linux-kernel - Re: MMC: Make the configuration memory resource optional

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]

Message-ID: <4A7A08FE.5000909@mnementh.co.uk>
Date:	Wed, 05 Aug 2009 23:34:38 +0100
From:	Ian Molton <ian@...menth.co.uk>
To:	Guennadi Liakhovetski <g.liakhovetski@....de>
CC:	Magnus Damm <magnus.damm@...il.com>,
	pHilipp Zabel <philipp.zabel@...il.com>,
	Paul Mundt <lethal@...ux-sh.org>,
	Mark Brown <broonie@...nsource.wolfsonmicro.com>,
	linux-kernel@...r.kernel.org, Pierre Ossman <drzeus@...eus.cx>,
	Magnus Damm <damm@...nsource.se>
Subject: Re: MMC: Make the configuration memory resource optional

Guennadi Liakhovetski wrote:

> How do we verify this? Do I need some "very fast" card, so that sd would 
> try to drive the clock beyond pdata->hclk / 2, which should then fail? 

One way would be a 'scope on the card pins, but since you havent said I 
am assuming you have not altered the clock-speed setting routine.

Here you can see why I didnt like the silently-drop-conf-accesses 
version of the patch.  If you had not made conf area accesses silently 
succeed, you'd have found one lurking in the set_clock function.

Since your controller has no CNF area, it was discarding the write to 
the 1:1 clock bit (that isnt  in its none-existant conf area) and thus 
you got the next lowest clock as a result (/2)

> Currently I'm using 24MHz (copied from the original driver), and the only 
> thing I know about the controller is its "Maximum operating frequency: 25 
> MHz."

Thats the HCLK frequency, not the card clock.

the TMIO MFD devices can divide this HCLK by anything from 512 to 2, and 
  MFD devices have a facility to disable the divider, yielding full HCLK 
speed as the card clock.

If you havent already found a 1:1 clock enable bit on your device, you 
might try looking for it - it'd yield a near linear factor-of-two speed 
increase.

-Ian
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/