From: Keir Fraser <keir.xen@gmail.com>
To: Jan Beulich <JBeulich@suse.com>,
Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: jun.nakajima@intel.com, George.Dunlap@eu.citrix.com,
jacob.shin@amd.com, eddie.dong@intel.com,
dietmar.hahn@ts.fujitsu.com, suravee.suthikulpanit@amd.com,
xen-devel <xen-devel@lists.xenproject.org>
Subject: Re: [PATCH v2 02/13] Set VCPU's is_running flag closer to when the VCPU is dispatched
Date: Wed, 25 Sep 2013 15:08:56 +0100 [thread overview]
Message-ID: <CE68AD08.5F07F%keir.xen@gmail.com> (raw)
In-Reply-To: <5243046802000078000F642A@nat28.tlf.novell.com>
On 25/09/2013 14:42, "Jan Beulich" <JBeulich@suse.com> wrote:
>>>> On 20.09.13 at 11:42, Boris Ostrovsky <boris.ostrovsky@oracle.com> wrote:
>> An interrupt handler happening during new VCPU scheduling may want to know
>> who was on the (physical) processor at the point of the interrupt. Just
>> looking at 'current' may not be accurate since there is a window of time when
>> 'current' points to new VCPU and its is_running flag is set but the VCPU has
>> not been dispatched yet. More importantly, on Intel processors, if the
>> handler
>> wants to examine certain state of an HVM VCPU (such as segment registers) the
>> VMCS pointer is not set yet.
>>
>> This patch will move setting the is_running flag to a later point.
>
> As said on v1 already - I'm all but convinced that this is not going to
> break something in subtle ways. Without you assuring us that you
> carefully inspected _all_ uses of this flag, I don't think this can go in.
It's very definitely not safe. It would break vcpu_sleep_sync() for example
-- which depends on ->is_running being set inside the scheduler lock.
Otherwise, if vcpu_sleep_nosync() occurs immediately after dropping the lock
in schedule() then vcpu_sleep_sync() can see !v->is_running even though v is
in the process of being activated to run.
It's a subtle flag, messing with it is unlikely ever to be the right answer.
-- Keir
> Jan
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel
next prev parent reply other threads:[~2013-09-25 14:09 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-20 9:41 [PATCH v2 00/13] x86/PMU: Xen PMU PV support Boris Ostrovsky
2013-09-20 9:42 ` [PATCH v2 01/13] Export hypervisor symbols Boris Ostrovsky
2013-09-23 19:42 ` Konrad Rzeszutek Wilk
2013-09-23 20:06 ` Boris Ostrovsky
2013-09-24 17:40 ` Konrad Rzeszutek Wilk
2013-09-25 13:15 ` Jan Beulich
2013-09-25 14:03 ` Boris Ostrovsky
2013-09-25 14:53 ` Jan Beulich
2013-09-20 9:42 ` [PATCH v2 02/13] Set VCPU's is_running flag closer to when the VCPU is dispatched Boris Ostrovsky
2013-09-25 13:42 ` Jan Beulich
2013-09-25 14:08 ` Keir Fraser [this message]
2013-09-20 9:42 ` [PATCH v2 03/13] x86/PMU: Stop AMD counters when called from vpmu_save_force() Boris Ostrovsky
2013-09-20 9:42 ` [PATCH v2 04/13] x86/VPMU: Minor VPMU cleanup Boris Ostrovsky
2013-09-20 9:42 ` [PATCH v2 05/13] intel/VPMU: Clean up Intel VPMU code Boris Ostrovsky
2013-09-23 11:42 ` Dietmar Hahn
2013-09-23 19:46 ` Konrad Rzeszutek Wilk
2013-09-25 13:55 ` Jan Beulich
2013-09-25 14:39 ` Boris Ostrovsky
2013-09-25 14:57 ` Jan Beulich
2013-09-25 15:37 ` Boris Ostrovsky
2013-09-20 9:42 ` [PATCH v2 06/13] x86/PMU: Add public xenpmu.h Boris Ostrovsky
2013-09-23 13:04 ` Dietmar Hahn
2013-09-23 13:16 ` Jan Beulich
2013-09-23 14:00 ` Boris Ostrovsky
2013-09-23 13:45 ` Boris Ostrovsky
2013-09-25 14:04 ` Jan Beulich
2013-09-25 15:59 ` Boris Ostrovsky
2013-09-25 16:08 ` Jan Beulich
2013-09-30 13:25 ` Boris Ostrovsky
2013-09-30 13:30 ` Jan Beulich
2013-09-30 13:55 ` Boris Ostrovsky
2013-09-20 9:42 ` [PATCH v2 07/13] x86/PMU: Make vpmu not HVM-specific Boris Ostrovsky
2013-09-25 14:05 ` Jan Beulich
2013-09-25 14:49 ` Boris Ostrovsky
2013-09-25 14:57 ` Jan Beulich
2013-09-20 9:42 ` [PATCH v2 08/13] x86/PMU: Interface for setting PMU mode and flags Boris Ostrovsky
2013-09-25 14:11 ` Jan Beulich
2013-09-25 14:55 ` Boris Ostrovsky
2013-09-20 9:42 ` [PATCH v2 09/13] x86/PMU: Initialize PMU for PV guests Boris Ostrovsky
2013-09-20 9:42 ` [PATCH v2 10/13] x86/PMU: Add support for PMU registes handling on " Boris Ostrovsky
2013-09-23 13:50 ` Dietmar Hahn
2013-09-25 14:23 ` Jan Beulich
2013-09-25 15:03 ` Boris Ostrovsky
2013-09-20 9:42 ` [PATCH v2 11/13] x86/PMU: Handle PMU interrupts for " Boris Ostrovsky
2013-09-25 14:33 ` Jan Beulich
2013-09-25 14:40 ` Andrew Cooper
2013-09-25 15:52 ` Boris Ostrovsky
2013-09-25 15:19 ` Boris Ostrovsky
2013-09-25 15:25 ` Jan Beulich
2013-09-20 9:42 ` [PATCH v2 12/13] x86/PMU: Save VPMU state for PV guests during context switch Boris Ostrovsky
2013-09-20 9:42 ` [PATCH v2 13/13] x86/PMU: Move vpmu files up from hvm directory Boris Ostrovsky
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=CE68AD08.5F07F%keir.xen@gmail.com \
--to=keir.xen@gmail.com \
--cc=George.Dunlap@eu.citrix.com \
--cc=JBeulich@suse.com \
--cc=boris.ostrovsky@oracle.com \
--cc=dietmar.hahn@ts.fujitsu.com \
--cc=eddie.dong@intel.com \
--cc=jacob.shin@amd.com \
--cc=jun.nakajima@intel.com \
--cc=suravee.suthikulpanit@amd.com \
--cc=xen-devel@lists.xenproject.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).