Hi Michael,

 

Did a little bit more checking.  The smallest block size appears to be 512 
bytes.  The more common block size is 1024 bytes to 2048 bytes for the larger 
cards. At the time of the document in 2003 they were saying that for a 4 mega 
byte area they allotted a 3 percent amount of cells or 234 512 byte blocks as a 
buffer or wear leveling area.  First they wear level over the 4 MB zone.  Then 
as sections hit the wear out limit they take them out of service and mark them 
as used up.  Once the 234 blocks are gone the part will fail critically.  At 
that point the part will not be able to hide the faults from you.  If you look 
at the fist page of SanDisk patent 6,594,183 it shows a nice diagram of the 
address translation table to block scheme.  The diagram is also on page 4 
figure 2.  There are some other SanDisk patents that clearly show the SD card 
arrangement in more detail, but the one noted has some good diagrams.  In some 
of the original SD cards they actually had a controller/interface chip coupled 
to an off the shelf flash chip.  Think of the controller as an address 
translation engine more than anything else.


Bobby

  ----- Original Message ----- 
  From: Michael Schnell 
  To: uClinux development list 
  Sent: Friday, March 27, 2009 6:42 AM
  Subject: Re: [uClinux-dev] SD card corruption upon reboot and de-power




    Not exactly.  The 4MB zone is just the area that they give 3 percent 
additional memory cells.  Each 4MB zone would have the extra 3 percent.  

  I can't imagine how this is supposed to work, as I don't believe that such a 
4MB area can be partly erased, (as small erase blocks result in a high hardware 
price. 

  But maybe there is some kind of trick that I don't understand yet. 

  -Michael



------------------------------------------------------------------------------


  _______________________________________________
  uClinux-dev mailing list
  uClinux-dev@uclinux.org
  http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
  This message was resent by uclinux-dev@uclinux.org
  To unsubscribe see:
  http://mailman.uclinux.org/mailman/options/uclinux-dev
_______________________________________________
uClinux-dev mailing list
uClinux-dev@uclinux.org
http://mailman.uclinux.org/mailman/listinfo/uclinux-dev
This message was resent by uclinux-dev@uclinux.org
To unsubscribe see:
http://mailman.uclinux.org/mailman/options/uclinux-dev

Reply via email to