diff mbox series

[RFC,v2,2/2] scsi: ufs: Fix deadlock while suspending ufs host

Message ID d50e7620c47109ea7664dd9ca4144fc0c7c8502d.1611803264.git.asutoshd@codeaurora.org
State New
Headers show
Series Fix deadlock in ufs | expand

Commit Message

Asutosh Das (asd) Jan. 28, 2021, 3:26 a.m. UTC 
During runtime-suspend of ufs host, the scsi devices are
already suspended and so are the queues associated with them.
But the ufs host sends SSU to wlun during its runtime-suspend.
During the process blk_queue_enter checks if the queue is not in
suspended state. If so, it waits for the queue to resume, and never
comes out of it.
The commit
(d55d15a33: scsi: block: Do not accept any requests while suspended)
adds the check if the queue is in suspended state in blk_queue_enter().

Fix this, by decoupling wlun scsi devices from block layer pm.
The runtime-pm for these devices would be managed by bsg and sg drivers.

Call trace:
 __switch_to+0x174/0x2c4
 __schedule+0x478/0x764
 schedule+0x9c/0xe0
 blk_queue_enter+0x158/0x228
 blk_mq_alloc_request+0x40/0xa4
 blk_get_request+0x2c/0x70
 __scsi_execute+0x60/0x1c4
 ufshcd_set_dev_pwr_mode+0x124/0x1e4
 ufshcd_suspend+0x208/0x83c
 ufshcd_runtime_suspend+0x40/0x154
 ufshcd_pltfrm_runtime_suspend+0x14/0x20
 pm_generic_runtime_suspend+0x28/0x3c
 __rpm_callback+0x80/0x2a4
 rpm_suspend+0x308/0x614
 rpm_idle+0x158/0x228
 pm_runtime_work+0x84/0xac
 process_one_work+0x1f0/0x470
 worker_thread+0x26c/0x4c8
 kthread+0x13c/0x320
 ret_from_fork+0x10/0x18

Signed-off-by: Asutosh Das <asutoshd@codeaurora.org>
Signed-off-by: Can Guo <cang@codeaurora.org>
Signed-off-by: Bao D. Nguyen <nguyenb@codeaurora.org>
---
 drivers/scsi/ufs/ufshcd.c | 18 ++----------------
 1 file changed, 2 insertions(+), 16 deletions(-)

Comments

Asutosh Das (asd) Jan. 28, 2021, 4:39 p.m. UTC | #1 
On Thu, Jan 28 2021 at 04:21 -0800, Avri Altman wrote:
>>
>> During runtime-suspend of ufs host, the scsi devices are
>> already suspended and so are the queues associated with them.
>> But the ufs host sends SSU to wlun during its runtime-suspend.
>Do you possible meant: "sends request-sense while clearing UAC to...."
>

The idea was to show that there's a scsi command that's sent during
suspend which may deadlock.

Yes, I agree to your comment and would change the message to reflect it in
the next version.

>
>Thanks,
>Avri
diff mbox series

Patch

diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index 9c691e4..b7e7f81 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -7217,16 +7217,6 @@  static void ufshcd_set_active_icc_lvl(struct ufs_hba *hba)
 	kfree(desc_buf);
 }
 
-static inline void ufshcd_blk_pm_runtime_init(struct scsi_device *sdev)
-{
-	scsi_autopm_get_device(sdev);
-	blk_pm_runtime_init(sdev->request_queue, &sdev->sdev_gendev);
-	if (sdev->rpm_autosuspend)
-		pm_runtime_set_autosuspend_delay(&sdev->sdev_gendev,
-						 RPM_AUTOSUSPEND_DELAY_MS);
-	scsi_autopm_put_device(sdev);
-}
-
 /**
  * ufshcd_scsi_add_wlus - Adds required W-LUs
  * @hba: per-adapter instance
@@ -7265,7 +7255,6 @@  static int ufshcd_scsi_add_wlus(struct ufs_hba *hba)
 		hba->sdev_ufs_device = NULL;
 		goto out;
 	}
-	ufshcd_blk_pm_runtime_init(hba->sdev_ufs_device);
 	scsi_device_put(hba->sdev_ufs_device);
 
 	hba->sdev_rpmb = __scsi_add_device(hba->host, 0, 0,
@@ -7274,17 +7263,14 @@  static int ufshcd_scsi_add_wlus(struct ufs_hba *hba)
 		ret = PTR_ERR(hba->sdev_rpmb);
 		goto remove_sdev_ufs_device;
 	}
-	ufshcd_blk_pm_runtime_init(hba->sdev_rpmb);
 	scsi_device_put(hba->sdev_rpmb);
 
 	sdev_boot = __scsi_add_device(hba->host, 0, 0,
 		ufshcd_upiu_wlun_to_scsi_wlun(UFS_UPIU_BOOT_WLUN), NULL);
-	if (IS_ERR(sdev_boot)) {
+	if (IS_ERR(sdev_boot))
 		dev_err(hba->dev, "%s: BOOT WLUN not found\n", __func__);
-	} else {
-		ufshcd_blk_pm_runtime_init(sdev_boot);
+	else
 		scsi_device_put(sdev_boot);
-	}
 	goto out;
 
 remove_sdev_ufs_device: