linux-kernel - Re: [PATCH v1 2/2] mmc: sdhci: Use the SW timer when the HW timer cannot meet the timeout value required by the device

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] [day] [month] [year] [list]

Date:   Thu, 30 Sep 2021 11:02:11 +0200
From:   Bean Huo <huobean@...il.com>
To:     Adrian Hunter <adrian.hunter@...el.com>,
        Ulf Hansson <ulf.hansson@...aro.org>
Cc:     Bean Huo <beanhuo@...ron.com>, linux-mmc@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1 2/2] mmc: sdhci: Use the SW timer when the HW timer
 cannot meet the timeout value required by the device

On Thu, 2021-09-30 at 11:59 +0300, Adrian Hunter wrote:
> On 30/09/2021 11:34, Bean Huo wrote:
> > Hi Adrian,
> > 
> > 
> > Thanks.
> > I want to give a short conclusion  for our discussion:
> > 
> > Based on your information, these sounds disable of HW timer timeout
> > interrupt will make eMMC host controller malfunction, in another
> > word,
> > the disable of timeout interrupt will make the eMMC host cannot
> > correctly provide the completion interrupt. And unless only when
> > the
> > SOC vendor signals that their SOC supports that the host side SW
> > can
> > disable this HW timeout interrupt, as TI does.
> > 
> > I studied the SDHCI Spec, and tried to see if there is this kind of
> > support statement, but not been found yet. I will check with other
> > SOC
> > vendors.
> > 
> > I have one more question, if you know, please give me your
> > information.
> > 
> > I did testing on HW timer bahevior in case CQE is on.  Currently,
> > we
> > always set the HW timer with the maximum timeout value if CQE is
> > on.
> > Based on my testing, the HW timer will never timeout when we enable
> > CQE. I changed the HW timer value to be lower, it is the same
> > result.
> > Do you know that the HW timer will be inactivated in case CQE is
> > on?  but its timeout interrupt is still enabled.
> 
> No I don't know how different CQE handle timeouts.

Thanks anyway.

Bean

> 
> > Kind regards,
> > Bean
> >