From: "H. Peter Anvin" <hpa@zytor.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andi Kleen <andi@firstfloor.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>,
Al Viro <viro@zeniv.linux.org.uk>
Subject: Re: Word-at-a-time dcache name accesses (was Re: .. anybody know of any filesystems that depend on the exact VFS 'namehash' implementation?)
Date: Fri, 02 Mar 2012 17:17:39 -0800 [thread overview]
Message-ID: <4F517133.5070504@zytor.com> (raw)
In-Reply-To: <CA+55aFwoifDB9YPC53KrFm1DZNVEzYL4+dfP96GE3rz8YCVaMg@mail.gmail.com>
On 03/02/2012 05:11 PM, Linus Torvalds wrote:
> On Fri, Mar 2, 2012 at 5:02 PM, H. Peter Anvin <hpa@zytor.com> wrote:
>>
>> Note that does mean we need a guard page after each and every
>> discontiguous RAM range, not just the last one. Raising that issue
>> since we have had serious bugs in that area in the past.
>
> Are you sure? I didn't think we even *mapped* things at that granularity.
>
> We only really need a guard page at the end of an actual end-of-ram
> where we no longer have page tables and/or could hit device space.
>
Yes of course. Note that I'm currently pushing for mapping only RAM
regions; we have a lot of bugs and workarounds related to mapping too
much, and the answer to fixing that should be pretty obvious.
> For robustness and actual deployment, I do think that yes, we do want
> to make it an explicit rule.
Definitely. Ideally those pages should be zeroed and mapped readonly.
-hpa
next prev parent reply other threads:[~2012-03-03 1:17 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-03-02 23:46 Word-at-a-time dcache name accesses (was Re: .. anybody know of any filesystems that depend on the exact VFS 'namehash' implementation?) Linus Torvalds
2012-03-03 0:02 ` Ted Ts'o
2012-03-03 0:17 ` david
2012-03-03 0:24 ` Linus Torvalds
2012-03-04 22:19 ` Matthew Wilcox
2012-03-04 23:27 ` Linus Torvalds
2012-03-03 0:17 ` Linus Torvalds
2012-03-03 0:38 ` H. Peter Anvin
2012-03-03 0:57 ` Linus Torvalds
2012-03-03 1:02 ` H. Peter Anvin
2012-03-03 1:11 ` Linus Torvalds
2012-03-03 1:17 ` H. Peter Anvin [this message]
2012-03-03 16:12 ` Word-at-a-time dcache name accesses Andi Kleen
2012-03-03 18:47 ` H. Peter Anvin
2012-03-03 20:10 ` Word-at-a-time dcache name accesses (was Re: .. anybody know of any filesystems that depend on the exact VFS 'namehash' implementation?) Linus Torvalds
2012-03-04 2:27 ` Word-at-a-time dcache name accesses (was Re: .. anybody know ofany " Tetsuo Handa
2012-03-04 4:31 ` Andi Kleen
2012-03-05 3:58 ` Word-at-a-time dcache name accesses (was Re: .. anybody know of any " Jason Garrett-Glaser
2012-03-05 5:38 ` Linus Torvalds
2012-03-27 4:42 ` Brian Gerst
2012-03-27 4:42 ` Brian Gerst
2012-03-27 5:02 ` Dave Jones
2012-03-27 5:31 ` Brian Gerst
2012-03-28 0:39 ` Linus Torvalds
2012-03-28 0:39 ` Linus Torvalds
2012-03-28 0:50 ` Linus Torvalds
2012-03-28 0:50 ` Linus Torvalds
2012-03-28 0:56 ` Brian Gerst
[not found] ` <CACvQF53YasSCUit2KoWDimgObknCz++aU90MesSfvAZTeUFQHw@mail.gmail.com>
2013-04-04 16:50 ` Lai Jiangshan
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=4F517133.5070504@zytor.com \
--to=hpa@zytor.com \
--cc=andi@firstfloor.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=torvalds@linux-foundation.org \
--cc=viro@zeniv.linux.org.uk \
/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.