@@ -8,7 +8,7 @@
static inline bool ufshcd_is_user_access_allowed(struct ufs_hba *hba)
{
- return !hba->shutting_down;
+ return !atomic_read(&hba->shutting_down);
}
void ufshcd_schedule_eh_work(struct ufs_hba *hba);
@@ -1729,12 +1729,10 @@ static ssize_t ufshcd_clkscale_enable_store(struct device *dev,
if (kstrtou32(buf, 0, &value))
return -EINVAL;
- down(&hba->host_sem);
- if (!ufshcd_is_user_access_allowed(hba)) {
- err = -EBUSY;
- goto out;
- }
+ if (!ufshcd_is_user_access_allowed(hba))
+ return -EBUSY;
+ down(&hba->host_sem);
value = !!value;
if (value == hba->clk_scaling.is_enabled)
goto out;
@@ -6416,8 +6414,7 @@ static void ufshcd_err_handling_unprepare(struct ufs_hba *hba)
static inline bool ufshcd_err_handling_should_stop(struct ufs_hba *hba)
{
- return (!hba->is_powered || hba->shutting_down ||
- !hba->ufs_device_wlun ||
+ return (!hba->is_powered || !hba->ufs_device_wlun ||
hba->ufshcd_state == UFSHCD_STATE_ERROR ||
(!(hba->saved_err || hba->saved_uic_err || hba->force_reset ||
ufshcd_is_link_broken(hba))));
@@ -6541,10 +6538,13 @@ static void ufshcd_err_handler(struct work_struct *work)
dev_info(hba->dev,
"%s started; HBA state %s; powered %d; shutting down %d; saved_err = %d; saved_uic_err = %d; force_reset = %d%s\n",
__func__, ufshcd_state_name[hba->ufshcd_state],
- hba->is_powered, hba->shutting_down, hba->saved_err,
+ hba->is_powered, atomic_read(&hba->shutting_down), hba->saved_err,
hba->saved_uic_err, hba->force_reset,
ufshcd_is_link_broken(hba) ? "; link is broken" : "");
+ if (!ufshcd_is_user_access_allowed(hba))
+ return;
+
down(&hba->host_sem);
spin_lock_irqsave(hba->host->host_lock, flags);
if (ufshcd_err_handling_should_stop(hba)) {
@@ -10194,10 +10194,7 @@ static void ufshcd_wl_shutdown(struct device *dev)
struct scsi_device *sdev = to_scsi_device(dev);
struct ufs_hba *hba = shost_priv(sdev->host);
- down(&hba->host_sem);
- hba->shutting_down = true;
- up(&hba->host_sem);
-
+ atomic_set(&hba->shutting_down, 1);
/* Turn on everything while shutting down */
ufshcd_rpm_get_sync(hba);
scsi_device_quiesce(sdev);
@@ -1034,7 +1034,7 @@ struct ufs_hba {
u16 ee_usr_mask;
struct mutex ee_ctrl_mutex;
bool is_powered;
- bool shutting_down;
+ atomic_t shutting_down;
struct semaphore host_sem;
/* Work Queues */
There is a deadlock when suspend, shutdown and resume are ocurred almost simultaneously. Here is the deadlock cause: 1. System enter suspend flow, it will call ufshcd_wl_suspend() to get host_sem, the host_sem is supposed to up in ufshcd_wl_resume: pm_suspend ufshcd_wl_suspend down(&hba->host_sem) 2. Shutdown is happened due to low battery during suspend progress, it will get ufshcd_wl device mutex and blocked in down host_sem in ufshcd_wl_shutdown() unfortunately. __switch_to+0x174/0x338 __schedule+0x608/0x9f0 schedule+0x7c/0xe8 schedule_timeout+0x44/0x1c8 __down_common+0xbc/0x238 __down+0x18/0x28 down+0x50/0x54 ufshcd_wl_shutdown+0x28/0xb4 device_shutdown+0x1a0/0x254 kernel_power_off+0x3c/0xf0 power_misc_routine_thread+0x814/0x970 [mt6375_battery] kthread+0x104/0x1d4 ret_from_fork+0x10/0x20 3. Meanwhile, suspend flow is waked up by a interrupt and go to resume work, the resume work will be blocked in getting ufshcd_wl device mutex. __switch_to+0x174/0x338 __schedule+0x608/0x9f0 schedule+0x7c/0xe8 schedule_preempt_disabled+0x24/0x40 __mutex_lock+0x408/0xdac __mutex_lock_slowpath+0x14/0x24 mutex_lock+0x40/0xec __device_resume+0x50/0x360 async_resume+0x24/0x3c async_run_entry_fn+0x44/0x118 process_one_work+0x1e4/0x43c worker_thread+0x25c/0x430 kthread+0x104/0x1d4 ret_from_fork+0x10/0x20 Fix the deadlock issue by define an atomic value of shutting_down to check if shutdown has been invoked. Signed-off-by: qiwu.chen <qiwu.chen@transsion.com> --- drivers/ufs/core/ufshcd-priv.h | 2 +- drivers/ufs/core/ufshcd.c | 21 +++++++++------------ include/ufs/ufshcd.h | 2 +- 3 files changed, 11 insertions(+), 14 deletions(-)