD_REG_WORD(®24->mailbox7) & BIT_8; + if ((IS_QLA27XX(ha) || IS_QLA28XX(ha)) && + RD_REG_WORD(®24->mailbox7) & BIT_8) + ha->isp_ops->mpi_fw_dump(vha, 1); ha->isp_ops->fw_dump(vha, 1); ha->flags.fw_init_done = 0; QLA_FW_STOPPED(ha);@@ -1375,20 +1408,7 @@ qla2x00_async_event(scsi_qla_host_t *vha, struct rsp_que *rsp, uint16_t *mb)case
MBA_IDC_AEN:
if (IS_QLA27XX(ha)||IS_QLA28XX(ha)) {
-ha->flags.fw_init_done= 0;
- ql_log(ql_log_warn, vha,0xffff,
- "MPI Heartbeat stop. Chip reset needed. MB0[%xh] MB1[%xh] MB2[%xh] MB3[%xh]\n",
- mb[0], mb[1], mb[2], mb[3]);
-
- if ((mb[1] &BIT_8) ||
- (mb[2] &BIT_8)) {
- ql_log(ql_log_warn, vha,0xd013,
- "MPI Heartbeat stop. FW dump needed\n"); -ha->fw_dump_mpi= 1;
-ha->isp_ops->fw_dump(vha, 1);
- }
-set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags); -qla2xxx_wake_dpc(vha);
+qla27xx_handle_8200_aen(vha, mb);
} else if (IS_QLA83XX(ha)) {
mb[4] =RD_REG_WORD(®24->mailbox4); mb[5] =RD_REG_WORD(®24->mailbox5);drivers/scsi/qla2xxx/qla_os.c +6
-0@@ -35,6 +35,11 @@ static int apidev_major;*/ struct kmem_cache *srb_cachep; +int ql2xfulldump_on_mpifail; +module_param (ql2xfulldump_on_mpifail, int,S_IRUGO |S_IWUSR ); +MODULE_PARM_DESC (ql2xfulldump_on_mpifail, + "Set this to take full dump on MPI hang."); + /** CT6 CTX allocation cache */ @@ -2518,6 +2523,7 @@ static struct isp_operations qla27xx_isp_ops = { .read_nvram = NULL, .write_nvram = NULL, .fw_dump = qla27xx_fwdump, + .mpi_fw_dump = qla27xx_mpi_fwdump, .beacon_on = qla24xx_beacon_on, .beacon_off = qla24xx_beacon_off, .beacon_blink = qla83xx_beacon_blink,drivers/scsi/qla2xxx/qla_tmpl.c +27
-0@@ -12,6 +12,33 @@ #define IOBASE(vha) IOBAR(ISPREG(vha)) #define INVALID_ENTRY ((struct qla27xx_fwdt_entry *)0xffffffffffffffffUL) +/* hardware_lock assumed held. */ +static void +qla27xx_write_remote_reg (struct scsi_qla_host *vha, + u32 addr, u32 data) +{ + char *reg = (char *)ISPREG(vha); + + ql_dbg(ql_dbg_misc, vha, 0xd300, + "%s: addr/data = %xh/%xh\n", __func__, addr, data); + + WRT_REG_DWORD(reg + IOBASE(vha), 0x40); + WRT_REG_DWORD(reg + 0xc4, data); + WRT_REG_DWORD(reg + 0xc0, addr); +} + +void +qla27xx_reset_mpi (scsi_qla_host_t *vha) +{ + ql_dbg(ql_dbg_misc + ql_dbg_verbose, vha, 0xd301, + "Entered %s.\n", __func__); + + qla27xx_write_remote_reg(vha, 0x104050, 0x40004); + qla27xx_write_remote_reg(vha, 0x10405c, 0x4); + + vha->hw->stat.num_mpi_reset++; +} +static inline void
qla27xx_insert16(uint16_t value, void *buf, ulong *len){drivers/scsi/qla2xxx/qla_tmpl.c +56
-0@@ -997,6 +1024,62 @@ qla27xx_fwdt_template_valid(void *p) return true; } +void +qla27xx_mpi_fwdump (scsi_qla_host_t *vha,int hardware_locked) +{ + ulong flags = 0; + bool need_mpi_reset = 1; + +#ifndef __CHECKER__ + if (!hardware_locked) + spin_lock_irqsave(&vha->hw->hardware_lock, flags); +#endif + if (!vha->hw->mpi_fw_dump) { + ql_log(ql_log_warn, vha, 0x02f3, "-> mpi_fwdump no buffer\n"); + } else if (vha->hw->mpi_fw_dumped) { + ql_log(ql_log_warn, vha, 0x02f4, + "-> MPI firmware already dumped (%p) -- ignoring request\n", + vha->hw->mpi_fw_dump); + } else { + struct fwdt *fwdt = &vha->hw->fwdt[1]; + ulong len; + void *buf = vha->hw->mpi_fw_dump; + + ql_log(ql_log_warn, vha, 0x02f5, "-> fwdt1 running...\n"); + if (!fwdt->template) { + ql_log(ql_log_warn, vha, 0x02f6, + "-> fwdt1 no template\n"); + goto bailout; + } + len = qla27xx_execute_fwdt_template(vha, fwdt->template, buf); + if (len == 0) { + goto bailout; + } else if (len != fwdt->dump_size) { + ql_log(ql_log_warn, vha, 0x02f7, + "-> fwdt1 fwdump residual=%+ld\n", + fwdt->dump_size - len); + } else { + need_mpi_reset = 0; + } + + vha->hw->mpi_fw_dump_len = len; + vha->hw->mpi_fw_dumped = 1; + + ql_log(ql_log_warn, vha, 0x02f8, + "-> MPI firmware dump saved to buffer (%lu/%p)\n", + vha->host_no, vha->hw->mpi_fw_dump); + qla2x00_post_uevent_work(vha, QLA_UEVENT_CODE_FW_DUMP); + } + +bailout: + if (need_mpi_reset) + qla27xx_reset_mpi(vha); +#ifndef __CHECKER__ + if (!hardware_locked) + spin_unlock_irqrestore(&vha->hw->hardware_lock, flags); +#endif +} +void
qla27xx_fwdump(scsi_qla_host_t *vha, int hardware_locked){drivers/scsi/qla2xxx/qla_tmpl.c +15
-19@@ -1015,30 +1098,26 @@ qla27xx_fwdump(scsi_qla_host_t *vha, int hardware_locked) vha->hw->fw_dump); } else { struct fwdt *fwdt = vha->hw->fwdt; - uint j; ulong len; void *buf = vha->hw->fw_dump; uint count = vha->hw->fw_dump_mpi ? 2 : 1; - for (j = 0; j < count; j++, fwdt++, buf += len) { - ql_log(ql_log_warn, vha, 0xd011, - "-> fwdt%u running...\n", j); - if (!fwdt->template) { - ql_log(ql_log_warn, vha, 0xd012, - "-> fwdt%u no template\n", j); - break; - } - len = qla27xx_execute_fwdt_template(vha, - fwdt->template, buf); - if (len == 0) { - goto bailout; - } else if (len != fwdt->dump_size) { - ql_log(ql_log_warn, vha, 0xd013, - "-> fwdt%u fwdump residual=%+ld\n", - j, fwdt->dump_size - len); - } + ql_log(ql_log_warn, vha, 0xd011, "-> fwdt0 running...\n"); + if (!fwdt->template) { + ql_log(ql_log_warn, vha, 0xd012, + "-> fwdt0 no template\n"); + goto bailout; } - vha->hw->fw_dump_len = buf - (void *)vha->hw->fw_dump; + len = qla27xx_execute_fwdt_template(vha, fwdt->template, buf); + if (len == 0) { + goto bailout; + } else if (len != fwdt->dump_size) { + ql_log(ql_log_warn, vha, 0xd013, + "-> fwdt0 fwdump residual=%+ld\n", + fwdt->dump_size - len); + } + + vha->hw->fw_dump_len = len; vha->hw->fw_dumped = 1; ql_log(ql_log_warn, vha, 0xd015,@@ -1048,7 +1127,6 @@ qla27xx_fwdump(scsi_qla_host_t *vha, int hardware_locked) }
Quoted 8 lines
Quoted 6 lines
bailout:
-vha->hw->fw_dump_mpi= 0;
#ifndef__CHECKER__
if (!hardware_locked)spin_unlock_irqrestore(&vha->hw->hardware_lock, flags);Reviewed-by Himanshu Madhani <[email protected]>
Nilesh,
FYI, I have reviewed 1/3 as well so please go ahead add my reviewed-by tag.
Thanks,
- Himanshu
Nilesh Javali
maintainerPATCH 2/3 qla2xxx: Fix hang when issuing nvme disconnect-all in NPIV.
In reply to Nilesh JavaliFrom: Arun Easi <[email protected]>
In NPIV environment, a NPIV host may use a queue pair created by base host or other NPIVs, so the check for a queue pair created by this NPIV is not correct, and can cause an abort to fail, which in turn means the NVME command not returned. This leads to hang in nvme_fc layer in nvme_fc_delete_association() which waits for all I/Os to be returned, which is seen as hang in the application.
--- drivers/scsi/qla2xxx/qla_mbx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/qla2xxx/qla_mbx.c +1 -1
@@ -3153,7 +3153,7 @@ qla24xx_abort_command(srb_t *sp)ql_dbg (ql_dbg_mbx + ql_dbg_verbose, vha,0x108c ,"Entered %s.\n" , __func__); -if (vha->flags.qpairs_available && sp->qpair) +if (sp->qpair) req = sp->qpair->req;else return QLA_FUNCTION_FAILED ;
Signature · 1 line
2.19.0.rc0
Re: [PATCH 2/3] qla2xxx: Fix hang when issuing nvme disconnect-all in NPIV.
Quoted 22 lines · Nilesh Javali
From: Arun Easi <[email protected]>
In NPIV environment, a NPIV host may use a queue pair created by base host or other NPIVs, so the check for a queue pair created by this NPIV is not correct, and can cause an abort to fail, which in turn means the NVME command not returned. This leads to hang in
nvme_fclayer innvme_fc_delete_association()which waits for all I/Os to be returned, which is seen as hang in the application.Signed-off-by Arun Easi <[email protected]>Signed-off-by Nilesh Javali <[email protected]>--- drivers/scsi/qla2xxx/qla_mbx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/scsi/qla2xxx/qla_mbx.c +0
-0
drivers/scsi/qla2xxx/qla_mbx.c qla24xx_abort_command(srb_t *sp) · line 3159
Quoted 8 lines
drivers/scsi/qla2xxx/qla_mbx.c +1
-1@@ -3153,7 +3153,7 @@ qla24xx_abort_command(srb_t *sp)ql_dbg (ql_dbg_mbx + ql_dbg_verbose, vha,0x108c ,"Entered %s.\n" , __func__); -if (vha->flags.qpairs_available && sp->qpair) +if (sp->qpair) req = sp->qpair->req;else return QLA_FUNCTION_FAILED ;
Nilesh Javali
maintainerPATCH 3/3 qla2xxx: delete all sessions before unregister local nvme port
In reply to Nilesh JavaliFrom: Quinn Tran <[email protected]>
Delete all sessions before unregistering local nvme port. This allows nvme layer to decrement all active rport count down to zero. Once the count is down to zero, nvme would call qla to continue with the npiv port deletion.
PID: 27448 TASK: ffff9e34b777c1c0 CPU: 0 COMMAND: "qaucli"
0 [ffff9e25e84abbd8] __schedule at ffffffff977858ca
1 [ffff9e25e84abc68] schedule at ffffffff97785d79
2 [ffff9e25e84abc78] schedule_timeout at ffffffff97783881
3 [ffff9e25e84abd28] wait_for_completion at ffffffff9778612d
4 [ffff9e25e84abd88] qla_nvme_delete at ffffffffc0e3024e [qla2xxx]
5 [ffff9e25e84abda8] qla24xx_vport_delete at ffffffffc0e024b9 [qla2xxx]
6 [ffff9e25e84abdf0] fc_vport_terminate at ffffffffc011c247 [scsi_transport_fc] 7 [ffff9e25e84abe28] store_fc_host_vport_delete at ffffffffc011cd94 [scsi_transport_fc] 8 [ffff9e25e84abe70] dev_attr_store at ffffffff974b376b
9 [ffff9e25e84abe80] sysfs_kf_write at ffffffff972d9a92
10 [ffff9e25e84abe90] kernfs_fop_write at ffffffff972d907b
11 [ffff9e25e84abec8] vfs_write at ffffffff9724c790
12 [ffff9e25e84abf08] sys_write at ffffffff9724d55f
13 [ffff9e25e84abf50] system_call_fastpath at ffffffff97792ed2
RIP: 00007fc0bd81a6fd RSP: 00007ffff78d9648 RFLAGS: 00010202
RAX: 0000000000000001 RBX: 0000000000000022 RCX: 00007ffff78d96e0
RDX: 0000000000000022 RSI: 00007ffff78d94e0 RDI: 0000000000000008
RBP: 00007ffff78d9440 R8: 0000000000000000 R9: 00007fc0bd48b2cd
R10: 0000000000000017 R11: 0000000000000293 R12: 0000000000000000
R13: 00005624e4dac840 R14: 00005624e4da9a10 R15: 0000000000000000
ORIG_RAX: 0000000000000001 CS: 0033 SS: 002b--- drivers/scsi/qla2xxx/qla_attr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/qla2xxx/qla_attr.c +1 -1
@@ -3055,11 +3055,11 @@ qla24xx_vport_delete(struct fc_vport *fc_vport) test_bit(FCPORT_UPDATE_NEEDED, &vha->dpc_flags))msleep (1000 ); -qla_nvme_delete (vha);qla24xx_disable_vp (vha);qla2x00_wait_for_sess_deletion (vha); +qla_nvme_delete (vha); vha->flags.delete_progress =1 ;qlt_remove_target (ha, vha);
Signature · 1 line
2.19.0.rc0
Re: [PATCH 3/3] qla2xxx: delete all sessions before unregister local nvme port
Quoted 44 lines · Nilesh Javali
From: Quinn Tran <[email protected]>
Delete all sessions before unregistering local nvme port. This allows nvme layer to decrement all active rport count down to zero. Once the count is down to zero, nvme would call qla to continue with the npiv port deletion.
PID: 27448 TASK:
ffff9e34b777c1c0CPU: 0 COMMAND: "qaucli"
0 [ffff9e25e84abbd8]__scheduleatffffffff977858ca
1 [ffff9e25e84abc68] schedule atffffffff97785d79
2 [ffff9e25e84abc78]schedule_timeoutatffffffff97783881
3 [ffff9e25e84abd28]wait_for_completionatffffffff9778612d
4 [ffff9e25e84abd88]qla_nvme_deleteatffffffffc0e3024e[qla2xxx]
5 [ffff9e25e84abda8]qla24xx_vport_deleteatffffffffc0e024b9[qla2xxx]
6 [ffff9e25e84abdf0]fc_vport_terminateatffffffffc011c247[scsi_transport_fc] 7 [ffff9e25e84abe28]store_fc_host_vport_deleteatffffffffc011cd94[scsi_transport_fc] 8 [ffff9e25e84abe70]dev_attr_storeatffffffff974b376b
9 [ffff9e25e84abe80]sysfs_kf_writeatffffffff972d9a92
10 [ffff9e25e84abe90]kernfs_fop_writeatffffffff972d907b
11 [ffff9e25e84abec8]vfs_writeatffffffff9724c790
12 [ffff9e25e84abf08]sys_writeatffffffff9724d55f
13 [ffff9e25e84abf50]system_call_fastpathatffffffff97792ed2RIP: 00007fc0bd81a6fd RSP: 00007ffff78d9648 RFLAGS: 00010202 RAX: 0000000000000001 RBX: 0000000000000022 RCX: 00007ffff78d96e0 RDX: 0000000000000022 RSI: 00007ffff78d94e0 RDI: 0000000000000008 RBP: 00007ffff78d9440 R8: 0000000000000000 R9: 00007fc0bd48b2cd R10: 0000000000000017 R11: 0000000000000293 R12: 0000000000000000 R13: 00005624e4dac840 R14: 00005624e4da9a10 R15: 0000000000000000 ORIG_RAX: 0000000000000001 CS: 0033 SS: 002bSigned-off-by Quinn Tran <[email protected]>Signed-off-by Nilesh Javali <[email protected]>--- drivers/scsi/qla2xxx/qla_attr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/scsi/qla2xxx/qla_attr.c +0
-0@@ -3055,11 +3055,11 @@ qla24xx_vport_delete(struct fc_vport *fc_vport) test_bit(FCPORT_UPDATE_NEEDED, &vha->dpc_flags))msleep (1000 );
drivers/scsi/qla2xxx/qla_attr.c qla24xx_vport_delete(struct fc_vport *fc_vport) · line 3065
Quoted 10 lines
drivers/scsi/qla2xxx/qla_attr.c +1
-1-qla_nvme_delete (vha);qla24xx_disable_vp (vha);qla2x00_wait_for_sess_deletion (vha); +qla_nvme_delete (vha); vha->flags.delete_progress =1 ;qlt_remove_target (ha, vha);
Nilesh Javali
maintainerHi Martin,
The patch 1/3 looks incomplete. I will send the v2 version of the series. Please ignore this series.
Thanks,
Nilesh
-----Original Message-----
From: Nilesh Javali <[email protected]>
Sent: Friday, March 27, 2020 3:57 PM
To: [email protected]; [email protected]
Cc: [email protected]; GR-QLogic-Storage-Upstream <[email protected]>
Subject: [PATCH 0/3] qla2xxx: Updates for the driver
Hi Martin,
Please apply the qla2xxx driver bug fixes to the scsi tree at your earliest convenience.
Thanks,
Nilesh
Arun Easi (2):
qla2xxx: Fix MPI failure AEN (8200) handling.
qla2xxx: Fix hang when issuing nvme disconnect-all in NPIV.
Quinn Tran (1):
qla2xxx: delete all sessions before unregister local nvme port
drivers/scsi/qla2xxx/qla_attr.c | 32 +++++++-- drivers/scsi/qla2xxx/qla_def.h | 13 +++- drivers/scsi/qla2xxx/qla_gbl.h | 3 + drivers/scsi/qla2xxx/qla_init.c | 2 + drivers/scsi/qla2xxx/qla_isr.c | 54 ++++++++++----- drivers/scsi/qla2xxx/qla_mbx.c | 2 +- drivers/scsi/qla2xxx/qla_os.c | 6 ++ drivers/scsi/qla2xxx/qla_tmpl.c | 118 ++++++++++++++++++++++++++------ 8 files changed, 187 insertions(+), 43 deletions(-)
Signature · 1 line
2.19.0.rc0