From: Matan Azrad <matan@mellanox.com>
To: "Tan, Jianfeng" <jianfeng.tan@intel.com>,
"Yigit, Ferruh" <ferruh.yigit@intel.com>
Cc: "Richardson, Bruce" <bruce.richardson@intel.com>,
"Ananyev, Konstantin" <konstantin.ananyev@intel.com>,
Thomas Monjalon <thomas@monjalon.net>,
"maxime.coquelin@redhat.com" <maxime.coquelin@redhat.com>,
"Burakov, Anatoly" <anatoly.burakov@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH 3/4] drivers/net: do not allocate rte_eth_dev_data privately
Date: Wed, 7 Mar 2018 06:00:44 +0000 [thread overview]
Message-ID: <AM4PR0501MB265753064BC2B7DDE6397B71D2D80@AM4PR0501MB2657.eurprd05.prod.outlook.com> (raw)
In-Reply-To: <ED26CBA2FAD1BF48A8719AEF02201E3651463A53@SHSMSX103.ccr.corp.intel.com>
Hi Jianfeng
From: Tan, Jianfeng, Sent: Tuesday, March 6, 2018 10:56 AM
> > -----Original Message-----
> > From: Matan Azrad [mailto:matan@mellanox.com]
> > Sent: Tuesday, March 6, 2018 2:08 PM
> > To: Tan, Jianfeng; Yigit, Ferruh
> > Cc: Richardson, Bruce; Ananyev, Konstantin; Thomas Monjalon;
> > maxime.coquelin@redhat.com; Burakov, Anatoly; dev@dpdk.org
> > Subject: RE: [dpdk-dev] [PATCH 3/4] drivers/net: do not allocate
> > rte_eth_dev_data privately
> >
> > Hi Jianfeng
> >
> > Please see a comment below.
> >
> > > From: Jianfeng Tan, Sent: Sunday, March 4, 2018 5:30 PM We
> > > introduced private rte_eth_dev_data to allow vdev to be created both
> > in
> > > primary process and secondary process(es). This is not friendly to
> > > multi- process model, for example, it leads to port id contention
> > > issue if two processes both find the data entry is free.
> > >
> > > And to get stats of primary vdev in secondary, we must allocate from
> > > the pre-defined array so that we can find it.
> > >
> > > Suggested-by: Bruce Richardson <bruce.richardson@intel.com>
> > > Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
> > > ---
> > > drivers/net/af_packet/rte_eth_af_packet.c | 25 +++++++------------------
> > > drivers/net/kni/rte_eth_kni.c | 13 ++-----------
> > > drivers/net/null/rte_eth_null.c | 17 +++--------------
> > > drivers/net/octeontx/octeontx_ethdev.c | 14 ++------------
> > > drivers/net/pcap/rte_eth_pcap.c | 18 +++---------------
> > > drivers/net/tap/rte_eth_tap.c | 9 +--------
> > > drivers/net/vhost/rte_eth_vhost.c | 17 ++---------------
> > > 7 files changed, 20 insertions(+), 93 deletions(-)
> > >
> > > diff --git a/drivers/net/af_packet/rte_eth_af_packet.c
> > > b/drivers/net/af_packet/rte_eth_af_packet.c
> > > index 57eccfd..2db692f 100644
> > > --- a/drivers/net/af_packet/rte_eth_af_packet.c
> > > +++ b/drivers/net/af_packet/rte_eth_af_packet.c
> > > @@ -564,25 +564,17 @@ rte_pmd_init_internals(struct rte_vdev_device
> > > *dev,
> > > RTE_LOG(ERR, PMD,
> > > "%s: no interface specified for AF_PACKET
> ethdev\n",
> > > name);
> > > - goto error_early;
> > > + return -1;
> > > }
> > >
> > > RTE_LOG(INFO, PMD,
> > > "%s: creating AF_PACKET-backed ethdev on numa socket
> %u\n",
> > > name, numa_node);
> > >
> > > - /*
> > > - * now do all data allocation - for eth_dev structure, dummy pci
> > > driver
> > > - * and internal (private) data
> > > - */
> > > - data = rte_zmalloc_socket(name, sizeof(*data), 0, numa_node);
> > > - if (data == NULL)
> > > - goto error_early;
> > > -
> > > *internals = rte_zmalloc_socket(name, sizeof(**internals),
> > > 0, numa_node);
> > > if (*internals == NULL)
> > > - goto error_early;
> > > + return -1;
> > >
> > > for (q = 0; q < nb_queues; q++) {
> > > (*internals)->rx_queue[q].map = MAP_FAILED; @@ -604,24
> > > +596,24 @@ rte_pmd_init_internals(struct rte_vdev_device *dev,
> > > RTE_LOG(ERR, PMD,
> > > "%s: I/F name too long (%s)\n",
> > > name, pair->value);
> > > - goto error_early;
> > > + return -1;
> > > }
> > > if (ioctl(sockfd, SIOCGIFINDEX, &ifr) == -1) {
> > > RTE_LOG(ERR, PMD,
> > > "%s: ioctl failed (SIOCGIFINDEX)\n",
> > > name);
> > > - goto error_early;
> > > + return -1;
> > > }
> > > (*internals)->if_name = strdup(pair->value);
> > > if ((*internals)->if_name == NULL)
> > > - goto error_early;
> > > + return -1;
> > > (*internals)->if_index = ifr.ifr_ifindex;
> > >
> > > if (ioctl(sockfd, SIOCGIFHWADDR, &ifr) == -1) {
> > > RTE_LOG(ERR, PMD,
> > > "%s: ioctl failed (SIOCGIFHWADDR)\n",
> > > name);
> > > - goto error_early;
> > > + return -1;
> > > }
> > > memcpy(&(*internals)->eth_addr, ifr.ifr_hwaddr.sa_data,
> ETH_ALEN);
> > >
> > > @@ -775,14 +767,13 @@ rte_pmd_init_internals(struct rte_vdev_device
> > > *dev,
> > >
> > > (*internals)->nb_queues = nb_queues;
> > >
> > > - rte_memcpy(data, (*eth_dev)->data, sizeof(*data));
> > > + data = (*eth_dev)->data;
> > > data->dev_private = *internals;
> > > data->nb_rx_queues = (uint16_t)nb_queues;
> > > data->nb_tx_queues = (uint16_t)nb_queues;
> > > data->dev_link = pmd_link;
> > > data->mac_addrs = &(*internals)->eth_addr;
> > >
> > > - (*eth_dev)->data = data;
> > > (*eth_dev)->dev_ops = &ops;
> > >
> > > return 0;
> > > @@ -802,8 +793,6 @@ rte_pmd_init_internals(struct rte_vdev_device
> > *dev,
> > > }
> > > free((*internals)->if_name);
> > > rte_free(*internals);
> > > -error_early:
> > > - rte_free(data);
> > > return -1;
> > > }
> > >
> >
> > I think you should remove the private rte_eth_dev_data freeing in
> > rte_pmd_af_packet_remove().
> > This is relevant to all the vdevs here.
>
> Ah, yes, you are correct. I will fix that in v2.
>
> >
> > Question:
> > Does the patch include all the vdevs which allocated private
> > rte_eth_dev_data?
>
> Yes, we are removing all private rte_eth_dev_data. If I missed some device,
> welcome to point out.
net/ring
> > If so, it may solve also part of the issue discussed here:
> >
> https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdp
> d
> >
> k.org%2Fdev%2Fpatchwork%2Fpatch%2F34047%2F&data=02%7C01%7Cmata
> n%40mell
> >
> anox.com%7C4143e70010774a15672708d583401618%7Ca652971c7d2e4d9ba6
> a4d149
> >
> 256f461b%7C0%7C0%7C636559233645410291&sdata=G1pYHEXENP3low8oziaI
> KsxiHB
> > mlEjV1f89LMZmnzvc%3D&reserved=0
>
> Yes, related. We now allocate rte_eth_dev_data which can be indexed by all
> primary/secondary processes.
>
> Thanks,
> Jianfeng
next prev parent reply other threads:[~2018-03-07 6:00 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-04 15:30 [dpdk-dev] [PATCH 0/4] allow procinfo and pdump on eth vdev Jianfeng Tan
2018-03-04 15:30 ` [dpdk-dev] [PATCH 1/4] eal: bring forward multi-process channel init Jianfeng Tan
2018-03-04 15:30 ` [dpdk-dev] [PATCH 2/4] bus/vdev: bus scan by multi-process channel Jianfeng Tan
2018-03-05 9:36 ` Burakov, Anatoly
2018-03-06 0:50 ` Tan, Jianfeng
2018-03-07 14:00 ` Burakov, Anatoly
2018-03-12 3:22 ` Tan, Jianfeng
2018-03-04 15:30 ` [dpdk-dev] [PATCH 3/4] drivers/net: do not allocate rte_eth_dev_data privately Jianfeng Tan
2018-03-06 6:07 ` Matan Azrad
2018-03-06 8:55 ` Tan, Jianfeng
2018-03-07 6:00 ` Matan Azrad [this message]
2018-03-07 6:10 ` Matan Azrad
2018-03-12 3:40 ` Tan, Jianfeng
2018-03-04 15:30 ` [dpdk-dev] [PATCH 4/4] drivers/net: share vdev data to secondary process Jianfeng Tan
2018-04-19 16:50 ` [dpdk-dev] [PATCH v3 0/5] allow procinfo and pdump on eth vdev Jianfeng Tan
2018-04-19 16:50 ` [dpdk-dev] [PATCH v3 1/5] eal: bring forward multi-process channel init Jianfeng Tan
2018-04-20 8:16 ` Burakov, Anatoly
2018-04-20 14:08 ` Tan, Jianfeng
2018-04-19 16:50 ` [dpdk-dev] [PATCH v3 2/5] bus/vdev: add lock on vdev device list Jianfeng Tan
2018-04-20 8:26 ` Burakov, Anatoly
2018-04-20 14:19 ` Tan, Jianfeng
2018-04-20 15:16 ` Burakov, Anatoly
2018-04-20 15:23 ` Tan, Jianfeng
2018-04-19 16:50 ` [dpdk-dev] [PATCH v3 3/5] bus/vdev: bus scan by multi-process channel Jianfeng Tan
2018-04-20 8:41 ` Burakov, Anatoly
2018-04-20 14:28 ` Tan, Jianfeng
2018-04-20 15:19 ` Burakov, Anatoly
2018-04-20 15:32 ` Tan, Jianfeng
2018-04-20 15:39 ` Burakov, Anatoly
2018-04-19 16:50 ` [dpdk-dev] [PATCH v3 4/5] drivers/net: not use private eth dev data Jianfeng Tan
2018-04-19 16:50 ` [dpdk-dev] [PATCH v3 5/5] drivers/net: share vdev data to secondary process Jianfeng Tan
2018-04-20 16:57 ` [dpdk-dev] [PATCH v4 0/5] allow procinfo and pdump on eth vdev Jianfeng Tan
2018-04-20 16:57 ` [dpdk-dev] [PATCH v4 1/5] eal: bring forward multi-process channel init Jianfeng Tan
2018-04-20 16:57 ` [dpdk-dev] [PATCH v4 2/5] bus/vdev: add lock on vdev device list Jianfeng Tan
2018-04-23 9:47 ` Burakov, Anatoly
2018-04-20 16:57 ` [dpdk-dev] [PATCH v4 3/5] bus/vdev: bus scan by multi-process channel Jianfeng Tan
2018-04-23 9:54 ` Burakov, Anatoly
2018-04-24 5:22 ` Tan, Jianfeng
2018-04-20 16:57 ` [dpdk-dev] [PATCH v4 4/5] drivers/net: not use private eth dev data Jianfeng Tan
2018-04-20 16:57 ` [dpdk-dev] [PATCH v4 5/5] drivers/net: share vdev data to secondary process Jianfeng Tan
2018-04-24 5:51 ` [dpdk-dev] [PATCH v5 0/5] allow procinfo and pdump on eth vdev Jianfeng Tan
2018-04-24 5:51 ` [dpdk-dev] [PATCH v5 1/5] eal: bring forward multi-process channel init Jianfeng Tan
2018-04-24 5:51 ` [dpdk-dev] [PATCH v5 2/5] bus/vdev: add lock on vdev device list Jianfeng Tan
2018-04-24 5:51 ` [dpdk-dev] [PATCH v5 3/5] bus/vdev: bus scan by multi-process channel Jianfeng Tan
2018-04-24 10:09 ` Thomas Monjalon
2018-04-24 5:51 ` [dpdk-dev] [PATCH v5 4/5] drivers/net: not use private eth dev data Jianfeng Tan
2018-04-24 5:51 ` [dpdk-dev] [PATCH v5 5/5] drivers/net: share vdev data to secondary process Jianfeng Tan
2018-04-24 10:32 ` [dpdk-dev] [PATCH v5 0/5] allow procinfo and pdump on eth vdev Thomas Monjalon
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=AM4PR0501MB265753064BC2B7DDE6397B71D2D80@AM4PR0501MB2657.eurprd05.prod.outlook.com \
--to=matan@mellanox.com \
--cc=anatoly.burakov@intel.com \
--cc=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@intel.com \
--cc=jianfeng.tan@intel.com \
--cc=konstantin.ananyev@intel.com \
--cc=maxime.coquelin@redhat.com \
--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).