Initially I noticed that the data written to the microSD was incorrect. The unique pattern of the error indicated that it was in some way related to 4 KB buffers. I only saw the error when writing to the microSD and not to the USB thumbdrive or in data sent out the ethernet interface. This to me hinted at a  microSD driver type problem. Changing from 2.6.21 to 2.6.24 improved the reliability (but did not eliminate the problem), so I never thought that the problem might be with the PXA270. 

Reading the link to the 'arm' thread that I posted, it seems that there may be an issue with the PXA270 (E#91) but I've seen no Gumstix confirmation. Only those with access to the Marvel documentation will know.

The patch that I was referring to in my previous email is targeted at the sd controller on the PXA270 (Gumstix Verdex and Verdex Pro). I don't believe that this will have any effect on the Gumstix Overo.

Greg

On Tue, Feb 10, 2009 at 9:27 AM, PPaFin <ppafin@gmail.com> wrote:
Hi,

Interesting issue, I've came across similar thing with our application
on Gumstix Overo. My application handles cp from /tmp to microsd and it
seems to get stuck after few hours.

So far I have been able to resolve following details of this case (dmesg):

Spurious irq 95: 0xffffffdf, please flush posted write for irq 56

Spurious irq 95: 0xffffffdf, please flush posted write for irq 56

Spurious irq 95: 0xffffffdf, please flush posted write for irq 56

and

end_request: I/O error, dev mmcblk0, sector 5099805

end_request: I/O error, dev mmcblk0, sector 5099813

=> So I don't know how this relates, but at least last example could be
something similar to your findings?

PPa

titan308 wrote:
> I've noticed that nobody seems to be overly concerned about this data
> error that I reported back in July 2008. For those on the list that have
> had microSD problems, the links below might be worth looking into:
>
> http://bec-systems.com/site/314/fix-for-pxa270-mmcsd-controller-write-corruption
>
> http://lists.arm.linux.org.uk/lurker/message/20090117.220838.f9e6472f.en.html
>
> Since posting the initial email, I switched to the buildroot
> configuration and upgraded the kernel to 2.6.24 and the error count went
> way down (~5% of all 'large' file copies had the error).  Basically, I
> had decided to compute md5sums before and after the copy. If an error
> occurred, I repeated the copy. Once, the file system on the microSD card
> got corrupted and I had to reformat the card.
>
> When I heard about the problem described by Cliff Brake in the link
> above, I set up a test fixture:
>
> Verdex XL6P, console-vx, netmicrosd, SanDisk USB thumbdrive (4 GB), and
> a SanDisk microSD (8 GB)
>
> For my test I copy a 600 + MB file from the USB thumbdrive to the
> microSD card, sync the image to insure all data is written to flash, and
> then compute an md5sum of the file on the microSD card. I compare the
> new md5sum to the known md5sum of the file taking note of any
> differences. I then delete the file from the microSD card and re-sync.
> Then the process gets repeated ...
>
> Results:
>
> With the non-patched 2.6.24 kernel I encountered 15 errors in 281
> copies. (I have no way of knowing if there was more than one dma buffer
> error in the copy).
>
> With the PATCHED 2.6.24 kernel I encountered 0 errors in 750 copies.
>
> I have implemented the modification in my project.
>
> BTW, the arm link above references a Marvel Errata #91, the latest
> Specification Update that I can find (Intel) has Errata #88. Can anyone
> confirm if this problem is known to Marvel? Does the problem exist with
> the processor used on the Verdex Pro?
>
> Greg
>
> P.S. I didn't mention in my original email, but the problem occurred
> with every file system that I used on the microSD card, not just FAT32.
>
>
>
> On Wed, Jul 23, 2008 at 10:49 AM,  <titan308@gmail.com
> <mailto:titan308@gmail.com>> wrote:
>
>     When I write a large file ( > 350 MB ) to a microSD card, I almost
>     always get one or more corrupted 4KB blocks of data in the file. The
>     file size is correct but an md5sum of the new file does not match
>     the original. Using 'cmp' I see where the data looks shifted by one
>     byte for about 4 KB then the data is correct again. The audio and
>     video files I initially wrote did not seem corrupted. I didn't see
>     the error until I copied a tarball image and something 'important'
>     got corrupted.
>
>     My setup is a Verdex XM4, console-vx, and netmicrosd. I first
>     noticed the problem with pre-built 308M versions of u-boot, uimage
>     (2.6.21), and rootfs (basic). I have since loaded the pre-built
>     images of 318M versions of each, trying both glibc and uclibc
>     combinations of the kernel and fs. (The pre-built kernel images are
>     still 2.6.21). I still have the problem.
>
>     My current test file is a tar image of the gumstix-oe directory (tar
>     cf gs-oe.tar gumstix-oe/). The file is about 390 MB.
>
>     I can use scp to write the file to the microSD card:
>
>     scp gs-oe.tar root@192.168.1.15:/media/card/test.tar
>
>     I've ruled out the ethernet interface by buying a microSD USB card
>     reader/adapter and copying the file directly to the microSD card
>     then with the microSD card in my gumstix:
>
>     dd if=/media/card/gs-oe.tar of=/media/card/test-2.tar
>
>     I've done this test several times (with various 'dd' block sizes)
>     and rarely did the file get written without an error.
>
>     I've tried several microSD cards:
>
>     SanDisk 1GB (2)
>     Kingston 2GB
>     SanDisk 8GB (SDHC)
>
>     I have tried a different netmicrosd board with the same results. I
>     have a 'new' netwifimicrosd board on order.
>
>     Any suggestions on how to resolve this would be greatly appreciated.
>
>      Greg
>
>
>
> ------------------------------------------------------------------------
>
> ------------------------------------------------------------------------------
> Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM)
> software. With Adobe AIR, Ajax developers can use existing skills and code to
> build responsive, highly engaging applications that combine the power of local
> resources and data with the reach of the web. Download the Adobe AIR SDK and
> Ajax docs to start building applications today-http://p.sf.net/sfu/adobe-com
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> gumstix-users mailing list
> gumstix-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/gumstix-users


------------------------------------------------------------------------------
Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM)
software. With Adobe AIR, Ajax developers can use existing skills and code to
build responsive, highly engaging applications that combine the power of local
resources and data with the reach of the web. Download the Adobe AIR SDK and
Ajax docs to start building applications today-http://p.sf.net/sfu/adobe-com
_______________________________________________
gumstix-users mailing list
gumstix-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/gumstix-users