From: Wen Congyang <wency@cn.fujitsu.com>
To: David Vrabel <david.vrabel@citrix.com>,
xen devel <xen-devel@lists.xen.org>
Cc: Ian Campbell <Ian.Campbell@citrix.com>,
Ian Jackson <Ian.Jackson@eu.citrix.com>,
Jiang Yunhong <yunhong.jiang@intel.com>,
Dong Eddie <eddie.dong@intel.com>,
Paul Durrant <paul.durrant@citrix.com>,
Yang Hongyang <yanghy@cn.fujitsu.com>,
Lai Jiangshan <laijs@cn.fujitsu.com>
Subject: Re: [RFC Patch v3 03/18] don't zero out ioreq page
Date: Fri, 12 Sep 2014 15:33:49 +0800 [thread overview]
Message-ID: <5412A1DD.9010707@cn.fujitsu.com> (raw)
In-Reply-To: <54099457.4040603@citrix.com>
On 09/05/2014 06:45 PM, David Vrabel wrote:
> On 05/09/14 10:10, Wen Congyang wrote:
>> ioreq page may contain some pending I/O requests, and we need to
>> handle the pending I/O req after migration.
>>
>> TODO:
>> 1. update qemu to handle the pending I/O req
>
> Under what conditions can there be pending I/O? A domain suspend should
> not complete until any I/O accesses are complete.
IIRC, I print ioreq.state after suspending vm, and ioreq.state can be
STATE_IOREQ_NONE, STATE_IOREQ_READY or STATE_IORESP_READY. If the
state is STATE_IOREQ_READY, we should kick vcpu event channel on resume
after migration.
But I test it again today, the state is always STATE_IOREQ_NONE...
If the state is always STATE_IOREQ_NONE, no need to touch hypervisor/Qemu
to handle pending I/O req(state is STATE_IOREQ_READY).
Thanks
Wen Congyang
>
> Suspending a domain that's part-way through an instruction seems mad to
> me -- how are you supposed to capture this sort of state?
>
> David
> .
>
next prev parent reply other threads:[~2014-09-12 7:33 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-09-05 9:10 [RFC Patch v3 00/18] Some bugfix patches Wen Congyang
2014-09-05 9:10 ` [RFC Patch v3 01/18] copy the correct page to memory Wen Congyang
2014-09-08 11:27 ` Ian Campbell
2014-09-08 11:58 ` Andrew Cooper
2014-09-05 9:10 ` [RFC Patch v3 02/18] csum the correct page Wen Congyang
2014-09-08 11:28 ` Ian Campbell
2014-09-05 9:10 ` [RFC Patch v3 03/18] don't zero out ioreq page Wen Congyang
2014-09-05 9:25 ` Paul Durrant
2014-09-05 9:33 ` Wen Congyang
2014-09-05 9:39 ` Paul Durrant
2014-09-05 10:45 ` David Vrabel
2014-09-12 7:33 ` Wen Congyang [this message]
2014-09-05 9:10 ` [RFC Patch v3 04/18] blktap2: dynamic allocate aio_requests to avoid -EBUSY error Wen Congyang
2014-09-08 11:34 ` Ian Campbell
2014-09-24 18:22 ` Shriram Rajagopalan
2014-09-05 9:10 ` [RFC Patch v3 05/18] blktap2: return the correct dev path Wen Congyang
2014-09-05 9:10 ` [RFC Patch v3 06/18] blktap2: use correct way to get free event id Wen Congyang
2014-09-05 9:10 ` [RFC Patch v3 07/18] blktap2: don't return negative " Wen Congyang
2014-09-05 9:10 ` [RFC Patch v3 08/18] blktap2: use correct way to define array Wen Congyang
2014-09-05 9:10 ` [RFC Patch v3 09/18] block-remus: fix memory leak Wen Congyang
2014-09-24 19:37 ` Shriram Rajagopalan
2014-09-25 5:23 ` Wen Congyang
2014-09-25 11:14 ` Shriram Rajagopalan
2014-09-26 2:29 ` Wen Congyang
2014-09-05 9:10 ` [RFC Patch v3 10/18] block-remus: pass uuid to the callback td_open Wen Congyang
2014-09-24 19:27 ` Shriram Rajagopalan
2014-09-05 9:10 ` [RFC Patch v3 11/18] block-remus: use correct way to get remus_image Wen Congyang
2014-09-24 19:26 ` Shriram Rajagopalan
2014-09-05 9:10 ` [RFC Patch v3 12/18] block-remus: fix bug in tdremus_close() Wen Congyang
2014-09-24 19:24 ` Shriram Rajagopalan
2014-09-05 9:10 ` [RFC Patch v3 13/18] don't call client_flush() when switching to unprotected mode Wen Congyang
2014-09-05 9:10 ` [RFC Patch v3 14/18] pass correct file to qemu if we use blktap2 Wen Congyang
2014-09-08 11:35 ` Ian Campbell
2014-09-05 9:10 ` [RFC Patch v3 15/18] support blktap remus in xl Wen Congyang
2014-09-08 11:39 ` Ian Campbell
2014-09-10 7:19 ` Wen Congyang
2014-09-10 10:04 ` Ian Campbell
2014-09-10 10:36 ` Wen Congyang
2014-09-05 9:10 ` [RFC Patch v3 16/18] update libxl__device_disk_from_xs_be() to support blktap device Wen Congyang
2014-09-08 11:42 ` Ian Campbell
2014-09-09 1:57 ` Wen Congyang
2014-09-11 7:58 ` Wen Congyang
2014-09-12 8:53 ` Wei Liu
2014-09-12 9:03 ` Wen Congyang
2014-09-12 10:35 ` Wei Liu
2014-09-05 9:11 ` [RFC Patch v3 17/18] read nictype from xenstore Wen Congyang
2014-09-08 11:41 ` Ian Campbell
2014-09-05 9:11 ` [RFC Patch v3 18/18] x86/hvm: Always set pending event injection when loading VMC[BS] state Wen Congyang
2014-09-10 15:06 ` Aravind Gopalakrishnan
2014-09-11 6:10 ` Wen Congyang
2014-09-11 10:35 ` Tim Deegan
2014-09-12 3:14 ` Wen Congyang
2014-09-12 15:43 ` Tim Deegan
2014-09-17 7:56 ` Wen Congyang
2014-09-17 14:29 ` Aravind Gopalakrishnan
2014-09-18 0:05 ` Aravind Gopalakrishnan
2014-09-18 0:05 ` Aravind Gopalakrishnan
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=5412A1DD.9010707@cn.fujitsu.com \
--to=wency@cn.fujitsu.com \
--cc=Ian.Campbell@citrix.com \
--cc=Ian.Jackson@eu.citrix.com \
--cc=david.vrabel@citrix.com \
--cc=eddie.dong@intel.com \
--cc=laijs@cn.fujitsu.com \
--cc=paul.durrant@citrix.com \
--cc=xen-devel@lists.xen.org \
--cc=yanghy@cn.fujitsu.com \
--cc=yunhong.jiang@intel.com \
/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).