Re: [EXT] [PATCH v1 1/6] scsi: ufs: Add device reset in link recovery path

[cang]

On 2019-11-04 05:48, Bean Huo (beanhuo) wrote:
> Hi, Can Guo
>
> > In order to recover from hibern8 exit failure, perform a reset in link 
> > recovery
> > path before issuing link start-up.
> >
> > Signed-off-by: Can Guo <cang@xxxxxxxxxxxxxx>
> > ---
> >  drivers/scsi/ufs/ufshcd.c | 3 +++
> >  1 file changed, 3 insertions(+)
> >
> > diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c 
> > index
> > c28c144..525f8e6 100644
> > --- a/drivers/scsi/ufs/ufshcd.c
> > +++ b/drivers/scsi/ufs/ufshcd.c
> > @@ -3859,6 +3859,9 @@ static int ufshcd_link_recovery(struct ufs_hba 
> > *hba)
> >  	ufshcd_set_eh_in_progress(hba);
> >  	spin_unlock_irqrestore(hba->host->host_lock, flags);
> >
> > +	/* Reset the attached device */
> > +	ufshcd_vops_device_reset(hba);
> > +
> >  	ret = ufshcd_host_reset_and_restore(hba);
> There is time consumption in reset,  It is true that reset can
> hide/solve some issues.
> I don't know if you experienced issue resulting from an absent reset
> in this case mentioned in
> Patch commit comment.

Hi Bean,

Yes, we did see some issues without this device reset here. For example,
link start-up failure and/or NOP-IN timeout during probe stage.

Best regards,
Can Guo.

> >  	spin_lock_irqsave(hba->host->host_lock, flags);
> > --
> > The Qualcomm Innovation Center, Inc. is a member of the Code Aurora 
> > Forum,
> > a Linux Foundation Collaborative Project