From: Joerg Roedel <joro@8bytes.org>
To: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: "ksummit-discuss@lists.linuxfoundation.org"
<ksummit-discuss@lists.linuxfoundation.org>
Subject: Re: [Ksummit-discuss] [CORE TOPIC] Device error handling / reporting / isolation
Date: Mon, 12 May 2014 16:58:06 +0200 [thread overview]
Message-ID: <20140512145806.GM12376@8bytes.org> (raw)
In-Reply-To: <1399666748.2166.68.camel@dabdike.int.hansenpartnership.com>
On Fri, May 09, 2014 at 01:19:08PM -0700, James Bottomley wrote:
> On Fri, 2014-05-09 at 20:13 +0000, Luck, Tony wrote:
> > Defending against buggy hardware is interesting from a RAS perspective.
> > You don't want a card with a stuck address line scribbling on memory
> > that you didn't want it to touch.
>
> But for a laptop or desktop kernel, how far do we want to go? In
> theory, once the iommu is turned on, it corrals the device, since access
> to non programmed addresses (those without IOTLB entries) produces a
> fault. Is there anything extra we need to do beyond turning on the
> IOMMU?
Especially for Laptops and Desktops proper fault handling is important.
Newer GPUS can use the IOMMU to directly access process address spaces
and support demand paging and CPU page-table layouts. Support for these
features in Linux is already being worked on, so handling faults in a
meaningful way is important there too.
Joerg
next prev parent reply other threads:[~2014-05-12 14:58 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-08 12:37 [Ksummit-discuss] [CORE TOPIC] Device error handling / reporting / isolation David Woodhouse
2014-05-08 18:03 ` Bjorn Helgaas
2014-05-08 20:00 ` Rafael J. Wysocki
2014-05-08 19:56 ` James Bottomley
2014-05-09 8:55 ` David Woodhouse
2014-05-09 11:31 ` Laurent Pinchart
2014-05-14 1:28 ` Benjamin Herrenschmidt
2014-05-09 17:48 ` Roland Dreier
2014-05-09 17:58 ` Matthew Wilcox
2014-05-09 18:08 ` Roland Dreier
2014-05-09 18:08 ` Roland Dreier
2014-05-14 1:40 ` Benjamin Herrenschmidt
2014-05-09 18:05 ` Will Deacon
2014-05-12 15:03 ` Joerg Roedel
2014-05-09 19:37 ` Josh Triplett
2014-05-09 19:44 ` David Woodhouse
2014-05-09 19:53 ` Roland Dreier
2014-05-09 20:13 ` Luck, Tony
2014-05-09 20:19 ` James Bottomley
2014-05-10 1:09 ` Laurent Pinchart
2014-05-11 22:43 ` Daniel Vetter
2014-05-12 15:07 ` Joerg Roedel
2014-05-12 15:35 ` Daniel Vetter
2014-05-12 16:16 ` Andy Lutomirski
2014-05-12 16:28 ` Joerg Roedel
2014-05-12 16:59 ` Laurent Pinchart
2014-05-12 17:15 ` Joerg Roedel
2014-05-12 17:11 ` Daniel Vetter
2014-05-12 17:40 ` Joerg Roedel
2014-05-13 10:06 ` Daniel Vetter
2014-05-12 17:04 ` Daniel Vetter
2014-05-13 11:27 ` David Woodhouse
2014-05-13 17:25 ` Daniel Vetter
2014-05-14 1:50 ` Benjamin Herrenschmidt
2014-05-14 20:09 ` Daniel Vetter
2014-05-15 1:08 ` Benjamin Herrenschmidt
2014-05-12 16:26 ` Joerg Roedel
2014-05-12 14:58 ` Joerg Roedel [this message]
2014-05-13 14:37 ` David Woodhouse
2014-05-14 1:46 ` Benjamin Herrenschmidt
2014-05-14 1:43 ` Benjamin Herrenschmidt
2014-05-14 1:42 ` Benjamin Herrenschmidt
2014-05-14 1:24 ` Benjamin Herrenschmidt
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=20140512145806.GM12376@8bytes.org \
--to=joro@8bytes.org \
--cc=James.Bottomley@HansenPartnership.com \
--cc=ksummit-discuss@lists.linuxfoundation.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.