arm/155214: [patch] MMC/SD IO slow on Atmel ARM with modern large SD cards

[Bernd Walter]

On Thu, Mar 03, 2011 at 12:51:30PM -0700, Warner Losh wrote:
> On 03/02/2011 17:56, Greg Ansley wrote:
> >On 3/2/11 7:40 PM, Ian Lepore wrote:
> >>The following reply was made to PR arm/155214; it has been noted by 
> >>GNATS.
> >>
> >>From: Ian Lepore<freebsd at damnhippie.dyndns.org>
> >>To: ticso at cicely.de
> >>Cc: FreeBSD-gnats-submit at freebsd.org
> >>Subject: Re: arm/155214: [patch] MMC/SD IO slow on Atmel ARM with modern
> >>  large SD cards
> >>Date: Wed, 02 Mar 2011 17:21:09 -0700
> >>  I have not tested with 4-bit enabled; I wasn't aware (but in 
> >>retrospect
> >>  I probably should have assumed) that the hardware bugs are different
> >>  with 4-bit enabled.  I'm not even sure our hardware design carries 
> >>all 4
> >>  lines to the card; I'll look at the schematics and if they're 
> >>connected
> >>  I'll see about testing that mode.  (And if they're not I'll see about
> >>  having our designers wire up all 4 lines on future designs.)
> >>
> >>  I also haven't tested with the SAM9-series, because I don't have that
> >>  hardware available.  (I hope to convince our hardware designers to
> >>  migrate us to SAM9 this year.)
> >>
> >With the current code (prepatch) 4bit mode is known to work at least 
> >on the SAM9G20 with kernel option AT91_MCI_HAS_4WIRE.  I'll be working 
> >on the SAM9G20 in the next few days and I can test the patch on both a 
> >RM9200 (1 bit only) and on a couple of SAM9G20 designs with 4bit 
> >hardware.
> 
> For some ancient history...
> 
> The Atmel Linux repo went back and forth on the 4-bit support in 
> RM9200.  When the SAM9260 was released, they added support there and 
> specifically disabled it for the RM9200.  When I asked about it, they 
> muttered something about silicon bugs.  After that, it ping ponged back 
> and forth between supported and not for a while.  I honestly don't know 
> where it settled finally.

Interesting.
I'm not aware of 4-bit silicon bugs, but of course there are boards, which
are 1-bit wired only.
And I'm aware that you need 100k pullup or pulldown to the data wires,
although my first prototype boards worked fine without them.
I don't remember the reason for them.

-- 
B.Walter <bernd at bwct.de> http://www.bwct.de
Modbus/TCP Ethernet I/O Baugruppen, ARM basierte FreeBSD Rechner uvm.