From: "Tian, Kevin" <kevin.tian@intel.com>
To: Jan Beulich <JBeulich@suse.com>, "Gao, Chao" <chao.gao@intel.com>
Cc: Andrew Cooper <andrew.cooper3@citrix.com>,
Xuquan <xuquan8@huawei.com>,
"osstest-admin@xenproject.org" <osstest-admin@xenproject.org>,
xen-devel <xen-devel@lists.xenproject.org>
Subject: Re: [xen-unstable test] 106504: regressions - FAIL
Date: Fri, 24 Mar 2017 07:48:08 +0000 [thread overview]
Message-ID: <AADFC41AFE54684AB9EE6CBC0274A5D190C7CEAE@SHSMSX101.ccr.corp.intel.com> (raw)
In-Reply-To: <58D280750200007800146399@prv-mh.provo.novell.com>
> From: Jan Beulich [mailto:JBeulich@suse.com]
> Sent: Wednesday, March 22, 2017 8:48 PM
>
> > 3. We read RTE 3 times. 1st happens when we set vIRR. 2nd happens when
> > pt_update_irq() returns. 3rd happens in pt_intr_post(). If guest
> > changes the vector in RTE during the window, it will also incur losing
> > or getting more periodic timer interrupt.
>
> Which raises the question whether latching the value read the first time
> would address the issue you demonstrate with the test case.
> Or alternatively deferring writes to take effect only once readers are done
> with their perhaps multiple accesses?
>
> Can you get in touch with your chipset folks to find out whether hardware
> has cases where multiple reads occur during the processing of a single event?
>
There is a similar case. For level-triggered interrupt, there is a "remote IRR"
bit in RTE which is set to 1 when LAPIC accepts the level interrupt sent by
IOAPIC. It's then cleared by EOI broadcast from LAPIC later, based on
matching interrupt vectors. If software happens to change the vector of
the said RTE in-between, "remote IRR" bit will never be cleared (it
expects an EOI with new vector now while actual EOI for previous injection
contains old vector).
Of course in our case pt timer is edge-interrupt, which shouldn't trigger
such multi-reads issue in real hardware. But anyway it's not a good behavior
to change RTE vector w/o stopping the interrupt source first...
Thanks
Kevin
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
next prev parent reply other threads:[~2017-03-24 7:48 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-07 5:52 [xen-unstable test] 106504: regressions - FAIL osstest service owner
2017-03-07 9:16 ` Jan Beulich
2017-03-07 4:24 ` Chao Gao
2017-03-07 14:11 ` Jan Beulich
2017-03-22 4:53 ` Chao Gao
2017-03-22 12:47 ` Jan Beulich
2017-03-22 6:13 ` Chao Gao
2017-03-22 13:40 ` Jan Beulich
2017-03-29 3:28 ` Xuquan (Quan Xu)
2017-03-28 20:48 ` Chao Gao
2017-03-24 7:48 ` Tian, Kevin [this message]
2017-03-24 8:17 ` Jan Beulich
2017-03-24 8:25 ` Tian, Kevin
[not found] ` <AADFC41AFE54684AB9EE6CBC0274A5D190C7CFB9@SHSMSX101.ccr.corp.intel.com>
2017-03-24 8:49 ` Tian, Kevin
2017-03-24 9:00 ` Andrew Cooper
2017-04-04 23:57 ` Chao Gao
2017-04-05 7:48 ` Jan Beulich
2017-04-05 1:49 ` Chao Gao
2017-04-07 8:56 ` Xuquan (Quan Xu)
2017-03-08 3:16 ` Xuquan (Quan Xu)
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=AADFC41AFE54684AB9EE6CBC0274A5D190C7CEAE@SHSMSX101.ccr.corp.intel.com \
--to=kevin.tian@intel.com \
--cc=JBeulich@suse.com \
--cc=andrew.cooper3@citrix.com \
--cc=chao.gao@intel.com \
--cc=osstest-admin@xenproject.org \
--cc=xen-devel@lists.xenproject.org \
--cc=xuquan8@huawei.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.