RE: [RFC 1/2] net: Add support to configure SR-IOV VF minimum and maximum Tx rate through ip tool.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



> -----Original Message-----
> From: netdev-owner@xxxxxxxxxxxxxxx [mailto:netdev-
> owner@xxxxxxxxxxxxxxx] On Behalf Of Ben Hutchings
> Sent: Sunday, March 09, 2014 10:53 PM
> To: Sucheta Chakraborty
> Cc: David Miller; netdev; Dept-HSG Linux NIC Dev
> Subject: Re: [RFC 1/2] net: Add support to configure SR-IOV VF minimum
> and maximum Tx rate through ip tool.
> 
> On Wed, 2014-03-05 at 02:33 -0500, Sucheta Chakraborty wrote:
> > o A new handler set_vf_rate for attr IFLA_VF_RATE has been introduced
> >   which takes 4 arguments:
> >   netdev, VF number, min_tx_rate, max_tx_rate
> >
> > o min_tx_rate puts lower limit on the VF bandwidth. VF is guaranteed
> >   to have a bandwidth of at least this value.
> >   max_tx_rate puts cap on the VF bandwidth. VF can have a bandwidth
> >   of up to this value.
> 
> Supposing I only want to set a minimum, how would I specify that there
> is no maximum?
> 

Sucheta> The current implementation allows user to send only
min_tx_rate as well. In that case, max_tx_rate is read back from driver
and passed as it is. If user has not done any setting with respect to
max_tx_rate till now, the default max returned by driver will be used.

> > o Idea is to deprecate IFLA_VF_TX_RATE and ndo_set_vf_tx_rate in
> future.
> >   And to have consistent display of rate values to user.
> [...]
> 
> You can't deprecate the userland interface, but the driver operation
> should be replaced at the same time.
> 
> - The rtnetlink core should combine userland settings from
>   IFLA_VF_TX_RATE and IFLA_VF_RATE.  If both are specified, I think
>   IFLA_VF_RATE should override.
> - If only IFLA_VF_TX_RATE is specified, the core should get the current
>   minimum before calling ndo_set_vf_rate.
> - Drivers that currently implement ndo_set_vf_tx_rate should be
>   converted to implement ndo_set_vf_rate and reject attempts to set a
>   minimum bandwidth greater than 0.
> 

Sucheta> Ok. By minimum bandwidth 0, did you mean driver specified
default for minimum BW? So, drivers that currently implement
ndo_set_vf_tx_rate should call ndo_set_vf_rate instead, with default
minimum BW set for the driver.

> Ben.
> 
> --
> Ben Hutchings
> I say we take off; nuke the site from orbit.  It's the only way to be
> sure.

Thanks for your comments.

Regards,
Sucheta.
��.n��������+%������w��{.n����z����ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f





[Index of Archives]     [Linux Kernel Discussion]     [TCP Instrumentation]     [Ethernet Bridging]     [Linux Wireless Networking]     [Linux WPAN Networking]     [Linux Host AP]     [Linux WPAN Networking]     [Linux Bluetooth Networking]     [Linux ATH6KL Networking]     [Linux Networking Users]     [Linux Coverity]     [VLAN]     [Git]     [IETF Annouce]     [Linux Assembly]     [Security]     [Bugtraq]     [Yosemite Information]     [MIPS Linux]     [ARM Linux Kernel]     [ARM Linux]     [Linux Virtualization]     [Linux IDE]     [Linux RAID]     [Linux SCSI]