From: "Zhang, Qi Z" <qi.z.zhang@intel.com>
To: "Ergin, Mesut A" <mesut.a.ergin@intel.com>,
"Xing, Beilei" <beilei.xing@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH 3/3] net/i40e: fix inadvertent override of vector RX allowance
Date: Sat, 25 May 2019 11:29:51 +0000 [thread overview]
Message-ID: <039ED4275CED7440929022BC67E706115338732B@SHSMSX103.ccr.corp.intel.com> (raw)
In-Reply-To: <3615B82CA151CF42A86EDDD9846A8B38C7A8E453@ORSMSX112.amr.corp.intel.com>
Hi Muset:
> -----Original Message-----
> From: Ergin, Mesut A
> Sent: Saturday, May 25, 2019 2:09 AM
> To: Zhang, Qi Z <qi.z.zhang@intel.com>; Xing, Beilei <beilei.xing@intel.com>
> Cc: dev@dpdk.org
> Subject: RE: [PATCH 3/3] net/i40e: fix inadvertent override of vector RX
> allowance
>
> Hi,
>
> > > > > i40e_rx_vec_dev_conf_condition_check_default(struct rte_eth_dev
> *dev)
> > > > > if (rxmode->offloads & DEV_RX_OFFLOAD_VLAN_EXTEND)
> > > > > return -1;
> > > > >
> > > > > + /* Should not override if vector was already disallowed */
> > > >
> > > > It is possible a device be reconfigured between
> > > > dev_stop/dev_start, vector mode may fit for the new configure, so
> > > > the old rx_vec_allowd should be ignored,
> > > >
> > >
> > > i40e_dev_configure would reset rx_vec_allowed already. Am I missing
> > > another reconfiguration path?
> >
> > Look at below scenario,
> >
> > 1. dev_configure (rx_vec_allowed is reset to true) 2. queue_setup (the
> > ring size is not power of 2) 3. dev_start (vector will not be selected
> > due to ring size, rx_vec_allowed set to
> > false)
> > 4. dev_stop
> > 5. queue_setup (this time, with power of 2 ring size) 6. dev_start
> > (assume vector path should be selected, and rx_vec_allowed should be
> > overwrite to true, but your patch will prevent it)
> >
> > Also, I may not get the point of the gap you observed, would you share
> > more detail scenario?
> >
> > Regards
> > Qi
>
> It seems like queue setup should reset vector_allowed flag, too. Because
> i40e_dev_rx_queue_setup_runtime is already doing it, though covering only
> half of the state space (i.e. device already started).
>
> Do you think this patch plus that would be a preferred approach?
First, I need to understand your gap, that might be obvious, but so far I just did not figure out it base your commit log yet, so please give me more heads up :)
Secondary, I think move below codes from dev_configure to dev_start might be a better solution
ad->rx_bulk_alloc_allowed = true;
ad->rx_vec_allowed = true;
ad->tx_simple_allowed = true;
ad->tx_vec_allowed = true;
Regards
Qi
>
> >
> > >
> > > Mesut
> > >
> > > > Regards
> > > > Qi
> > > >
> > > > > + if (!ad->rx_vec_allowed)
> > > > > + return -1;
> > > > > +
> > > > > /**
> > > > > * Vector mode is allowed only when number of Rx queue
> > > > > * descriptor is power of 2.
> > > > > --
> > > > > 2.7.4
prev parent reply other threads:[~2019-05-25 11:29 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-16 4:28 [dpdk-dev] [PATCH 0/3] net/i40e: improve rte_flow offload with MARK + RSS Mesut Ali Ergin
2019-05-16 4:28 ` Mesut Ali Ergin
2019-05-16 4:28 ` [dpdk-dev] [PATCH 1/3] net/i40e: add support for MARK + RSS action in rte_flow Mesut Ali Ergin
2019-05-16 4:28 ` Mesut Ali Ergin
2019-05-22 12:30 ` Zhang, Qi Z
2019-05-16 4:28 ` [dpdk-dev] [PATCH 2/3] net/i40e: add runtime option to disable vector rx Mesut Ali Ergin
2019-05-16 4:28 ` Mesut Ali Ergin
2019-05-16 8:17 ` Maxime Coquelin
2019-05-16 8:17 ` Maxime Coquelin
2019-05-16 21:26 ` Ergin, Mesut A
2019-05-20 8:29 ` Ananyev, Konstantin
2019-05-20 17:53 ` Ergin, Mesut A
2019-05-20 23:11 ` Ananyev, Konstantin
2019-05-21 16:33 ` Ergin, Mesut A
2019-05-21 17:20 ` Ananyev, Konstantin
2019-05-21 20:55 ` Ergin, Mesut A
2019-05-22 11:01 ` Ananyev, Konstantin
2019-05-22 14:05 ` Thomas Monjalon
2019-05-22 14:32 ` Zhang, Qi Z
2019-05-23 13:10 ` Jerin Jacob Kollanukkaran
2019-05-23 17:54 ` Ergin, Mesut A
2019-05-16 4:28 ` [dpdk-dev] [PATCH 3/3] net/i40e: fix inadvertent override of vector RX allowance Mesut Ali Ergin
2019-05-16 4:28 ` Mesut Ali Ergin
2019-05-16 8:17 ` Maxime Coquelin
2019-05-16 8:17 ` Maxime Coquelin
2019-05-16 20:57 ` Ergin, Mesut A
2019-05-22 12:42 ` Zhang, Qi Z
2019-05-23 18:25 ` Ergin, Mesut A
2019-05-24 2:39 ` Zhang, Qi Z
2019-05-24 18:08 ` Ergin, Mesut A
2019-05-25 11:29 ` Zhang, Qi Z [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=039ED4275CED7440929022BC67E706115338732B@SHSMSX103.ccr.corp.intel.com \
--to=qi.z.zhang@intel.com \
--cc=beilei.xing@intel.com \
--cc=dev@dpdk.org \
--cc=mesut.a.ergin@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).