[PATCH v2] ufs: core: wlun resume SSU(Acitve) fail recovery

[peter.wang at mediatek.com]

From: Peter Wang <peter.wang at mediatek.com>

When wlun resume SSU(Active) timeout, scsi try eh_host_reset_handler.
But ufshcd_eh_host_reset_handler hang at wait flush_work(&hba->eh_work).
And ufshcd_err_handler hang at wait rpm resume.
Do link recovery only in this case. Below is IO hang stack dump.

<ffffffdd78e02b34> schedule+0x110/0x204
<ffffffdd78e0be60> schedule_timeout+0x98/0x138
<ffffffdd78e040e8> wait_for_common_io+0x130/0x2d0
<ffffffdd77d6a000> blk_execute_rq+0x10c/0x16c
<ffffffdd78126d90> __scsi_execute+0xfc/0x278
<ffffffdd7813891c> ufshcd_set_dev_pwr_mode+0x1c8/0x40c
<ffffffdd78137d1c> __ufshcd_wl_resume+0xf0/0x5cc
<ffffffdd78137ae0> ufshcd_wl_runtime_resume+0x40/0x18c
<ffffffdd78136108> scsi_runtime_resume+0x88/0x104
<ffffffdd7809a4f8> __rpm_callback+0x1a0/0xaec
<ffffffdd7809b624> rpm_resume+0x7e0/0xcd0
<ffffffdd7809a788> __rpm_callback+0x430/0xaec
<ffffffdd7809b644> rpm_resume+0x800/0xcd0
<ffffffdd780a0778> pm_runtime_work+0x148/0x198

<ffffffdd78e02b34> schedule+0x110/0x204
<ffffffdd78e0be10> schedule_timeout+0x48/0x138
<ffffffdd78e03d9c> wait_for_common+0x144/0x2dc
<ffffffdd7758bba4> __flush_work+0x3d0/0x508
<ffffffdd7815572c> ufshcd_eh_host_reset_handler+0x134/0x3a8
<ffffffdd781216f4> scsi_try_host_reset+0x54/0x204
<ffffffdd78120594> scsi_eh_ready_devs+0xb30/0xd48
<ffffffdd7812373c> scsi_error_handler+0x260/0x874

<ffffffdd78e02b34> schedule+0x110/0x204
<ffffffdd7809af64> rpm_resume+0x120/0xcd0
<ffffffdd7809fde8> __pm_runtime_resume+0xa0/0x17c
<ffffffdd7815193c> ufshcd_err_handling_prepare+0x40/0x430
<ffffffdd7814cce8> ufshcd_err_handler+0x1c4/0xd4c

Signed-off-by: Peter Wang <peter.wang at mediatek.com>
---
 drivers/ufs/core/ufshcd.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
index e18c9f4463ec..de20cf587fa1 100644
--- a/drivers/ufs/core/ufshcd.c
+++ b/drivers/ufs/core/ufshcd.c
@@ -7363,9 +7363,27 @@ static int ufshcd_eh_host_reset_handler(struct scsi_cmnd *cmd)
 	int err = SUCCESS;
 	unsigned long flags;
 	struct ufs_hba *hba;
+	struct device *dev;
 
 	hba = shost_priv(cmd->device->host);
 
+	/*
+	 * If __ufshcd_wl_suspend get fail and runtime_status = RPM_RESUMING,
+	 * do link recovery only. Because schedule eh work will get dead lock
+	 * in ufshcd_rpm_get_sync to wait wlun resume, but wlun resume get
+	 * error and wait eh work finish.
+	 */
+	dev = &hba->ufs_device_wlun->sdev_gendev;
+	if (dev->power.runtime_status == RPM_RESUMING) {
+		err = ufshcd_link_recovery(hba);
+		if (err) {
+			dev_err(hba->dev, "WL Device PM: status:%d, err:%d\n",
+				dev->power.runtime_status,
+				dev->power.runtime_error);
+		}
+		return err;
+	}
+
 	spin_lock_irqsave(hba->host->host_lock, flags);
 	hba->force_reset = true;
 	ufshcd_schedule_eh_work(hba);
-- 
2.18.0