[PATCH 03/11] scsi: hisi_sas: check host frozen before calling "done" function

From: John Garry
Date: Wed May 02 2018 - 11:59:28 EST


From: Xiang Chen <chenxiang66@xxxxxxxxxxxxx>

When the host is frozen in SCSI EH state, at any point after
the LLDD sets SAS_TASK_STATE_DONE for the sas_task task state,
libsas may free the task; see sas_scsi_find_task().

This puts the LLDD in a difficult position, in that once it
sets SAS_TASK_STATE_DONE for the task state it should not
reference the sas_task again. But the LLDD needs will check
the sas_task indirectly in calling
task->task_done()->sas_scsi_task_done() or sas_ata_task_done()
(to check if the host is frozen state actually).

And the LLDD cannot set SAS_TASK_STATE_DONE for the task
state after task->task_done() is called (as the sas_task is
free'd at this point).

This situation would seem to be a problem made by libsas.

To work around, check in the LLDD whether the host is in
frozen state to ensure it is ok to call task->task_done()
function. If in the frozen state, we rely on SCSI EH and
libsas to free the sas_task directly.

We do not do this for the following IO types:
- SMP - they are managed in libsas directly, outside SCSI EH
- Any internally originated IO, for similar reason

Signed-off-by: Xiang Chen <chenxiang66@xxxxxxxxxxxxx>
Signed-off-by: John Garry <john.garry@xxxxxxxxxx>
---
drivers/scsi/hisi_sas/hisi_sas.h | 1 +
drivers/scsi/hisi_sas/hisi_sas_main.c | 3 +++
drivers/scsi/hisi_sas/hisi_sas_v2_hw.c | 14 ++++++++++++++
drivers/scsi/hisi_sas/hisi_sas_v3_hw.c | 14 ++++++++++++++
4 files changed, 32 insertions(+)

diff --git a/drivers/scsi/hisi_sas/hisi_sas.h b/drivers/scsi/hisi_sas/hisi_sas.h
index d413d05..147cfaf 100644
--- a/drivers/scsi/hisi_sas/hisi_sas.h
+++ b/drivers/scsi/hisi_sas/hisi_sas.h
@@ -196,6 +196,7 @@ struct hisi_sas_slot {
dma_addr_t cmd_hdr_dma;
struct work_struct abort_slot;
struct timer_list internal_abort_timer;
+ bool is_internal;
};

struct hisi_sas_tmf_task {
diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c
index 52746e2..8f8e642 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_main.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_main.c
@@ -382,6 +382,8 @@ static int hisi_sas_task_prep(struct sas_task *task, struct hisi_sas_dq
slot->cmd_hdr = &cmd_hdr_base[dlvry_queue_slot];
slot->task = task;
slot->port = port;
+ if (is_tmf)
+ slot->is_internal = true;
task->lldd_task = slot;
INIT_WORK(&slot->abort_slot, hisi_sas_slot_abort);

@@ -1486,6 +1488,7 @@ static int hisi_sas_query_task(struct sas_task *task)
slot->cmd_hdr = &cmd_hdr_base[dlvry_queue_slot];
slot->task = task;
slot->port = port;
+ slot->is_internal = true;
task->lldd_task = slot;

slot->buf = dma_pool_alloc(hisi_hba->buffer_pool,
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index 8ca0044..6dda6eb 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -2380,18 +2380,21 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
struct device *dev = hisi_hba->dev;
struct task_status_struct *ts;
struct domain_device *device;
+ struct sas_ha_struct *ha;
enum exec_status sts;
struct hisi_sas_complete_v2_hdr *complete_queue =
hisi_hba->complete_hdr[slot->cmplt_queue];
struct hisi_sas_complete_v2_hdr *complete_hdr =
&complete_queue[slot->cmplt_queue_slot];
unsigned long flags;
+ bool is_internal = slot->is_internal;

if (unlikely(!task || !task->lldd_task || !task->dev))
return -EINVAL;

ts = &task->task_status;
device = task->dev;
+ ha = device->port->ha;
sas_dev = device->lldd_dev;

spin_lock_irqsave(&task->task_state_lock, flags);
@@ -2523,6 +2526,17 @@ static void slot_err_v2_hw(struct hisi_hba *hisi_hba,
task->task_state_flags |= SAS_TASK_STATE_DONE;
spin_unlock_irqrestore(&task->task_state_lock, flags);

+ if (!is_internal && (task->task_proto != SAS_PROTOCOL_SMP)) {
+ spin_lock_irqsave(&device->done_lock, flags);
+ if (test_bit(SAS_HA_FROZEN, &ha->state)) {
+ spin_unlock_irqrestore(&device->done_lock, flags);
+ dev_info(dev, "slot complete: task(%p) ignored\n ",
+ task);
+ return sts;
+ }
+ spin_unlock_irqrestore(&device->done_lock, flags);
+ }
+
if (task->task_done)
task->task_done(task);

diff --git a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
index 7346110..5c0d968 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
@@ -1571,18 +1571,21 @@ static irqreturn_t fatal_axi_int_v3_hw(int irq_no, void *p)
struct device *dev = hisi_hba->dev;
struct task_status_struct *ts;
struct domain_device *device;
+ struct sas_ha_struct *ha;
enum exec_status sts;
struct hisi_sas_complete_v3_hdr *complete_queue =
hisi_hba->complete_hdr[slot->cmplt_queue];
struct hisi_sas_complete_v3_hdr *complete_hdr =
&complete_queue[slot->cmplt_queue_slot];
unsigned long flags;
+ bool is_internal = slot->is_internal;

if (unlikely(!task || !task->lldd_task || !task->dev))
return -EINVAL;

ts = &task->task_status;
device = task->dev;
+ ha = device->port->ha;
sas_dev = device->lldd_dev;

spin_lock_irqsave(&task->task_state_lock, flags);
@@ -1700,6 +1703,17 @@ static irqreturn_t fatal_axi_int_v3_hw(int irq_no, void *p)
task->task_state_flags |= SAS_TASK_STATE_DONE;
spin_unlock_irqrestore(&task->task_state_lock, flags);

+ if (!is_internal && (task->task_proto != SAS_PROTOCOL_SMP)) {
+ spin_lock_irqsave(&device->done_lock, flags);
+ if (test_bit(SAS_HA_FROZEN, &ha->state)) {
+ spin_unlock_irqrestore(&device->done_lock, flags);
+ dev_info(dev, "slot complete: task(%p) ignored\n ",
+ task);
+ return sts;
+ }
+ spin_unlock_irqrestore(&device->done_lock, flags);
+ }
+
if (task->task_done)
task->task_done(task);

--
1.9.1