From: "Wu, Feng" <feng.wu@intel.com>
To: Jan Beulich <JBeulich@suse.com>
Cc: "Tian, Kevin" <kevin.tian@intel.com>,
"keir@xen.org" <keir@xen.org>,
"george.dunlap@eu.citrix.com" <george.dunlap@eu.citrix.com>,
"andrew.cooper3@citrix.com" <andrew.cooper3@citrix.com>,
"dario.faggioli@citrix.com" <dario.faggioli@citrix.com>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>,
"Wu, Feng" <feng.wu@intel.com>
Subject: Re: [PATCH v2 3/4] VMX: Assign the right value to 'NDST' field in a concern case
Date: Tue, 31 May 2016 10:27:34 +0000 [thread overview]
Message-ID: <E959C4978C3B6342920538CF579893F019700A43@SHSMSX103.ccr.corp.intel.com> (raw)
In-Reply-To: <57486F0402000078000EF404@prv-mh.provo.novell.com>
> -----Original Message-----
> From: Jan Beulich [mailto:JBeulich@suse.com]
> Sent: Friday, May 27, 2016 10:00 PM
> To: Wu, Feng <feng.wu@intel.com>
> Cc: andrew.cooper3@citrix.com; dario.faggioli@citrix.com;
> george.dunlap@eu.citrix.com; Tian, Kevin <kevin.tian@intel.com>; xen-
> devel@lists.xen.org; konrad.wilk@oracle.com; keir@xen.org
> Subject: Re: [PATCH v2 3/4] VMX: Assign the right value to 'NDST' field in a
> concern case
>
> >>> On 26.05.16 at 15:39, <feng.wu@intel.com> wrote:
> > Normally, in vmx_cpu_block() 'NDST' filed should have the same
> > value with 'dest' or 'MASK_INSR(dest, PI_xAPIC_NDST_MASK)' depending
> > on whether x2apic is enabled. However, in the following scenario,
> > 'NDST' has different value:
> >
> > 'vcpu_block' hook gets assigned in vmx_pi_hooks_assign(), but all
> > other three PI hooks have not been assigned or not been excuted yet.
> > And during this interval, we are running in vmx_vcpu_block(), then
> > 'NDST' may have different value.
>
> How about simply assigning vcpu_block last? Or alternatively
> holding pi_blocking_list_lock around all four assignments? Or
> (maybe in addition to one of these) writing something sensible in
> vmx_pi_hooks_assign() before doing the hook assignments?
The problem is vmx_vcpu_block() can still get called first, since
the other ones might not get called yet.
> Of course much depends on whether the ASSERT() you replace was
> meant just as a sanity check, or instead logic elsewhere relies on
> NDST having the right value already before getting into
> vmx_vcpu_block().
The point is we need to make sure NDST have the right value
after leaving vmx_vcpu_block().
>
> Also, rather than saying twice that NDST has a different value in
> that case, how about stating what exact value it has then and
> why that's not what we want/need? (Same goes for the code
> comment then.)
Sure, will add this in the next version.
>
> > This patch fix this concern case.
>
> Here as well as in earlier patches - do you perhaps mean "corner
> case"?
Sorry for the typo.
Thanks,
Feng
>
> Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
next prev parent reply other threads:[~2016-05-31 10:27 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-26 13:39 [PATCH v2 0/4] VMX: Properly handle pi descriptor and per-cpu blocking list Feng Wu
2016-05-26 13:39 ` [PATCH v2 1/4] VMX: Properly handle pi when all the assigned devices are removed Feng Wu
2016-05-27 13:43 ` Jan Beulich
2016-05-31 10:22 ` Wu, Feng
2016-05-31 11:52 ` Jan Beulich
2016-06-03 5:12 ` Wu, Feng
2016-06-03 9:45 ` Jan Beulich
2016-05-26 13:39 ` [PATCH v2 2/4] VMX: Cleanup PI per-cpu blocking list when vcpu is destroyed Feng Wu
2016-05-27 13:49 ` Jan Beulich
2016-05-31 10:22 ` Wu, Feng
2016-05-31 11:54 ` Jan Beulich
2016-05-26 13:39 ` [PATCH v2 3/4] VMX: Assign the right value to 'NDST' field in a concern case Feng Wu
2016-05-27 14:00 ` Jan Beulich
2016-05-31 10:27 ` Wu, Feng [this message]
2016-05-31 11:58 ` Jan Beulich
2016-06-03 5:23 ` Wu, Feng
2016-06-03 9:57 ` Jan Beulich
2016-06-22 18:00 ` George Dunlap
2016-06-24 9:08 ` Wu, Feng
2016-05-26 13:39 ` [PATCH v2 4/4] VMX: fixup PI descritpor when cpu is offline Feng Wu
2016-05-27 14:56 ` Jan Beulich
2016-05-31 10:31 ` Wu, Feng
2016-06-22 18:33 ` George Dunlap
2016-06-24 6:34 ` Wu, Feng
2016-05-26 17:20 ` [PATCH v2 0/4] VMX: Properly handle pi descriptor and per-cpu blocking list Dario Faggioli
2016-05-31 10:19 ` Wu, Feng
2016-06-22 21:33 ` Dario Faggioli
2016-06-24 6:33 ` Wu, Feng
2016-06-24 10:29 ` Dario Faggioli
2016-06-24 13:42 ` Wu, Feng
2016-06-24 13:49 ` George Dunlap
2016-06-24 14:36 ` Dario Faggioli
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=E959C4978C3B6342920538CF579893F019700A43@SHSMSX103.ccr.corp.intel.com \
--to=feng.wu@intel.com \
--cc=JBeulich@suse.com \
--cc=andrew.cooper3@citrix.com \
--cc=dario.faggioli@citrix.com \
--cc=george.dunlap@eu.citrix.com \
--cc=keir@xen.org \
--cc=kevin.tian@intel.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).