From: David Marchand <david.marchand@redhat.com>
To: Ferruh Yigit <ferruh.yigit@intel.com>
Cc: Andrew Rybchenko <arybchenko@solarflare.com>,
Bruce Richardson <bruce.richardson@intel.com>,
"A.McLoughlin" <aideen.mcloughlin@intel.com>,
Wenzhuo Lu <wenzhuo.lu@intel.com>,
Jingjing Wu <jingjing.wu@intel.com>,
Bernard Iremonger <bernard.iremonger@intel.com>,
dev <dev@dpdk.org>, dpdk stable <stable@dpdk.org>,
Thomas Monjalon <thomas@monjalon.net>,
Stephen Hemminger <stephen@networkplumber.org>
Subject: Re: [dpdk-stable] [dpdk-dev] [PATCH] app/testpmd: fix doubling of 'total TX dropped'
Date: Tue, 23 Jul 2019 13:16:06 +0200 [thread overview]
Message-ID: <CAJFAV8zENMUz10ich-nN0UJruNGrFtSPZoNWBLtU6QeEEAHdKA@mail.gmail.com> (raw)
In-Reply-To: <ef0d49bd-8885-1702-5481-40d3394eab46@intel.com>
On Tue, Jul 16, 2019 at 5:29 PM Ferruh Yigit <ferruh.yigit@intel.com> wrote:
>
> On 7/16/2019 4:00 PM, Andrew Rybchenko wrote:
> > On 7/16/19 5:28 PM, Bruce Richardson wrote:
> >> On Tue, Jul 16, 2019 at 03:23:03PM +0100, Ferruh Yigit wrote:
> >>> On 7/16/2019 1:23 PM, Andrew Rybchenko wrote:
> >>>> On 7/15/19 5:53 PM, Ferruh Yigit wrote:
> >>>>> On 7/12/2019 9:32 AM, A.McLoughlin wrote:
> >>>>>> The 'Accumulated forward statistics for all ports' incorrectly displayed
> >>>>>> double the actual value for 'total_tx_dropped'. This was because 2
> >>>>>> lines in the same function both incremented total_tx_dropped every time
> >>>>>> a packet was dropped. I removed one of these lines to fix this issue.
> >>>>>>
> >>>>>> Fixes: 53324971a14e ("app/testpmd: display/clear forwarding stats on demand")
> >>>>>> Cc: david.marchand@redhat.com
> >>>>>> Cc: stable@dpdk.org
> >>>>>>
> >>>>>> Signed-off-by: A.McLoughlin <aideen.mcloughlin@intel.com>
> >>>>>> ---
> >>>>>> app/test-pmd/testpmd.c | 1 -
> >>>>>> 1 file changed, 1 deletion(-)
> >>>>>>
> >>>>>> diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
> >>>>>> index 3ed3523b7..c41bada50 100644
> >>>>>> --- a/app/test-pmd/testpmd.c
> >>>>>> +++ b/app/test-pmd/testpmd.c
> >>>>>> @@ -1555,7 +1555,6 @@ fwd_stats_display(void)
> >>>>>> total_recv += stats.ipackets;
> >>>>>> total_xmit += stats.opackets;
> >>>>>> total_rx_dropped += stats.imissed;
> >>>>>> - total_tx_dropped += ports_stats[pt_id].tx_dropped;
> >>>>>> total_tx_dropped += stats.oerrors;
> >>>>>> total_rx_nombuf += stats.rx_nombuf;
> >>>>>>
> >>>>>>
> >>>>>
> >>>>> Hi Aideen,
> >>>>>
> >>>>> Indeed 'rte_eth_stats->oerrors' and 'tx_dropped' are different values,
> >>>>>
> >>>>> in testpmd, 'TX-total' is taken as "total_xmit + total_tx_dropped", from this
> >>>>> description it may be fair to say
> >>>>> "total_tx_dropped = oerrors + tx_dropped"
> >>>>>
> >>>>> This is easier to see in HW devices, 'oerrors' is the packets sent to HW but HW
> >>>>> reported failure for some reason, so these packets not transmitted to the medium.
> >>>>> 'tx_dropped' is mostly calculated by application, driver returns packets that
> >>>>> can't able to sent to HW, so application can re-try to send or free them and
> >>>>> increase 'tx_dropped' counter.
> >>>>>
> >>>>>
> >>>>> The problem is in the virtual drivers, the packets not able to sent are
> >>>>> calculated as 'oerrors' and tx_burst functions returns the number of the
> >>>>> successfully sent packets which cause application calculate remaining ones as
> >>>>> 'tx_dropped' which cause the duplication.
> >>>>
> >>>> I don't understand how it is. Tx burst returns a number of owned packets
> >>>> (either successfully transmitted or internally dropped/freed). If it is
> >>>> smaller than number of packets in request, other packets are either
> >>>> retried or calculated as tx_dropped.
> >>>
> >>> Virtual PMDs, at least the ones I checked, calculating not sent packets as
> >>> error, also application calculates them as tx_dropped.
> >>>
> >>> Like:
> >>> https://git.dpdk.org/dpdk/tree/drivers/net/ring/rte_eth_ring.c?h=v19.08-rc1#n97
> >>>
> >> That is probably incorrect to do. Virtual PMDs should behave as real ones
> >> do as far as possible. I think we should change them to not count as errors
> >> any that are not handled by the driver, provided those are returned to the
> >> app.
> >
> > I agree with Bruce. It is a bug in ring PMD. If so, too fast attempts to
> > transmit packets will blow up err_pkts counter.
> >
>
> +1, as far as I can see following PMDs requires fixing:
> ring
> kni
> pcap
> tap
I already have fixes for some of them (found when I started to look at
the stats).
--
David Marchand
prev parent reply other threads:[~2019-07-23 11:16 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-12 8:32 [dpdk-stable] " A.McLoughlin
2019-07-12 13:33 ` Iremonger, Bernard
2019-07-15 14:53 ` [dpdk-stable] [dpdk-dev] " Ferruh Yigit
2019-07-16 12:23 ` Andrew Rybchenko
2019-07-16 14:23 ` Ferruh Yigit
2019-07-16 14:28 ` Bruce Richardson
2019-07-16 15:00 ` Andrew Rybchenko
2019-07-16 15:29 ` Ferruh Yigit
2019-07-23 11:16 ` David Marchand [this message]
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=CAJFAV8zENMUz10ich-nN0UJruNGrFtSPZoNWBLtU6QeEEAHdKA@mail.gmail.com \
--to=david.marchand@redhat.com \
--cc=aideen.mcloughlin@intel.com \
--cc=arybchenko@solarflare.com \
--cc=bernard.iremonger@intel.com \
--cc=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
--cc=jingjing.wu@intel.com \
--cc=stable@dpdk.org \
--cc=stephen@networkplumber.org \
--cc=thomas@monjalon.net \
--cc=wenzhuo.lu@intel.com \
/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).