RE: [PATCH RESEND 2/2] scsi: ufs: Update L4 attributes on manual hibern8 exit in Cadence UFS.

From: Alim Akhtar
Date: Tue Nov 26 2019 - 22:42:45 EST

Next message: Sean Christopherson: "Re: [PATCH] KVM: SVM: Fix "error" isn't initialized"
Previous message: Mr. Ahmed: "ENQUIRY"
Next in thread: Vignesh Raghavendra: "Re: [PATCH RESEND 2/2] scsi: ufs: Update L4 attributes on manual hibern8 exit in Cadence UFS."
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

> -----Original Message-----
> From: Vignesh Raghavendra <vigneshr@xxxxxx>
> Sent: 21 November 2019 16:26
> To: Alim Akhtar <alim.akhtar@xxxxxxxxx>; sheebab <sheebab@xxxxxxxxxxx>
> Cc: Alim Akhtar <alim.akhtar@xxxxxxxxxxx>; Avri Altman
> <avri.altman@xxxxxxx>; Pedro Sousa <pedrom.sousa@xxxxxxxxxxxx>; James
> E.J. Bottomley <jejb@xxxxxxxxxxxxx>; Martin K. Petersen
> <martin.petersen@xxxxxxxxxx>; Stanley Chu <stanley.chu@xxxxxxxxxxxx>;
> Bean Huo (beanhuo) <beanhuo@xxxxxxxxxx>; yuehaibing@xxxxxxxxxx; linux-
> scsi@xxxxxxxxxxxxxxx; open list <linux-kernel@xxxxxxxxxxxxxxx>; linux-
> block@xxxxxxxxxxxxxxx; rafalc@xxxxxxxxxxx; mparab@xxxxxxxxxxx
> Subject: Re: [PATCH RESEND 2/2] scsi: ufs: Update L4 attributes on manual
> hibern8 exit in Cadence UFS.
>
>
>
> On 20/11/19 9:50 PM, Alim Akhtar wrote:
> > Hi Sheebab
> >
> > On Tue, Nov 19, 2019 at 12:38 PM sheebab <sheebab@xxxxxxxxxxx> wrote:
> >>
> >> Backup L4 attributes duirng manual hibern8 entry and restore the L4
> >> attributes on manual hibern8 exit as per JESD220C.
> >>
> > Can you point me to the relevant section on the spec?
> >
>
> Per JESD 220C 9.4 UniPro/UFS Control Interface (Control Plane):
>
> "NOTE After exit from Hibernate all UniPro Transport Layer attributes (including
> L4 T_PeerDeviceID,
>
> L4 T_PeerCPortID, L4 T_ConnectionState, etc.) will be reset to their reset values.
> All required attributes
>
> must be restored properly on both ends before communication can resume."
>
> But its not clear whether SW needs to restore these attributes or hardware
>
Thanks Vignesh for pointing out the spec section, yes it is not clear, one way to confirm this is just by read L4 attributes before
And after hinern8 entry/exit.
(at least in the current platform it is not being done)
AFA this patch is concerns, this looks ok to me.
@ Avri , any thought on this?

> Regards
> Vignesh
>
> >> Signed-off-by: sheebab <sheebab@xxxxxxxxxxx>
> >> ---
> >> drivers/scsi/ufs/cdns-pltfrm.c | 97
> >> +++++++++++++++++++++++++++++++++++++++++-
> >> 1 file changed, 95 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/drivers/scsi/ufs/cdns-pltfrm.c
> >> b/drivers/scsi/ufs/cdns-pltfrm.c index adbbd60..5510567 100644
> >> --- a/drivers/scsi/ufs/cdns-pltfrm.c
> >> +++ b/drivers/scsi/ufs/cdns-pltfrm.c
> >> @@ -19,6 +19,14 @@
> >>
> >> #define CDNS_UFS_REG_HCLKDIV 0xFC
> >> #define CDNS_UFS_REG_PHY_XCFGD1 0x113C
> >> +#define CDNS_UFS_MAX 12
> >> +
> >> +struct cdns_ufs_host {
> >> + /**
> >> + * cdns_ufs_dme_attr_val - for storing L4 attributes
> >> + */
> >> + u32 cdns_ufs_dme_attr_val[CDNS_UFS_MAX];
> >> +};
> >>
> >> /**
> >> * cdns_ufs_enable_intr - enable interrupts @@ -47,6 +55,77 @@
> >> static void cdns_ufs_disable_intr(struct ufs_hba *hba, u32 intrs) }
> >>
> >> /**
> >> + * cdns_ufs_get_l4_attr - get L4 attributes on local side
> >> + * @hba: per adapter instance
> >> + *
> >> + */
> >> +static void cdns_ufs_get_l4_attr(struct ufs_hba *hba) {
> >> + struct cdns_ufs_host *host = ufshcd_get_variant(hba);
> >> +
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_PEERDEVICEID),
> >> + &host->cdns_ufs_dme_attr_val[0]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_PEERCPORTID),
> >> + &host->cdns_ufs_dme_attr_val[1]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_TRAFFICCLASS),
> >> + &host->cdns_ufs_dme_attr_val[2]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_PROTOCOLID),
> >> + &host->cdns_ufs_dme_attr_val[3]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_CPORTFLAGS),
> >> + &host->cdns_ufs_dme_attr_val[4]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_TXTOKENVALUE),
> >> + &host->cdns_ufs_dme_attr_val[5]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_RXTOKENVALUE),
> >> + &host->cdns_ufs_dme_attr_val[6]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_LOCALBUFFERSPACE),
> >> + &host->cdns_ufs_dme_attr_val[7]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_PEERBUFFERSPACE),
> >> + &host->cdns_ufs_dme_attr_val[8]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_CREDITSTOSEND),
> >> + &host->cdns_ufs_dme_attr_val[9]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_CPORTMODE),
> >> + &host->cdns_ufs_dme_attr_val[10]);
> >> + ufshcd_dme_get(hba, UIC_ARG_MIB(T_CONNECTIONSTATE),
> >> + &host->cdns_ufs_dme_attr_val[11]);
> >> +}
> >> +
> >> +/**
> >> + * cdns_ufs_set_l4_attr - set L4 attributes on local side
> >> + * @hba: per adapter instance
> >> + *
> >> + */
> >> +static void cdns_ufs_set_l4_attr(struct ufs_hba *hba) {
> >> + struct cdns_ufs_host *host = ufshcd_get_variant(hba);
> >> +
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_CONNECTIONSTATE), 0);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_PEERDEVICEID),
> >> + host->cdns_ufs_dme_attr_val[0]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_PEERCPORTID),
> >> + host->cdns_ufs_dme_attr_val[1]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_TRAFFICCLASS),
> >> + host->cdns_ufs_dme_attr_val[2]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_PROTOCOLID),
> >> + host->cdns_ufs_dme_attr_val[3]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_CPORTFLAGS),
> >> + host->cdns_ufs_dme_attr_val[4]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_TXTOKENVALUE),
> >> + host->cdns_ufs_dme_attr_val[5]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_RXTOKENVALUE),
> >> + host->cdns_ufs_dme_attr_val[6]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_LOCALBUFFERSPACE),
> >> + host->cdns_ufs_dme_attr_val[7]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_PEERBUFFERSPACE),
> >> + host->cdns_ufs_dme_attr_val[8]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_CREDITSTOSEND),
> >> + host->cdns_ufs_dme_attr_val[9]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_CPORTMODE),
> >> + host->cdns_ufs_dme_attr_val[10]);
> >> + ufshcd_dme_set(hba, UIC_ARG_MIB(T_CONNECTIONSTATE),
> >> + host->cdns_ufs_dme_attr_val[11]); }
> >> +
> >> +/**
> >> * Sets HCLKDIV register value based on the core_clk
> >> * @hba: host controller instance
> >> *
> >> @@ -134,6 +213,7 @@ static void cdns_ufs_hibern8_notify(struct ufs_hba
> *hba, enum uic_cmd_dme cmd,
> >> * before manual hibernate entry.
> >> */
> >> cdns_ufs_enable_intr(hba, UFSHCD_UIC_HIBERN8_MASK);
> >> + cdns_ufs_get_l4_attr(hba);
> >> }
> >> if (status == POST_CHANGE && cmd == UIC_CMD_DME_HIBER_EXIT) {
> >> /**
> >> @@ -141,6 +221,7 @@ static void cdns_ufs_hibern8_notify(struct ufs_hba
> *hba, enum uic_cmd_dme cmd,
> >> * after manual hibern8 exit.
> >> */
> >> cdns_ufs_disable_intr(hba, UFSHCD_UIC_HIBERN8_MASK);
> >> + cdns_ufs_set_l4_attr(hba);
> >> }
> >> }
> >>
> >> @@ -245,15 +326,27 @@ static int cdns_ufs_pltfrm_probe(struct
> platform_device *pdev)
> >> const struct of_device_id *of_id;
> >> struct ufs_hba_variant_ops *vops;
> >> struct device *dev = &pdev->dev;
> >> + struct cdns_ufs_host *host;
> >> + struct ufs_hba *hba;
> >>
> >> of_id = of_match_node(cdns_ufs_of_match, dev->of_node);
> >> vops = (struct ufs_hba_variant_ops *)of_id->data;
> >>
> >> /* Perform generic probe */
> >> err = ufshcd_pltfrm_init(pdev, vops);
> >> - if (err)
> >> + if (err) {
> >> dev_err(dev, "ufshcd_pltfrm_init() failed %d\n",
> >> err);
> >> -
> >> + goto out;
> >> + }
> >> + host = devm_kzalloc(dev, sizeof(*host), GFP_KERNEL);
> >> + if (!host) {
> >> + err = -ENOMEM;
> >> + dev_err(dev, "%s: no memory for cdns host\n", __func__);
> >> + goto out;
> >> + }
> >> + hba = platform_get_drvdata(pdev);
> >> + ufshcd_set_variant(hba, host);
> >> +out:
> >> return err;
> >> }
> >>
> >> --
> >> 2.7.4
> >>
> >
> >
>
> --
> Regards
> Vignesh

Next message: Sean Christopherson: "Re: [PATCH] KVM: SVM: Fix "error" isn't initialized"
Previous message: Mr. Ahmed: "ENQUIRY"
Next in thread: Vignesh Raghavendra: "Re: [PATCH RESEND 2/2] scsi: ufs: Update L4 attributes on manual hibern8 exit in Cadence UFS."
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]