From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Thomas Monjalon <thomas@monjalon.net>
Cc: dev@dpdk.org, David Marchand <david.marchand@redhat.com>,
Andrew Rybchenko <arybchenko@solarflare.com>
Subject: Re: [dpdk-dev] [PATCH 00/12] rxq q_errors[] statistics fixes
Date: Fri, 12 Apr 2019 16:57:07 +0100 [thread overview]
Message-ID: <70d587b4-2906-4c33-cb10-bca07e83b4b9@intel.com> (raw)
In-Reply-To: <2700501.70IsncsQEY@xps>
On 4/12/2019 4:45 PM, Thomas Monjalon wrote:
> 12/04/2019 17:38, Ferruh Yigit:
>> On 4/12/2019 4:07 PM, Thomas Monjalon wrote:
>>> 11/03/2019 18:22, Ferruh Yigit:
>>>> On 3/4/2019 11:18 AM, David Marchand wrote:
>>>>> According to the api, the q_errors[] per queue statistic is for reception
>>>>> errors not transmit errors.
>>>>> This is a first cleanup on statistics before looking at oerrors.
>>>>>
>>>>
>>>> Yes, the patchset looks aligned with the API documentation [1].
>>>>
>>>> What can be the solution after cleanup? We can merge this cleanup and solution
>>>> next to each-other to not leave a gap?
>>>
>>> I think we should merge those fixes in 19.05-rc2.
>>>
>>> It seems there is a lot more work to achieve on stats, so better
>>> to start without waiting for the full picture.
>>>
>>
>> The problem is "q_errors" is available only for Rx queues, and David's patch is
>> preventing drivers to put Tx error stats into "q_errors" field.
>>
>> But it is clear that there is a need for a field for Tx queues errors. David has
>> another patch to using xstats for this. But I believe xstats is making solution
>> confusing, and now approach is unbalanced for Rx and Tx queues.
>>
>> I am for adding a new field for Tx queues "q_errors", and this will make getting
>> stats and David's patch very simple.
>>
>> The problem with the new fields is it breaks the ABI, but we already increased
>> the ABIVER for ethdev this release, I believe this is very good timing for this fix.
>
> If changing the stats API, we should increase the number of stats per queue:
> #define RTE_ETHDEV_QUEUE_STAT_CNTRS 16
> What about 128 queues per port? 256?
We have 5 uint64_t using this [1], it will be 6 if we add new one.
So having 256 queues, will cost 12K memory, this is not a big number.
Is there any other concern of having large array other than possible memory waste?
[1]
uint64_t q_ipackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
uint64_t q_opackets[RTE_ETHDEV_QUEUE_STAT_CNTRS];
uint64_t q_ibytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
uint64_t q_obytes[RTE_ETHDEV_QUEUE_STAT_CNTRS];
uint64_t q_errors[RTE_ETHDEV_QUEUE_STAT_CNTRS];
next prev parent reply other threads:[~2019-04-12 15:57 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-04 11:18 David Marchand
2019-03-04 11:18 ` [dpdk-dev] [PATCH 01/12] net/af_packet: fix incorrect rxq errors stat David Marchand
2019-03-04 11:18 ` [dpdk-dev] [PATCH 02/12] net/avp: " David Marchand
2019-03-04 12:18 ` Legacy, Allain
2019-03-04 11:18 ` [dpdk-dev] [PATCH 03/12] net/bnxt: " David Marchand
2019-03-04 11:18 ` [dpdk-dev] [PATCH 04/12] net/cxgbe: " David Marchand
2019-03-04 11:18 ` [dpdk-dev] [PATCH 05/12] net/kni: " David Marchand
2019-03-04 11:18 ` [dpdk-dev] [PATCH 06/12] net/mlx4: " David Marchand
2019-03-05 8:19 ` Shahaf Shuler
2019-03-04 11:18 ` [dpdk-dev] [PATCH 07/12] net/mlx5: " David Marchand
2019-03-05 8:18 ` Shahaf Shuler
2019-03-04 11:18 ` [dpdk-dev] [PATCH 08/12] net/null: " David Marchand
2019-03-04 11:18 ` [dpdk-dev] [PATCH 09/12] net/pcap: " David Marchand
2019-03-04 11:18 ` [dpdk-dev] [PATCH 10/12] net/ring: " David Marchand
2019-03-04 11:18 ` [dpdk-dev] [PATCH 11/12] net/szedata2: " David Marchand
2019-03-04 11:18 ` [dpdk-dev] [PATCH 12/12] net/tap: " David Marchand
2019-03-04 13:58 ` Wiles, Keith
2019-03-11 17:22 ` [dpdk-dev] [PATCH 00/12] rxq q_errors[] statistics fixes Ferruh Yigit
2019-03-11 18:09 ` David Marchand
2019-03-14 15:12 ` David Marchand
2019-03-14 15:12 ` David Marchand
2019-03-14 15:13 ` [dpdk-dev] [RFC PATCH 1/2] ethdev: introduce internal rxq/txq stats API David Marchand
2019-03-14 15:13 ` David Marchand
2019-03-14 15:13 ` [dpdk-dev] [RFC PATCH 2/2] net/af_packet: convert to new " David Marchand
2019-03-14 15:13 ` David Marchand
2019-03-15 13:30 ` [dpdk-dev] [RFC PATCH 1/2] ethdev: introduce internal " David Marchand
2019-03-15 13:30 ` David Marchand
2019-03-19 17:18 ` Ferruh Yigit
2019-03-19 17:18 ` Ferruh Yigit
2019-03-19 17:54 ` Stephen Hemminger
2019-03-19 17:54 ` Stephen Hemminger
2019-04-12 13:18 ` Thomas Monjalon
2019-04-12 13:18 ` Thomas Monjalon
2019-03-26 9:29 ` David Marchand
2019-03-26 9:29 ` David Marchand
2019-04-12 13:29 ` Thomas Monjalon
2019-04-12 13:29 ` Thomas Monjalon
2019-04-12 14:32 ` David Marchand
2019-04-12 14:32 ` David Marchand
2019-04-12 16:05 ` Stephen Hemminger
2019-04-12 16:05 ` Stephen Hemminger
2019-04-12 15:07 ` [dpdk-dev] [PATCH 00/12] rxq q_errors[] statistics fixes Thomas Monjalon
2019-04-12 15:07 ` Thomas Monjalon
2019-04-12 15:38 ` Ferruh Yigit
2019-04-12 15:38 ` Ferruh Yigit
2019-04-12 15:45 ` Thomas Monjalon
2019-04-12 15:45 ` Thomas Monjalon
2019-04-12 15:57 ` Ferruh Yigit [this message]
2019-04-12 15:57 ` Ferruh Yigit
2019-05-28 21:38 ` Yigit, Ferruh
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=70d587b4-2906-4c33-cb10-bca07e83b4b9@intel.com \
--to=ferruh.yigit@intel.com \
--cc=arybchenko@solarflare.com \
--cc=david.marchand@redhat.com \
--cc=dev@dpdk.org \
--cc=thomas@monjalon.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).