Re: [dpdk-dev] [PATCH] vhost: batch used descriptors chains write-back with packed ring

[Maxime Coquelin <maxime.coquelin@redhat.com>]

On 12/12/18 7:53 PM, Michael S. Tsirkin wrote:
> On Wed, Dec 12, 2018 at 05:34:31PM +0100, Maxime Coquelin wrote:
>> Hi Ilya,
>>
>> On 12/12/18 4:23 PM, Ilya Maximets wrote:
>>> On 12.12.2018 11:24, Maxime Coquelin wrote:
>>>> Instead of writing back descriptors chains in order, let's
>>>> write the first chain flags last in order to improve batching.
>>>>
>>>> With Kernel's pktgen benchmark, ~3% performance gain is measured.
>>>>
>>>> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
>>>> ---
>>>>    lib/librte_vhost/virtio_net.c | 39 +++++++++++++++++++++--------------
>>>>    1 file changed, 24 insertions(+), 15 deletions(-)
>>>>
>>>
>>> Hi.
>>> I made some rough testing on my ARMv8 system with this patch and v1 of it.
>>> Here is the performance difference with current master:
>>>       v1: +1.1 %
>>>       v2: -3.6 %
>>>
>>> So, write barriers are quiet heavy in practice.
>>
>> Thanks for testing it on ARM. Indeed, SMP WMB is heavier on ARM.
> 
> Besides your ideas for improving packed rings, maybe we should switch to
> load_acquite/store_release?
> 
> See
> 	virtio: use smp_load_acquire/smp_store_release
> 
> which worked fine but as I only tested on x86 did not result in any gains.
> 

Thanks for the pointer.
We'll look into it for v19.05, as -rc1 for v19.02 is planned for end of
week, so it will be too late to introduce such changes.

Regards,
Maxime