From: "H. Peter Anvin" <hpa@zytor.com>
To: Pavel Machek <pavel@ucw.cz>
Cc: Karel Zak <kzak@redhat.com>,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
util-linux-ng@vger.kernel.org
Subject: Re: [ANNOUNCE] util-linux-ng v2.17 (stable)
Date: Mon, 11 Jan 2010 15:26:15 -0800 [thread overview]
Message-ID: <4B4BB397.5090500@zytor.com> (raw)
In-Reply-To: <20100111201734.GA11674@elf.ucw.cz>
On 01/11/2010 12:17 PM, Pavel Machek wrote:
>>
>> Uhm, that's just plain wrong.
>>
>> It doesn't matter if there is a "special mapping layer" -- if you're
>> crossing multiple erase blocks you're still having more churn in
>> your flash translation layer, with more wear on the device, and
>> lower performance than if you didn't.
>
> Eraseblocks really should not matter. It is not as if each logical
> sector belongs to one eraseblock....
>
> (OTOH, maybe the eraseblock *groups* that are basis for wear-leveling
> do, or maybe firmware is doing something really really strange.)
> Pavel
Maybe they "should not" matter, but they *do* matter. In most existing
FTLs, each logical sector *does* belong to one erase block, although
which particular erase block that is of course moves around. However,
the invariant that matters though -- and the reason alignment matters --
is that for most FTLs, the *offset* of any particular logical sector
within the erase block it currently belongs to is invariant, i.e. the
FTL operates on physical sectors which are the same size as the erase
blocks.
-hpa
WARNING: multiple messages have this Message-ID (diff)
From: "H. Peter Anvin" <hpa-YMNOUZJC4hwAvxtiuMwx3w@public.gmane.org>
To: Pavel Machek <pavel-+ZI9xUNit7I@public.gmane.org>
Cc: Karel Zak <kzak-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
util-linux-ng-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: [ANNOUNCE] util-linux-ng v2.17 (stable)
Date: Mon, 11 Jan 2010 15:26:15 -0800 [thread overview]
Message-ID: <4B4BB397.5090500@zytor.com> (raw)
In-Reply-To: <20100111201734.GA11674-I/5MKhXcvmPrBKCeMvbIDA@public.gmane.org>
On 01/11/2010 12:17 PM, Pavel Machek wrote:
>>
>> Uhm, that's just plain wrong.
>>
>> It doesn't matter if there is a "special mapping layer" -- if you're
>> crossing multiple erase blocks you're still having more churn in
>> your flash translation layer, with more wear on the device, and
>> lower performance than if you didn't.
>
> Eraseblocks really should not matter. It is not as if each logical
> sector belongs to one eraseblock....
>
> (OTOH, maybe the eraseblock *groups* that are basis for wear-leveling
> do, or maybe firmware is doing something really really strange.)
> Pavel
Maybe they "should not" matter, but they *do* matter. In most existing
FTLs, each logical sector *does* belong to one erase block, although
which particular erase block that is of course moves around. However,
the invariant that matters though -- and the reason alignment matters --
is that for most FTLs, the *offset* of any particular logical sector
within the erase block it currently belongs to is invariant, i.e. the
FTL operates on physical sectors which are the same size as the erase
blocks.
-hpa
--
To unsubscribe from this list: send the line "unsubscribe util-linux-ng" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2010-01-11 23:26 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-08 9:33 [ANNOUNCE] util-linux-ng v2.17 (stable) Karel Zak
2010-01-08 9:33 ` Karel Zak
2010-01-08 21:43 ` H. Peter Anvin
2010-01-08 22:40 ` Andreas Dilger
2010-01-08 22:40 ` Andreas Dilger
2010-01-08 23:33 ` H. Peter Anvin
2010-01-08 23:33 ` H. Peter Anvin
2010-01-11 6:36 ` Martin K. Petersen
2010-01-11 6:36 ` Martin K. Petersen
2010-01-11 7:27 ` H. Peter Anvin
2010-01-11 7:27 ` H. Peter Anvin
2010-01-11 14:05 ` Pavel Machek
2010-01-11 14:05 ` Pavel Machek
2010-01-11 16:52 ` H. Peter Anvin
2010-01-11 20:17 ` Pavel Machek
2010-01-11 20:17 ` Pavel Machek
2010-01-11 23:26 ` H. Peter Anvin [this message]
2010-01-11 23:26 ` H. Peter Anvin
2010-01-12 13:33 ` Artem Bityutskiy
2010-01-12 13:33 ` Artem Bityutskiy
2010-01-12 13:35 ` Artem Bityutskiy
2010-01-12 13:35 ` Artem Bityutskiy
2010-01-11 19:17 ` Artem Bityutskiy
2010-01-11 19:17 ` Artem Bityutskiy
2010-01-12 7:19 ` Jörn Engel
2010-01-12 8:19 ` H. Peter Anvin
2010-01-12 8:19 ` H. Peter Anvin
2010-01-12 9:37 ` Jörn Engel
2010-01-12 9:37 ` Jörn Engel
2010-01-12 16:20 ` H. Peter Anvin
2010-01-12 16:20 ` H. Peter Anvin
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=4B4BB397.5090500@zytor.com \
--to=hpa@zytor.com \
--cc=kzak@redhat.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pavel@ucw.cz \
--cc=util-linux-ng@vger.kernel.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.