[dpdk-dev] [RFC 0/4] mempool: avoid objects allocations across pages

[Vamsi Krishna Attunuru]

> -----Original Message-----
> From: Olivier Matz <olivier.matz at 6wind.com>
> Sent: Friday, July 19, 2019 7:09 PM
> To: Vamsi Krishna Attunuru <vattunuru at marvell.com>; dev at dpdk.org
> Cc: Andrew Rybchenko <arybchenko at solarflare.com>; Thomas Monjalon
> <thomas at monjalon.net>; Anatoly Burakov <anatoly.burakov at intel.com>; Jerin
> Jacob Kollanukkaran <jerinj at marvell.com>; Kiran Kumar Kokkilagadda
> <kirankumark at marvell.com>; Ferruh Yigit <ferruh.yigit at intel.com>
> Subject: [RFC 0/4] mempool: avoid objects allocations across pages
> 
> When IOVA mode is VA, a mempool can be created with objects that are not
> physically contiguous, which breaks KNI.
> 
> To solve this, this patchset changes the default behavior of mempool populate
> function, to prevent objects from being located across pages.
> 
> Olivier Matz (4):
>   mempool: clarify default populate function
>   mempool: unalign size when calculating required mem amount
>   mempool: introduce function to get mempool page size
>   mempool: prevent objects from being across pages
> 
>  lib/librte_mempool/rte_mempool.c             | 106 +++++++++++----------------
>  lib/librte_mempool/rte_mempool.h             |   8 +-
>  lib/librte_mempool/rte_mempool_ops.c         |   4 +-
>  lib/librte_mempool/rte_mempool_ops_default.c |  39 +++++++++-
>  4 files changed, 90 insertions(+), 67 deletions(-)
> 
> ---
> 
> Hi,
> 
> > @Olivier,
> > Any suggestions..?
> 
> I took some time to go a bit deeper. I still think we can change the default
> behavior to avoid objects to be located accross pages. But it is more complex
> that I expected.
> 
> I made a draft patchset, that, in short:
> - cleans/renames variables
> - removes the optimistic full iova contiguous allocation
> - changes return value of calc_mem_size to return the unaligned size,
>   therefore the allocation is smaller in case of big hugepages
> - changes rte_mempool_op_populate_default() to prevent allocation
>   of objects accross multiple pages
> 
> Andrew, Anatoly, did I miss something?
> Vamsi, can you check if it solves your issue?
> 
Thanks Olivier for your time, it solves our problem with the KNI in VA mode. 
Since the change set could not go in 19.08 and as you suggested, 
implemented mempool populate without a flag and used the same for kni application. 
Sending v8 with these changes.

> Anyway, even if validate the patchset it and make it work, I'm afraid this is not
> something that could go in 19.08.
> 
> The only alternative I see is a specific mempool allocation function when used in
> iova=va mode + kni, as you proposed previously.
> 
> It can probably be implemented without adding a flag, starting from
> rte_mempool_create(), and replacing rte_mempool_populate_default(mp) by
> something else: allocate pages one by one, and call
> rte_mempool_populate_iova() for each of them.
> 
> Hope it helps. Unfortunately, I may not have too much time to spend on it in the
> coming days.
> 
> Regards,
> Olivier