linux-kernel - Re: [PATCH] ufs: core: Disable auto h8 before ssu

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:   Fri, 14 Oct 2022 18:15:33 +0800
From:   高严凯 <gaoyankaigeren@...il.com>
To:     Bean Huo <huobean@...il.com>
Cc:     Avri Altman <Avri.Altman@....com>,
        Alim Akhtar <alim.akhtar@...sung.com>,
        Bart Van Assche <bvanassche@....org>,
        "James E . J . Bottomley" <jejb@...ux.ibm.com>,
        Bean Huo <beanhuo@...ron.com>,
        Adrian Hunter <adrian.hunter@...el.com>,
        "linux-scsi@...r.kernel.org" <linux-scsi@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] ufs: core: Disable auto h8 before ssu

hi Bean
Yes, This is the time from when the system enters idle to when auto h8
is issued.
This patch is just to ensure that there is no continuous action of
"auto h8 exit" before dme h8 enter, it has nothing to do with idle
time.
The reason I want to ensure this order is because when there is an
action of "auto h8 exit", most devices take a certain amount of time
to get the state ready.
Immediately after the action of "auto h8 exit" is completed and "dme
h8 enter" is followed, the device will be abnormal.
And I think it's more of a common patch for all platforms, Patch can
adapt different vendors,and align to ufshcd_auto_hibern8_enable in
__ufshcd_wl_resume.

Bean Huo <huobean@...il.com> 于2022年10月14日周五 13:48写道：
>
> On Fri, 2022-10-14 at 10:44 +0800, 高严凯 wrote:
> > " UFSHCI has idle time before bringing Link into hibern8 mode, " I
> > think  ufs put the link to h8 during the time
>
>
> no,  it is ufs controller. see the ufshci spec:
>
> "Auto-Hibern8 Idle Timer Value (AH8ITV): This is the timer that UFS
> subsystem must be idle before UFS host *controller* may put UniPro link
> into Hibernate state autonomously. The idle timer value is multiplied
> by the indicated timer scale to yield an absolute timer value. ....
>
> ...
> Any non-zero value will enable Auto-Hibernate idle timer.
> UFS host controller shall put Unipro link out of Hibernate state when
> the link communication is required. The mechanism to decide when the
> Unipro link needs to become active is host controller specific
> implementation, and is transparent to the software."
>
>