[Bug-ddrescue] Re: Algorithm used in ddrescue

[Antonio Diaz Diaz]

Hello Nik,

address@hidden wrote:
> first of all - thanks for your great work with ddrescue. It has helped
> me a lot.

Glad to hear this.


> There is one question which I had about the algorithm/strategy which
> ddrescue used. This question has been posted to the mailing list two
> years ago but has not been answered yet. You can find it here:

Yes. Sometimes I receive too many emails and forget to answer some of them.


> Could you kindly do me the favor and answer this question? This would
> help me :)

Of course.

> How does ddrescue work on defective sectors? When does it decide to
> skip sectors and much more important: how does it determine the size
> of sectors to skip or jump ahead?

You'll need to look at copy_non_tried() in rescuebook.cc to answer this. 
The important points are:
  1) while copying the non-tried parts of the input file, ddrescue 
jumps forward every time it finds a bad sector.
  2) the skipped size is increased until a good sector is found or the 
end of the current block is reached.


> Also I have read something about backward reading capabilities but
> haven't seen ddrescue actually reading backwards. Does the current
> ddrescue read backwards? If so - when?

From 
http://www.gnu.org/software/ddrescue/manual/ddrescue_manual.html#Algorithm

"3) Read backwards one sector at a time the non-trimmed blocks, until a 
bad sector is found. For each non-trimmed block, mark the bad sector 
found as bad-sector and mark the rest of that block as non-split."


Best regards,
Antonio.