Scott Dwyer wrote:
I just did some testing, and in short, I can confirm that if read speed
slows down after encountering errors, even when reading another good
area, that closing and reopening the device does bring the read speed
back up where it should be. So I am confirming that it is in the kernel,
and not ddrescue. Although I would like to point out that for me this
issue does not happen when using the --direct option.
Thanks Scott. All versions of ddrescue reset skip_size when moving to a
new area, so your findings indeed confirm thar the slowdown is not
produced by ddrescue.
>> Do you have the "reopen" patch that I can try on my 18pre2 copy
>> here so that I can see how the performance changes?
>
> No, I do not have a patch for ddrescue. I wrote an independent program
> for myself to test and prove the theory. Antonio will have to make any
> changes to ddrescue as he wishes.
I'll make a hacked version of ddrescue and post the URL here as soon as
it is done. If it proves useful, a command line option can be added to
activate the "reopen" behaviour.
Best regards,
Antonio.
_______________________________________________
Bug-ddrescue mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/bug-ddrescue
