From: Vadim Fedorenko <vadim.fedorenko@linux.dev>
To: Fan Gong <gongfan1@huawei.com>, Zhu Yikai <zhuyikai1@h-partners.com>
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
"David S. Miller" <davem@davemloft.net>,
Eric Dumazet <edumazet@google.com>,
Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
Simon Horman <horms@kernel.org>,
Andrew Lunn <andrew+netdev@lunn.ch>,
linux-doc@vger.kernel.org, Jonathan Corbet <corbet@lwn.net>,
Bjorn Helgaas <helgaas@kernel.org>, luosifu <luosifu@huawei.com>,
Xin Guo <guoxin09@huawei.com>,
Shen Chenyang <shenchenyang1@hisilicon.com>,
Zhou Shuai <zhoushuai28@huawei.com>, Wu Like <wulike1@huawei.com>,
Shi Jing <shijing34@huawei.com>,
Meny Yossefi <meny.yossefi@huawei.com>,
Gur Stavi <gur.stavi@huawei.com>, Lee Trager <lee@trager.us>,
Michael Ellerman <mpe@ellerman.id.au>,
Suman Ghosh <sumang@marvell.com>,
Przemek Kitszel <przemyslaw.kitszel@intel.com>,
Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Subject: Re: [PATCH net-next v01 10/12] hinic3: Add Rss function
Date: Tue, 26 Aug 2025 18:06:42 +0100 [thread overview]
Message-ID: <53206f29-7da8-4145-aef0-7bdacef3bb55@linux.dev> (raw)
In-Reply-To: <13ffd1d836eb7aa6563ad93bf5fa5196afdf0053.1756195078.git.zhuyikai1@h-partners.com>
On 26/08/2025 10:05, Fan Gong wrote:
> Initialize rss functions. Configure rss hash data and HW resources.
>
> Co-developed-by: Xin Guo <guoxin09@huawei.com>
> Signed-off-by: Xin Guo <guoxin09@huawei.com>
> Co-developed-by: Zhu Yikai <zhuyikai1@h-partners.com>
> Signed-off-by: Zhu Yikai <zhuyikai1@h-partners.com>
> Signed-off-by: Fan Gong <gongfan1@huawei.com>
> ---
> drivers/net/ethernet/huawei/hinic3/Makefile | 1 +
> .../net/ethernet/huawei/hinic3/hinic3_main.c | 9 +-
> .../huawei/hinic3/hinic3_mgmt_interface.h | 55 +++
> .../huawei/hinic3/hinic3_netdev_ops.c | 18 +
> .../ethernet/huawei/hinic3/hinic3_nic_dev.h | 5 +
> .../net/ethernet/huawei/hinic3/hinic3_rss.c | 359 ++++++++++++++++++
> .../net/ethernet/huawei/hinic3/hinic3_rss.h | 14 +
> 7 files changed, 460 insertions(+), 1 deletion(-)
[...]
> +static int alloc_rss_resource(struct net_device *netdev)
> +{
> + struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
> + static const u8 default_rss_key[L2NIC_RSS_KEY_SIZE] = {
> + 0x6d, 0x5a, 0x56, 0xda, 0x25, 0x5b, 0x0e, 0xc2,
> + 0x41, 0x67, 0x25, 0x3d, 0x43, 0xa3, 0x8f, 0xb0,
> + 0xd0, 0xca, 0x2b, 0xcb, 0xae, 0x7b, 0x30, 0xb4,
> + 0x77, 0xcb, 0x2d, 0xa3, 0x80, 0x30, 0xf2, 0x0c,
> + 0x6a, 0x42, 0xb7, 0x3b, 0xbe, 0xac, 0x01, 0xfa};
> +
> + nic_dev->rss_hkey = kzalloc(L2NIC_RSS_KEY_SIZE, GFP_KERNEL);
no need to request zero'ed allocation if you are going to overwrite it
completely on the very next line.
> + if (!nic_dev->rss_hkey)
> + return -ENOMEM;
> +
> + memcpy(nic_dev->rss_hkey, default_rss_key, L2NIC_RSS_KEY_SIZE);
I would better move this line after both allocations when the code flow
has no way to fail.
> + nic_dev->rss_indir = kcalloc(L2NIC_RSS_INDIR_SIZE, sizeof(u32),
> + GFP_KERNEL);
why do you allocate L2NIC_RSS_INDIR_SIZE of u32 when the HW table has
le16 type for the entry?
> + if (!nic_dev->rss_indir) {
> + kfree(nic_dev->rss_hkey);
> + nic_dev->rss_hkey = NULL;
> + return -ENOMEM;
> + }
> +
> + return 0;
> +}
> +
> +static int hinic3_rss_set_indir_tbl(struct hinic3_hwdev *hwdev,
> + const u32 *indir_table)
> +{
> + struct l2nic_cmd_rss_set_indir_tbl *indir_tbl;
> + struct hinic3_cmd_buf *cmd_buf;
> + __le64 out_param;
> + int err;
> + u32 i;
> +
> + cmd_buf = hinic3_alloc_cmd_buf(hwdev);
> + if (!cmd_buf) {
> + dev_err(hwdev->dev, "Failed to allocate cmd buf\n");
> + return -ENOMEM;
> + }
> +
> + cmd_buf->size = cpu_to_le16(sizeof(struct l2nic_cmd_rss_set_indir_tbl));
> + indir_tbl = cmd_buf->buf;
> + memset(indir_tbl, 0, sizeof(*indir_tbl));
> +
> + for (i = 0; i < L2NIC_RSS_INDIR_SIZE; i++)
> + indir_tbl->entry[i] = cpu_to_le16((u16)indir_table[i]);
> +
> + hinic3_cmdq_buf_swab32(indir_tbl, sizeof(*indir_tbl));
> +
> + err = hinic3_cmdq_direct_resp(hwdev, MGMT_MOD_L2NIC,
> + L2NIC_UCODE_CMD_SET_RSS_INDIR_TBL,
> + cmd_buf, &out_param);
> + if (err || out_param != 0) {
no need for "!= 0"
> + dev_err(hwdev->dev, "Failed to set rss indir table\n");
> + err = -EFAULT;
> + }
> +
> + hinic3_free_cmd_buf(hwdev, cmd_buf);
> +
> + return err;
> +}
[...]
> +static int hinic3_rss_cfg_hash_key(struct hinic3_hwdev *hwdev, u8 opcode,
> + u8 *key)
> +{
> + struct l2nic_cmd_cfg_rss_hash_key hash_key = {};
> + struct mgmt_msg_params msg_params = {};
> + int err;
> +
> + hash_key.func_id = hinic3_global_func_id(hwdev);
> + hash_key.opcode = opcode;
> +
> + if (opcode == MGMT_MSG_CMD_OP_SET)
> + memcpy(hash_key.key, key, L2NIC_RSS_KEY_SIZE);
here you copy hash key to a stack allocated structure ...
> +
> + mgmt_msg_params_init_default(&msg_params, &hash_key, sizeof(hash_key));
... which is copied to another stack allocated structure ...
> +
> + err = hinic3_send_mbox_to_mgmt(hwdev, MGMT_MOD_L2NIC,
> + L2NIC_CMD_CFG_RSS_HASH_KEY, &msg_params);
> + if (err || hash_key.msg_head.status) {
> + dev_err(hwdev->dev, "Failed to %s hash key, err: %d, status: 0x%x\n",
> + opcode == MGMT_MSG_CMD_OP_SET ? "set" : "get",
> + err, hash_key.msg_head.status);
> + return -EINVAL;
> + }
> +
> + if (opcode == MGMT_MSG_CMD_OP_GET)
> + memcpy(key, hash_key.key, L2NIC_RSS_KEY_SIZE);
> +
> + return 0;
> +}
> +
> +static int hinic3_rss_set_hash_key(struct hinic3_hwdev *hwdev, const u8 *key)
> +{
> + u8 hash_key[L2NIC_RSS_KEY_SIZE];
> +
> + memcpy(hash_key, key, L2NIC_RSS_KEY_SIZE);
... but it was already copied to stack allocated buffer ...
> +
> + return hinic3_rss_cfg_hash_key(hwdev, MGMT_MSG_CMD_OP_SET, hash_key);
> +}
> +
> +static int hinic3_set_hw_rss_parameters(struct net_device *netdev, u8 rss_en)
> +{
> + struct hinic3_nic_dev *nic_dev = netdev_priv(netdev);
> + int err;
> +
> + err = hinic3_rss_set_hash_key(nic_dev->hwdev, nic_dev->rss_hkey);
... which is previously copied from static array.
It's 4 copies in total to configure one simple thing. Looks like too
much of copying with no good reason
> + if (err)
> + return err;
> +
next prev parent reply other threads:[~2025-08-26 17:06 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-08-26 9:05 [PATCH net-next v01 00/12] net: hinic3: Add a driver for Huawei 3rd gen NIC - sw and hw initialization Fan Gong
2025-08-26 9:05 ` [PATCH net-next v01 01/12] hinic3: HW initialization Fan Gong
2025-08-26 11:37 ` Vadim Fedorenko
2025-08-26 19:52 ` ALOK TIWARI
2025-08-26 9:05 ` [PATCH net-next v01 02/12] hinic3: HW management interfaces Fan Gong
2025-08-26 9:05 ` [PATCH net-next v01 03/12] hinic3: HW common function initialization Fan Gong
2025-08-26 9:05 ` [PATCH net-next v01 04/12] hinic3: HW capability initialization Fan Gong
2025-08-26 14:17 ` Vadim Fedorenko
2025-08-26 9:05 ` [PATCH net-next v01 05/12] hinic3: Command Queue flush interfaces Fan Gong
2025-08-26 9:05 ` [PATCH net-next v01 06/12] hinic3: Nic_io initialization Fan Gong
2025-08-26 15:49 ` Vadim Fedorenko
2025-08-26 9:05 ` [PATCH net-next v01 07/12] hinic3: Queue pair resource initialization Fan Gong
2025-08-26 16:08 ` Vadim Fedorenko
2025-08-26 20:05 ` ALOK TIWARI
2025-08-26 9:05 ` [PATCH net-next v01 08/12] hinic3: Queue pair context initialization Fan Gong
2025-08-26 16:42 ` Vadim Fedorenko
2025-08-26 9:05 ` [PATCH net-next v01 09/12] hinic3: Tx & Rx configuration Fan Gong
2025-08-26 9:05 ` [PATCH net-next v01 10/12] hinic3: Add Rss function Fan Gong
2025-08-26 17:06 ` Vadim Fedorenko [this message]
2025-08-26 17:49 ` Jakub Kicinski
2025-08-26 17:54 ` Eric Dumazet
2025-08-27 9:50 ` Fan Gong
2025-08-26 9:05 ` [PATCH net-next v01 11/12] hinic3: Add port management Fan Gong
2025-08-26 9:05 ` [PATCH net-next v01 12/12] hinic3: Fix missing napi->dev in netif_queue_set_napi Fan Gong
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=53206f29-7da8-4145-aef0-7bdacef3bb55@linux.dev \
--to=vadim.fedorenko@linux.dev \
--cc=andrew+netdev@lunn.ch \
--cc=christophe.jaillet@wanadoo.fr \
--cc=corbet@lwn.net \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=gongfan1@huawei.com \
--cc=guoxin09@huawei.com \
--cc=gur.stavi@huawei.com \
--cc=helgaas@kernel.org \
--cc=horms@kernel.org \
--cc=kuba@kernel.org \
--cc=lee@trager.us \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luosifu@huawei.com \
--cc=meny.yossefi@huawei.com \
--cc=mpe@ellerman.id.au \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=przemyslaw.kitszel@intel.com \
--cc=shenchenyang1@hisilicon.com \
--cc=shijing34@huawei.com \
--cc=sumang@marvell.com \
--cc=wulike1@huawei.com \
--cc=zhoushuai28@huawei.com \
--cc=zhuyikai1@h-partners.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).