linux-mmc.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jisheng Zhang <Jisheng.Zhang@synaptics.com>
To: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ulf Hansson <ulf.hansson@linaro.org>,
	Chunyan Zhang <zhang.lyra@gmail.com>,
	"linux-mmc@vger.kernel.org" <linux-mmc@vger.kernel.org>
Subject: Re: [PATCH] mmc: sdhci: Use Auto CMD Auto Select only when v4_mode is true
Date: Thu, 15 Oct 2020 14:24:06 +0800	[thread overview]
Message-ID: <20201015142353.4ef890ed@xhacker.debian> (raw)
In-Reply-To: <f1d6c1cf-b74a-701f-5136-e6be7bd4bfa9@intel.com>

On Thu, 15 Oct 2020 08:57:05 +0300 Adrian Hunter wrote:

> 
> 
> On 15/10/20 5:38 am, Jisheng Zhang wrote:
> > On Thu, 15 Oct 2020 10:12:07 +0800 Jisheng Zhang wrote:
> >
> >  
> >>
> >> On Wed, 14 Oct 2020 22:44:50 +0300 Adrian Hunter wrote:
> >>  
> >>>
> >>>
> >>> + Chunyan
> >>>
> >>> On 14/10/20 1:32 pm, Jisheng Zhang wrote:  
> >>>> Auto CMD Auto Select can only be used when v4_mode is enabled.  
> >>>
> >>> The SDHCI spec. doesn't seem to say that.  AFAICS it refers only to v4.1 not
> >>> v4 mode.  
> >>
> >> 4.10 defines the "Auto CMD Auto Select" mode, v4 mode bit in SDHCI_HOST_CONTROL2
> >> selects V3 compatible or V4 compatible mode, I think the v4 here includes
> >> v4.0, v4.1 and v4.2, so if we want to use the mode we have to ensure v4_mode  
> >
> > So if we want to use the "Auto CMD Auto Select" mode, we have to ensure v4 mode
> > is enabled.  
> 
> But that is not exactly what the SDHCI spec. says.  It is quite explicit
> about which registers and bit fields are affected by "Host Version 4 Enable =1".
> 

Just my humble opinion, this is implied, my logic look like:

Host Version 4 Enable == 0 => only V3 compatible mode \
                                                       => v4 mode is must for auto cmd auto select
No "Auto CMD Auto Select" definition in v3 spec      /

> So the question is whether this is standard or a quirk of your controller.
> 

other v4 controllers can do the same benchmark test after removing
sdhci_enable_v4_mode() in the controller's probe.


Thanks

  reply	other threads:[~2020-10-15  6:25 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-13 11:08 eMMC performance regression caused by 427b6514d0953bf Jisheng Zhang
2020-10-13 12:32 ` Ulf Hansson
2020-10-13 13:23   ` Adrian Hunter
2020-10-14 10:32     ` [PATCH] mmc: sdhci: Use Auto CMD Auto Select only when v4_mode is true Jisheng Zhang
2020-10-14 19:44       ` Adrian Hunter
2020-10-15  2:12         ` Jisheng Zhang
2020-10-15  2:38           ` Jisheng Zhang
2020-10-15  5:57             ` Adrian Hunter
2020-10-15  6:24               ` Jisheng Zhang [this message]
2020-10-15  8:50                 ` Adrian Hunter
2020-10-14 10:42     ` eMMC performance regression caused by 427b6514d0953bf Jisheng Zhang

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20201015142353.4ef890ed@xhacker.debian \
    --to=jisheng.zhang@synaptics.com \
    --cc=adrian.hunter@intel.com \
    --cc=linux-mmc@vger.kernel.org \
    --cc=ulf.hansson@linaro.org \
    --cc=zhang.lyra@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).