Hi Anderson,

Could you please check about below thread?

https://patchwork.kernel.org/patch/6357751/

Your patch is same as my old one.

On 11/30/17, Bjorn Andersson bjorn.andersson@linaro.org wrote:

Occasionally the following error message can be seen in the logs of Qualcomm devices using UFS:

EXT4-fs (sda9): Delayed block allocation failed for inode 685600 at logical offset 1086 with max blocks 3 with error 121 EXT4-fs (sda9): This should not happen!! Data will be lost

This is caused by a failing WRITE_SAME command, which per the JEDEC UFS specification is not a supported. Set the no_write_same flag on the ufshcd SCSI host to let the SCSI layer know this.

Fixes: 7a3e97b0dc4b ("[SCSI] ufshcd: UFS Host controller driver") Cc: stable@vger.kernel.org Signed-off-by: Bjorn Andersson bjorn.andersson@linaro.org

---

drivers/scsi/ufs/ufshcd.c | 1 + 1 file changed, 1 insertion(+)

diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c index 88c086f5c4e3..e5b1efd1dafd 100644 --- a/drivers/scsi/ufs/ufshcd.c +++ b/drivers/scsi/ufs/ufshcd.c @@ -6515,6 +6515,7 @@ static struct scsi_host_template ufshcd_driver_template = { .can_queue = UFSHCD_CAN_QUEUE, .max_host_blocked = 1, .track_queue_depth = 1,

• .no_write_same = 1,

};

static int ufshcd_config_vreg_load(struct device *dev, struct ufs_vreg

*vreg,

2.15.0

WBR, Kyuho Choi