+1

Writing specs is useless if you are not going to put the engineering effort
in.

Let's see FreeDOS 1.2 released sometime soon.
I don't want to sound harsh, but "drawing specs" is easy, anybody can do
that. Do some code that actually works, then we're talking.

Mateusz



On 25/09/2015 03:44, Mercury Thirteen wrote:
> I don't see where we need multitasking for NAS use. A program could be
> made to both handle incoming requests while serving data and doing other
> tasks, eliminating the need for a proper multitasking kernel. Even if
> that was the case, the bloat of the Linux kernel would make it
> prohibitive in certain applications. Sure, you could compile your own
> slimmed version specific for the task, but DOS' lightweight requirements
> and comparatively speediness make it the clear out-of-the-box winner.
>
> An improved FAT was something I planned for eventual use with the 32-bit
> kernel we're creating. Perhaps it's best to keep it that way since
> regular DOS (as Eric points out) is not that well suited for the job
> being sans-multitasking as it is.
>
> I will draw up a spec as I said when I get the time. After that,
> implementation is up to the rest of the community. We could just as
> easily go with FAT+ or not advance the filesystem at all.
>
> On 9/24/2015 9:21 PM, JAYDEN CHARBONNEAU wrote:
>> First thing I noticed (This may be just me.),is that we need more
>> memory for the OS environment.Normally,when I boot FreeDOS on ANY
>> computer (Be it modern or old),the memory is always 601 MB free.More
>> memory would be needed for a bigger file system and multi-tasking.
>>
>> On Thu, Sep 24, 2015 at 5:54 PM, Eric Auer <e.a...@jpberlin.de
>> <mailto:e.a...@jpberlin.de>> wrote:
>>
>>
>>     Hi Mercury,
>>
>>     so you want to run a NAS or home automation on DOS?
>>
>>     For NAS, you need a multitasking OS, not DOS. For
>>     home automation, which limitation of FAT would be
>>     a problem? Same for other light embedded devices.
>>
>>     Flash does not give good performance for FAT, but
>>     embedded devices would have been free to use one
>>     of many available Linux filesystems. But did not.
>>
>>     Of course the question can be extended: What if an
>>     existing nicer-than-FAT filesystem is used more in
>>     DOS? Have a look at what already EXISTS for Linux,
>>     then have a look at the source code to check which
>>     filesystems are 1. simple enough to make a "light"
>>     DOS driver possible (some might even be so simple
>>     that booting DOS from them is feasible, but only a
>>     really popular filesystem may get kernel drivers),
>>     2. better than FAT in some way (e.g. more speed on
>>     flash storage, better space allocation or LFN in a
>>     less insane way than VFAT) but 3. not putting lots
>>     of code into features which mean nothing for DOS,
>>     such as ACL based file permissions or extreme file
>>     or disk size support beyond existing FAT32 API or
>>     "network redirector" API expressible number range.
>>
>>     Looking forward to your review of existing FS-es!
>>
>>     Of course with an outlook towards which properties
>>     a not-yet-existing FS could have to be even nicer
>>     for use within a DOS based storage "ecosystem".
>>
>>     Cheers, Eric
>>
>>     PS: By "light", I mean a driver which is not 100s
>>     of kilobytes in size and which can be fast with a
>>     bit of DOS RAM and XMS instead of needing 500 MB
>>     of DPMI RAM and protected mode implementation :-)
>>
>>
>>
>>     > NAS devices, home automation computers and other similar devices
are
>>     > becoming increasingly common, and offering a filesystem finally
>>     capable
>>     > of handling the sizes of modern hard drives could be a welcome
>>     > improvement for them, and just may help get FreeDOS used in a
>>     wider market.
>>     >
>>     > How do we know this isn't a chicken-and-egg problem? Maybe all the
>>     > devices only use the proprietary exFAT because there was no open
>>     > alternative. Maybe, had there been one available, we would all...



------------------------------------------------------------------------------
_______________________________________________
Freedos-devel mailing list
Freedos-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freedos-devel
------------------------------------------------------------------------------
_______________________________________________
Freedos-devel mailing list
Freedos-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/freedos-devel

Reply via email to