Text 2488, 57 rader
Skriven 2011-03-25 17:13:35 av Jonathan de Boyne Pollard
Ärende: Re: formatting to FAT32
===============================
Gecko/20110303 Thunderbird/3.1.9
UTC)
comp.os.os2.utilities:113
From: Jonathan de Boyne Pollard <J.deBoynePollard-newsgroups@NTLWorld.COM>

>> It's not finished, yet. Yes, it will format and check FAT volumes.
>>
> BTW, does 'check' including fixing FAT32 ?
>

Aside from the part that deals with the FATs themselves, the checking 
code is bitness-agnostic.  So it should work the same on all bitnesses.

> The current FAT32 check often leads to a msg like: Please run u$ 
> Scandisk to fix your partition :-/
>

Not knowing what error causes that, or even what the real message is 
(which what you write clearly is not), I cannot address specifics.  The 
checker recognizes the errors that I know how to recognize, and 
automatically fixes the ones that I know how to automatically fix.  
There's a whole repair policy section of the reference manual 
documenting what errors are recognized and fixed.  It's fairly long.

In general, the target problem is metadata corruption errors, and things 
that IBM's CHKDSK cannot handle at all such as orphaned long filenames, 
erroneously lowercased 8.3 names, and dot entries in the root 
directory.  (CHKDSK enters an infinite loop on that latter.)  Physical 
I/O problems reading/writing the medium are an entirely different kettle 
of fish, and not the target problem.

>> But I'd like to have HPFS support as well. That's mostly just a 
>> placeholder at the moment.
>>
> Well, for first test version, we will all be quite happy with only 
> FAT32 :-)
> Anyway, current HPFS formatting is dead slow, so a better version 
> would surely be nice.
>

It may be that that's as fast as it is possible to format an HPFS 
volume.  I don't remember it being particularly slow.  (It has been a 
long time since I had to format an HPFS volume.)  I don't know how much 
has to be done to high-level format an HPFS volume, since I haven't 
written the code to do that, yet.  It's probably a little more than to 
high-level format a FAT volume (which pretty much only involves writing 
3 things: a BPB, the FATs, and the root directory).  There are a few 
more data structures to write with HPFS, and they are a little more 
evenly spread throughout a volume.  But off the top of my head I cannot 
think of a reason why high-level formatting FAT and HPFS should be 
greatly different from each other as far as execution times are 
concerned.  From what I remember when, years ago, I last formatted HPFS 
disc volumes, they weren't.


--- Internet Rex 2.31
 * Origin: virginmedia.com (1:261/20.999)