scsi: pm80xx: Add controller SCSI host fatal error uevents

Add pm80xx_fatal_error_uevent_emit() which is called when the pm80xx
driver encouters a fatal error. The uevent has the following additional
custom key/value pair sets:

 - DRIVER: driver name, pm80xx in this case
 - HBA_NUM: the scsi host id of the device
 - EVENT_TYPE: to indicate a fatal error
 - REPORTED_BY: either driver or firmware

The uevent is anchored to the kernel object that represents the SCSI
controller, which includes other useful core variables, such as, ACTION,
DEVPATH, SUBSYSTEM, and more.

The fatal_error_uevent_emit() function is called when the controller
fatal error state changes. Since this doesn't happen often for a
specific SCSI host, there is no worries of a uevent storm.

Signed-off-by: Salomon Dushimirimana <salomondush@google.com>
Link: https://lore.kernel.org/r/20250616190018.2136260-1-salomondush@google.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

diff --git a/drivers/scsi/pm8001/pm8001_sas.h b/drivers/scsi/pm8001/pm8001_sas.h
index 315f6a7523f09897147e57b23e4ed6bd51d68aec..334485bb2c12d3aeabbcb36ca623cb6c690c01ed 100644 (file)
--- a/drivers/scsi/pm8001/pm8001_sas.h
+++ b/drivers/scsi/pm8001/pm8001_sas.h
@@ -170,6 +170,14 @@ struct forensic_data {
 #define SPCV_MSGU_CFG_TABLE_TRANSFER_DEBUG_INFO  0x80
 #define MAIN_MERRDCTO_MERRDCES                  0xA0/* DWORD 0x28) */
 
+/**
+ * enum fatal_error_reporter: Indicates the originator of the fatal error
+ */
+enum fatal_error_reporter {
+       REPORTER_DRIVER,
+       REPORTER_FIRMWARE,
+};
+
 struct pm8001_dispatch {
        char *name;
        int (*chip_init)(struct pm8001_hba_info *pm8001_ha);
@@ -715,6 +723,8 @@ ssize_t pm80xx_get_non_fatal_dump(struct device *cdev,
                struct device_attribute *attr, char *buf);
 ssize_t pm8001_get_gsm_dump(struct device *cdev, u32, char *buf);
 int pm80xx_fatal_errors(struct pm8001_hba_info *pm8001_ha);
+void pm80xx_fatal_error_uevent_emit(struct pm8001_hba_info *pm8001_ha,
+       enum fatal_error_reporter error_reporter);
 void pm8001_free_dev(struct pm8001_device *pm8001_dev);
 /* ctl shared API */
 extern const struct attribute_group *pm8001_host_groups[];

diff --git a/drivers/scsi/pm8001/pm80xx_hwi.c b/drivers/scsi/pm8001/pm80xx_hwi.c
index c4074f062d93111f64ae2acdc8a916e27b034636..c1bae995a412844081682da71af96bec447ce7d7 100644 (file)
--- a/drivers/scsi/pm8001/pm80xx_hwi.c
+++ b/drivers/scsi/pm8001/pm80xx_hwi.c
@@ -1551,6 +1551,52 @@ static int mpi_uninit_check(struct pm8001_hba_info *pm8001_ha)
        return 0;
 }
 
+/**
+ * pm80xx_fatal_error_uevent_emit - emits a single fatal error uevent
+ * @pm8001_ha: our hba card information
+ * @error_reporter: reporter of fatal error
+ */
+void pm80xx_fatal_error_uevent_emit(struct pm8001_hba_info *pm8001_ha,
+       enum fatal_error_reporter error_reporter)
+{
+       struct kobj_uevent_env *env;
+
+       pm8001_dbg(pm8001_ha, FAIL, "emitting fatal error uevent");
+
+       env = kzalloc(sizeof(struct kobj_uevent_env), GFP_KERNEL);
+       if (!env)
+               return;
+
+       if (add_uevent_var(env, "DRIVER=%s", DRV_NAME))
+               goto exit;
+
+       if (add_uevent_var(env, "HBA_NUM=%u", pm8001_ha->id))
+               goto exit;
+
+       if (add_uevent_var(env, "EVENT_TYPE=FATAL_ERROR"))
+               goto exit;
+
+       switch (error_reporter) {
+       case REPORTER_DRIVER:
+               if (add_uevent_var(env, "REPORTED_BY=DRIVER"))
+                       goto exit;
+               break;
+       case REPORTER_FIRMWARE:
+               if (add_uevent_var(env, "REPORTED_BY=FIRMWARE"))
+                       goto exit;
+               break;
+       default:
+               if (add_uevent_var(env, "REPORTED_BY=OTHER"))
+                       goto exit;
+               break;
+       }
+
+       kobject_uevent_env(&pm8001_ha->shost->shost_dev.kobj, KOBJ_CHANGE, env->envp);
+
+exit:
+       kfree(env);
+}
+
 /**
  * pm80xx_fatal_errors - returns non-zero *ONLY* when fatal errors
  * @pm8001_ha: our hba card information
@@ -1580,6 +1626,7 @@ pm80xx_fatal_errors(struct pm8001_hba_info *pm8001_ha)
                        "Fatal error SCRATCHPAD1 = 0x%x SCRATCHPAD2 = 0x%x SCRATCHPAD3 = 0x%x SCRATCHPAD_RSVD0 = 0x%x SCRATCHPAD_RSVD1 = 0x%x\n",
                                scratch_pad1, scratch_pad2, scratch_pad3,
                                scratch_pad_rsvd0, scratch_pad_rsvd1);
+               pm80xx_fatal_error_uevent_emit(pm8001_ha, REPORTER_DRIVER);
                ret = 1;
        }
 
@@ -4039,6 +4086,7 @@ static int process_oq(struct pm8001_hba_info *pm8001_ha, u8 vec)
                        pm8001_dbg(pm8001_ha, FAIL,
                                   "Firmware Fatal error! Regval:0x%x\n",
                                   regval);
+                       pm80xx_fatal_error_uevent_emit(pm8001_ha, REPORTER_FIRMWARE);
                        pm8001_handle_event(pm8001_ha, NULL, IO_FATAL_ERROR);
                        print_scratchpad_registers(pm8001_ha);
                        return ret;