From: Eric Dumazet <dada1@cosmosbay.com>
To: Grzegorz Kulewski <kangur@polcom.net>
Cc: Mikulas Patocka <mikulas@artax.karlin.mff.cuni.cz>,
linux-kernel@vger.kernel.org
Subject: Re: New filesystem for Linux
Date: Fri, 03 Nov 2006 00:10:20 +0100 [thread overview]
Message-ID: <454A7ADC.4030203@cosmosbay.com> (raw)
In-Reply-To: <Pine.LNX.4.63.0611022346450.14187@alpha.polcom.net>
Grzegorz Kulewski a écrit :
> Hi,
>
> On Thu, 2 Nov 2006, Mikulas Patocka wrote:
>> As my PhD thesis, I am designing and writing a filesystem, and it's
>> now in a state that it can be released. You can download it from
>> http://artax.karlin.mff.cuni.cz/~mikulas/spadfs/
>
> "Disk that can atomically write one sector (512 bytes) so that the sector
> contains either old or new content in case of crash."
>
> Well, maybe I am completly wrong but as far as I understand no disk
> currently will provide such requirement. Disks can have (after halted
> write):
> - old data,
> - new data,
> - nothing (unreadable sector - result of not full write and disk
> internal checksum failute for that sector, happens especially often if
> you have frequent power outages).
>
I believe some vendors have such devices. Mikulas called them 'disk', but it's
in fact a (disk(s), controler, ram, battery)
Some controlers are even able to write into flash memory the un-written data
when/if the battery/power is about to fail. When power goes up, controler can
finaly do the writes on disks.
> And possibly some broken drives may also return you something that they
> think is good data but really is not (shouldn't happen since both disks
> and cables should be protected by checksums, but hey... you can never be
> absolutely sure especially on very big storages).
>
> So... isn't this making your filesystem a little flawed in design?
Well... even RAM can fail :) In this case isnt linux flawed in design ?
next prev parent reply other threads:[~2006-11-02 23:12 UTC|newest]
Thread overview: 99+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-11-02 21:52 New filesystem for Linux Mikulas Patocka
2006-11-02 22:32 ` Gabriel C
2006-11-03 1:22 ` Mikulas Patocka
2006-11-03 1:41 ` Andrew Morton
2006-11-03 17:14 ` Oleg Verych
2006-11-03 17:09 ` Mikulas Patocka
2006-11-03 17:36 ` Oleg Verych
2006-11-03 18:14 ` Mikulas Patocka
2006-11-03 19:08 ` Adrian Bunk
2006-11-03 19:32 ` Oleg Verych
2006-11-03 19:00 ` Alan Cox
2006-11-03 19:14 ` Andi Kleen
2006-11-03 2:09 ` Gabriel C
2006-11-03 8:26 ` Jan Engelhardt
2006-11-03 11:52 ` Mikulas Patocka
2006-11-03 11:59 ` Mikulas Patocka
2006-11-03 12:50 ` Jan Engelhardt
2006-11-03 18:48 ` Mikulas Patocka
2006-11-03 21:51 ` Jan Engelhardt
2006-11-03 11:47 ` Mikulas Patocka
2006-11-02 22:53 ` Eric Dumazet
2006-11-03 1:28 ` Mikulas Patocka
2006-11-03 1:43 ` Andrew Morton
2006-11-04 18:40 ` Mikulas Patocka
2006-11-04 19:07 ` Eric Dumazet
2006-11-04 19:39 ` Tomasz Torcz
2006-11-05 1:58 ` Alan Cox
2006-11-05 2:09 ` Patrick McFarland
2006-11-05 13:03 ` Maurizio Lombardi
2006-11-05 20:16 ` H. Peter Anvin
2006-11-02 22:54 ` Grzegorz Kulewski
2006-11-02 23:10 ` Eric Dumazet [this message]
2006-11-02 23:19 ` Mikulas Patocka
2006-11-02 23:29 ` Grzegorz Kulewski
2006-11-03 1:34 ` Mikulas Patocka
2006-11-03 20:30 ` Christoph Lameter
2006-11-04 18:46 ` Mikulas Patocka
2006-11-05 12:02 ` Theodore Tso
2006-11-03 22:00 ` Oleg Verych
2006-11-03 22:42 ` Mikulas Patocka
2006-11-03 0:57 ` Nigel Cunningham
2006-11-03 13:05 ` Ric Wheeler
2006-11-06 2:42 ` Phillip Susi
2006-11-04 19:59 ` Albert Cahalan
2006-11-04 21:01 ` Jan-Benedict Glaw
2006-11-05 16:37 ` Albert Cahalan
2006-11-04 23:38 ` Mikulas Patocka
2006-11-04 23:46 ` Kyle Moffett
2006-11-05 20:26 ` H. Peter Anvin
2006-11-05 21:27 ` Rene Herman
2006-11-05 21:51 ` H. Peter Anvin
2006-11-06 0:36 ` Rene Herman
2006-11-05 21:49 ` Pavel Machek
2006-11-05 1:57 ` Alan Cox
2006-11-05 11:14 ` James Courtier-Dutton
2006-11-05 11:27 ` Brad Campbell
2006-11-05 12:37 ` Alan Cox
2006-11-06 2:48 ` Phillip Susi
2006-11-05 16:22 ` Albert Cahalan
2006-11-05 17:18 ` Mikulas Patocka
2006-11-05 18:14 ` Alan Cox
2006-11-05 18:18 ` Mikulas Patocka
2006-11-05 19:14 ` Alan Cox
2006-11-02 23:15 ` Linus Torvalds
2006-11-03 20:02 ` Paul E. McKenney
2006-11-02 23:41 ` Andi Kleen
2006-11-03 1:45 ` Mikulas Patocka
2006-11-03 13:47 ` Nikita Danilov
2006-11-03 14:39 ` Mikulas Patocka
2006-11-02 23:59 ` Jörn Engel
2006-11-03 1:19 ` Mikulas Patocka
2006-11-03 10:19 ` Jörn Engel
2006-11-03 11:56 ` Mikulas Patocka
2006-11-03 12:21 ` Jörn Engel
2006-11-03 13:31 ` Mikulas Patocka
2006-11-03 13:48 ` Jörn Engel
2006-11-03 14:19 ` Mikulas Patocka
2006-11-03 14:53 ` Jörn Engel
2006-11-03 19:01 ` Mikulas Patocka
2006-11-04 10:46 ` Jörn Engel
2006-11-04 18:50 ` Mikulas Patocka
2006-11-06 21:19 ` Jörn Engel
2006-11-03 19:51 ` Adrian Bunk
2006-11-03 19:00 ` dean gaudet
2006-11-04 10:53 ` Jörn Engel
2006-11-04 11:13 ` dean gaudet
2006-11-04 20:07 ` Jörn Engel
2006-11-04 18:52 ` Mikulas Patocka
2006-11-04 18:56 ` Grzegorz Kulewski
2006-11-04 19:18 ` Mikulas Patocka
2006-11-04 17:37 ` Gautham R Shenoy
2006-11-04 18:27 ` Eric Dumazet
2006-11-05 22:33 ` Paul E. McKenney
2006-11-05 0:52 ` Linus Torvalds
2006-11-05 4:14 ` Mikulas Patocka
2006-11-05 8:34 ` Willy Tarreau
2006-11-05 11:31 ` Jan Engelhardt
2006-11-05 14:48 ` Bruno Cesar Ribas
-- strict thread matches above, loose matches on Subject: below --
2006-11-06 17:40 Al Boldi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=454A7ADC.4030203@cosmosbay.com \
--to=dada1@cosmosbay.com \
--cc=kangur@polcom.net \
--cc=linux-kernel@vger.kernel.org \
--cc=mikulas@artax.karlin.mff.cuni.cz \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox