[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <9ef62d97-e72b-4b66-b535-a1de8fe43d57@ti.com>
Date: Mon, 2 Sep 2024 11:05:45 +0530
From: "Anwar, Md Danish" <a0501179@...com>
To: Roger Quadros <rogerq@...nel.org>, MD Danish Anwar <danishanwar@...com>,
Andrew Lunn <andrew@...n.ch>, Dan Carpenter <dan.carpenter@...aro.org>,
Jan
Kiszka <jan.kiszka@...mens.com>,
Javier Carrasco
<javier.carrasco.cruz@...il.com>,
Jacob Keller <jacob.e.keller@...el.com>,
Diogo Ivo <diogo.ivo@...mens.com>, Simon Horman <horms@...nel.org>,
Richard
Cochran <richardcochran@...il.com>,
Paolo Abeni <pabeni@...hat.com>, Jakub
Kicinski <kuba@...nel.org>,
Eric Dumazet <edumazet@...gle.com>,
"David S.
Miller" <davem@...emloft.net>
CC: <linux-kernel@...r.kernel.org>, <netdev@...r.kernel.org>,
<linux-arm-kernel@...ts.infradead.org>, <srk@...com>,
Vignesh Raghavendra
<vigneshr@...com>
Subject: Re: [PATCH net-next v3 4/6] net: ti: icssg-prueth: Enable HSR Tx
Packet duplication offload
On 8/30/2024 7:00 PM, Roger Quadros wrote:
>
>
> On 28/08/2024 12:18, MD Danish Anwar wrote:
>> From: Ravi Gunasekaran <r-gunasekaran@...com>
>>
>> The HSR stack allows to offload its Tx packet duplication functionality to
>> the hardware. Enable this offloading feature for ICSSG driver
>>
>> Signed-off-by: Ravi Gunasekaran <r-gunasekaran@...com>
>> Signed-off-by: MD Danish Anwar <danishanwar@...com>
>> ---
>> drivers/net/ethernet/ti/icssg/icssg_common.c | 13 ++++++++++---
>> drivers/net/ethernet/ti/icssg/icssg_prueth.c | 5 +++--
>> drivers/net/ethernet/ti/icssg/icssg_prueth.h | 2 ++
>> 3 files changed, 15 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c
>> index b9d8a93d1680..2d6d8648f5a9 100644
>> --- a/drivers/net/ethernet/ti/icssg/icssg_common.c
>> +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c
>> @@ -660,14 +660,15 @@ enum netdev_tx icssg_ndo_start_xmit(struct sk_buff *skb, struct net_device *ndev
>> {
>> struct cppi5_host_desc_t *first_desc, *next_desc, *cur_desc;
>> struct prueth_emac *emac = netdev_priv(ndev);
>> + struct prueth *prueth = emac->prueth;
>> struct netdev_queue *netif_txq;
>> struct prueth_tx_chn *tx_chn;
>> dma_addr_t desc_dma, buf_dma;
>> + u32 pkt_len, dst_tag_id;
>> int i, ret = 0, q_idx;
>> bool in_tx_ts = 0;
>> int tx_ts_cookie;
>> void **swdata;
>> - u32 pkt_len;
>> u32 *epib;
>>
>> pkt_len = skb_headlen(skb);
>> @@ -712,9 +713,15 @@ enum netdev_tx icssg_ndo_start_xmit(struct sk_buff *skb, struct net_device *ndev
>>
>> /* set dst tag to indicate internal qid at the firmware which is at
>> * bit8..bit15. bit0..bit7 indicates port num for directed
>> - * packets in case of switch mode operation
>> + * packets in case of switch mode operation and port num 0
>> + * for undirected packets in case of HSR offload mode
>> */
>> - cppi5_desc_set_tags_ids(&first_desc->hdr, 0, (emac->port_id | (q_idx << 8)));
>> + dst_tag_id = emac->port_id | (q_idx << 8);
>> +
>> + if (prueth->is_hsr_offload_mode && (ndev->features & NETIF_F_HW_HSR_DUP))
>> + dst_tag_id = PRUETH_UNDIRECTED_PKT_DST_TAG;
>> +
>> + cppi5_desc_set_tags_ids(&first_desc->hdr, 0, dst_tag_id);
>> k3_udma_glue_tx_dma_to_cppi5_addr(tx_chn->tx_chn, &buf_dma);
>> cppi5_hdesc_attach_buf(first_desc, buf_dma, pkt_len, buf_dma, pkt_len);
>> swdata = cppi5_hdesc_get_swdata(first_desc);
>> diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.c b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
>> index f4fd346fe6f5..b60efe7bd7a7 100644
>> --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.c
>> +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.c
>> @@ -41,7 +41,8 @@
>> #define DEFAULT_PORT_MASK 1
>> #define DEFAULT_UNTAG_MASK 1
>>
>> -#define NETIF_PRUETH_HSR_OFFLOAD_FEATURES NETIF_F_HW_HSR_FWD
>> +#define NETIF_PRUETH_HSR_OFFLOAD_FEATURES (NETIF_F_HW_HSR_FWD | \
>> + NETIF_F_HW_HSR_DUP)
>
> You mentioned that these 2 features can't be enabled individually.
>
Not these two but NETIF_F_HW_HSR_TAG_INS and NETIF_F_HW_HSR_DUP needs to
be enabled together. NETIF_F_HW_HSR_TAG_INS is added in patch 6/6.
2) Inorder to enable hsr-tag-ins-offload, hsr-dup-offload
must also be enabled as these are tightly coupled in
the firmware implementation.
> So better to squash this with previous patch and use ndo_fix_features() to make sure both
> are set or cleared together.
>
I will squash patch 6/6 with this patch so that both are added in the
same patch. I will use ndo_fix_features() to make sure they re set /
unset together. I will add this also in this patch. Let me know if this
sounds good to you.
>>
>> /* CTRLMMR_ICSSG_RGMII_CTRL register bits */
>> #define ICSSG_CTRL_RGMII_ID_MODE BIT(24)
>> @@ -897,7 +898,7 @@ static int prueth_netdev_init(struct prueth *prueth,
>> ndev->ethtool_ops = &icssg_ethtool_ops;
>> ndev->hw_features = NETIF_F_SG;
>> ndev->features = ndev->hw_features;
>> - ndev->hw_features |= NETIF_F_HW_HSR_FWD;
>> + ndev->hw_features |= NETIF_PRUETH_HSR_OFFLOAD_FEATURES;
>>
>> netif_napi_add(ndev, &emac->napi_rx, icssg_napi_rx_poll);
>> hrtimer_init(&emac->rx_hrtimer, CLOCK_MONOTONIC,
>> diff --git a/drivers/net/ethernet/ti/icssg/icssg_prueth.h b/drivers/net/ethernet/ti/icssg/icssg_prueth.h
>> index a4b025fae797..e110a5f92684 100644
>> --- a/drivers/net/ethernet/ti/icssg/icssg_prueth.h
>> +++ b/drivers/net/ethernet/ti/icssg/icssg_prueth.h
>> @@ -59,6 +59,8 @@
>>
>> #define IEP_DEFAULT_CYCLE_TIME_NS 1000000 /* 1 ms */
>>
>> +#define PRUETH_UNDIRECTED_PKT_DST_TAG 0
>> +
>> /* Firmware status codes */
>> #define ICSS_HS_FW_READY 0x55555555
>> #define ICSS_HS_FW_DEAD 0xDEAD0000 /* lower 16 bits contain error code */
>
--
Thanks and Regards,
Md Danish Anwar
Powered by blists - more mailing lists