DPDK patches and discussions
 help / color / mirror / Atom feed
From: Manoj Viswanath <manoj.viswanath@gmail.com>
To: Bruce Richardson <bruce.richardson@intel.com>
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] Ports not detected by IGB_UIO in DPDK 1.7.1 in QEMU_KVM environment
Date: Fri, 7 Nov 2014 20:31:34 +0530	[thread overview]
Message-ID: <CAC1b25pY-wbEeVicBEVhYChOqjWXNRHO51wQwc8q4uo28DGO2g@mail.gmail.com> (raw)
In-Reply-To: <20141105101246.GA9856@bricha3-MOBL3>

Hi Bruce,

I was not doing anything specific for binding the NICs to IGB_UIO (like
invoking "dpdk_nic_bind.py" script explicitly) when using my application
with DPDK 1.6.0. The e1000 devices assigned via virt-manager to the VM were
automatically getting picked up and initialized by IGB_UIO within each VM.

The same is not working with DPDK 1.7.1 now.

I tried exporting the "dpdk_nic_bind.py" script into my VM (running DPDK
1.7.1) and tried to check the status. The emulated devices were shown as
neither bound to kernel nor to IGB_UIO as evident from below output:-

<--------------------------------------------------------------------------------------------------->
Network devices using DPDK-compatible driver
============================================
<none>

Network devices using kernel driver
===================================
0000:00:03.0 'Virtio network device' if= drv=virtio-pci unused=igb_uio

Other network devices
=====================
0000:00:04.0 '82540EM Gigabit Ethernet Controller' unused=igb_uio
0000:00:05.0 '82540EM Gigabit Ethernet Controller' unused=igb_uio
<--------------------------------------------------------------------------------------------------->

When i tried to forcefully bind the NICs using the "--bind=igb_uio" option
of the script, i am getting the following sequence of error log
continuously:-
<--------------------------------------------------------------------------------------------------->
........
........
EAL: Error reading from file descriptor 31: Input/output error
EAL: Error reading from file descriptor 32: Input/output error
........
........
<--------------------------------------------------------------------------------------------------->

Kindly advise.

Thanks in advance !

Regards,
Manoj

On Wed, Nov 5, 2014 at 3:42 PM, Bruce Richardson <bruce.richardson@intel.com
> wrote:

> On Wed, Nov 05, 2014 at 03:28:13PM +0530, Manoj Viswanath wrote:
> > Hi,
> >
> > I have a DPDK application running on QEMU-KVM environment using DPDK
> 1.6.0.
> > I am trying to port the same to DPDK version 1.7.1.
> >
> > I am using Virt-manager GUI to assign e1000 emulated port to the VM. This
> > works fine in DPDK 1.6.0. The device is identified by IGB_UIO and
> > initialized by my application as expected.
> >
> > However in case of DPDK 1.7.1, the emulated e1000 devices do not seem to
> be
> > recognized.
> > Following is my analysis:
> >
> > 1. The API pci_get_uio_dev() is returning ERROR. This is called from
> > pci_uio_map_resource() in the flow of PCI PROBE [rte_eal_pci_probe()].
> >
> > 2. Due to this, the PCI device is not getting mapped to the correct
> driver
> > (EM
> > ​ Driver​
> > ).
> >
> > 3. The reason for the error in [1] appears to be that "uio" sub-directory
> > doesn't seem to be correctly created for interfaces assigned to this VM.
> >
> > 4. Upon further analysis i found that IGB_UIO probe function
> > ["igbuio_pci_probe()"] is not getting triggered indicating the port has
> *not
> > been assigned* to the IGB_UIO.
> >
> > Kindly refer to the attachments:-
> > - "Output of sys-bus-pci-devices" - indicating "uio" subdirectory not
> > created for PCI devices in case of DPDK 1.7.1
> > - "Output of lspci -v" - indicating device not bound to driver in case of
> > DPDK 1.7.1
> > - IGB_UIO init log snippet - indicating PCI devices not detected and
> > initialized by IGB_UIO in case of DPDK 1.7.1
> > - CONFIG file used for DPDK compilation
> >
> > Not sure what has changed between 1.6.0 and 1.7.1 which is impacting
> this.
> >
> > Could someone throw light in this regard as to what i may be missing ?
> >
> > Thanks in advance.
> >
> > Regards,
> > Manoj
>
> Hi Manoj,
>
> can you perhaps give some details on how you were binding the device to
> the uio
> module both for 1.6 and for 1.7?
>
> /Bruce
>

  reply	other threads:[~2014-11-07 14:52 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CAC1b25pX3R2y_Cjp5UAdL6Bozrf01OkG6U6LGQvJ2aUTDXGvSA@mail.gmail.com>
2014-11-05 10:12 ` Bruce Richardson
2014-11-07 15:01   ` Manoj Viswanath [this message]
     [not found]     ` <20141107153053.GA10376@bricha3-MOBL3>
2014-11-07 17:56       ` Manoj Viswanath
2014-11-10 10:58         ` Bruce Richardson
2014-11-11 13:48           ` Manoj Viswanath
2014-11-12 14:12             ` Bruce Richardson

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=CAC1b25pY-wbEeVicBEVhYChOqjWXNRHO51wQwc8q4uo28DGO2g@mail.gmail.com \
    --to=manoj.viswanath@gmail.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.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).