From: luoben <luoben@linux.alibaba.com>
To: Alex Williamson <alex.williamson@redhat.com>
Cc: tglx@linutronix.de, linux-kernel@vger.kernel.org,
tao.ma@linux.alibaba.com, gerry@linux.alibaba.com,
nanhai.zou@linux.alibaba.com, linyunsheng@huawei.com
Subject: Re: [PATCH v3 0/3] genirq/vfio: Introduce update_irq_devid and optimize VFIO irq ops
Date: Tue, 20 Aug 2019 23:51:23 +0800 [thread overview]
Message-ID: <4b2c76e3-91da-09b6-65f4-535828373c04@linux.alibaba.com> (raw)
In-Reply-To: <20190820092209.0c89effd@x1.home>
在 2019/8/20 下午11:22, Alex Williamson 写道:
> On Tue, 20 Aug 2019 12:03:50 +0800
> luoben <luoben@linux.alibaba.com> wrote:
>
>> 在 2019/8/20 上午4:51, Alex Williamson 写道:
>>> On Thu, 15 Aug 2019 21:02:58 +0800
>>> Ben Luo <luoben@linux.alibaba.com> wrote:
>>>
>>>> Currently, VFIO takes a lot of free-then-request-irq actions whenever
>>>> a VM (with device passthru via VFIO) sets irq affinity or mask/unmask
>>>> irq. Those actions only change the cookie data of irqaction or even
>>>> change nothing. The free-then-request-irq not only adds more latency,
>>>> but also increases the risk of losing interrupt, which may lead to a
>>>> VM hung forever in waiting for IO completion
>>> What guest environment is generating this? Typically I don't see that
>>> Windows or Linux guests bounce the interrupt configuration much.
>>> Thanks,
>>>
>>> Alex
>> By tracing centos5u8 on host, I found it keep masking and unmasking
>> interrupt like this:
>>
>> [1566032533709879] index:28 irte_hi:000000010004a601
>> irte_lo:adb54bc000b98001
>> [1566032533711242] index:28 irte_hi:0000000000000000
>> irte_lo:0000000000000000
>> [1566032533711258] index:28 irte_hi:000000000004a601
>> irte_lo:00003fff00ac002d
>> [1566032533711269] index:28 irte_hi:000000000004a601
>> irte_lo:00003fff00ac002d
> [snip]
>> "[1566032533720007]" is timestamp in μs, so centos5u8 tiggers 30+ irte
>> modification within 10ms
> Ok, that matches my understanding that only very old guests behave in
> this manner. It's a curious case to optimize as RHEL5 is in extended
> life-cycle support, with regular maintenance releases ending 2+ years
> ago. Thanks,
>
> Alex
But repeatedly set interrupt affinity in a new version guest can also
cause the problem.
Thanks,
Ben
prev parent reply other threads:[~2019-08-20 15:51 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-15 13:02 [PATCH v3 0/3] genirq/vfio: Introduce update_irq_devid and optimize VFIO irq ops Ben Luo
2019-08-15 13:02 ` [PATCH v3 1/3] genirq: enhance error recovery code in free irq Ben Luo
2019-08-15 14:20 ` Thomas Gleixner
2019-08-15 13:03 ` [PATCH v3 2/3] genirq: introduce update_irq_devid() Ben Luo
2019-08-15 14:58 ` Thomas Gleixner
2019-08-19 5:31 ` luoben
2019-08-19 8:18 ` Thomas Gleixner
2019-08-15 13:03 ` [PATCH v3 3/3] vfio_pci: make use of update_irq_devid and optimize irq ops Ben Luo
2019-08-15 16:45 ` Thomas Gleixner
[not found] ` <7a3606ad-8fa6-45d5-b5a4-ee3f07893a25@linux.alibaba.com>
2019-08-20 15:27 ` Liu, Jiang
2019-08-22 13:41 ` luoben
2019-08-19 20:51 ` [PATCH v3 0/3] genirq/vfio: Introduce update_irq_devid and optimize VFIO " Alex Williamson
2019-08-20 4:03 ` luoben
2019-08-20 15:22 ` Alex Williamson
2019-08-20 15:51 ` luoben [this message]
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=4b2c76e3-91da-09b6-65f4-535828373c04@linux.alibaba.com \
--to=luoben@linux.alibaba.com \
--cc=alex.williamson@redhat.com \
--cc=gerry@linux.alibaba.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linyunsheng@huawei.com \
--cc=nanhai.zou@linux.alibaba.com \
--cc=tao.ma@linux.alibaba.com \
--cc=tglx@linutronix.de \
/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).