From: "Jan Beulich" <JBeulich@suse.com>
To: Haozhong Zhang <haozhong.zhang@intel.com>
Cc: Juergen Gross <JGross@suse.com>,
Kevin Tian <kevin.tian@intel.com>,
Stefano Stabellini <sstabellini@kernel.org>,
Wei Liu <wei.liu2@citrix.com>,
George Dunlap <George.Dunlap@eu.citrix.com>,
Andrew Cooper <andrew.cooper3@citrix.com>,
Ian Jackson <ian.jackson@eu.citrix.com>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>,
Jun Nakajima <jun.nakajima@intel.com>,
Xiao Guangrong <guangrong.xiao@linux.intel.com>
Subject: Re: [RFC Design Doc v2] Add vNVDIMM support for Xen
Date: Wed, 03 Aug 2016 03:47:44 -0600 [thread overview]
Message-ID: <57A1D9E002000078001021C9@prv-mh.provo.novell.com> (raw)
In-Reply-To: <20160803093713.5ydxavzsmj2x5353@hz-desktop>
>>> On 03.08.16 at 11:37, <haozhong.zhang@intel.com> wrote:
> On 08/03/16 02:45, Jan Beulich wrote:
>> >>> On 03.08.16 at 08:54, <haozhong.zhang@intel.com> wrote:
>> > On 08/02/16 08:46, Jan Beulich wrote:
>> >> >>> On 18.07.16 at 02:29, <haozhong.zhang@intel.com> wrote:
>> >> > (4) Because the reserved area is now used by Xen hypervisor, it
>> >> > should not be accessible by Dom0 any more. Therefore, if a host
>> >> > pmem device is recorded by Xen hypervisor, Xen will unmap its
>> >> > reserved area from Dom0. Our design also needs to extend Linux
>> >> > NVDIMM driver to "balloon out" the reserved area after it
>> >> > successfully reports a pmem device to Xen hypervisor.
>> >>
>> >> ... "balloon out" ... _after_? That'd be unsafe.
>> >>
>> >
>> > Before ballooning is accomplished, the pmem driver does not create any
>> > device node under /dev/ and hence no one except the pmem drive can
>> > access the reserved area on pmem, so I think it's okey to balloon
>> > after reporting.
>>
>> Right now Dom0 isn't allowed to access any memory in use by Xen
>> (and not explicitly shared), and I don't think we should deviate
>> from that model for pmem.
>
> In this design, Xen hypervisor unmaps the reserved area from Dom0 so
> that Dom0 cannot access the reserved area afterwards. And "balloon" is
> in fact not a memory ballooning, because Linux kernel never allocates
> from pmem like normal ram. In my current implementation, it's just to
> remove the reserved area from a resource struct covering pmem.
Ah, in that case please either use a different term, or explain what
"balloon out" is meant to mean in this context.
>> >> > 4.2.3 Get Host Machine Address (SPA) of Host pmem Files
>> >> >
>> >> > Before a pmem file is assigned to a domain, we need to know the host
>> >> > SPA ranges that are allocated to this file. We do this work in xl.
>> >> >
>> >> > If a pmem device /dev/pmem0 is given, xl will read
>> >> > /sys/block/pmem0/device/{resource,size} respectively for the start
>> >> > SPA and size of the pmem device.
>> >> >
>> >> > If a pre-allocated file /mnt/dax/file is given,
>> >> > (1) xl first finds the host pmem device where /mnt/dax/file is. Then
>> >> > it uses the method above to get the start SPA of the host pmem
>> >> > device.
>> >> > (2) xl then uses fiemap ioctl to get the extend mappings of
>> >> > /mnt/dax/file, and adds the corresponding physical offsets and
>> >> > lengths in each mapping entries to above start SPA to get the SPA
>> >> > ranges pre-allocated for this file.
>> >>
>> >> Remind me again: These extents never change, not even across
>> >> reboot? I think this would be good to be written down here explicitly.
>> >
>> > Yes
>> >
>> >> Hadn't there been talk of using labels to be able to allow a guest to
>> >> own the exact same physical range again after reboot or guest or
>> >> host?
>> >
>> > You mean labels in NVDIMM label storage area? As defined in Intel
>> > NVDIMM Namespace Specification, labels are used to specify
>> > namespaces. For a pmem interleave set (possible cross several dimms),
>> > at most one pmem namespace (and hence at most one label) is
>> > allowed. Therefore, labels can not be used to partition pmem.
>>
>> Okay. But then how do particular ranges get associated with the
>> owning guest(s)? Merely by SPA would seem rather fragile to me.
>>
>
> By using the file name, e.g. if I specify vnvdimm = [ 'file=/mnt/dax/foo' ]
> in a domain config file, SPA occupied by /mnt/dax/foo are mapped to
> the domain. If the same file is used every time the domain is created,
> the same virtual device will be seen by that domain.
So what if the file got deleted and re-created in between? Since
I don't think you can specify the SPAs to use when creating such
a file, such an operation would be quite different from removing
and re-adding e.g. a specific PCI device (to be used by a guest)
on a host (while the guest is not running).
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
next prev parent reply other threads:[~2016-08-03 9:47 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-18 0:29 [RFC Design Doc v2] Add vNVDIMM support for Xen Haozhong Zhang
2016-07-18 8:36 ` Tian, Kevin
2016-07-18 9:01 ` Zhang, Haozhong
2016-07-19 0:58 ` Tian, Kevin
2016-07-19 2:10 ` Zhang, Haozhong
2016-07-19 1:57 ` Bob Liu
2016-07-19 2:40 ` Haozhong Zhang
2016-08-02 14:46 ` Jan Beulich
2016-08-03 6:54 ` Haozhong Zhang
2016-08-03 8:45 ` Jan Beulich
2016-08-03 9:37 ` Haozhong Zhang
2016-08-03 9:47 ` Jan Beulich [this message]
2016-08-03 10:08 ` Haozhong Zhang
2016-08-03 10:18 ` Jan Beulich
2016-08-03 21:25 ` Konrad Rzeszutek Wilk
2016-08-03 23:16 ` Konrad Rzeszutek Wilk
2016-08-04 1:51 ` Haozhong Zhang
2016-08-04 8:52 ` Haozhong Zhang
2016-08-04 9:25 ` Jan Beulich
2016-08-04 9:35 ` Haozhong Zhang
2016-08-04 14:51 ` Konrad Rzeszutek Wilk
2016-08-04 14:51 ` Konrad Rzeszutek Wilk
2016-08-05 6:25 ` Haozhong Zhang
2016-08-05 13:29 ` Konrad Rzeszutek Wilk
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=57A1D9E002000078001021C9@prv-mh.provo.novell.com \
--to=jbeulich@suse.com \
--cc=George.Dunlap@eu.citrix.com \
--cc=JGross@suse.com \
--cc=andrew.cooper3@citrix.com \
--cc=guangrong.xiao@linux.intel.com \
--cc=haozhong.zhang@intel.com \
--cc=ian.jackson@eu.citrix.com \
--cc=jun.nakajima@intel.com \
--cc=kevin.tian@intel.com \
--cc=sstabellini@kernel.org \
--cc=wei.liu2@citrix.com \
--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).