From: Rasesh Mody <rasesh.mody@qlogic.com>
To: Chas Williams <3chas3@gmail.com>, "dev@dpdk.org" <dev@dpdk.org>,
"Bruce Richardson" <bruce.richardson@intel.com>
Cc: "Charles \(Chas\) Williams" <ciwillia@brocade.com>
Subject: Re: [dpdk-dev] [PATCH 2/2] bnx2x: Determine rx/tx queue sizes sooner
Date: Fri, 4 Mar 2016 22:28:57 +0000 [thread overview]
Message-ID: <2552F74A0BCCBE4DBE2AD218C81B2811086B1539@avmb3.qlogic.org> (raw)
In-Reply-To: <1451522271-16924-2-git-send-email-3chas3@gmail.com>
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Chas Williams
> Sent: Wednesday, December 30, 2015 4:38 PM
>
> From: "Charles (Chas) Williams" <ciwillia@brocade.com>
>
> The VF needs to determine the queues sizes before .dev_infos_get so that it
> can hint to the upper layer the proper sizes. Move
> bnx2x_vf_get_resources() to .eth_dev_init and probe with the guesses from
> bnx2x_init_rte().
>
> Signed-off-by: Chas Williams <3chas3@gmail.com>
> ---
Acked-by: Rasesh Mody <rasesh.mody@qlogic.com>
The change looks good.
Thanks!
Rasesh
> drivers/net/bnx2x/bnx2x.c | 9 ++++++--
> drivers/net/bnx2x/bnx2x_ethdev.c | 46 ++++++++++++++++++++++--------
> ----------
> drivers/net/bnx2x/bnx2x_vfpf.c | 2 ++
> 3 files changed, 34 insertions(+), 23 deletions(-)
>
> diff --git a/drivers/net/bnx2x/bnx2x.c b/drivers/net/bnx2x/bnx2x.c index
> 67af5da..f003875 100644
> --- a/drivers/net/bnx2x/bnx2x.c
> +++ b/drivers/net/bnx2x/bnx2x.c
> @@ -9578,8 +9578,13 @@ static int bnx2x_pci_get_caps(struct bnx2x_softc
> *sc)
>
> static void bnx2x_init_rte(struct bnx2x_softc *sc) {
> - sc->max_tx_queues = 128;
> - sc->max_rx_queues = 128;
> + if (IS_VF(sc)) {
> + sc->max_tx_queues = BNX2X_VF_MAX_QUEUES_PER_VF;
> + sc->max_rx_queues = BNX2X_VF_MAX_QUEUES_PER_VF;
> + } else {
> + sc->max_tx_queues = 128;
> + sc->max_rx_queues = 128;
> + }
> }
>
> #define FW_HEADER_LEN 104
> diff --git a/drivers/net/bnx2x/bnx2x_ethdev.c
> b/drivers/net/bnx2x/bnx2x_ethdev.c
> index 69df02e..fe8cfd0 100644
> --- a/drivers/net/bnx2x/bnx2x_ethdev.c
> +++ b/drivers/net/bnx2x/bnx2x_ethdev.c
> @@ -87,7 +87,6 @@ bnx2x_dev_configure(struct rte_eth_dev *dev) {
> struct bnx2x_softc *sc = dev->data->dev_private;
> int mp_ncpus = sysconf(_SC_NPROCESSORS_CONF);
> - int ret;
>
> PMD_INIT_FUNC_TRACE();
>
> @@ -121,25 +120,6 @@ bnx2x_dev_configure(struct rte_eth_dev *dev)
> return -ENXIO;
> }
>
> - if (IS_VF(sc)) {
> - if (bnx2x_dma_alloc(sc, sizeof(struct bnx2x_vf_mbx_msg),
> - &sc->vf2pf_mbox_mapping, "vf2pf_mbox",
> - RTE_CACHE_LINE_SIZE) != 0)
> - return -ENOMEM;
> -
> - sc->vf2pf_mbox = (struct bnx2x_vf_mbx_msg *)sc-
> >vf2pf_mbox_mapping.vaddr;
> - if (bnx2x_dma_alloc(sc, sizeof(struct bnx2x_vf_bulletin),
> - &sc->pf2vf_bulletin_mapping, "vf2pf_bull",
> - RTE_CACHE_LINE_SIZE) != 0)
> - return -ENOMEM;
> -
> - sc->pf2vf_bulletin = (struct bnx2x_vf_bulletin *)sc-
> >pf2vf_bulletin_mapping.vaddr;
> -
> - ret = bnx2x_vf_get_resources(sc, sc->num_queues, sc-
> >num_queues);
> - if (ret)
> - return ret;
> - }
> -
> return 0;
> }
>
> @@ -466,6 +446,7 @@ bnx2x_common_dev_init(struct rte_eth_dev
> *eth_dev, int is_vf)
> ret = bnx2x_attach(sc);
> if (ret) {
> PMD_DRV_LOG(ERR, "bnx2x_attach failed (%d)", ret);
> + return ret;
> }
>
> eth_dev->data->mac_addrs = (struct ether_addr *)sc-
> >link_params.mac_addr; @@ -479,7 +460,30 @@
> bnx2x_common_dev_init(struct rte_eth_dev *eth_dev, int is_vf)
> PMD_DRV_LOG(INFO, "portID=%d vendorID=0x%x deviceID=0x%x",
> eth_dev->data->port_id, pci_dev->id.vendor_id,
> pci_dev->id.device_id);
>
> - return ret;
> + if (IS_VF(sc)) {
> + if (bnx2x_dma_alloc(sc, sizeof(struct bnx2x_vf_mbx_msg),
> + &sc->vf2pf_mbox_mapping,
> "vf2pf_mbox",
> + RTE_CACHE_LINE_SIZE) != 0)
> + return -ENOMEM;
> +
> + sc->vf2pf_mbox = (struct bnx2x_vf_mbx_msg *)
> + sc->vf2pf_mbox_mapping.vaddr;
> +
> + if (bnx2x_dma_alloc(sc, sizeof(struct bnx2x_vf_bulletin),
> + &sc->pf2vf_bulletin_mapping,
> "vf2pf_bull",
> + RTE_CACHE_LINE_SIZE) != 0)
> + return -ENOMEM;
> +
> + sc->pf2vf_bulletin = (struct bnx2x_vf_bulletin *)
> + sc->pf2vf_bulletin_mapping.vaddr;
> +
> + ret = bnx2x_vf_get_resources(sc, sc->max_tx_queues,
> + sc->max_rx_queues);
> + if (ret)
> + return ret;
> + }
> +
> + return 0;
> }
>
> static int
> diff --git a/drivers/net/bnx2x/bnx2x_vfpf.c
> b/drivers/net/bnx2x/bnx2x_vfpf.c index 34b6360..bce734f 100644
> --- a/drivers/net/bnx2x/bnx2x_vfpf.c
> +++ b/drivers/net/bnx2x/bnx2x_vfpf.c
> @@ -282,6 +282,8 @@ int bnx2x_vf_get_resources(struct bnx2x_softc *sc,
> uint8_t tx_count, uint8_t rx_
> sc->igu_sb_cnt = sc_resp.resc.num_sbs;
> sc->igu_base_sb = sc_resp.resc.hw_sbs[0] & 0xFF;
> sc->igu_dsb_id = -1;
> + sc->max_tx_queues = sc_resp.resc.num_txqs;
> + sc->max_rx_queues = sc_resp.resc.num_rxqs;
>
> sc->link_params.chip_id = sc->devinfo.chip_id;
> sc->doorbell_size = sc_resp.db_size;
> --
> 2.5.0
next prev parent reply other threads:[~2016-03-04 22:28 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-31 0:37 [dpdk-dev] [PATCH 1/2] bnx2x: fix error handling in bnx2x_loop_obtain_resources() Chas Williams
2015-12-31 0:37 ` [dpdk-dev] [PATCH 2/2] bnx2x: Determine rx/tx queue sizes sooner Chas Williams
2016-03-04 22:28 ` Rasesh Mody [this message]
2016-03-10 10:32 ` Bruce Richardson
2016-02-08 13:51 ` [dpdk-dev] [PATCH 1/2] bnx2x: fix error handling in bnx2x_loop_obtain_resources() Bruce Richardson
2016-02-08 15:53 ` Charles (Chas) Williams
2016-02-08 16:03 ` Bruce Richardson
2016-03-04 22:28 ` Rasesh Mody
2016-03-10 10:33 ` Bruce Richardson
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=2552F74A0BCCBE4DBE2AD218C81B2811086B1539@avmb3.qlogic.org \
--to=rasesh.mody@qlogic.com \
--cc=3chas3@gmail.com \
--cc=bruce.richardson@intel.com \
--cc=ciwillia@brocade.com \
--cc=dev@dpdk.org \
/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).