[PATCH v3 04/33] net/ena: sub-optimal configuration notifications support

[Brandes, Shai]

> -----Original Message-----
> From: Ferruh Yigit <ferruh.yigit at amd.com>
> Sent: Friday, March 8, 2024 7:23 PM
> To: Brandes, Shai <shaibran at amazon.com>
> Cc: dev at dpdk.org
> Subject: RE: [EXTERNAL] [PATCH v3 04/33] net/ena: sub-optimal
> configuration notifications support
> 
> CAUTION: This email originated from outside of the organization. Do not click
> links or open attachments unless you can confirm the sender and know the
> content is safe.
> 
> 
> 
> On 3/6/2024 12:24 PM, shaibran at amazon.com wrote:
> > From: Shai Brandes <shaibran at amazon.com>
> >
> > ENA device will send asynchronous notifications to the driver in order
> > to notify users about sub-optimal configurations and refer them to
> > public AWS documentation for further action.
> >
> 
> Hi Shai,
> 
> This is an interesting feature, I am curious, is there more public detail
> provided by AWS on how it detects sub-optimal configuration and what are
> the possible types of the notifications?
> 
[Brandes, Shai] This is only a framework to allow notifications to the user. Currently, the only notification the device supports relate to sub-optimal configuration when enabling ena-express feature.
The public documentation for it was not published yet, but it currently contains only two codes, indicating the user that it is better to run with normal-llq when working with ena-express and an option to increase the Tx queue depth when working with ena-express to double the default size on specific hardwares that have a larger bar (known only in run-time)


> > Signed-off-by: Shai Brandes <shaibran at amazon.com>
> > Reviewed-by: Amit Bernstein <amitbern at amazon.com>
> > ---
> >  doc/guides/rel_notes/release_24_03.rst        |  1 +
> >  .../net/ena/base/ena_defs/ena_admin_defs.h    | 11 +++++++-
> >  drivers/net/ena/ena_ethdev.c                  | 26 +++++++++++++++++--
> >  3 files changed, 35 insertions(+), 3 deletions(-)
> >
> > diff --git a/doc/guides/rel_notes/release_24_03.rst
> > b/doc/guides/rel_notes/release_24_03.rst
> > index fb66d67d32..f47073c7dc 100644
> > --- a/doc/guides/rel_notes/release_24_03.rst
> > +++ b/doc/guides/rel_notes/release_24_03.rst
> > @@ -104,6 +104,7 @@ New Features
> >  * **Updated Amazon ena (Elastic Network Adapter) net driver.**
> >
> >    * Removed the reporting of `rx_overruns` errors from xstats and instead
> updated `imissed` stat with its value.
> > +  * Added support for sub-optimal configuration notifications from the
> device.
> >
> >  * **Updated Atomic Rules' Arkville driver.**
> >
> > diff --git a/drivers/net/ena/base/ena_defs/ena_admin_defs.h
> > b/drivers/net/ena/base/ena_defs/ena_admin_defs.h
> > index fa43e22918..4172916551 100644
> > --- a/drivers/net/ena/base/ena_defs/ena_admin_defs.h
> > +++ b/drivers/net/ena/base/ena_defs/ena_admin_defs.h
> > @@ -1214,7 +1214,8 @@ enum ena_admin_aenq_group {
> >       ENA_ADMIN_NOTIFICATION                      = 3,
> >       ENA_ADMIN_KEEP_ALIVE                        = 4,
> >       ENA_ADMIN_REFRESH_CAPABILITIES              = 5,
> > -     ENA_ADMIN_AENQ_GROUPS_NUM                   = 6,
> > +     ENA_ADMIN_CONF_NOTIFICATIONS                = 6,
> > +     ENA_ADMIN_AENQ_GROUPS_NUM                   = 7,
> >  };
> >
> >  enum ena_admin_aenq_notification_syndrome { @@ -1251,6 +1252,14
> @@
> > struct ena_admin_aenq_keep_alive_desc {
> >       uint32_t rx_overruns_high;
> >  };
> >
> > +struct ena_admin_aenq_conf_notifications_desc {
> > +     struct ena_admin_aenq_common_desc aenq_common_desc;
> > +
> > +     uint64_t notifications_bitmap;
> > +
> > +     uint64_t reserved;
> > +};
> > +
> >  struct ena_admin_ena_mmio_req_read_less_resp {
> >       uint16_t req_id;
> >
> > diff --git a/drivers/net/ena/ena_ethdev.c
> > b/drivers/net/ena/ena_ethdev.c index d3f395a832..3157237c0d 100644
> > --- a/drivers/net/ena/ena_ethdev.c
> > +++ b/drivers/net/ena/ena_ethdev.c
> > @@ -36,6 +36,10 @@
> >
> >  #define ENA_MIN_RING_DESC    128
> >
> > +#define BITS_PER_BYTE 8
> > +
> > +#define BITS_PER_TYPE(type) (sizeof(type) * BITS_PER_BYTE)
> > +
> >
> 
> 'CHAR_BIT' macro can be used here, but I can see there are multiple drivers
> defining similar macros.So no need to update this patch, but to record that
> this is something to address DPDK wide.
> 
> If ena team volunteers to tackle this update, it is welcomed ;)
[Brandes, Shai] sure, can be done