[dpdk-stable] [v2] app/testpmd: revert setting default RSS

[Ophir Munk]

> -----Original Message-----
> From: Ilya Maximets [mailto:i.maximets at samsung.com]
> Sent: Friday, November 09, 2018 12:43 PM
> To: Ophir Munk <ophirmu at mellanox.com>; Wenzhuo Lu
> <wenzhuo.lu at intel.com>; Jingjing Wu <jingjing.wu at intel.com>; Bernard
> Iremonger <bernard.iremonger at intel.com>; dev at dpdk.org; Adrien
> Mazarguil <adrien.mazarguil at 6wind.com>
> Cc: Asaf Penso <asafp at mellanox.com>; Shahaf Shuler
> <shahafs at mellanox.com>; Thomas Monjalon <thomas at monjalon.net>;
> Olga Shern <olgas at mellanox.com>; stable at dpdk.org; Ferruh Yigit
> <ferruh.yigit at intel.com>
> Subject: Re: [v2] app/testpmd: revert setting default RSS
> 
> On 09.11.2018 12:38, Ophir Munk wrote:
> > This reverts the patch that enables default RSS action by setting
> > key=NULL and key_len=0.
> > In current testpmd implementation a key pointer must exist if
> > key_len!=0. For example, the following flow rule will cause a
> > segmentation fault:
> > flow create 0 <pattern> actions rss queues 0 1 end key_len 40 / end
> 
> Maybe it's better to check that 'key_len' and 'key' passed both or none?

I agree. However I don't see this option easily added to current testpmd flow implementation.
Adrien - how would you recommend adding this check?
Please note that currently if no key and no key_len are specified - testpmd still assign a dummy string.

> 
> BTW, is there any profit from the 'key_len' argument for testpmd?
> Can we just always use the size of the passed 'key' and drop the configurable
> from the user interface?
> 

If you just specify key without key_len then the key length is calculated implicitly from the key itself. So this is an already implemented feature. You can still use key_len (with different values) maybe for special case handling / debugging in the PMD.

> Best regards, Ilya Maximets.