From: Yuanhan Liu <yuanhan.liu@linux.intel.com>
To: Thomas Monjalon <thomas.monjalon@6wind.com>
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] [Question] How pmd virtio works without UIO?
Date: Thu, 24 Dec 2015 11:30:27 +0800 [thread overview]
Message-ID: <20151224033027.GY18863@yliu-dev.sh.intel.com> (raw)
In-Reply-To: <2150122.EUSmcjyJtA@xps13>
On Wed, Dec 23, 2015 at 11:26:17PM +0100, Thomas Monjalon wrote:
> 2015-12-23 10:09, Yuanhan Liu:
> > On Wed, Dec 23, 2015 at 09:55:54AM +0800, Peter Xu wrote:
> > > On Tue, Dec 22, 2015 at 04:38:30PM +0000, Xie, Huawei wrote:
> > > > On 12/22/2015 7:39 PM, Peter Xu wrote:
> > > > > I tried to unbind one of the virtio net device, I see the PCI entry
> > > > > still there.
> > > > >
> > > > > Before unbind:
> > > > >
> > > > > [root@vm proc]# lspci -k -s 00:03.0
> > > > > 00:03.0 Ethernet controller: Red Hat, Inc Virtio network device
> > > > > Subsystem: Red Hat, Inc Device 0001
> > > > > Kernel driver in use: virtio-pci
> > > > > [root@vm proc]# cat /proc/ioports | grep c060-c07f
> > > > > c060-c07f : 0000:00:03.0
> > > > > c060-c07f : virtio-pci
> > > > >
> > > > > After unbind:
> > > > >
> > > > > [root@vm proc]# lspci -k -s 00:03.0
> > > > > 00:03.0 Ethernet controller: Red Hat, Inc Virtio network device
> > > > > Subsystem: Red Hat, Inc Device 0001
> > > > > [root@vm proc]# cat /proc/ioports | grep c060-c07f
> > > > > c060-c07f : 0000:00:03.0
> > > > >
> > > > > So... does this means that it is an alternative to black list
> > > > > solution?
> > > > Oh, we could firstly check if this port is manipulated by kernel driver
> > > > in virtio_resource_init/eth_virtio_dev_init, as long as it is not too late.
> >
> > Why can't we simply quit at pci_scan_one, once finding that it's not
> > bond to uio (or similar stuff)? That would be generic enough, that we
> > don't have to do similar checks for each new pmd driver.
> >
> > Or, am I missing something?
>
> UIO is not needed to make virtio works (without interrupt support).
> Sometimes it may be required to avoid using kernel modules.
While dig the git history, I found the commit:
commit da978dfdc43b59e290a46d7ece5fd19ce79a1162
Author: Ouyang Changchun <changchun.ouyang@intel.com>
Date: Mon Feb 9 09:14:06 2015 +0800
virtio: use port IO to get PCI resource
Make virtio not require UIO for some security reasons, this is to match
6WIND's virtio-net-pmd.
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Huawei Xie <huawei.xie@intel.com>
The commit log is not well written, giving no explanation about the
"some security reasons".
Anyway, I see it now that it's kind of a design.
Note that my first patch set about enabling virtio 1.0 [0] sets the
RTE_PCI_DRV_NEED_MAPPING flag, which somehow implies that uio is a
must, otherwise, eal init would fail at pci_map_device().
So that my pathset breaks the un-documented rule, and I need fix it.
How about adding a wrapper, say rte_pci_map_device(), and exporting
it, so that virtio pmd driver could map resources when necessary?
[0]: http://dpdk.org/dev/patchwork/bundle/yliu/virtio-1.0-pmd/
> > > I guess there might be two problems? Which are:
> > >
> > > 1. How user avoid DPDK taking over virtio devices that they do not
> > > want for IO (chooses which device to use)
> >
> > Isn't that what's the 'binding/unbinding' for?
>
> Binding is, sometimes, required.
We may need fix the doc then. As the doc says it's a must:
3.6. Binding and Unbinding Network Ports to/from the Kernel Modules
Instead, all ports that are to be used by an DPDK application
==> must be bound to the uio_pci_generic, igb_uio or vfio-pci
module before the application is run. Any network ports under
Linux* control will be ignored by the DPDK poll-mode drivers
and cannot be used by the application.
--yliu
> But does it mean DPDK should use every available ports?
> That's the default and may be configured with blacklist/whitelist.
>
> > > 2. Driver conflict between virtio PMD in DPDK, and virtio-pci in
> > > kernel (happens on every virtio device that DPDK uses)
> >
> > If you unbinded the kernel driver first, which is the suggested (or
> > must?) way to use DPDK, that will not happen.
next prev parent reply other threads:[~2015-12-24 3:28 UTC|newest]
Thread overview: 92+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-22 3:50 Peter Xu
2015-12-22 7:00 ` Yuanhan Liu
2015-12-22 8:23 ` Peter Xu
2015-12-22 8:32 ` Yuanhan Liu
2015-12-22 9:56 ` Peter Xu
2015-12-22 10:47 ` Xie, Huawei
2015-12-22 10:53 ` Xie, Huawei
2015-12-22 11:39 ` Peter Xu
2015-12-22 14:31 ` Xie, Huawei
2015-12-22 16:38 ` Xie, Huawei
2015-12-23 1:55 ` Peter Xu
2015-12-23 2:09 ` Yuanhan Liu
2015-12-23 2:38 ` Peter Xu
2015-12-23 22:26 ` Thomas Monjalon
2015-12-24 3:30 ` Yuanhan Liu [this message]
2015-12-24 17:56 ` Stephen Hemminger
2015-12-23 2:01 ` Yuanhan Liu
2015-12-23 2:41 ` Peter Xu
2015-12-23 2:58 ` Yuanhan Liu
2015-12-23 5:13 ` Xie, Huawei
2015-12-23 22:20 ` Thomas Monjalon
2015-12-24 18:38 ` [dpdk-dev] [PATCH 0/4] check if any kernel driver is manipulating the virtio device Huawei Xie
2015-12-24 18:38 ` [dpdk-dev] [PATCH 1/4] eal: make the comment more accurate Huawei Xie
2015-12-24 18:38 ` [dpdk-dev] [PATCH 2/4] eal: set kdrv to RTE_KDRV_NONE if kernel driver isn't manipulating the device Huawei Xie
2015-12-28 20:24 ` David Marchand
2015-12-24 18:38 ` [dpdk-dev] [PATCH 3/4] virtio: return 1 to tell the upper layer we don't take over this device Huawei Xie
2015-12-28 5:25 ` Yuanhan Liu
2015-12-28 5:38 ` Xie, Huawei
2015-12-24 18:38 ` [dpdk-dev] [PATCH 4/4] virtio: check if any kernel driver is manipulating the device Huawei Xie
2015-12-28 5:26 ` Yuanhan Liu
2015-12-28 5:29 ` Xie, Huawei
2016-01-04 9:02 ` Xie, Huawei
2016-01-04 17:29 ` Stephen Hemminger
2016-01-05 14:44 ` Panu Matilainen
2015-12-28 3:08 ` [dpdk-dev] [PATCH 0/4] check if any kernel driver is manipulating the virtio device Peter Xu
2016-01-03 17:56 ` [dpdk-dev] [PATCH v2 0/4] fix the issue that DPDK takes over virtio device blindly Huawei Xie
2016-01-03 17:56 ` [dpdk-dev] [PATCH v2 1/4] eal: make the comment more accurate Huawei Xie
2016-01-03 17:56 ` [dpdk-dev] [PATCH v2 2/4] eal: set kdrv to RTE_KDRV_NONE if kernel driver isn't manipulating the device Huawei Xie
2016-01-03 17:56 ` [dpdk-dev] [PATCH v2 3/4] virtio: return 1 to tell the upper layer we don't take over this device Huawei Xie
2016-01-03 17:56 ` [dpdk-dev] [PATCH v2 4/4] virtio: check if any kernel driver is manipulating the virtio device Huawei Xie
2016-01-04 17:24 ` Stephen Hemminger
2016-01-04 17:56 ` Xie, Huawei
2016-01-05 1:56 ` Yuanhan Liu
2016-01-07 13:17 ` Ferruh Yigit
2016-01-07 14:17 ` Panu Matilainen
2016-01-27 12:43 ` Thomas Monjalon
2016-01-04 17:25 ` [dpdk-dev] [PATCH v2 0/4] fix the issue that DPDK takes over virtio device blindly Stephen Hemminger
2016-01-12 3:02 ` Xie, Huawei
2016-01-12 4:23 ` Santosh Shukla
2016-01-12 5:16 ` Xie, Huawei
2016-01-13 12:17 ` Santosh Shukla
2016-01-27 15:21 ` [dpdk-dev] [PATCH v3 " Huawei Xie
2016-01-27 15:21 ` [dpdk-dev] [PATCH v3 1/4] eal: make the comment more accurate Huawei Xie
2016-01-27 15:21 ` [dpdk-dev] [PATCH v3 2/4] eal: set kdrv to RTE_KDRV_NONE if kernel driver isn't manipulating the device Huawei Xie
2016-01-27 15:21 ` [dpdk-dev] [PATCH v3 3/4] virtio: return 1 to tell the upper layer we don't take over this device Huawei Xie
2016-01-27 15:21 ` [dpdk-dev] [PATCH v3 4/4] virtio: check if kernel driver is manipulating the virtio device Huawei Xie
2016-01-28 9:55 ` Panu Matilainen
2016-01-29 7:40 ` [dpdk-dev] [PATCH v3 0/4] fix the issue that DPDK takes over virtio device blindly Yuanhan Liu
2016-02-24 12:43 ` Thomas Monjalon
2016-02-26 6:09 ` Xie, Huawei
2016-02-26 8:40 ` David Marchand
2016-02-26 9:00 ` Xie, Huawei
2016-02-26 1:53 ` [dpdk-dev] [PATCH v4 " Huawei Xie
2016-02-26 1:53 ` [dpdk-dev] [PATCH v4 1/4] eal: make the comment more accurate Huawei Xie
2016-02-29 8:48 ` David Marchand
2016-02-26 1:53 ` [dpdk-dev] [PATCH v4 2/4] eal: set kdrv to RTE_KDRV_NONE if kernel driver isn't managing the device Huawei Xie
2016-02-29 2:34 ` Xie, Huawei
2016-02-29 8:46 ` David Marchand
2016-02-29 9:00 ` Xie, Huawei
2016-02-29 9:05 ` David Marchand
2016-02-26 1:53 ` [dpdk-dev] [PATCH v4 3/4] eal: call pci_ioport_map when " Huawei Xie
2016-02-29 9:02 ` David Marchand
2016-02-26 1:53 ` [dpdk-dev] [PATCH v4 4/4] virtio: return 1 to tell the upper layer we don't take over this device Huawei Xie
2016-02-29 13:15 ` Santosh Shukla
2016-03-01 7:16 ` Thomas Monjalon
2016-03-01 7:53 ` Xie, Huawei
2016-03-01 8:20 ` Thomas Monjalon
2016-03-01 8:39 ` Xie, Huawei
2016-03-01 9:55 ` Thomas Monjalon
2016-03-01 10:08 ` Xie, Huawei
2016-03-08 17:00 ` Xie, Huawei
2016-03-08 23:01 ` Thomas Monjalon
2016-03-08 15:33 ` [dpdk-dev] [PATCH v5 0/6] fix the issue that DPDK takes over virtio device blindly Huawei Xie
2016-03-08 15:33 ` [dpdk-dev] [PATCH v5 1/6] eal: make the comment more accurate Huawei Xie
2016-03-08 15:33 ` [dpdk-dev] [PATCH v5 2/6] eal: RTE_KDRV_NONE means kernel driver isn't managing the device Huawei Xie
2016-03-08 15:33 ` [dpdk-dev] [PATCH v5 3/6] eal: use new RTE_ARCH_X86 for x86 arch Huawei Xie
2016-03-09 23:04 ` Thomas Monjalon
2016-03-08 15:33 ` [dpdk-dev] [PATCH v5 4/6] eal: simple code rework Huawei Xie
2016-03-08 15:33 ` [dpdk-dev] [PATCH v5 5/6] eal: map IO port when kernel driver isn't managing the device Huawei Xie
2016-03-08 15:33 ` [dpdk-dev] [PATCH v5 6/6] virtio: return 1 to tell the upper layer we don't take over this device Huawei Xie
2016-03-09 23:23 ` Thomas Monjalon
2016-03-09 23:35 ` [dpdk-dev] [PATCH v5 0/6] fix the issue that DPDK takes over virtio device blindly Thomas Monjalon
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=20151224033027.GY18863@yliu-dev.sh.intel.com \
--to=yuanhan.liu@linux.intel.com \
--cc=dev@dpdk.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).