Date: Sun, 27 Apr 1997 12:20:44 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: tony@dell.com (Tony Overfield) Cc: terry@lambert.org, ksmm@cybercom.net, joa@kuebart.stuttgart.netsurf.de, hackers@freebsd.org Subject: Re: VFAT 32 support in msdosfs Message-ID: <199704271920.MAA09008@phaeton.artisoft.com> In-Reply-To: <3.0.1.32.19970426195829.007af2b0@bugs.us.dell.com> from "Tony Overfield" at Apr 26, 97 07:58:29 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > I'm not sure that an OEMSR2 INT 21 is capable of identifying and
> > booting from a "VFAT" drive at all.
>
> This is nonsense.
Well, I did say I wasn't sure.
> FAT32 is a superset of FAT16/FAT12. The OEMSR2 version of io.sys
> supports all three. There is not a special version of io.sys
> which only supports FAT32 partitions.
>
> io.sys does not "do" VFAT at all. VFAT "happens" when the
> protected mode VFAT driver is loaded. This is why real-mode
> Windows 95 (DOS mode) cannot see or use long filenames.
>
> Of course, io.dos/msdos.dos are irrelevant, since they aren't even
> used when booting Windows 95, they're used when booting your
> *previous* DOS, which might be MS-DOS 5.0.
>
> Why you think that INT 21 is related to any of this remains a
> mystery to me.
Because the Schulman book claims that they are used in the boot process
in order to locate and load windows components prior to switching to
protected mode. This is backed by my personal experience in IFS-land,
where an installed FS must be able to convert a real mode handle to
a protected mode IFS open file to support TSR's and other FS-using
crap loaded from config.sys and/or autoexec.bat.
If the FS layout is different in VFAT32 (it is), then they would need
to be able to traverse the FS, both for running DOS stuff ("Realmode
DOS Prompt" versions of the MSDOS command.com shortcut, and the "shutdown
to DOS prompt"), and for starting up windows from the FS.
Since VFAT32 "solves" the cluster allocation/fragmentation problem,
and it is the default boot FS, it logically follows that the INT 21
installed during the boot phase would have to have VFAT32 specific
knowledge.
The only thing I wasn't sure of (and which you cleared up; thanks) was
whether or not it *only* had VFAT32 specific knowledge.
Regards,
Terry Lambert
terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199704271920.MAA09008>