[ovs-dev] [PATCH v4 2/3] netdev-dpdk: Add custom stat for vhost tx retries.

Ian Stokes ian.stokes at intel.com
Thu Jul 4 15:13:16 UTC 2019


On 7/4/2019 12:18 PM, Ilya Maximets wrote:
> On 04.07.2019 14:06, David Marchand wrote:
>>
>>
>> On Thu, Jul 4, 2019 at 12:42 PM Ilya Maximets <i.maximets at samsung.com <mailto:i.maximets at samsung.com>> wrote:
>>
>>      On 03.07.2019 18:03, Ian Stokes wrote:
>>      > On 7/2/2019 1:32 AM, Kevin Traynor wrote:
>>      >> vhost tx retries may occur, and it can be a sign that
>>      >> the guest is not optimally configured.
>>      >>
>>      >> Add a custom stat so a user will know if vhost tx retries are
>>      >> occurring and hence give a hint that guest config should be
>>      >> examined.
>>      >>
>>      >
>>      > Thanks Kevin, tests ok for me.
>>      >
>>      > Just a general comment on the design. In comparison to the previous approach proposed there seems to be more required with the custom stat approach below.
>>      >
>>      > This may be ok as the retry is very OVS DPDK specific and doesn;t come dor lets say DPDK (unlike the XTATS).
>>      >
>>      > @Ilya, whats your thoughts? Is this approach preferable for you rather than adding it to the general stats for all devices? (in which case I agree, they will not be used for dpdk types so will not be of use).
>>
>>      I think, It's better to keep this in custom stats section since
>>      no other port types are going to use it in a near future.
>>
>>      Have a few style/naming comments inline.
>>
>>      >
>>      > One other minor comment below.
>>      >
>>      >> Signed-off-by: Kevin Traynor <ktraynor at redhat.com <mailto:ktraynor at redhat.com>>
>>      >> ---
>>      >>   Documentation/topics/dpdk/vhost-user.rst |  5 ++++
>>      >>   lib/netdev-dpdk.c                        | 38 +++++++++++++++++++++++-
>>      >>   2 files changed, 42 insertions(+), 1 deletion(-)
>>      >>
>>      >> diff --git a/Documentation/topics/dpdk/vhost-user.rst b/Documentation/topics/dpdk/vhost-user.rst
>>      >> index 5f393aced..368f44520 100644
>>      >> --- a/Documentation/topics/dpdk/vhost-user.rst
>>      >> +++ b/Documentation/topics/dpdk/vhost-user.rst
>>      >> @@ -521,4 +521,9 @@ The guest should also have sufficient cores dedicated for consuming and
>>      >>   processing packets at the required rate.
>>      >>   +The amount of Tx retries on a vhost-user or vhost-user-client interface can be
>>      >> +shown with::
>>      >> +
>>      >> +  $ ovs-vsctl get Interface dpdkvhostclient0 statistics:tx_retries
>>      >> +
>>      >>   vhost-user Dequeue Zero Copy (experimental)
>>      >>   -------------------------------------------
>>      >> diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
>>      >> index a380b6ffe..d3e02d389 100644
>>      >> --- a/lib/netdev-dpdk.c
>>      >> +++ b/lib/netdev-dpdk.c
>>      >> @@ -139,4 +139,10 @@ BUILD_ASSERT_DECL((MAX_NB_MBUF / ROUND_DOWN_POW2(MAX_NB_MBUF / MIN_NB_MBUF))
>>      >>   #define XSTAT_RX_JABBER_ERRORS           "rx_jabber_errors"
>>      >>   +/* Size of vHost custom stats. */
>>      >> +#define VHOST_STAT_TX_RETRIES_SIZE    1
>>
>>      I think, it's better to rename to something like VHOST_CUSTOM_STATS_SIZE.
>>
>>
>> +1
>>
>>
>>
>>      >> +
>>      >> +/* Name of vHost custom stats. */
>>
>>      s/Name/Names/ ?
>>
>>      >> +#define VHOST_STAT_TX_RETRIES    "tx_retries" > +
>>      >
>>      > The alignment of the defines above look odd as they are not aligned with the existing DPDK XSTAT name definitions. It's only minor, as Kevin is on leave I think this will be ok to change on commit if there are no objections.
>>
>>      Sure.
>>
>>      >
>>      > Regards
>>      > Ian
>>      >>   #define SOCKET0              0
>>      >>   @@ -434,7 +440,9 @@ struct netdev_dpdk {
>>      >>       PADDED_MEMBERS(CACHE_LINE_SIZE,
>>      >>           struct netdev_stats stats;
>>      >> +        /* Custom stat for retries when unable to transmit. */
>>      >> +        uint64_t tx_retries;
>>      >>           /* Protects stats */
>>      >>           rte_spinlock_t stats_lock;
>>      >> -        /* 44 pad bytes here. */
>>      >> +        /* 4 pad bytes here. */
>>      >>       );
>>      >>   @@ -1190,4 +1198,6 @@ common_construct(struct netdev *netdev, dpdk_port_t port_no,
>>      >>       dev->rte_xstats_ids_size = 0;
>>      >>   +    dev->tx_retries = 0;
>>      >> +
>>      >>       return 0;
>>      >>   }
>>      >> @@ -2381,4 +2391,5 @@ __netdev_dpdk_vhost_send(struct netdev *netdev, int qid,
>>      >>       netdev_dpdk_vhost_update_tx_counters(&dev->stats, pkts, total_pkts,
>>      >>                                            cnt + dropped);
>>      >> +    dev->tx_retries += MIN(retries, VHOST_ENQ_RETRY_NUM);
>>      >>       rte_spinlock_unlock(&dev->stats_lock);
>>      >>   @@ -2818,4 +2829,27 @@ netdev_dpdk_get_custom_stats(const struct netdev *netdev,
>>      >>   }
>>      >>   +static int
>>      >> +netdev_dpdk_vhost_get_custom_stats(const struct netdev *netdev,
>>      >> +                                   struct netdev_custom_stats *custom_stats)
>>      >> +{
>>      >> +    struct netdev_dpdk *dev = netdev_dpdk_cast(netdev);
>>      >> +
>>      >> +    ovs_mutex_lock(&dev->mutex);
>>      >> +
>>      >> +    custom_stats->size = VHOST_STAT_TX_RETRIES_SIZE;
>>      >> +    custom_stats->counters = (struct netdev_custom_counter *)
>>      >> +                                 xzalloc(sizeof(struct netdev_custom_counter));
>>
>>      Since we have a _SIZE, we, probably, need to use it while allocation.
>>      With this and a few coding-style fixes, this line should look like:
>>
>>          custom_stats->counters = xcalloc(VHOST_STAT_TX_RETRIES_SIZE,
>>                                           sizeof *custom_stats->counters);
>>
>>
>> Can I suggest?
>>      custom_stats->counters = xcalloc(custom_stats->size,
>>                                       sizeof *custom_stats->counters);
> 
> Sure. This looks fine for me too.
> 
>>
>>      >> +    ovs_strlcpy(custom_stats->counters->name, VHOST_STAT_TX_RETRIES,
>>      >> +                NETDEV_CUSTOM_STATS_NAME_SIZE);
>>      >> +
>>      >> +    rte_spinlock_lock(&dev->stats_lock);
>>      >> +    custom_stats->counters->value = dev->tx_retries;
>>
>>
>>
>> I have a patch after this that will add another stats, so I'd rather see custom_stats->counters[0]. than custom_stats->counters->
> 
> +1.
> This is also in line with my previous comment about having a _SIZE.

Just to clarify above, maybe I misunderstood what you meant, but are you 
suggesting

custom_stats->counters[0] = dev->tx_retries;

I don't think that will work as it will be an incorrect type assignment 
right?

Regards
Ian


More information about the dev mailing list