From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Thomas Monjalon <thomas.monjalon@6wind.com>,
Jianfeng Tan <jianfeng.tan@intel.com>
Cc: dev@dpdk.org, stephen@networkplumber.org, jing.d.chen@intel.com
Subject: Re: [dpdk-dev] [PATCH] igb_uio: stop device when closing /dev/uioX
Date: Fri, 31 Mar 2017 15:16:16 +0100 [thread overview]
Message-ID: <6e872ad0-a212-2324-41ad-66ef6e6a30d3@intel.com> (raw)
In-Reply-To: <1729096.OcIPJKQlaJ@xps13>
On 3/30/2017 9:22 PM, Thomas Monjalon wrote:
> 2016-12-02 16:45, Jianfeng Tan:
>> Depends-on: http://dpdk.org/dev/patchwork/patch/17493/
>
> The above patch is marked as rejected.
> Do we want to reject also this patch?
Above patch rejected because it may have backward compatibility issues
with the older DPDK (1.8 and older if I remember correct) with newer
kernels.
Plan was checking if this patch can be updated to remove dependency, and
if not accepting above patch.
But as far as I can see not worked on it. We may drop this if Jianfeng
don't have any plan to work on it.
>
>> When a DPDK application grab a PCI device, device and driver work
>> together to Rx/Tx packets. If the DPDK app crashes or gets killed,
>> there's no chance for DPDK driver to stop the device, which means
>> rte_eth_dev_stop() will not be called.
>>
>> This could result in problems. In virtio's case, the device (the
>> vhost backend), will keep working. If packets come, the vhost will
>> copy them into the vring, which is negotiated with the previous DPDK
>> app. But the resources, especially hugepages, are recycled by VM
>> kernel. What's worse, the memory could be allocated for other usage,
>> and re-written. This leads to an uncertain situation. Like this post
>> has reported, https://bugs.launchpad.net/qemu/+bug/1558175, QEMU's
>> vhost finds out wrong value, and exits the whole QEMU process.
>>
>> To make it right, we need to restart (or reset) the device and make
>> the device go into the initial state, when uio-generic or igb_uio
>> recycles the PCI device. There's a chance in uio framework to disable
>> devices when /dev/uioX gets closed. Here we enable the pci device
>> in open() hook and disable it in release() hook.
>>
>> However, if device is not enabled in probe() phase any more, we can
>> not register irq in probe() through uio_register_device(). To address
>> that, we invoke request_irq() to register callback directly.
>>
>> Similar change needs to be done in uio-generic driver. And vfio-pci
>> seems to have done that already.
>>
>> Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
>
next prev parent reply other threads:[~2017-03-31 14:16 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-09-01 2:16 [dpdk-dev] [RFC] igb_uio: deprecate iomem and ioport mapping Jianfeng Tan
2016-09-02 12:31 ` Ferruh Yigit
2016-09-02 12:59 ` Tan, Jianfeng
2016-09-09 9:06 ` David Marchand
2016-09-09 9:31 ` Tan, Jianfeng
2016-09-22 5:44 ` [dpdk-dev] [PATCH] doc: remove iomem and ioport handling in igb_uio Jianfeng Tan
2016-09-30 10:13 ` Ferruh Yigit
2016-11-11 2:12 ` Remy Horton
2016-11-13 8:55 ` Thomas Monjalon
2016-12-02 16:28 ` [dpdk-dev] [PATCH] igb_uio: deprecate iomem and ioport mapping Jianfeng Tan
2016-12-02 16:45 ` [dpdk-dev] [PATCH] igb_uio: stop device when closing /dev/uioX Jianfeng Tan
2017-03-30 20:22 ` Thomas Monjalon
2017-03-31 14:16 ` Ferruh Yigit [this message]
2016-12-02 23:47 ` [dpdk-dev] [RFC] igb_uio: deprecate iomem and ioport mapping Stephen Hemminger
2016-12-05 7:04 ` Tan, Jianfeng
2017-01-05 15:23 ` Ferruh Yigit
2017-01-06 1:52 ` Tan, Jianfeng
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=6e872ad0-a212-2324-41ad-66ef6e6a30d3@intel.com \
--to=ferruh.yigit@intel.com \
--cc=dev@dpdk.org \
--cc=jianfeng.tan@intel.com \
--cc=jing.d.chen@intel.com \
--cc=stephen@networkplumber.org \
--cc=thomas.monjalon@6wind.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).