RE: [PATCH] scsi: ufs: no sw reset after last linkstartup fail

[Seunghui Lee]

> -----Original Message-----
> From: Avri Altman <Avri.Altman@xxxxxxx>
> Sent: Tuesday, July 5, 2022 3:51 PM
> To: Seunghui Lee <sh043.lee@xxxxxxxxxxx>; alim.akhtar@xxxxxxxxxxx;
> bvanassche@xxxxxxx; jejb@xxxxxxxxxxxxx; martin.petersen@xxxxxxxxxx; linux-
> scsi@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx
> Cc: grant.jung@xxxxxxxxxxx; jt77.jang@xxxxxxxxxxx;
> dh0421.hwang@xxxxxxxxxxx; junwoo80.lee@xxxxxxxxxxx; jangsub.yi@xxxxxxxxxxx;
> cw9316.lee@xxxxxxxxxxx; sh8267.baek@xxxxxxxxxxx; wkon.kim@xxxxxxxxxxx;
> seunghwan.hyun@xxxxxxxxxxx
> Subject: RE: [PATCH] scsi: ufs: no sw reset after last linkstartup fail
> 
> 
> > As we know, the ufs host controller is disabled and enabled in the
> > ufshcd_hba_enable(). So, host controller registers are clear after sw
> reset.
> >
> > Once the host fails to link startup 3 times, all host's controller can
> > be clear due to executing last ufshcd_hba_enable()(s/w reset).
> > That's why we need to skip last sw reset to get valid host register
> values.
> >
> > Before
> > (ufshcd_dme_link_startup() -> ufshcd_hba_enable() -> retries--) * 3
> times.
> >
> > After
> > (ufshcd_dme_link_startup() -> if retries is 0, break. ->
> > ufshcd_hba_enable()
> > -> retries--) * 3 times.
> > In this case, ufshcd_print_host_state(), ufshcd_print_pwr_info(),
> > ufshcd_print_evt_hist() are valid.
> If you find it necessary to elaborate the commit log, why not do just that?
> Also maybe make the title more indicative of what you want to achieve:
> "skip last hci reset to get valid register values" or something.
> 
> Thanks,
> Avri

Thank you for your comment.
I'll modify the commit log and update it again.

> 
> 
> >
> > > -----Original Message-----
> > > From: Seunghui Lee <sh043.lee@xxxxxxxxxxx>
> > > Sent: Thursday, June 30, 2022 1:48 PM
> > > To: alim.akhtar@xxxxxxxxxxx; avri.altman@xxxxxxx;
> > bvanassche@xxxxxxx;
> > > jejb@xxxxxxxxxxxxx; martin.petersen@xxxxxxxxxx; linux-
> > scsi@xxxxxxxxxxxxxxx;
> > > linux-kernel@xxxxxxxxxxxxxxx
> > > Cc: grant.jung@xxxxxxxxxxx; jt77.jang@xxxxxxxxxxx;
> > > dh0421.hwang@xxxxxxxxxxx; junwoo80.lee@xxxxxxxxxxx;
> > jangsub.yi@xxxxxxxxxxx;
> > > cw9316.lee@xxxxxxxxxxx; sh8267.baek@xxxxxxxxxxx;
> > wkon.kim@xxxxxxxxxxx;
> > > Seunghui Lee <sh043.lee@xxxxxxxxxxx>
> > > Subject: [PATCH] scsi: ufs: no sw reset after last linkstartup fail
> > >
> > > Host driver resets the host(ufshcd_hba_enable) after last
> > > linkstartup command failed. All of the member or host dump after
> > > linkstartup fail are reset value because of sw reset.
> > >
> > > Signed-off-by: Junwoo Lee <junwoo80.lee@xxxxxxxxxxx>
> > > Signed-off-by: Seunghui Lee <sh043.lee@xxxxxxxxxxx>
> > > ---
> > >  drivers/ufs/core/ufshcd.c | 2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
> > > index
> > > 7c1d7bb9c579..2cdc14675443 100644
> > > --- a/drivers/ufs/core/ufshcd.c
> > > +++ b/drivers/ufs/core/ufshcd.c
> > > @@ -4753,7 +4753,7 @@ static int ufshcd_link_startup(struct ufs_hba
> > *hba)
> > >                * but we can't be sure if the link is up until link startup
> > >                * succeeds. So reset the local Uni-Pro and try again.
> > >                */
> > > -             if (ret && ufshcd_hba_enable(hba)) {
> > > +             if (ret && retries && ufshcd_hba_enable(hba)) {
> > >                       ufshcd_update_evt_hist(hba,
> > >                                              UFS_EVT_LINK_STARTUP_FAIL,
> > >                                              (u32)ret);
> > > --
> > > 2.29.0
> >