linux-kernel - RE: [PATCH v2] scsi: ufs: modify Tactive time setting conditions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 10 Jan 2022 11:05:16 +0900
From:   서호영 <hy50.seo@...sung.com>
To:     "'Avri Altman'" <Avri.Altman@....com>,
        <linux-scsi@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <alim.akhtar@...sung.com>, <jejb@...ux.ibm.com>,
        <martin.petersen@...cle.com>, <beanhuo@...ron.com>,
        <asutoshd@...eaurora.org>, <cang@...eaurora.org>,
        <bvanassche@....org>
Subject: RE: [PATCH v2] scsi: ufs: modify Tactive time setting conditions

Hi,
If register "UFS_DEVICE_QUIRK_HOST_PA_TACTIVATE" quirk,
using this Tact time function.
And I mean all Samsung device use UFS_DEVICE_QUIRK_HOST_PA_TACTIVATE quirk below this. so I told that.
static struct ufs_dev_fix ufs_fixups[] = {                    
         /* UFS cards deviations table */                      
         UFS_FIX(UFS_VENDOR_MICRON, UFS_ANY_MODEL,             
                 UFS_DEVICE_QUIRK_DELAY_BEFORE_LPM |           
                 UFS_DEVICE_QUIRK_SWAP_L2P_ENTRY_FOR_HPB_READ),
         UFS_FIX(UFS_VENDOR_SAMSUNG, UFS_ANY_MODEL,            
                 UFS_DEVICE_QUIRK_DELAY_BEFORE_LPM |           
                 UFS_DEVICE_QUIRK_HOST_PA_TACTIVATE |

Other vendors can use it if necessary.
And for stability, the device tact time must be longer than host tact time.
(I already check Major device vendors)
That's reason why use this function.
But the Host tact time and Device tact time same, stability may not be satisfied.
So I changed this way

Thanks,
HOYOUNG.

> -----Original Message-----
> From: Avri Altman [mailto:Avri.Altman@....com]
> Sent: Friday, January 7, 2022 7:57 PM
> To: SEO HOYOUNG; linux-scsi@...r.kernel.org; linux-kernel@...r.kernel.org;
> alim.akhtar@...sung.com; jejb@...ux.ibm.com; martin.petersen@...cle.com;
> beanhuo@...ron.com; asutoshd@...eaurora.org; cang@...eaurora.org;
> bvanassche@....org
> Subject: RE: [PATCH v2] scsi: ufs: modify Tactive time setting conditions
> 
> > The Tactive time determine the waiting time before burst at hibern8
> > exit and is determined by H/W at linkup state However, in the case of
> > samsung devices, guided host's Tactive time
> > +100us for stability.
> > If the HCI's Tactive time is equal or greater than the device,
> > +100us should be set.
> This way you are changing this for everyone - not just for Samsung.
> e.g. Qualcomm are using this quirk as well for WDC devices.
> 
> Thanks,
> Avri
> 
> >
> > Signed-off-by: SEO HOYOUNG <hy50.seo@...sung.com>
> > ---
> >  drivers/scsi/ufs/ufshcd.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
> > index 1049e41abd5b..460d2b440d2e 100644
> > --- a/drivers/scsi/ufs/ufshcd.c
> > +++ b/drivers/scsi/ufs/ufshcd.c
> > @@ -7815,7 +7815,7 @@ static int
> > ufshcd_quirk_tune_host_pa_tactivate(struct
> > ufs_hba *hba)
> >         peer_pa_tactivate_us = peer_pa_tactivate *
> >                              gran_to_us_table[peer_granularity - 1];
> >
> > -       if (pa_tactivate_us > peer_pa_tactivate_us) {
> > +       if (pa_tactivate_us >= peer_pa_tactivate_us) {
> >                 u32 new_peer_pa_tactivate;
> >
> >                 new_peer_pa_tactivate = pa_tactivate_us /
> > --
> > 2.26.0