Re: 1571 blocks free

From: silverdr_at_wfmh.org.pl
Date: Sun, 26 May 2013 21:03:37 +0200
Message-Id: <ACED48EC-EADE-42E4-9A8C-92C2AE7A53DB@wfmh.org.pl>
On 2013-05-26, at 20:14, Spiro Trikaliotis wrote:

> Hello,
> 
> only an educated guess: ;)
> 
> * On Sun, May 26, 2013 at 07:46:27PM +0200 silverdr@wfmh.org.pl wrote:
>> Does anyone know (without digging the rom disassembly) if 1571, when
>> calculating "BLOCKS FREE" value, takes into account what is stored in
>> second BAM sector (53,0) for track 53? All sectors of this track
>> should be allocated anyway, which means that under normal
>> circumstances (a fully valid and non-custom, non-corrupted disk) it
>> should not matter. But what if someone decided to make some use of
>> those wasted sectors and modified the BAM accordingly?
> 
> I can only tell for the 2031/1540/1541 out of my head: The 1541
> calculates the blocks free that are actually free. That is, it takes the
> BAM into account, and does not use any pre-knowledge (other than knowing
> that 18/0 contains the BAM, that is).

AFAIR 1541 reads the "blocks free" bytes from the BAM area and sums them up but it skips the directory track when calculating them so free blocks from track 18 are not included.

> I would expect the 1571 to behave accordingly. But: I do not know for
> sure, it is only a guess.

1571 supposedly does the same for track 18 but it wastes almost whole track 53 and allocates in the BAM. Hence the question: why bother allocating if those blocks are not taken into account anyway? Maybe because in 1571 "blocks free" AND/OR other routines don't know that this track is now used for extended BAM?

>> Also: will it issue ILLEGAL T/S if a file is chained to track 53?
> 
> Again: The 2031/1540/1541 do not perform any such checks. Thus, chaining
> a file to 18/x is no problem there. Again, I would expect the 1571 to
> behave accordingly.

I think so too.

> Having said all this: It should be rather easy to test this, or am I
> missing something?

No, it should be relatively easy. I have to prepare a 1571 disk with some modifications on it. But since I am debugging something else now, therefore I hoped that maybe someone knows the facts already and an authoritative answer would come by the time I am done with the task at hand ;-)

-- 
SD!


       Message was sent through the cbm-hackers mailing list
Received on 2013-05-26 20:00:03

Archive generated by hypermail 2.2.0.