Message-Id:	<5.0.2.1.0.20010113120711.02772e00@pop5.banet.net>
X-Sender:	usbanet DOT farley3 AT pop5 DOT banet DOT net
X-Mailer:	QUALCOMM Windows Eudora Version 5.0.2
Date:	Sat, 13 Jan 2001 12:38:47 -0500
To:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
From:	"Peter J. Farley III" <pjfarley AT banet DOT net>
Subject:	Re: Fw: Patch for statfs.c
Cc:	ST001906 AT HRZ1 DOT HRZ DOT TU-Darmstadt DOT De, djgpp-workers AT delorie DOT com,
	Martin Stromberg <eplmst AT lu DOT erisoft DOT se>, ceo AT nbensacomputers DOT com
In-Reply-To:	<2593-Sat13Jan2001094434+0200-eliz@is.elta.co.il>
References:	<5 DOT 0 DOT 2 DOT 1 DOT 0 DOT 20010112211047 DOT 00a54a50 AT pop5 DOT banet DOT net>
	<5 DOT 0 DOT 2 DOT 1 DOT 0 DOT 20010111204358 DOT 0368ac40 AT pop5 DOT banet DOT net>
	<5 DOT 0 DOT 2 DOT 1 DOT 0 DOT 20010110233939 DOT 0275e8a0 AT pop5 DOT banet DOT net>
	<5 DOT 0 DOT 2 DOT 1 DOT 0 DOT 20010111204358 DOT 0368ac40 AT pop5 DOT banet DOT net>
	<5 DOT 0 DOT 2 DOT 1 DOT 0 DOT 20010112211047 DOT 00a54a50 AT pop5 DOT banet DOT net>
Mime-Version:	1.0
Reply-To:	djgpp-workers AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	djgpp-workers AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

At 09:44 AM 1/13/01 +0200, Eli Zaretskii wrote:
<Snipped>
 >??? Another mystery?  The binary from the old fil316b.zip is dated
 >April 18, 1997, and its size is 73216.  At least that's what I have 
on
 >two of my machines, and also in the old fil316b.zip.

This is the fil316b.zip that I have from 1997:

G:\DJGPP\V2GNU>dir fil316b.zip

  Volume in drive G is SOFTWARE
  Volume Serial Number is 390C-4791
  Directory of G:\DJGPP\V2GNU

FIL316B  ZIP       996,076  08-28-97  8:56p FIL316B.ZIP
          1 file(s)        996,076 bytes
          0 dir(s)      30,736,384 bytes free

Inside that archive, the executables are all dated 08-26-1997 
03:08PM.  The fil316s.zip I have from that era is 11 months newer 
(07-28-1998), but the newest contents in the source zip are dated 
04-18-1997 08:00PM (the README file).  The README in the binary zip I 
have is identical to the one in the source zip, and dated the same, 
04-18-1997 08:00PM.  Apparently, I downloaded just the binary zip at 
first, and only later downloaded the source zip.  I am *assuming* I got 
the binary from either the AOL or funet.fi mirrors (both of which I was 
using a lot at that time) or directly from the simtelnet site, but I 
have no way at this point of knowing exactly from where I downloaded 
it.

 >> >Since the old binary seems to work better for CDs, I think it is
 >> >worth our while to try to understand why.
 >>
 >> I'm not sure those results are "better" in any sense I can
 >> quantify.  They *are* different (and lower) than the current 
statfs.c
 >> returns.  And let's not forget that Corel Linux/fileutils 4.1 
returns
 >> a third different value, somewhere between the 1997 df value and
 >> the current statfs.c/AX1510 value.
 >
 >I'm confused.  Didn't you, or someone else, say that the old df.exe
 >agreed with Windows Explorer?

Ah, yes, so I did, and so it does.

 >If we have more than 2 different answers, and they are all similar,
 >then I agree that it is a waste of time looking for the reasons for
 >these small differences.

We do indeed have more than two answers --

8/97 df == WinExplorer < Corel Linux 1.2/df v4.1 < statfs.c/AX1510

However, we also have this relationship:

Total size returned by WinExplorer == AX7303 == 8/97 df

I'm beginning to suspect that the "intermediate" 8/97 version of df 
uses AX7303 and not AX1510 to get its results, though there seems no 
way to prove it at this point.

If you would privately email me your April 1997 copy of fil316b.zip as 
an attachment, I will test that one out also to see what results it 
gives.

---------------------------------------------------------
Peter J. Farley III (pjfarley AT dorsai DOT org OR
                      pjfarley AT banet DOT net)

- Raw text -