[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1606195221.17338.6.camel@mtkswgap22>
Date: Tue, 24 Nov 2020 13:20:21 +0800
From: Stanley Chu <stanley.chu@...iatek.com>
To: Can Guo <cang@...eaurora.org>
CC: <asutoshd@...eaurora.org>, <nguyenb@...eaurora.org>,
<hongwus@...eaurora.org>, <ziqichen@...eaurora.org>,
<rnayak@...eaurora.org>, <linux-scsi@...r.kernel.org>,
<kernel-team@...roid.com>, <saravanak@...gle.com>,
<salyzyn@...gle.com>, Alim Akhtar <alim.akhtar@...sung.com>,
Avri Altman <avri.altman@....com>,
"James E.J. Bottomley" <jejb@...ux.ibm.com>,
"Martin K. Petersen" <martin.petersen@...cle.com>,
Bean Huo <beanhuo@...ron.com>,
Bart Van Assche <bvanassche@....org>,
open list <linux-kernel@...r.kernel.org>,
<kuohong.wang@...iatek.com>
Subject: Re: [PATCH] scsi: ufs: Don't disable core_clk_unipro if the link is
active
Hi Can,
On Mon, 2020-11-23 at 21:05 -0800, Can Guo wrote:
> If we want to disable clocks but still keep the link active, both ref_clk
> and core_clk_unipro should be skipped.
>
"core_clk_unipro" seems used by ufs-qcom only and not defined in the UFS
platform binding document: ufshcd_pltfrm.txt.
Could you please add the definition first and then it would be
reasonable to be used in common driver?
Or, how about add a flag in struct ufs_clk_info indicating if this clock
needs be ON to keep the link active? The flag could be set properly by
vendor initialization functions. In this way, we can also remove the
hard-coded "ref_clk" in __ufshcd_setup_clocks().
Thanks.
Stanley Chu
> Signed-off-by: Can Guo <cang@...eaurora.org>
>
> diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
> index a7857f6..69c2e91 100644
> --- a/drivers/scsi/ufs/ufshcd.c
> +++ b/drivers/scsi/ufs/ufshcd.c
> @@ -222,7 +222,7 @@ static int ufshcd_clear_tm_cmd(struct ufs_hba *hba, int tag);
> static void ufshcd_hba_exit(struct ufs_hba *hba);
> static int ufshcd_probe_hba(struct ufs_hba *hba, bool async);
> static int __ufshcd_setup_clocks(struct ufs_hba *hba, bool on,
> - bool skip_ref_clk);
> + bool keep_link_active);
> static int ufshcd_setup_clocks(struct ufs_hba *hba, bool on);
> static int ufshcd_uic_hibern8_enter(struct ufs_hba *hba);
> static inline void ufshcd_add_delay_before_dme_cmd(struct ufs_hba *hba);
> @@ -1710,7 +1710,6 @@ static void ufshcd_gate_work(struct work_struct *work)
> if (!ufshcd_is_link_active(hba))
> ufshcd_setup_clocks(hba, false);
> else
> - /* If link is active, device ref_clk can't be switched off */
> __ufshcd_setup_clocks(hba, false, true);
>
> /*
> @@ -7991,7 +7990,7 @@ static int ufshcd_init_hba_vreg(struct ufs_hba *hba)
> }
>
> static int __ufshcd_setup_clocks(struct ufs_hba *hba, bool on,
> - bool skip_ref_clk)
> + bool keep_link_active)
> {
> int ret = 0;
> struct ufs_clk_info *clki;
> @@ -8009,7 +8008,13 @@ static int __ufshcd_setup_clocks(struct ufs_hba *hba, bool on,
>
> list_for_each_entry(clki, head, list) {
> if (!IS_ERR_OR_NULL(clki->clk)) {
> - if (skip_ref_clk && !strcmp(clki->name, "ref_clk"))
> + /*
> + * To keep link active, ref_clk and core_clk_unipro
> + * should be kept ON.
> + */
> + if (keep_link_active &&
> + (!strcmp(clki->name, "ref_clk") ||
> + !strcmp(clki->name, "core_clk_unipro")))
> continue;
>
> clk_state_changed = on ^ clki->enabled;
> @@ -8580,7 +8585,6 @@ static int ufshcd_suspend(struct ufs_hba *hba, enum ufs_pm_op pm_op)
> if (!ufshcd_is_link_active(hba))
> ufshcd_setup_clocks(hba, false);
> else
> - /* If link is active, device ref_clk can't be switched off */
> __ufshcd_setup_clocks(hba, false, true);
>
> if (ufshcd_is_clkgating_allowed(hba)) {
Powered by blists - more mailing lists