linux-kernel - Re: [Sdhci-devel] SDHCI driver runnining in aspire one, allows to write to R/O SD cards

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: <49F62DF4.5070709@damore.org>
Date:	Mon, 27 Apr 2009 15:13:08 -0700
From:	Garrett D'Amore <garrett@...ore.org>
To:	Maxim Levitsky <maximlevitsky@...il.com>
CC:	Robert Hancock <hancockrwd@...il.com>, pierre@...man.eu,
	sdhci-devel@...ts.ossman.eu,
	linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [Sdhci-devel] SDHCI driver runnining in aspire one, allows to
 write to R/O SD cards

Maxim Levitsky wrote:
> On Mon, 2009-04-27 at 11:12 -0700, Garrett D'Amore wrote:
>   
>> As an extra data point, my sdhost driver in OpenSolaris works fine on 
>> the AA1.. it does see that the cards are locked, and refuses to allow 
>> them to be mounted read/write.
>>
>> This leads me to believe that the original linux problem is one of two 
>> things:
>>
>> * either a bug in the linux code, or
>> * busted circuit/hardware on a specific unit (i.e. a damaged unit)
>>
>> I believe that if the Linux code was buggy here, the problem would be 
>> reproducible on all platforms, not just the AA1.  So my inclination, 
>> barring independent confirmation, is that most likely there are one ore 
>> more defective units out there.
>>
>> Note that the AA1 also has buggy BIOS -- OpenSolaris can't "see" the 
>> cards (registers aren't properly initialized) unless they are present 
>> when the unit first powers up.
>>     
>
> But, why it works in windows?
>   

The Microsoft PCI code base probably is more willing to "work around" 
BIOSes that don't properly configure PCI configuration registers... this 
is an outstanding RFE for Solaris.

The Windows code base also has some other unrelated quirks in it -- it 
uses DAT3 for card presence detection instead of the card detect bit in 
the sdhci controller, for example.   (It also doesn't support MMC, which 
is to blame for some of the bastardized controller hacks that mfgrs like 
Ricoh have done to separate their MMC and SDcard handling into separate 
PCI functions.  Yes, that's a hardware workaround for a software bug!  
Go figure -- its what happens when one software company has a virtual 
monopoly...)

    -- Garrett
> Regards,
> 	Maxim Levitsky
>
>   

--
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/