From: George Dunlap <George.Dunlap@eu.citrix.com>
To: Gordan Bobic <gordan@bobich.net>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>
Subject: Re: Bug: Limitation of <=2GB RAM in domU persists with 4.3.0
Date: Thu, 25 Jul 2013 20:18:08 +0100 [thread overview]
Message-ID: <CAFLBxZZjvWchAr3SkcWOScgNxz5onZW8DX8H+hm_UzC-sFDJjw@mail.gmail.com> (raw)
In-Reply-To: <51F051F1.5050806@bobich.net>
On Wed, Jul 24, 2013 at 11:15 PM, Gordan Bobic <gordan@bobich.net> wrote:
> Attached are the logs (loglvl=all) and configs for 2GB (working) and 8GB
> (screen corruption + domU crash + sometimes dom0 crashing with it).
>
> I can see in the xl-dmesg log in 8GB case that there is memory remapping
> going on to allow for the lowmem MMIO hole, but it doesn't seem to help.
Gordan,
There's a possibility that it's actually got nothing to do with
relocation, but with bugs in your hardware. Can you try:
* Set the guest memory to 3600
* Boot the guest, and check to make sure that xl dmesg shows does
*not* relocate memory?
* Report whether it crashes?
If it's a bug in the hardware, I would expect to see that memory was
not relocated, but that the system will lock up anyway.
Can you also do lspci -vvv in dom0 before assigning the device and
attach the output?
The hardware bug we've seen is this: In order for the IOMMU to work
properly, *all* DMA transactions must be passed up to the root bridge
so the IOMMU can translate the addresses from guest address to host
address. Unfortunately, an awful lot of bridges will not do this
properly, which means that the address is not translated properly,
which means that if a *guest* memory address overlaps the a *host*
MMIO range, badness ensues. There's nothing we can do about this in
Xen other than make the guest MMIO hole the same size as the host MMIO
hole.
Thanks,
-George
next prev parent reply other threads:[~2013-07-25 19:18 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-23 22:34 Bug: Limitation of <=2GB RAM in domU persists with 4.3.0 Gordan Bobic
2013-07-24 14:08 ` Konrad Rzeszutek Wilk
2013-07-24 14:17 ` Gordan Bobic
2013-07-24 16:06 ` Konrad Rzeszutek Wilk
2013-07-24 16:14 ` Gordan Bobic
2013-07-24 16:31 ` Konrad Rzeszutek Wilk
2013-07-24 17:26 ` Gordan Bobic
2013-07-24 22:15 ` Gordan Bobic
2013-07-25 19:18 ` George Dunlap [this message]
2013-07-25 21:48 ` Gordan Bobic
2013-07-25 22:23 ` Gordan Bobic
2013-07-26 0:21 ` Ian Campbell
2013-07-26 1:15 ` Andrew Bobulsky
2013-07-26 9:28 ` Gordan Bobic
2013-07-26 13:11 ` Gordan Bobic
2013-07-31 17:53 ` George Dunlap
2013-07-31 17:56 ` Andrew Cooper
2013-07-31 19:36 ` Gordan Bobic
2013-07-31 19:35 ` Gordan Bobic
2013-08-01 9:15 ` George Dunlap
2013-08-01 13:10 ` Fabio Fantoni
2013-08-02 14:43 ` George Dunlap
2013-07-28 10:26 ` Konrad Rzeszutek Wilk
2013-07-28 21:24 ` Gordan Bobic
2013-07-28 23:17 ` Konrad Rzeszutek Wilk
2013-07-28 23:30 ` Gordan Bobic
2013-07-29 9:53 ` Ian Campbell
2013-07-26 9:23 ` Gordan Bobic
2013-07-29 11:14 ` Ian Campbell
2013-07-29 18:04 ` Konrad Rzeszutek Wilk
2013-09-03 13:53 ` Gordan Bobic
2013-09-03 14:59 ` Konrad Rzeszutek Wilk
2013-09-03 19:47 ` HVM support for e820_host (Was: Bug: Limitation of <=2GB RAM in domU persists with 4.3.0) Gordan Bobic
2013-09-03 20:35 ` Gordan Bobic
2013-09-03 20:49 ` Gordan Bobic
2013-09-03 21:10 ` Konrad Rzeszutek Wilk
2013-09-03 21:24 ` Gordan Bobic
2013-09-03 21:30 ` Konrad Rzeszutek Wilk
2013-09-04 0:18 ` Gordan Bobic
2013-09-04 14:08 ` Konrad Rzeszutek Wilk
2013-09-04 14:23 ` Gordan Bobic
2013-09-04 18:00 ` Konrad Rzeszutek Wilk
2013-09-03 21:08 ` Konrad Rzeszutek Wilk
2013-09-04 9:21 ` Gordan Bobic
2013-09-04 11:01 ` Gordan Bobic
2013-09-04 13:11 ` Gordan Bobic
2013-09-04 20:18 ` Gordan Bobic
2013-09-05 2:04 ` Konrad Rzeszutek Wilk
2013-09-05 9:41 ` Gordan Bobic
2013-09-05 10:00 ` Gordan Bobic
2013-09-05 12:36 ` Konrad Rzeszutek Wilk
2013-09-05 10:26 ` Gordan Bobic
2013-09-05 12:38 ` Konrad Rzeszutek Wilk
2013-09-05 21:13 ` Gordan Bobic
2013-09-05 21:29 ` Gordan Bobic
2013-09-05 21:46 ` Gordan Bobic
2013-09-05 22:23 ` Konrad Rzeszutek Wilk
2013-09-05 22:42 ` Gordan Bobic
2013-09-06 13:09 ` Konrad Rzeszutek Wilk
2013-09-06 14:09 ` Gordan Bobic
2013-09-05 22:45 ` Gordan Bobic
2013-09-05 23:01 ` Konrad Rzeszutek Wilk
2013-09-06 12:23 ` Gordan Bobic
2013-09-06 13:20 ` Konrad Rzeszutek Wilk
2013-09-06 14:45 ` Gordan Bobic
2013-09-05 22:33 ` Gordan Bobic
2013-09-06 13:04 ` Konrad Rzeszutek Wilk
2013-09-06 13:34 ` Gordan Bobic
2013-09-06 14:32 ` Konrad Rzeszutek Wilk
2013-09-06 16:30 ` Gordan Bobic
2013-09-06 19:54 ` Gordan Bobic
2013-09-10 13:35 ` Konrad Rzeszutek Wilk
2013-09-10 15:04 ` Gordan Bobic
2013-07-25 21:26 ` Bug: Limitation of <=2GB RAM in domU persists with 4.3.0 Gordan Bobic
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=CAFLBxZZjvWchAr3SkcWOScgNxz5onZW8DX8H+hm_UzC-sFDJjw@mail.gmail.com \
--to=george.dunlap@eu.citrix.com \
--cc=andrew.cooper3@citrix.com \
--cc=gordan@bobich.net \
--cc=xen-devel@lists.xen.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).