Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 28 Aug 2013 09:41:31 -0300
From:      Luiz Otavio O Souza <lists.br@gmail.com>
To:        Ian Lepore <ian@freebsd.org>
Cc:        "freebsd-arm@freebsd.org" <freebsd-arm@freebsd.org>
Subject:   Re: arm/181601: Sporadic failure of root mount on ARM/Raspberry
Message-ID:  <CAB=2f8wbfQ1myBssBkUs-yh8k7Ji-j8MeUEOZGGG%2Bt6CXkjv%2BQ@mail.gmail.com>
In-Reply-To: <201308281230.r7SCU0k5093956@freefall.freebsd.org>
References:  <201308281230.r7SCU0k5093956@freefall.freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On 28 August 2013 09:30, Ian Lepore <ian@freebsd.org> wrote:

[...]
 > mmcsd0: 8GB <SDHC NCard 1.0 SN 1075839384 MFG 05/2013 by 130 JT> at mmc0
50.0MHz/4bit/65535-block
 > WARNING: WITNESS option enabled, expect reduced performance.
 > WARNING: DIAGNOSTIC option enabled, expect reduced performance.
 > mmcsd0: Error indicated: 1 Timeout
 > mmcsd0: Error indicated: 1 Timeout
[...]

>
>  We have long had a problem with mysterious sdcard timeout errors on RPi
>  that doesn't happen on other hardware with sdhci controllers.  Until
>  now, it was thought that these timeouts always occurred shortly after
>  the controller was initialized by the OS.  The timeouts would affect the
>  early card-detection sequences; we worked around them by adding
>  automatic retries to the mmc code that identifies and initializes cards.
>
>  This error appears to be a timeout that occurs after the card init
>  sequences are done (the errors are reported by mmcsd0, not mmc0).
>

Yes, i've seen this as well and for me, every time it fails it thinks my SD
is HS capable while it isn't. This make it fail later.

Setting hw.bcm2835.sdhci.hs=0 on loader.conf works fine as a workaround.

Luiz



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAB=2f8wbfQ1myBssBkUs-yh8k7Ji-j8MeUEOZGGG%2Bt6CXkjv%2BQ>