* [PATCH v4 18/23] net/sxe2: support SFP module info and EEPROM access
From: liujie5 @ 2026-06-19 8:11 UTC (permalink / raw)
To: stephen; +Cc: dev, Jie Liu
In-Reply-To: <20260618082723.571054-21-liujie5@linkdatatechnology.com>
From: Jie Liu <liujie5@linkdatatechnology.com>
This patch implements 'get_module_info' and 'get_module_eeprom'
ops for the sxe2 PMD. These interfaces allow applications to retrieve
the type of the plugged-in optical module and read its internal
EEPROM data.
The implementation utilizes the shared SFP header definitions to
parse the module ID, connector type, and encoding. It supports
reading the standard 256-byte EEPROM maps (SFF-8472 for SFP and
SFF-8636 for QSFP) via hardware-specific access commands.
Key features:
- Identify module types (SFP/SFP+/QSFP/QSFP28).
- Support standard EEPROM data retrieval for diagnostic tools.
- Add boundary checks to ensure safe I2C memory access.
Signed-off-by: Jie Liu <liujie5@linkdatatechnology.com>
---
drivers/net/sxe2/sxe2_cmd_chnl.c | 46 +++++
drivers/net/sxe2/sxe2_cmd_chnl.h | 3 +
drivers/net/sxe2/sxe2_drv_cmd.h | 18 ++
drivers/net/sxe2/sxe2_ethdev.c | 298 +++++++++++++++++++++++++++++++
drivers/net/sxe2/sxe2_ethdev.h | 9 +
5 files changed, 374 insertions(+)
diff --git a/drivers/net/sxe2/sxe2_cmd_chnl.c b/drivers/net/sxe2/sxe2_cmd_chnl.c
index 926eaee062..43e8c59487 100644
--- a/drivers/net/sxe2/sxe2_cmd_chnl.c
+++ b/drivers/net/sxe2/sxe2_cmd_chnl.c
@@ -1833,3 +1833,49 @@ int32_t sxe2_drv_srcvsi_prune_config(struct sxe2_adapter *adapter,
return ret;
}
+
+int32_t sxe2_drv_sfp_eeprom_read(struct sxe2_adapter *adapter, struct sxe2_sfp_read_info *sfp_info)
+{
+ int32_t ret = -1;
+ struct sxe2_drv_sfp_req req = {0};
+ struct sxe2_drv_sfp_resp *resp = NULL;
+ struct sxe2_drv_cmd_params cmd = {0};
+
+ resp = rte_zmalloc("read sfp data", sizeof(*resp) + sfp_info->len, 0);
+ if (!resp) {
+ PMD_LOG_ERR(DRV, "Alloc memory failed");
+ ret = -ENOMEM;
+ goto l_end;
+ }
+
+ req.is_wr = false;
+ req.is_qsfp = sfp_info->is_qsfp;
+ req.page_cnt = rte_cpu_to_le_16(sfp_info->page_cnt);
+ req.offset = rte_cpu_to_le_16(sfp_info->offset);
+ req.data_len = rte_cpu_to_le_16(sfp_info->len);
+ req.bus_addr = rte_cpu_to_le_16(sfp_info->bus_addr);
+
+ PMD_DEV_LOG_INFO(adapter, DRV, "is_qsfp=%u, page_cnt=%u, offset=%u, datalen=%u, "
+ "bus_addr=%u", sfp_info->is_qsfp, sfp_info->page_cnt, sfp_info->offset,
+ sfp_info->len, sfp_info->bus_addr);
+
+ sxe2_drv_cmd_params_fill(adapter, &cmd, SXE2_DRV_CMD_OPT_EEP_GET,
+ &req, sizeof(req),
+ resp, sizeof(*resp) + sfp_info->len);
+ ret = sxe2_drv_cmd_exec(adapter->cdev, &cmd);
+ if (ret) {
+ PMD_DEV_LOG_ERR(adapter, DRV, "Failed to read sfp, ret=%d", ret);
+ goto l_end;
+ }
+
+ ret = 0;
+ rte_memcpy(sfp_info->data, resp->data, sfp_info->len);
+
+l_end:
+ if (resp) {
+ rte_free(resp);
+ resp = NULL;
+ }
+
+ return ret;
+}
diff --git a/drivers/net/sxe2/sxe2_cmd_chnl.h b/drivers/net/sxe2/sxe2_cmd_chnl.h
index 97007c7cfa..988d4b458b 100644
--- a/drivers/net/sxe2/sxe2_cmd_chnl.h
+++ b/drivers/net/sxe2/sxe2_cmd_chnl.h
@@ -167,4 +167,7 @@ int32_t sxe2_drv_flow_fnav_query_stat(struct sxe2_adapter *adapter,
int32_t sxe2_drv_srcvsi_prune_config(struct sxe2_adapter *adapter,
uint16_t *vsi_list, uint16_t vsi_cnt, bool set);
+int32_t sxe2_drv_sfp_eeprom_read(struct sxe2_adapter *adapter,
+ struct sxe2_sfp_read_info *sfp_info);
+
#endif /* SXE2_CMD_CHNL_H */
diff --git a/drivers/net/sxe2/sxe2_drv_cmd.h b/drivers/net/sxe2/sxe2_drv_cmd.h
index 3e0b70ab02..3fabf351af 100644
--- a/drivers/net/sxe2/sxe2_drv_cmd.h
+++ b/drivers/net/sxe2/sxe2_drv_cmd.h
@@ -633,6 +633,24 @@ struct __rte_aligned(4) __rte_packed_begin sxe2_drv_udp_tunnel_resp {
uint8_t rsv;
} __rte_packed_end;
+struct __rte_aligned(4) __rte_packed_begin sxe2_drv_sfp_req {
+ uint8_t is_wr;
+ uint8_t is_qsfp;
+ uint16_t bus_addr;
+ uint16_t page_cnt;
+ uint16_t offset;
+ uint16_t data_len;
+ uint16_t rvd;
+ uint8_t data[];
+} __rte_packed_end;
+
+struct __rte_aligned(4) __rte_packed_begin sxe2_drv_sfp_resp {
+ uint8_t is_wr;
+ uint8_t is_qsfp;
+ uint16_t data_len;
+ uint8_t data[];
+} __rte_packed_end;
+
enum sxe2_drv_cmd_module {
SXE2_DRV_CMD_MODULE_HANDSHAKE = 0,
SXE2_DRV_CMD_MODULE_DEV = 1,
diff --git a/drivers/net/sxe2/sxe2_ethdev.c b/drivers/net/sxe2/sxe2_ethdev.c
index 493b720f0b..cadb926e5d 100644
--- a/drivers/net/sxe2/sxe2_ethdev.c
+++ b/drivers/net/sxe2/sxe2_ethdev.c
@@ -40,6 +40,7 @@
#include "sxe2_ethdev_repr.h"
#include "sxe2vf_regs.h"
#include "sxe2_switchdev.h"
+#include "sxe2_msg.h"
#define SXE2_PCI_VENDOR_ID_1 0x1ff2
#define SXE2_PCI_DEVICE_ID_PF_1 0x10b1
@@ -123,6 +124,10 @@ static int32_t sxe2_udp_tunnel_port_del(struct rte_eth_dev *dev,
struct rte_eth_udp_tunnel *tunnel_udp);
static int32_t sxe2_fw_version_string_get(struct rte_eth_dev *dev,
char *fw_version, size_t fw_size);
+static int32_t sxe2_get_module_info(struct rte_eth_dev *dev,
+ struct rte_eth_dev_module_info *info);
+static int32_t sxe2_get_module_eeprom(struct rte_eth_dev *dev,
+ struct rte_dev_eeprom_info *info);
static const struct eth_dev_ops sxe2_eth_dev_ops = {
.dev_configure = sxe2_dev_configure,
@@ -187,6 +192,9 @@ static const struct eth_dev_ops sxe2_eth_dev_ops = {
.fw_version_get = sxe2_fw_version_string_get,
.get_monitor_addr = sxe2_get_monitor_addr,
+
+ .get_module_info = sxe2_get_module_info,
+ .get_module_eeprom = sxe2_get_module_eeprom,
};
static int32_t sxe2_dev_configure(struct rte_eth_dev *dev)
@@ -292,6 +300,296 @@ static int32_t sxe2_dev_start(struct rte_eth_dev *dev)
return ret;
}
+static int32_t sxe2_sfp_type_get(struct sxe2_adapter *adapter, uint8_t *type)
+{
+ int32_t ret = -1;
+ struct sxe2_sfp_read_info sfp_info;
+
+ memset(&sfp_info, 0, sizeof(sfp_info));
+ sfp_info.bus_addr = SXE2_SFP_E2P_I2C_7BIT_ADDR0;
+ sfp_info.len = 1;
+ sfp_info.data = type;
+ sfp_info.offset = 0;
+ sfp_info.page_cnt = 0;
+ sfp_info.is_qsfp = false;
+
+ ret = sxe2_drv_sfp_eeprom_read(adapter, &sfp_info);
+ if (ret)
+ goto l_end;
+
+ ret = 0;
+ PMD_LOG_INFO(DRV, "Get sfp type success, type=%u", *type);
+
+l_end:
+ return ret;
+}
+
+static int32_t sxe2_sfp_module_info_get(struct sxe2_adapter *adapter,
+ struct rte_eth_dev_module_info *info)
+{
+ int32_t ret = -1;
+ bool page_swap = false;
+ uint8_t sff8472_rev = 0;
+ uint8_t addr_mode = 0;
+ struct sxe2_sfp_read_info sfp_info;
+
+ memset(&sfp_info, 0, sizeof(sfp_info));
+ sfp_info.bus_addr = SXE2_SFP_E2P_I2C_7BIT_ADDR0;
+ sfp_info.is_qsfp = false;
+ sfp_info.len = 1;
+ sfp_info.data = &sff8472_rev;
+ sfp_info.offset = SXE2_MODULE_SFF_8472_COMP;
+ sfp_info.page_cnt = 0;
+
+ ret = sxe2_drv_sfp_eeprom_read(adapter, &sfp_info);
+ if (ret) {
+ ret = -EIO;
+ PMD_LOG_ERR(DRV, "Failed to read 8472 protocol, ret=%d", ret);
+ goto l_end;
+ }
+
+ sfp_info.data = &addr_mode;
+ sfp_info.offset = SXE2_MODULE_SFF_8472_SWAP;
+
+ ret = sxe2_drv_sfp_eeprom_read(adapter, &sfp_info);
+ if (ret) {
+ ret = -EIO;
+ PMD_LOG_ERR(DRV, "Failed to read A2 page, ret=%d", ret);
+ goto l_end;
+ }
+
+ if (addr_mode & SXE2_MODULE_SFF_ADDR_MODE) {
+ PMD_LOG_ERR(DRV, "address change required to access page 0xA2, "
+ "but not supported. please report the module "
+ "type to the driver maintainers.");
+ page_swap = true;
+ }
+
+ PMD_LOG_INFO(DRV, "Read sfp module info, sff_8472=%u, a2_page=%u, swap_page=%d",
+ sff8472_rev, addr_mode, page_swap);
+
+ if (sff8472_rev == SXE2_MODULE_SFF_8472_UNSUP ||
+ page_swap ||
+ !(addr_mode & SXE2_MODULE_SFF_DDM_IMPLEMENTED)) {
+ info->type = SXE2_MODULE_SFF_8079;
+ info->eeprom_len = SXE2_MODULE_SFF_8079_LEN;
+ } else {
+ info->type = SXE2_MODULE_SFF_8472;
+ info->eeprom_len = SXE2_MODULE_SFF_8472_LEN;
+ }
+
+ ret = 0;
+
+l_end:
+ return ret;
+}
+
+static int32_t
+sxe2_qsfp_module_info_get(struct sxe2_adapter *adapter, struct rte_eth_dev_module_info *info)
+{
+ int32_t ret = -1;
+ uint8_t sff8636_rev = 0;
+ struct sxe2_sfp_read_info sfp_info;
+
+ memset(&sfp_info, 0, sizeof(sfp_info));
+ sfp_info.bus_addr = SXE2_SFP_E2P_I2C_7BIT_ADDR0;
+ sfp_info.is_qsfp = true;
+ sfp_info.len = 1;
+ sfp_info.data = &sff8636_rev;
+ sfp_info.offset = SXE2_MODULE_REVISION_ADDR;
+ sfp_info.page_cnt = 0;
+
+ ret = sxe2_drv_sfp_eeprom_read(adapter, &sfp_info);
+ if (ret) {
+ ret = -EIO;
+ PMD_LOG_ERR(DRV, "Failed to read 8636 protocol, ret=%d", ret);
+ goto l_end;
+ }
+
+ if (sff8636_rev > 0x02) {
+ info->type = SXE2_MODULE_SFF_8636;
+ info->eeprom_len = SXE2_MODULE_SFF_8636_MAX_LEN;
+ } else {
+ info->type = SXE2_MODULE_SFF_8436;
+ info->eeprom_len = SXE2_MODULE_SFF_8436_MAX_LEN;
+ }
+
+l_end:
+ return ret;
+}
+
+static int32_t
+sxe2_get_module_info(struct rte_eth_dev *dev, struct rte_eth_dev_module_info *info)
+{
+ int32_t ret = -1;
+ uint8_t type = 0;
+ struct sxe2_adapter *adapter = dev->data->dev_private;
+
+ ret = sxe2_sfp_type_get(adapter, &type);
+ if (ret) {
+ ret = -EIO;
+ PMD_LOG_ERR(DRV, "Failed to read sfp type, ret=%d", ret);
+ goto l_end;
+ }
+
+ switch (type) {
+ case SXE2_MODULE_SFF_SFP_TYPE:
+ ret = sxe2_sfp_module_info_get(adapter, info);
+ if (ret)
+ goto l_end;
+ break;
+ case SXE2_MODULE_TYPE_QSFP_PLUS:
+ case SXE2_MODULE_TYPE_QSFP28:
+ ret = sxe2_qsfp_module_info_get(adapter, info);
+ if (ret)
+ goto l_end;
+ break;
+ default:
+ ret = -ENXIO;
+ PMD_LOG_ERR(DRV, "Invalid sfp type, type=%d.", type);
+ goto l_end;
+ }
+
+ PMD_LOG_INFO(DRV, "sfp eeprom type=%x, eeprom len=%d.", info->type, info->eeprom_len);
+
+l_end:
+ return ret;
+}
+
+static int32_t
+sxe2_get_sfp_eeprom(struct sxe2_adapter *adapter, struct sxe2_sfp_read_info *sfp_info)
+{
+ int32_t ret = -1;
+ uint16_t ori_len = sfp_info->len;
+ uint16_t ori_offset = sfp_info->offset;
+
+ if ((ori_len + ori_offset) > SXE2_SFP_EEP_LEN_MAX) {
+ sfp_info->len = (uint16_t)(SXE2_SFP_EEP_LEN_MAX - ori_offset);
+ ret = sxe2_drv_sfp_eeprom_read(adapter, sfp_info);
+ if (ret)
+ goto l_end;
+ sfp_info->bus_addr = SXE2_SFP_E2P_I2C_7BIT_ADDR1;
+ sfp_info->len = (uint16_t)(ori_len - (SXE2_SFP_EEP_LEN_MAX - ori_offset));
+ sfp_info->data = (uint8_t *)(sfp_info->data) + (SXE2_SFP_EEP_LEN_MAX - ori_offset);
+ sfp_info->offset = 0;
+ sfp_info->page_cnt = 0;
+ ret = sxe2_drv_sfp_eeprom_read(adapter, sfp_info);
+ } else {
+ ret = sxe2_drv_sfp_eeprom_read(adapter, sfp_info);
+ }
+
+l_end:
+ if (ret)
+ PMD_LOG_ERR(DRV, "Failed to read sfp.");
+ return ret;
+}
+
+static int32_t
+sxe2_get_qsfp_eeprom(struct sxe2_adapter *adapter,
+ struct sxe2_sfp_read_info *sfp_info)
+{
+ int32_t ret = -1;
+ uint16_t ori_len = sfp_info->len;
+ uint16_t ori_offset = sfp_info->offset;
+ uint16_t read_len = 0;
+ uint16_t remain_len = 0;
+
+ if ((ori_len + ori_offset) > SXE2_SFP_EEP_LEN_MAX) {
+ sfp_info->len = (uint16_t)(SXE2_SFP_EEP_LEN_MAX - ori_offset);
+ ret = sxe2_drv_sfp_eeprom_read(adapter, sfp_info);
+ if (ret)
+ goto l_end;
+
+ do {
+ read_len = read_len + sfp_info->len;
+ sfp_info->data = (uint8_t *)(sfp_info->data) + sfp_info->len;
+ sfp_info->offset = SXE2_QSFP_PAGE_OFST_START;
+ sfp_info->page_cnt++;
+ remain_len = (uint16_t)(ori_len - read_len);
+ sfp_info->len = (remain_len > SXE2_QSFP_PAGE_OFST_START) ?
+ SXE2_QSFP_PAGE_OFST_START : remain_len;
+ ret = sxe2_drv_sfp_eeprom_read(adapter, sfp_info);
+ if (ret)
+ goto l_end;
+ } while (remain_len > SXE2_QSFP_PAGE_OFST_START);
+ } else {
+ ret = sxe2_drv_sfp_eeprom_read(adapter, sfp_info);
+ }
+
+l_end:
+ if (ret)
+ PMD_LOG_ERR(DRV, "Failed to read sfp.");
+ return ret;
+}
+
+static int32_t
+sxe2_get_module_eeprom(struct rte_eth_dev *dev, struct rte_dev_eeprom_info *info)
+{
+ int32_t ret = -1;
+ uint8_t type = 0;
+ struct sxe2_adapter *adapter = dev->data->dev_private;
+ struct sxe2_sfp_read_info sfp_info;
+
+ memset(&sfp_info, 0, sizeof(sfp_info));
+
+ if (!info || !info->length || !info->data ||
+ info->offset >= SXE2_SFP_EEP_LEN_MAX) {
+ ret = -EINVAL;
+ goto l_end;
+ }
+
+ PMD_LOG_INFO(DRV, "Dump sfp eeprom info offset=0x%x, len=0x%x.",
+ info->offset, info->length);
+
+ ret = sxe2_sfp_type_get(adapter, &type);
+ if (ret) {
+ ret = -EIO;
+ PMD_LOG_ERR(DRV, "Failed to read sfp type, ret=%d", ret);
+ goto l_end;
+ }
+
+ sfp_info.bus_addr = SXE2_SFP_E2P_I2C_7BIT_ADDR0;
+ sfp_info.len = info->length;
+ sfp_info.data = info->data;
+ sfp_info.offset = info->offset;
+ sfp_info.page_cnt = 0;
+
+ switch (type) {
+ case SXE2_MODULE_SFF_SFP_TYPE:
+ if (info->length > SXE2_SFP_EEP_LEN_MAX * 2) {
+ ret = -EINVAL;
+ PMD_LOG_ERR(DRV, "sfp read size[%u] > eeprom max size[%d], ret=%d",
+ info->length, SXE2_SFP_EEP_LEN_MAX * 2, ret);
+ goto l_end;
+ }
+ sfp_info.is_qsfp = false;
+ ret = sxe2_get_sfp_eeprom(adapter, &sfp_info);
+ if (ret)
+ goto l_end;
+ break;
+ case SXE2_MODULE_TYPE_QSFP_PLUS:
+ case SXE2_MODULE_TYPE_QSFP28:
+ if (info->length > SXE2_MODULE_SFF_8636_MAX_LEN) {
+ ret = -EINVAL;
+ PMD_LOG_ERR(DRV, "sfp read size[%u] > eeprom max size[%d], ret=%d",
+ info->length, SXE2_SFP_EEP_LEN_MAX * 2, ret);
+ goto l_end;
+ }
+ sfp_info.is_qsfp = true;
+ ret = sxe2_get_qsfp_eeprom(adapter, &sfp_info);
+ if (ret)
+ goto l_end;
+ break;
+ default:
+ ret = -ENXIO;
+ PMD_LOG_ERR(DRV, "Invalid sfp type, type=%d.", type);
+ goto l_end;
+ }
+
+l_end:
+ return ret;
+}
+
static enum sxe2_udp_tunnel_protocol
sxe2_udp_tunnel_type_rte_to_sxe2(enum rte_eth_tunnel_type rte_type)
{
diff --git a/drivers/net/sxe2/sxe2_ethdev.h b/drivers/net/sxe2/sxe2_ethdev.h
index e2e002bdb2..b63238f423 100644
--- a/drivers/net/sxe2/sxe2_ethdev.h
+++ b/drivers/net/sxe2/sxe2_ethdev.h
@@ -274,6 +274,15 @@ struct sxe2_sched_hw_cap {
uint8_t adj_lvl;
};
+struct sxe2_sfp_read_info {
+ uint8_t *data;
+ uint16_t offset;
+ uint16_t len;
+ uint16_t bus_addr;
+ uint16_t page_cnt;
+ bool is_qsfp;
+};
+
struct sxe2_link_context {
rte_spinlock_t link_lock;
bool link_up;
--
2.52.0
^ permalink raw reply related
* [PATCH v4 19/23] net/sxe2: implement private dump info
From: liujie5 @ 2026-06-19 8:11 UTC (permalink / raw)
To: stephen; +Cc: dev, Jie Liu
In-Reply-To: <20260618082723.571054-21-liujie5@linkdatatechnology.com>
From: Jie Liu <liujie5@linkdatatechnology.com>
This patch implements the 'eth_dev_priv_dump' ops for the sxe2 PMD.
This interface allows applications to dump driver-specific internal
state and configuration information to a file stream.
The output includes:
- capabilities.
- device base info.
- device args info.
- device filter info.
- reprenstor info.
Signed-off-by: Jie Liu <liujie5@linkdatatechnology.com>
---
drivers/net/sxe2/meson.build | 1 +
drivers/net/sxe2/sxe2_dump.c | 287 ++++++++++++++++++++++++++++
drivers/net/sxe2/sxe2_dump.h | 12 ++
drivers/net/sxe2/sxe2_ethdev.c | 3 +
drivers/net/sxe2/sxe2_ethdev_repr.c | 3 +
5 files changed, 306 insertions(+)
create mode 100644 drivers/net/sxe2/sxe2_dump.c
create mode 100644 drivers/net/sxe2/sxe2_dump.h
diff --git a/drivers/net/sxe2/meson.build b/drivers/net/sxe2/meson.build
index 65286299aa..d653d071a9 100644
--- a/drivers/net/sxe2/meson.build
+++ b/drivers/net/sxe2/meson.build
@@ -77,4 +77,5 @@ sources += files(
'sxe2_flow_parse_action.c',
'sxe2_flow_parse_pattern.c',
'sxe2_flow_parse_engine.c',
+ 'sxe2_dump.c',
)
diff --git a/drivers/net/sxe2/sxe2_dump.c b/drivers/net/sxe2/sxe2_dump.c
new file mode 100644
index 0000000000..d43473e083
--- /dev/null
+++ b/drivers/net/sxe2/sxe2_dump.c
@@ -0,0 +1,287 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright (C), 2025, Wuxi Stars Micro System Technologies Co., Ltd.
+ */
+
+#include <rte_malloc.h>
+#include <arpa/inet.h>
+
+#include "sxe2_common_log.h"
+#include "sxe2_ethdev.h"
+#include "sxe2_dump.h"
+#include "sxe2_stats.h"
+
+static void
+sxe2_dump_dev_feature_capability(FILE *file, struct rte_eth_dev *dev)
+{
+ uint32_t i;
+ struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+ const struct {
+ uint32_t cap_bit;
+ const char *name;
+ } caps_name[] = {
+ {SXE2_DEV_CAPS_OFFLOAD_L2, "L2"},
+ {SXE2_DEV_CAPS_OFFLOAD_VLAN, "VLAN"},
+ {SXE2_DEV_CAPS_OFFLOAD_IPSEC, "IPSEC"},
+ {SXE2_DEV_CAPS_OFFLOAD_RSS, "RSS"},
+ {SXE2_DEV_CAPS_OFFLOAD_FNAV, "FNAV"},
+ {SXE2_DEV_CAPS_OFFLOAD_TM, "TM"},
+ {SXE2_DEV_CAPS_OFFLOAD_PTP, "PTP"},
+ };
+ if (adapter->is_dev_repr)
+ goto l_end;
+
+ fprintf(file, " - Dev Capability:\n");
+ for (i = 0; i < RTE_DIM(caps_name); i++) {
+ fprintf(file, "\t -- support %s: %s\n", caps_name[i].name,
+ (adapter->cap_flags & caps_name[i].cap_bit) ? "Yes" :
+ "No");
+ }
+l_end:
+ return;
+}
+
+static void
+sxe2_dump_device_basic_info(FILE *file, struct rte_eth_dev *dev)
+{
+ struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+
+ fprintf(file,
+ " - Device Base Info:\n"
+ "\t -- name: %s\n"
+ "\t -- pf_idx: %u port_idx: %u\n"
+ "\t -- tx_mode_flags: 0x%x rx_mode_flags: 0x%x\n"
+ "\t -- flow_isolate_cfg: 0x%x flow_isolated: 0x%x\n"
+ "\t -- dev_type: 0x%x is_switchdev: 0x%x\n"
+ "\t -- is_dev_repr: 0x%x dev_port_id: 0x%x\n"
+ "\t -- dev_flags: 0x%x\n"
+ "\t -- intr_conf lsc: %u rxq: %u rmv: %u\n",
+ dev->data->name,
+ adapter->pf_idx, adapter->port_idx,
+ adapter->tx_mode_flags, adapter->rx_mode_flags,
+ adapter->flow_isolate_cfg, adapter->flow_isolated,
+ adapter->dev_type, adapter->switchdev_info.is_switchdev,
+ adapter->is_dev_repr, adapter->dev_port_id,
+ dev->data->dev_flags,
+ dev->data->dev_conf.intr_conf.lsc,
+ dev->data->dev_conf.intr_conf.rxq,
+ dev->data->dev_conf.intr_conf.rmv);
+}
+
+static void
+sxe2_dump_dev_args_info(FILE *file, struct rte_eth_dev *dev)
+{
+ struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+
+ if (adapter->is_dev_repr)
+ goto l_end;
+
+ fprintf(file,
+ " - Device Args Info:\n"
+ "\t -- no_sched_mode: %s\n"
+ "\t -- flow-duplicate-pattern: %u\n"
+ "\t -- fnav-stat-type: %u\n"
+ "\t -- sched_layer_mode: %u\n"
+ "\t -- rx_low_latency: %s\n"
+ "\t -- function-flow-direct: %s\n",
+ adapter->devargs.no_sched_mode ? "On" : "Off",
+ adapter->devargs.flow_dup_pattern_mode,
+ adapter->devargs.fnav_stat_type,
+ adapter->devargs.sched_layer_mode,
+ adapter->devargs.rx_low_latency ? "On" : "Off",
+ adapter->devargs.func_flow_direct_en ? "On" : "Off");
+l_end:
+ return;
+}
+
+static void sxe2_dump_filter_info(FILE *file, struct rte_eth_dev *dev)
+{
+ struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+ struct sxe2_mac_filter *mac_entry;
+ struct sxe2_mac_filter *next_mac_entry;
+ struct sxe2_vlan_filter *vlan_entry;
+ struct sxe2_vlan_filter *next_vlan_entry;
+
+ if (adapter->is_dev_repr)
+ goto l_end;
+
+ fprintf(file,
+ " - Device Filter Info:\n"
+ "\t -- cur_promisc:0x%x hw_promisc:0x%x\n"
+ "\t -- unicast_num: %u multicast_num: %u\n"
+ "\t -- vlan_num: %u filter_on: %u hw_filter_on: %u\n"
+ "\t -- vlan max_cnt: %u cnt: %u\n"
+ "\t -- tpid: 0x%x vid: 0x%x\n"
+ "\t -- vlan_outer_insert: 0x%x vlan_outer_strip: 0x%x\n"
+ "\t -- vlan_inner_insert: 0x%x vlan_inner_strip: 0x%x\n",
+ adapter->filter_ctxt.cur_promisc_flags,
+ adapter->filter_ctxt.hw_promisc_flags,
+ adapter->filter_ctxt.uc_num,
+ adapter->filter_ctxt.mc_num,
+ adapter->filter_ctxt.vlan_num,
+ adapter->filter_ctxt.vlan_info.filter_on,
+ adapter->filter_ctxt.vlan_info.hw_filter_on,
+ adapter->filter_ctxt.vlan_info.max_cnt,
+ adapter->filter_ctxt.vlan_info.cnt,
+ adapter->filter_ctxt.vlan_info.tpid,
+ adapter->filter_ctxt.vlan_info.vid,
+ adapter->filter_ctxt.vlan_info.outer_insert,
+ adapter->filter_ctxt.vlan_info.outer_strip,
+ adapter->filter_ctxt.vlan_info.inner_insert,
+ adapter->filter_ctxt.vlan_info.inner_strip);
+
+ if (adapter->filter_ctxt.uc_num > 0) {
+ fprintf(file,
+ "\t -- Unicast entry:\n");
+ RTE_TAILQ_FOREACH_SAFE(mac_entry, &adapter->filter_ctxt.uc_list, next,
+ next_mac_entry) {
+ fprintf(file,
+ "\t -- addr: %02x:%02x:%02x:%02x:%02x:%02x hw status:%u "
+ "default:%u\n",
+ mac_entry->mac_addr.addr_bytes[0],
+ mac_entry->mac_addr.addr_bytes[1],
+ mac_entry->mac_addr.addr_bytes[2],
+ mac_entry->mac_addr.addr_bytes[3],
+ mac_entry->mac_addr.addr_bytes[4],
+ mac_entry->mac_addr.addr_bytes[5],
+ mac_entry->hw_config,
+ mac_entry->default_config);
+ }
+ }
+
+ if (adapter->filter_ctxt.mc_num > 0) {
+ fprintf(file,
+ "\t -- Multicast entry:\n");
+ RTE_TAILQ_FOREACH_SAFE(mac_entry, &adapter->filter_ctxt.mc_list,
+ next, next_mac_entry) {
+ fprintf(file,
+ "\t -- addr: %02x:%02x:%02x:%02x:%02x:%02x "
+ "hw status:%u default:%u\n",
+ mac_entry->mac_addr.addr_bytes[0],
+ mac_entry->mac_addr.addr_bytes[1],
+ mac_entry->mac_addr.addr_bytes[2],
+ mac_entry->mac_addr.addr_bytes[3],
+ mac_entry->mac_addr.addr_bytes[4],
+ mac_entry->mac_addr.addr_bytes[5],
+ mac_entry->hw_config,
+ mac_entry->default_config);
+ }
+ }
+
+ if (adapter->filter_ctxt.vlan_num > 0) {
+ fprintf(file,
+ "\t -- Vlan entry:\n");
+ RTE_TAILQ_FOREACH_SAFE(vlan_entry, &adapter->filter_ctxt.vlan_list,
+ next, next_vlan_entry) {
+ fprintf(file,
+ "\t -- vlan tpid:0x%04x vid:0x%04x prio:%d "
+ "hw status:%u default:%u\n",
+ vlan_entry->vlan_info.tpid,
+ vlan_entry->vlan_info.vid,
+ vlan_entry->vlan_info.prio,
+ vlan_entry->hw_config,
+ vlan_entry->default_config);
+ }
+ }
+l_end:
+ return;
+}
+
+static const char *sxe2_vsi_id_str(uint16_t vsi_id, char *buf, size_t len)
+{
+ if (vsi_id == SXE2_INVALID_VSI_ID)
+ return "NA";
+
+ snprintf(buf, len, "%u", vsi_id);
+ return buf;
+}
+
+static void
+sxe2_dump_switchdev_info(FILE *file, struct rte_eth_dev *dev)
+{
+ struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+ uint32_t idx;
+ char k_vsi_buf[16];
+ char u_vsi_buf[16];
+
+ if (adapter->is_dev_repr && adapter->repr_priv_data) {
+ fprintf(file,
+ " - Reprenstor Info:\n"
+ "\t -- repr_id: %u\n"
+ "\t -- repr_q_id: %u\n"
+ "\t -- repr_pf_id: %u\n"
+ "\t -- repr_vf_id: %u\n"
+ "\t -- repr_vf_vsi_id: %u\n"
+ "\t -- repr_vf_k_vsi_id: %s\n"
+ "\t -- repr_vf_u_vsi_id: %s\n",
+ adapter->repr_priv_data->repr_id,
+ adapter->repr_priv_data->repr_q_id,
+ adapter->repr_priv_data->repr_pf_id,
+ adapter->repr_priv_data->repr_vf_id,
+ adapter->repr_priv_data->repr_vf_vsi_id,
+ sxe2_vsi_id_str(adapter->repr_priv_data->repr_vf_k_vsi_id,
+ k_vsi_buf, sizeof(k_vsi_buf)),
+ sxe2_vsi_id_str(adapter->repr_priv_data->repr_vf_u_vsi_id,
+ u_vsi_buf, sizeof(u_vsi_buf)));
+ goto l_end;
+ }
+ if (adapter->switchdev_info.is_switchdev) {
+ fprintf(file,
+ " - Switchdev Info:\n"
+ "\t -- primary:0x%x\n"
+ "\t -- representor: 0x%x\n"
+ "\t -- port_name_type: 0x%x\n"
+ "\t -- nb_vf: %u nb_repr_vf: %u\n",
+ adapter->switchdev_info.primary,
+ adapter->switchdev_info.representor,
+ adapter->switchdev_info.port_name_type,
+ adapter->repr_ctxt.nb_vf,
+ adapter->repr_ctxt.nb_repr_vf);
+ if (adapter->repr_ctxt.nb_vf > 0) {
+ fprintf(file,
+ "\t -- vf entry:\n");
+ for (idx = 0; idx < adapter->repr_ctxt.nb_vf; idx++) {
+ fprintf(file,
+ "\t -- func_id:%u vsi_type:%u kernel_vsi_id:%u dpdk_vsi_id:%u\n",
+ adapter->repr_ctxt.repr_vf_id[idx].func_id,
+ adapter->repr_ctxt.repr_vf_id[idx].vsi_type,
+ adapter->repr_ctxt.repr_vf_id[idx].kernel_vsi_id,
+ adapter->repr_ctxt.repr_vf_id[idx].dpdk_vsi_id);
+ }
+ }
+ }
+
+l_end:
+ return;
+}
+
+int32_t sxe2_eth_dev_priv_dump(struct rte_eth_dev *dev, FILE *file)
+{
+ char *buf = NULL;
+ size_t size = 0;
+ FILE *str;
+ int32_t ret = -1;
+
+ str = open_memstream(&buf, &size);
+ if (!str) {
+ PMD_LOG_ERR(DRV, "fopen fail.");
+ goto l_end;
+ }
+
+ sxe2_dump_dev_feature_capability(str, dev);
+ sxe2_dump_device_basic_info(str, dev);
+ sxe2_dump_dev_args_info(str, dev);
+ sxe2_dump_filter_info(str, dev);
+ sxe2_dump_switchdev_info(str, dev);
+
+ (void)fflush(str);
+
+ (void)fwrite(buf, 1, size, file);
+ (void)fflush(file);
+
+ ret = 0;
+
+ (void)fclose(str);
+ free(buf);
+l_end:
+ return ret;
+}
diff --git a/drivers/net/sxe2/sxe2_dump.h b/drivers/net/sxe2/sxe2_dump.h
new file mode 100644
index 0000000000..05d6db9b3d
--- /dev/null
+++ b/drivers/net/sxe2/sxe2_dump.h
@@ -0,0 +1,12 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright (C), 2025, Wuxi Stars Micro System Technologies Co., Ltd.
+ */
+
+#ifndef __SXE2_DUMP_H__
+#define __SXE2_DUMP_H__
+
+#include <ethdev_driver.h>
+
+int32_t sxe2_eth_dev_priv_dump(struct rte_eth_dev *dev, FILE *file);
+
+#endif /* __SXE2_DUMP_H__ */
diff --git a/drivers/net/sxe2/sxe2_ethdev.c b/drivers/net/sxe2/sxe2_ethdev.c
index cadb926e5d..6b60b6d75f 100644
--- a/drivers/net/sxe2/sxe2_ethdev.c
+++ b/drivers/net/sxe2/sxe2_ethdev.c
@@ -37,6 +37,7 @@
#include "sxe2_host_regs.h"
#include "sxe2_switchdev.h"
#include "sxe2_ioctl_chnl_func.h"
+#include "sxe2_dump.h"
#include "sxe2_ethdev_repr.h"
#include "sxe2vf_regs.h"
#include "sxe2_switchdev.h"
@@ -195,6 +196,8 @@ static const struct eth_dev_ops sxe2_eth_dev_ops = {
.get_module_info = sxe2_get_module_info,
.get_module_eeprom = sxe2_get_module_eeprom,
+
+ .eth_dev_priv_dump = sxe2_eth_dev_priv_dump,
};
static int32_t sxe2_dev_configure(struct rte_eth_dev *dev)
diff --git a/drivers/net/sxe2/sxe2_ethdev_repr.c b/drivers/net/sxe2/sxe2_ethdev_repr.c
index 15b839bb74..f32318b731 100644
--- a/drivers/net/sxe2/sxe2_ethdev_repr.c
+++ b/drivers/net/sxe2/sxe2_ethdev_repr.c
@@ -11,6 +11,7 @@
#include "sxe2_txrx.h"
#include "sxe2_switchdev.h"
#include "sxe2_mp.h"
+#include "sxe2_dump.h"
#include "sxe2_stats.h"
#include "sxe2_flow.h"
@@ -236,6 +237,8 @@ static const struct eth_dev_ops sxe2_switchdev_repr_dev_ops = {
.allmulticast_enable = sxe2_repr_allmulti_enable,
.allmulticast_disable = sxe2_repr_allmulti_disable,
+ .eth_dev_priv_dump = sxe2_eth_dev_priv_dump,
+
.stats_get = sxe2_stats_info_get,
.stats_reset = sxe2_stats_info_reset,
.xstats_get = sxe2_xstats_info_get,
--
2.52.0
^ permalink raw reply related
* [PATCH v4 20/23] net/sxe2: add mbuf validation in Tx debug mode
From: liujie5 @ 2026-06-19 8:11 UTC (permalink / raw)
To: stephen; +Cc: dev, Jie Liu
In-Reply-To: <20260618082723.571054-21-liujie5@linkdatatechnology.com>
From: Jie Liu <liujie5@linkdatatechnology.com>
Introduce the `sxe2_txrx_check_mbuf` helper function to validate outgoing
mbufs when `RTE_ETHDEV_DEBUG_TX` is enabled. This helps developers catch
malformed mbufs (e.g., invalid segment lengths, bad offload flags, or
unaligned buffers) before passing them to the hardware rings, avoiding
potential hardware hangs or silent packet drops.
The validation is fully wrapped inside `RTE_ETHDEV_DEBUG_TX` conditional
compilation blocks to ensure zero performance overhead in standard
production builds.
Signed-off-by: Jie Liu <liujie5@linkdatatechnology.com>
---
drivers/net/sxe2/meson.build | 1 +
drivers/net/sxe2/sxe2_txrx.c | 8 +-
drivers/net/sxe2/sxe2_txrx_check_mbuf.c | 595 ++++++++++++++++++++++++
drivers/net/sxe2/sxe2_txrx_check_mbuf.h | 38 ++
4 files changed, 640 insertions(+), 2 deletions(-)
create mode 100644 drivers/net/sxe2/sxe2_txrx_check_mbuf.c
create mode 100644 drivers/net/sxe2/sxe2_txrx_check_mbuf.h
diff --git a/drivers/net/sxe2/meson.build b/drivers/net/sxe2/meson.build
index d653d071a9..2405213809 100644
--- a/drivers/net/sxe2/meson.build
+++ b/drivers/net/sxe2/meson.build
@@ -78,4 +78,5 @@ sources += files(
'sxe2_flow_parse_pattern.c',
'sxe2_flow_parse_engine.c',
'sxe2_dump.c',
+ 'sxe2_txrx_check_mbuf.c',
)
diff --git a/drivers/net/sxe2/sxe2_txrx.c b/drivers/net/sxe2/sxe2_txrx.c
index 575c17fa3f..6584d9c454 100644
--- a/drivers/net/sxe2/sxe2_txrx.c
+++ b/drivers/net/sxe2/sxe2_txrx.c
@@ -13,6 +13,7 @@
#include "sxe2_txrx_common.h"
#include "sxe2_txrx_vec.h"
#include "sxe2_txrx_poll.h"
+#include "sxe2_txrx_check_mbuf.h"
#include "sxe2_ethdev.h"
#include "sxe2_common_log.h"
#include "sxe2_osal.h"
@@ -120,13 +121,11 @@ uint16_t sxe2_tx_pkts_prepare(void *tx_queue,
rte_errno = -EINVAL;
goto l_end;
}
-#ifdef RTE_ETHDEV_DEBUG_TX
ret = rte_validate_tx_offload(mbuf);
if (ret != 0) {
rte_errno = -ret;
goto l_end;
}
-#endif
ret = rte_net_intel_cksum_prepare(mbuf);
if (ret != 0) {
rte_errno = -ret;
@@ -137,6 +136,11 @@ uint16_t sxe2_tx_pkts_prepare(void *tx_queue,
rte_errno = -ret;
goto l_end;
}
+ ret = sxe2_txrx_check_mbuf(mbuf);
+ if (ret != 0) {
+ rte_errno = -ret;
+ goto l_end;
+ }
}
l_end:
return i;
diff --git a/drivers/net/sxe2/sxe2_txrx_check_mbuf.c b/drivers/net/sxe2/sxe2_txrx_check_mbuf.c
new file mode 100644
index 0000000000..7d316ae652
--- /dev/null
+++ b/drivers/net/sxe2/sxe2_txrx_check_mbuf.c
@@ -0,0 +1,595 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright (C), 2025, Wuxi Stars Micro System Technologies Co., Ltd.
+ */
+
+#include <rte_common.h>
+#include <rte_net.h>
+#include <rte_vect.h>
+#include <rte_malloc.h>
+#include <rte_memzone.h>
+#include <ethdev_driver.h>
+#include <rte_geneve.h>
+
+#include "sxe2_txrx_check_mbuf.h"
+#include "sxe2_common_log.h"
+
+#define TX_IPPROTO_IPIP 4
+#define TX_IPPROTO_GRE 47
+#define GRE_CHECKSUM_PRESENT 0x8000
+#define GRE_KEY_PRESENT 0x2000
+#define GRE_SEQUENCE_PRESENT 0x1000
+#define GRE_EXT_LEN 4
+#define GRE_SUPPORTED_FIELDS (GRE_CHECKSUM_PRESENT | GRE_KEY_PRESENT | GRE_SEQUENCE_PRESENT)
+
+
+static uint16_t vxlan_gpe_udp_port = RTE_VXLAN_GPE_DEFAULT_PORT;
+static uint16_t geneve_udp_port = RTE_GENEVE_DEFAULT_PORT;
+
+static inline int32_t check_mbuf_len(struct offload_info *info, struct rte_mbuf *m)
+{
+ int32_t ret = 0;
+ if (m->ol_flags & RTE_MBUF_F_TX_TUNNEL_MASK) {
+ if (info->outer_l2_len != m->outer_l2_len) {
+ PMD_LOG_ERR(TX, "outer_l2_len error in mbuf. Original "
+ "length:%u calculated length:%u", m->outer_l2_len,
+ info->outer_l2_len);
+ ret = -1;
+ goto end;
+ }
+ if (info->outer_l3_len != m->outer_l3_len) {
+ PMD_LOG_ERR(TX, "outer_l3_len error in mbuf. Original "
+ "length:%u calculated length:%u", m->outer_l3_len,
+ info->outer_l3_len);
+ ret = -1;
+ goto end;
+ }
+ }
+
+ if (info->l2_len != m->l2_len) {
+ PMD_LOG_ERR(TX, "l2_len error in mbuf. Original "
+ "length:%u calculated length:%u", m->l2_len, info->l2_len);
+ ret = -1;
+ goto end;
+ }
+ if (info->l3_len != m->l3_len) {
+ PMD_LOG_ERR(TX, "l3_len error in mbuf. Original "
+ "length:%u calculated length:%u", m->l3_len, info->l3_len);
+ ret = -1;
+ goto end;
+ }
+ if (info->l4_len != m->l4_len) {
+ PMD_LOG_ERR(TX, "l4_len error in mbuf. Original "
+ "length:%u calculated length:%u", m->l4_len, info->l4_len);
+ ret = -1;
+ goto end;
+ }
+ ret = 0;
+
+end:
+ return ret;
+}
+
+static inline int32_t check_ether_type(struct offload_info *info, struct rte_mbuf *m)
+{
+ int32_t ret = 0;
+
+ if (m->ol_flags & RTE_MBUF_F_TX_TUNNEL_MASK) {
+ if (info->outer_ethertype == rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV4)) {
+ if (!(m->ol_flags & RTE_MBUF_F_TX_OUTER_IPV4)) {
+ PMD_LOG_ERR(TX, "Outer ethernet type is ipv4, "
+ "tx offload missing `RTE_MBUF_F_TX_OUTER_IPV4` flag");
+ ret = -1;
+ goto end;
+ }
+ if (m->ol_flags & RTE_MBUF_F_TX_OUTER_IPV6) {
+ PMD_LOG_ERR(TX, "Outer ethernet type is ipv4, tx "
+ "offload contains wrong `RTE_MBUF_F_TX_OUTER_IPV6` flag");
+ ret = -1;
+ goto end;
+ }
+ } else if (info->outer_ethertype == rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV6)) {
+ if (!(m->ol_flags & RTE_MBUF_F_TX_OUTER_IPV6)) {
+ PMD_LOG_ERR(TX, "Outer ethernet type is ipv6, "
+ "tx offload missing `RTE_MBUF_F_TX_OUTER_IPV6` flag");
+ ret = -1;
+ goto end;
+ }
+ if (m->ol_flags & RTE_MBUF_F_TX_OUTER_IPV4) {
+ PMD_LOG_ERR(TX, "Outer ethernet type is ipv6, tx "
+ "offload contains wrong `RTE_MBUF_F_TX_OUTER_IPV4` flag");
+ ret = -1;
+ goto end;
+ }
+ }
+ }
+
+ if (info->ethertype == rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV4)) {
+ if (!(m->ol_flags & RTE_MBUF_F_TX_IPV4)) {
+ PMD_LOG_ERR(TX, "Ethernet type is ipv4, tx offload "
+ "missing `RTE_MBUF_F_TX_IPV4` flag.");
+ ret = -1;
+ goto end;
+ }
+ if (m->ol_flags & RTE_MBUF_F_TX_IPV6) {
+ PMD_LOG_ERR(TX, "Ethernet type is ipv4, tx "
+ "offload contains wrong `RTE_MBUF_F_TX_IPV6` flag");
+ ret = -1;
+ goto end;
+ }
+ } else if (info->ethertype == rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV6)) {
+ if (!(m->ol_flags & RTE_MBUF_F_TX_IPV6)) {
+ PMD_LOG_ERR(TX, "Ethernet type is ipv6, tx offload "
+ "missing `RTE_MBUF_F_TX_IPV6` flag.");
+ ret = -1;
+ goto end;
+ }
+ if (m->ol_flags & RTE_MBUF_F_TX_IPV4) {
+ PMD_LOG_ERR(TX, "Ethernet type is ipv6, tx offload "
+ "contains wrong `RTE_MBUF_F_TX_IPV4` flag");
+ ret = -1;
+ goto end;
+ }
+ }
+ ret = 0;
+
+end:
+ return ret;
+}
+
+static inline void parse_ipv4(struct rte_ipv4_hdr *ipv4_hdr, struct offload_info *info)
+{
+ struct rte_tcp_hdr *tcp_hdr;
+
+ info->l3_len = rte_ipv4_hdr_len(ipv4_hdr);
+ info->l4_proto = ipv4_hdr->next_proto_id;
+
+ if (info->l4_proto == IPPROTO_TCP) {
+ tcp_hdr = (struct rte_tcp_hdr *)((char *)ipv4_hdr + info->l3_len);
+ info->l4_len = (tcp_hdr->data_off & 0xf0) >> 2;
+ } else if (info->l4_proto == IPPROTO_UDP) {
+ info->l4_len = sizeof(struct rte_udp_hdr);
+ } else {
+ info->l4_len = 0;
+ }
+}
+
+static inline void parse_ipv6(struct rte_ipv6_hdr *ipv6_hdr, struct offload_info *info)
+{
+ struct rte_tcp_hdr *tcp_hdr;
+
+ info->l3_len = sizeof(struct rte_ipv6_hdr);
+ info->l4_proto = ipv6_hdr->proto;
+
+ if (info->l4_proto == IPPROTO_TCP) {
+ tcp_hdr = (struct rte_tcp_hdr *)((char *)ipv6_hdr + info->l3_len);
+ info->l4_len = (tcp_hdr->data_off & 0xf0) >> 2;
+ } else if (info->l4_proto == IPPROTO_UDP) {
+ info->l4_len = sizeof(struct rte_udp_hdr);
+ } else {
+ info->l4_len = 0;
+ }
+}
+
+static inline void parse_ethernet(struct rte_ether_hdr *eth_hdr, struct offload_info *info)
+{
+ struct rte_ipv4_hdr *ipv4_hdr;
+ struct rte_ipv6_hdr *ipv6_hdr;
+ struct rte_vlan_hdr *vlan_hdr;
+
+ info->l2_len = sizeof(struct rte_ether_hdr);
+ info->ethertype = eth_hdr->ether_type;
+
+ while (info->ethertype == rte_cpu_to_be_16(RTE_ETHER_TYPE_VLAN) ||
+ info->ethertype == rte_cpu_to_be_16(RTE_ETHER_TYPE_QINQ)) {
+ vlan_hdr = (struct rte_vlan_hdr *)
+ ((char *)eth_hdr + info->l2_len);
+ info->l2_len += sizeof(struct rte_vlan_hdr);
+ info->ethertype = vlan_hdr->eth_proto;
+ }
+
+ switch (info->ethertype) {
+ case RTE_STATIC_BSWAP16(RTE_ETHER_TYPE_IPV4):
+ ipv4_hdr = (struct rte_ipv4_hdr *)((char *)eth_hdr + info->l2_len);
+ parse_ipv4(ipv4_hdr, info);
+ break;
+ case RTE_STATIC_BSWAP16(RTE_ETHER_TYPE_IPV6):
+ ipv6_hdr = (struct rte_ipv6_hdr *)((char *)eth_hdr + info->l2_len);
+ parse_ipv6(ipv6_hdr, info);
+ break;
+ default:
+ info->l4_len = 0;
+ info->l3_len = 0;
+ info->l4_proto = 0;
+ break;
+ }
+}
+
+static inline void update_tunnel_outer(struct offload_info *info)
+{
+ info->is_tunnel = 1;
+ info->outer_ethertype = info->ethertype;
+ info->outer_l2_len = info->l2_len;
+ info->outer_l3_len = info->l3_len;
+ info->outer_l4_proto = info->l4_proto;
+}
+
+static inline void parse_gtp(struct rte_udp_hdr *udp_hdr, struct offload_info *info)
+{
+ struct rte_ipv4_hdr *ipv4_hdr;
+ struct rte_ipv6_hdr *ipv6_hdr;
+ struct rte_gtp_hdr *gtp_hdr;
+ uint8_t gtp_len = sizeof(*gtp_hdr);
+ uint8_t ip_ver;
+
+ if (udp_hdr->dst_port != rte_cpu_to_be_16(RTE_GTPC_UDP_PORT) &&
+ udp_hdr->src_port != rte_cpu_to_be_16(RTE_GTPC_UDP_PORT) &&
+ udp_hdr->dst_port != rte_cpu_to_be_16(RTE_GTPU_UDP_PORT))
+ goto end;
+
+ update_tunnel_outer(info);
+ info->l2_len = 0;
+
+ gtp_hdr = (struct rte_gtp_hdr *)((char *)udp_hdr + sizeof(*udp_hdr));
+
+ if (gtp_hdr->msg_type == 0xff) {
+ ip_ver = *(uint8_t *)((char *)udp_hdr + sizeof(*udp_hdr) + sizeof(*gtp_hdr));
+ ip_ver = (ip_ver) & 0xf0;
+
+ if (ip_ver == RTE_GTP_TYPE_IPV4) {
+ ipv4_hdr = (struct rte_ipv4_hdr *)((char *)gtp_hdr + gtp_len);
+ info->ethertype = rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV4);
+ parse_ipv4(ipv4_hdr, info);
+ } else if (ip_ver == RTE_GTP_TYPE_IPV6) {
+ ipv6_hdr = (struct rte_ipv6_hdr *)((char *)gtp_hdr + gtp_len);
+ info->ethertype = rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV6);
+ parse_ipv6(ipv6_hdr, info);
+ }
+ } else {
+ info->ethertype = 0;
+ info->l4_len = 0;
+ info->l3_len = 0;
+ info->l4_proto = 0;
+ }
+
+ info->l2_len += RTE_ETHER_GTP_HLEN;
+
+end:
+ return;
+}
+
+static inline void parse_vxlan(struct rte_udp_hdr *udp_hdr, struct offload_info *info)
+{
+ struct rte_ether_hdr *eth_hdr;
+
+ if (udp_hdr->dst_port != rte_cpu_to_be_16(RTE_VXLAN_DEFAULT_PORT))
+ goto end;
+
+ update_tunnel_outer(info);
+
+ eth_hdr = (struct rte_ether_hdr *)((char *)udp_hdr +
+ sizeof(struct rte_udp_hdr) + sizeof(struct rte_vxlan_hdr));
+
+ parse_ethernet(eth_hdr, info);
+ info->l2_len += RTE_ETHER_VXLAN_HLEN;
+
+end:
+ return;
+}
+
+static inline void parse_vxlan_gpe(struct rte_udp_hdr *udp_hdr, struct offload_info *info)
+{
+ struct rte_ether_hdr *eth_hdr;
+ struct rte_ipv4_hdr *ipv4_hdr;
+ struct rte_ipv6_hdr *ipv6_hdr;
+ struct rte_vxlan_gpe_hdr *vxlan_gpe_hdr;
+ uint8_t vxlan_gpe_len = sizeof(*vxlan_gpe_hdr);
+
+ if (udp_hdr->dst_port != rte_cpu_to_be_16(vxlan_gpe_udp_port))
+ goto end;
+
+ vxlan_gpe_hdr = (struct rte_vxlan_gpe_hdr *)((char *)udp_hdr + sizeof(struct rte_udp_hdr));
+
+ if (!vxlan_gpe_hdr->proto || vxlan_gpe_hdr->proto == RTE_VXLAN_GPE_TYPE_IPV4) {
+ update_tunnel_outer(info);
+
+ ipv4_hdr = (struct rte_ipv4_hdr *)((char *)vxlan_gpe_hdr + vxlan_gpe_len);
+
+ parse_ipv4(ipv4_hdr, info);
+ info->ethertype = rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV4);
+ info->l2_len = 0;
+
+ } else if (vxlan_gpe_hdr->proto == RTE_VXLAN_GPE_TYPE_IPV6) {
+ update_tunnel_outer(info);
+
+ ipv6_hdr = (struct rte_ipv6_hdr *)((char *)vxlan_gpe_hdr + vxlan_gpe_len);
+
+ info->ethertype = rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV6);
+ parse_ipv6(ipv6_hdr, info);
+ info->l2_len = 0;
+
+ } else if (vxlan_gpe_hdr->proto == RTE_VXLAN_GPE_TYPE_ETH) {
+ update_tunnel_outer(info);
+
+ eth_hdr = (struct rte_ether_hdr *)((char *)vxlan_gpe_hdr + vxlan_gpe_len);
+
+ parse_ethernet(eth_hdr, info);
+ } else {
+ goto end;
+ }
+
+ info->l2_len += RTE_ETHER_VXLAN_GPE_HLEN;
+
+end:
+ return;
+}
+
+static inline void parse_geneve(struct rte_udp_hdr *udp_hdr, struct offload_info *info)
+{
+ struct rte_ether_hdr *eth_hdr;
+ struct rte_ipv4_hdr *ipv4_hdr;
+ struct rte_ipv6_hdr *ipv6_hdr;
+ struct rte_geneve_hdr *geneve_hdr;
+ uint16_t geneve_len;
+
+ if (udp_hdr->dst_port != rte_cpu_to_be_16(geneve_udp_port))
+ goto end;
+
+ geneve_hdr = (struct rte_geneve_hdr *)((char *)udp_hdr + sizeof(struct rte_udp_hdr));
+ geneve_len = sizeof(struct rte_geneve_hdr) + geneve_hdr->opt_len * 4;
+ if (!geneve_hdr->proto || geneve_hdr->proto == rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV4)) {
+ update_tunnel_outer(info);
+ ipv4_hdr = (struct rte_ipv4_hdr *)((char *)geneve_hdr + geneve_len);
+ parse_ipv4(ipv4_hdr, info);
+ info->ethertype = rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV4);
+ info->l2_len = 0;
+ } else if (geneve_hdr->proto == rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV6)) {
+ update_tunnel_outer(info);
+ ipv6_hdr = (struct rte_ipv6_hdr *)((char *)geneve_hdr + geneve_len);
+ info->ethertype = rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV6);
+ parse_ipv6(ipv6_hdr, info);
+ info->l2_len = 0;
+
+ } else if (geneve_hdr->proto == rte_cpu_to_be_16(RTE_GENEVE_TYPE_ETH)) {
+ update_tunnel_outer(info);
+ eth_hdr = (struct rte_ether_hdr *)((char *)geneve_hdr + geneve_len);
+ parse_ethernet(eth_hdr, info);
+ } else {
+ goto end;
+ }
+
+ info->l2_len += (sizeof(struct rte_udp_hdr) + sizeof(struct rte_geneve_hdr) +
+ ((struct rte_geneve_hdr *)geneve_hdr)->opt_len * 4);
+
+end:
+ return;
+}
+
+static inline void parse_gre(struct simple_gre_hdr *gre_hdr, struct offload_info *info)
+{
+ struct rte_ether_hdr *eth_hdr;
+ struct rte_ipv4_hdr *ipv4_hdr;
+ struct rte_ipv6_hdr *ipv6_hdr;
+ uint8_t gre_len = 0;
+
+ gre_len += sizeof(struct simple_gre_hdr);
+
+ if (gre_hdr->flags & rte_cpu_to_be_16(GRE_KEY_PRESENT))
+ gre_len += GRE_EXT_LEN;
+ if (gre_hdr->flags & rte_cpu_to_be_16(GRE_SEQUENCE_PRESENT))
+ gre_len += GRE_EXT_LEN;
+ if (gre_hdr->flags & rte_cpu_to_be_16(GRE_CHECKSUM_PRESENT))
+ gre_len += GRE_EXT_LEN;
+
+ if (gre_hdr->proto == rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV4)) {
+ update_tunnel_outer(info);
+
+ ipv4_hdr = (struct rte_ipv4_hdr *)((char *)gre_hdr + gre_len);
+
+ parse_ipv4(ipv4_hdr, info);
+ info->ethertype = rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV4);
+ info->l2_len = 0;
+
+ } else if (gre_hdr->proto == rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV6)) {
+ update_tunnel_outer(info);
+
+ ipv6_hdr = (struct rte_ipv6_hdr *)((char *)gre_hdr + gre_len);
+
+ info->ethertype = rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV6);
+ parse_ipv6(ipv6_hdr, info);
+ info->l2_len = 0;
+
+ } else if (gre_hdr->proto == rte_cpu_to_be_16(RTE_ETHER_TYPE_TEB)) {
+ update_tunnel_outer(info);
+
+ eth_hdr = (struct rte_ether_hdr *)((char *)gre_hdr + gre_len);
+
+ parse_ethernet(eth_hdr, info);
+ } else {
+ goto end;
+ }
+
+ info->l2_len += gre_len;
+
+end:
+ return;
+}
+
+static inline void parse_encap_ip(void *encap_ip, struct offload_info *info)
+{
+ struct rte_ipv4_hdr *ipv4_hdr = encap_ip;
+ struct rte_ipv6_hdr *ipv6_hdr = encap_ip;
+ uint8_t ip_version;
+
+ ip_version = ((ipv4_hdr->version_ihl & 0xf0) >> 4);
+
+ if (ip_version != 4 && ip_version != 6)
+ goto end;
+
+ info->is_tunnel = 1;
+ info->outer_ethertype = info->ethertype;
+ info->outer_l2_len = info->l2_len;
+ info->outer_l3_len = info->l3_len;
+
+ if (ip_version == 4) {
+ parse_ipv4(ipv4_hdr, info);
+ info->ethertype = rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV4);
+ } else {
+ parse_ipv6(ipv6_hdr, info);
+ info->ethertype = rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV6);
+ }
+ info->l2_len = 0;
+
+end:
+ return;
+}
+
+__rte_unused int32_t sxe2_txrx_check_mbuf(struct rte_mbuf *m)
+{
+ int32_t ret = 0;
+ struct rte_ether_hdr *eth_hdr;
+ void *l3_hdr = NULL;
+ struct offload_info info = {0};
+ uint64_t ol_flags = m->ol_flags;
+ uint64_t tunnel_type = ol_flags & RTE_MBUF_F_TX_TUNNEL_MASK;
+
+ eth_hdr = rte_pktmbuf_mtod(m, struct rte_ether_hdr *);
+ parse_ethernet(eth_hdr, &info);
+ l3_hdr = (char *)eth_hdr + info.l2_len;
+ if (info.l4_proto == IPPROTO_UDP) {
+ struct rte_udp_hdr *udp_hdr;
+
+ udp_hdr = (struct rte_udp_hdr *)((char *)l3_hdr + info.l3_len);
+ if ((info.l2_len + info.l3_len + sizeof(struct rte_udp_hdr)) > m->data_len) {
+ PMD_LOG_ERR(TX, "UDP header exceeds mbuf data length");
+ ret = -1;
+ goto end;
+ }
+ parse_gtp(udp_hdr, &info);
+ if (info.is_tunnel) {
+ if (!tunnel_type) {
+ PMD_LOG_ERR(TX, "gtp tunnel packet missing tx "
+ "offload missing `RTE_MBUF_F_TX_TUNNEL_GTP` flag");
+ ret = -1;
+ goto end;
+ }
+ if (tunnel_type != RTE_MBUF_F_TX_TUNNEL_GTP) {
+ PMD_LOG_ERR(TX, "gtp tunnel packet, tx offload has wrong "
+ "`%s` flag correct is `RTE_MBUF_F_TX_TUNNEL_GTP` flag",
+ rte_get_tx_ol_flag_name(tunnel_type));
+ ret = -1;
+ goto end;
+ }
+ goto check_len;
+ }
+ parse_vxlan_gpe(udp_hdr, &info);
+ if (info.is_tunnel) {
+ if (!tunnel_type) {
+ PMD_LOG_ERR(TX, "vxlan gpe tunnel packet missing tx "
+ "offload missing `RTE_MBUF_F_TX_TUNNEL_VXLAN_GPE` flag");
+ ret = -1;
+ goto end;
+ }
+ if (tunnel_type != RTE_MBUF_F_TX_TUNNEL_VXLAN_GPE) {
+ PMD_LOG_ERR(TX, "vxlan gpe tunnel packet, tx offload has "
+ "wrong `%s` flag correct is `RTE_MBUF_F_TX_TUNNEL_VXLAN_GPE` flag",
+ rte_get_tx_ol_flag_name(tunnel_type));
+ ret = -1;
+ goto end;
+ }
+ goto check_len;
+ }
+ parse_vxlan(udp_hdr, &info);
+ if (info.is_tunnel) {
+ if (!tunnel_type) {
+ PMD_LOG_ERR(TX, "vxlan tunnel packet missing tx "
+ "offload missing `RTE_MBUF_F_TX_TUNNEL_VXLAN` flag");
+ ret = -1;
+ goto end;
+ }
+ if (tunnel_type != RTE_MBUF_F_TX_TUNNEL_VXLAN) {
+ PMD_LOG_ERR(TX, "vxlan tunnel packet, tx offload has "
+ "wrong `%s` flag correct is `RTE_MBUF_F_TX_TUNNEL_VXLAN` flag",
+ rte_get_tx_ol_flag_name(tunnel_type));
+ ret = -1;
+ goto end;
+ }
+ goto check_len;
+ }
+ parse_geneve(udp_hdr, &info);
+ if (info.is_tunnel) {
+ if (!tunnel_type) {
+ PMD_LOG_ERR(TX, "geneve tunnel packet missing tx "
+ "offload missing `RTE_MBUF_F_TX_TUNNEL_GENEVE` flag");
+ ret = -1;
+ goto end;
+ }
+ if (tunnel_type != RTE_MBUF_F_TX_TUNNEL_GENEVE) {
+ PMD_LOG_ERR(TX, "geneve tunnel packet, tx offload has "
+ "wrong `%s` flag correct is `RTE_MBUF_F_TX_TUNNEL_GENEVE` flag",
+ rte_get_tx_ol_flag_name(tunnel_type));
+ ret = -1;
+ goto end;
+ }
+ goto check_len;
+ }
+
+ if (unlikely(RTE_ETH_IS_TUNNEL_PKT(m->packet_type) != 0)) {
+ PMD_LOG_ERR(TX, "Unknown tunnel packet UDP dst port:%u",
+ udp_hdr->dst_port);
+ ret = -1;
+ goto end;
+ }
+ } else if (info.l4_proto == TX_IPPROTO_GRE) {
+ struct simple_gre_hdr *gre_hdr;
+
+ gre_hdr = (struct simple_gre_hdr *)((char *)l3_hdr + info.l3_len);
+ parse_gre(gre_hdr, &info);
+ if (info.is_tunnel) {
+ if (!tunnel_type) {
+ PMD_LOG_ERR(TX, "gre tunnel packet missing tx "
+ "offload missing `RTE_MBUF_F_TX_TUNNEL_GRE` flag.");
+ ret = -1;
+ goto end;
+ }
+ if (tunnel_type != RTE_MBUF_F_TX_TUNNEL_GRE) {
+ PMD_LOG_ERR(TX, "gre tunnel packet, tx offload has "
+ "wrong `%s` flag, correct is `RTE_MBUF_F_TX_TUNNEL_GRE` flag",
+ rte_get_tx_ol_flag_name(tunnel_type));
+ ret = -1;
+ goto end;
+ }
+ goto check_len;
+ }
+ } else if (info.l4_proto == TX_IPPROTO_IPIP) {
+ void *encap_ip_hdr;
+
+ encap_ip_hdr = (char *)l3_hdr + info.l3_len;
+ parse_encap_ip(encap_ip_hdr, &info);
+ if (info.is_tunnel) {
+ if (!tunnel_type) {
+ PMD_LOG_ERR(TX, "Ipip tunnel packet missing tx "
+ "offload missing `RTE_MBUF_F_TX_TUNNEL_IPIP` flag");
+ ret = -1;
+ goto end;
+ }
+ if (tunnel_type != RTE_MBUF_F_TX_TUNNEL_IPIP) {
+ PMD_LOG_ERR(TX, "Ipip tunnel packet, tx offload has "
+ "wrong `%s` flag, correct is `RTE_MBUF_F_TX_TUNNEL_IPIP` flag",
+ rte_get_tx_ol_flag_name(tunnel_type));
+ ret = -1;
+ goto end;
+ }
+ goto check_len;
+ }
+ }
+
+check_len:
+ if (check_mbuf_len(&info, m) != 0) {
+ ret = -1;
+ goto end;
+ }
+ ret = check_ether_type(&info, m);
+
+end:
+ return ret;
+}
diff --git a/drivers/net/sxe2/sxe2_txrx_check_mbuf.h b/drivers/net/sxe2/sxe2_txrx_check_mbuf.h
new file mode 100644
index 0000000000..98197f85d9
--- /dev/null
+++ b/drivers/net/sxe2/sxe2_txrx_check_mbuf.h
@@ -0,0 +1,38 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright (C), 2025, Wuxi Stars Micro System Technologies Co., Ltd.
+ */
+
+#ifndef __SXE2_TXRX_CHECK_MBUF_H__
+#define __SXE2_TXRX_CHECK_MBUF_H__
+
+#include <rte_common.h>
+#include <rte_net.h>
+#include <rte_vect.h>
+#include <rte_malloc.h>
+#include <rte_memzone.h>
+#include <ethdev_driver.h>
+
+struct offload_info {
+ uint16_t ethertype;
+ uint8_t gso_enable;
+ uint16_t l2_len;
+ uint16_t l3_len;
+ uint16_t l4_len;
+ uint8_t l4_proto;
+ uint8_t is_tunnel;
+ uint16_t outer_ethertype;
+ uint16_t outer_l2_len;
+ uint16_t outer_l3_len;
+ uint8_t outer_l4_proto;
+ uint16_t tso_segsz;
+ uint16_t tunnel_tso_segsz;
+ uint32_t pkt_len;
+};
+
+struct simple_gre_hdr {
+ uint16_t flags;
+ uint16_t proto;
+};
+
+__rte_unused int32_t sxe2_txrx_check_mbuf(struct rte_mbuf *m);
+#endif /* __SXE2_TXRX_CHECK_MBUF_H__ */
--
2.52.0
^ permalink raw reply related
* [PATCH v4 21/23] common/sxe2: add callback for memory event handling
From: liujie5 @ 2026-06-19 8:12 UTC (permalink / raw)
To: stephen; +Cc: dev, Jie Liu
In-Reply-To: <20260618082723.571054-21-liujie5@linkdatatechnology.com>
From: Jie Liu <liujie5@linkdatatechnology.com>
During memory hotplug events, the SXE2 driver needs to track memory
segment layout changes to maintain internal DMA mappings. However,
existing memseg walk functions (rte_memseg_walk) acquire memory locks
and cannot be called from within memory event callbacks, leading to
potential deadlocks.
The implementation follows the standard rte_memseg_walk_t prototype,
processing each memseg to update driver-specific data structures.
Signed-off-by: Jie Liu <liujie5@linkdatatechnology.com>
---
drivers/common/sxe2/sxe2_common.c | 110 ++++++++++++++++++++++++++
drivers/common/sxe2/sxe2_common.h | 2 +
drivers/common/sxe2/sxe2_ioctl_chnl.c | 2 +-
3 files changed, 113 insertions(+), 1 deletion(-)
diff --git a/drivers/common/sxe2/sxe2_common.c b/drivers/common/sxe2/sxe2_common.c
index c000a55cd0..5c5db85f29 100644
--- a/drivers/common/sxe2/sxe2_common.c
+++ b/drivers/common/sxe2/sxe2_common.c
@@ -196,6 +196,102 @@ static int32_t sxe2_parse_representor(const char *key, const char *value, void *
PMD_LOG_INFO(COM, "representor arg %s: \"%s\".", key, value);
+l_end:
+ return ret;
+}
+static int32_t sxe2_dma_mem_map(struct sxe2_common_device *cdev,
+ const void *addr, size_t len, bool do_map)
+{
+ struct rte_memseg_list *msl;
+ struct rte_memseg *ms;
+ size_t cur_len = 0;
+ int32_t ret = 0;
+
+ msl = rte_mem_virt2memseg_list(addr);
+ if (msl == NULL) {
+ ret = -EINVAL;
+ PMD_LOG_ERR(COM, "Invalid virt addr=%p.", addr);
+ goto l_end;
+ }
+
+ if ((uintptr_t)addr != RTE_ALIGN((uintptr_t)addr, msl->page_sz) ||
+ (len != RTE_ALIGN(len, msl->page_sz))) {
+ ret = -EINVAL;
+ PMD_LOG_ERR(COM, "Addr=%p and len=%zu not align page size=%" PRIu64 ".",
+ addr, len, msl->page_sz);
+ goto l_end;
+ }
+
+ /* memsegs are contiguous in memory */
+ ms = rte_mem_virt2memseg(addr, msl);
+ while (cur_len < len) {
+ /* some memory segments may have invalid IOVA */
+ if (ms->iova == RTE_BAD_IOVA) {
+ PMD_LOG_WARN(COM, "Memory segment at %p has bad IOVA, skipping.",
+ ms->addr);
+ goto next;
+ }
+ if (do_map)
+ sxe2_drv_dev_dma_map(cdev, ms->addr_64,
+ ms->iova, ms->len);
+ else
+ sxe2_drv_dev_dma_unmap(cdev, ms->iova);
+
+next:
+ cur_len += ms->len;
+ ++ms;
+ }
+
+l_end:
+ return ret;
+}
+
+RTE_EXPORT_INTERNAL_SYMBOL(sxe2_common_mem_event_cb)
+void
+sxe2_common_mem_event_cb(enum rte_mem_event type,
+ const void *addr, size_t size, void *arg __rte_unused)
+{
+ struct sxe2_common_device *cdev = NULL;
+
+ if (rte_eal_process_type() != RTE_PROC_PRIMARY)
+ goto l_end;
+
+ pthread_mutex_lock(&sxe2_common_devices_list_lock);
+ switch (type) {
+ case RTE_MEM_EVENT_FREE:
+ TAILQ_FOREACH(cdev, &sxe2_common_devices_list, next)
+ (void)sxe2_dma_mem_map(cdev, addr, size, 0);
+ break;
+ case RTE_MEM_EVENT_ALLOC:
+ TAILQ_FOREACH(cdev, &sxe2_common_devices_list, next)
+ (void)sxe2_dma_mem_map(cdev, addr, size, 1);
+ break;
+ default:
+ break;
+ }
+ pthread_mutex_unlock(&sxe2_common_devices_list_lock);
+l_end:
+ return;
+}
+
+static int32_t sxe2_memseg_walk_cb(const struct rte_memseg_list *msl,
+ const struct rte_memseg *ms, void *arg)
+{
+ struct sxe2_common_device *cdev = arg;
+ int32_t ret = 0;
+
+ if (msl->external && !msl->heap)
+ goto l_end;
+
+ if (ms->iova == RTE_BAD_IOVA)
+ goto l_end;
+
+ ret = sxe2_drv_dev_dma_map(cdev, ms->addr_64, ms->iova, ms->len);
+ if (ret != 0) {
+ PMD_LOG_ERR(COM, "Fail to memseg dma map.");
+ goto l_end;
+ }
+
l_end:
return ret;
}
@@ -220,6 +316,18 @@ static int32_t sxe2_common_device_setup(struct sxe2_common_device *cdev)
goto l_close_dev;
}
+ rte_mcfg_mem_read_lock();
+ ret = rte_memseg_walk_thread_unsafe(sxe2_memseg_walk_cb, cdev);
+ if (ret) {
+ PMD_LOG_ERR(COM, "Fail to walk memseg, ret=%d", ret);
+ rte_mcfg_mem_read_unlock();
+ goto l_close_dev;
+ }
+ rte_mcfg_mem_read_unlock();
+
+ (void)rte_mem_event_callback_register("SXE2_MEM_EVENT_CB",
+ sxe2_common_mem_event_cb, NULL);
+
goto l_end;
l_close_dev:
@@ -251,6 +359,7 @@ static struct sxe2_common_device *sxe2_common_device_alloc(
}
cdev->dev = rte_dev;
cdev->class_type = class_type;
+ cdev->config.cmd_fd = SXE2_CMD_FD_INVALID;
cdev->config.kernel_reset = false;
pthread_mutex_init(&cdev->config.lock, NULL);
@@ -631,6 +740,7 @@ static int32_t sxe2_common_pci_id_table_update(const struct rte_pci_id *id_table
updated_table = calloc(num_ids, sizeof(*updated_table));
if (!updated_table) {
+ ret = -ENOMEM;
PMD_LOG_ERR(COM, "Failed to allocate memory for PCI ID table");
goto l_end;
}
diff --git a/drivers/common/sxe2/sxe2_common.h b/drivers/common/sxe2/sxe2_common.h
index b02b6317da..efc8d3585a 100644
--- a/drivers/common/sxe2/sxe2_common.h
+++ b/drivers/common/sxe2/sxe2_common.h
@@ -14,6 +14,8 @@
#define SXE2_COMMON_PCI_DRIVER_NAME "sxe2_pci"
+#define SXE2_CMD_FD_INVALID (-1)
+
#define SXE2_CDEV_TO_CMD_FD(cdev) \
((cdev)->config.cmd_fd)
diff --git a/drivers/common/sxe2/sxe2_ioctl_chnl.c b/drivers/common/sxe2/sxe2_ioctl_chnl.c
index 173d8d57ae..a233a78136 100644
--- a/drivers/common/sxe2/sxe2_ioctl_chnl.c
+++ b/drivers/common/sxe2/sxe2_ioctl_chnl.c
@@ -110,7 +110,7 @@ sxe2_drv_dev_close(struct sxe2_common_device *cdev)
if (fd >= 0)
close(fd);
PMD_LOG_INFO(COM, "closed device fd=%d", fd);
- SXE2_CDEV_TO_CMD_FD(cdev) = -1;
+ SXE2_CDEV_TO_CMD_FD(cdev) = SXE2_CMD_FD_INVALID;
}
RTE_EXPORT_INTERNAL_SYMBOL(sxe2_drv_dev_handshake)
--
2.52.0
^ permalink raw reply related
* [PATCH v4 22/23] net/sxe2: add private devargs parsing
From: liujie5 @ 2026-06-19 8:12 UTC (permalink / raw)
To: stephen; +Cc: dev, Jie Liu
In-Reply-To: <20260618082723.571054-21-liujie5@linkdatatechnology.com>
From: Jie Liu <liujie5@linkdatatechnology.com>
Introduce runtime private device arguments (devargs) support for the
sxe2 PMD. This allows users to customize driver behavior at startup
without recompiling the source code.
The parameters are parsed using the standard 'rte_kvargs' library during
the PCI/vdev probing phase. Documentation for these parameters is also
updated.
Signed-off-by: Jie Liu <liujie5@linkdatatechnology.com>
---
drivers/net/sxe2/sxe2_cmd_chnl.c | 21 +++
drivers/net/sxe2/sxe2_cmd_chnl.h | 3 +
drivers/net/sxe2/sxe2_drv_cmd.h | 17 ++
drivers/net/sxe2/sxe2_dump.c | 15 ++
drivers/net/sxe2/sxe2_ethdev.c | 272 +++++++++++++++++++++++++++++--
drivers/net/sxe2/sxe2_ethdev.h | 6 +
drivers/net/sxe2/sxe2_flow.c | 9 +-
drivers/net/sxe2/sxe2_irq.c | 30 ++++
drivers/net/sxe2/sxe2_rx.c | 12 ++
9 files changed, 369 insertions(+), 16 deletions(-)
diff --git a/drivers/net/sxe2/sxe2_cmd_chnl.c b/drivers/net/sxe2/sxe2_cmd_chnl.c
index 43e8c59487..b09989fe50 100644
--- a/drivers/net/sxe2/sxe2_cmd_chnl.c
+++ b/drivers/net/sxe2/sxe2_cmd_chnl.c
@@ -99,6 +99,27 @@ int32_t sxe2_drv_dev_info_get(struct sxe2_adapter *adapter,
return ret;
}
+int32_t sxe2_drv_fc_state_get(struct sxe2_adapter *adapter,
+ struct sxe2_drv_vsi_fc_get_resp *dev_fc_state_resp)
+{
+ int32_t ret = 0;
+ struct sxe2_common_device *cdev = adapter->cdev;
+ struct sxe2_drv_cmd_params param = {0};
+ struct sxe2_drv_vsi_fc_get_req req = {0};
+
+ req.vsi_id = adapter->vsi_ctxt.main_vsi->vsi_id;
+ sxe2_drv_cmd_params_fill(adapter, ¶m, SXE2_DRV_CMD_VSI_FC_GET,
+ &req, sizeof(req),
+ dev_fc_state_resp,
+ sizeof(*dev_fc_state_resp));
+ ret = sxe2_drv_cmd_exec(cdev, ¶m);
+ if (ret) {
+ PMD_DEV_LOG_ERR(adapter, DRV, "get fc state failed, ret=%d", ret);
+ ret = -EIO;
+ }
+ return ret;
+}
+
int32_t sxe2_drv_dev_fw_info_get(struct sxe2_adapter *adapter,
struct sxe2_drv_dev_fw_info_resp *dev_fw_info_resp)
{
diff --git a/drivers/net/sxe2/sxe2_cmd_chnl.h b/drivers/net/sxe2/sxe2_cmd_chnl.h
index 988d4b458b..d63caad526 100644
--- a/drivers/net/sxe2/sxe2_cmd_chnl.h
+++ b/drivers/net/sxe2/sxe2_cmd_chnl.h
@@ -99,6 +99,9 @@ int32_t sxe2_drv_vsi_stats_reset(struct sxe2_adapter *adapter);
int32_t sxe2_drv_queue_info_get_update(struct sxe2_adapter *adapter,
struct eth_queue_stats *qstats);
+int32_t sxe2_drv_fc_state_get(struct sxe2_adapter *adapter,
+ struct sxe2_drv_vsi_fc_get_resp *dev_fc_state_resp);
+
int32_t sxe2_drv_rxq_mapping_set(struct rte_eth_dev *eth_dev, uint16_t queue_id, uint8_t pool_idx);
int32_t sxe2_drv_txq_mapping_set(struct rte_eth_dev *eth_dev, uint16_t queue_id, uint8_t pool_idx);
diff --git a/drivers/net/sxe2/sxe2_drv_cmd.h b/drivers/net/sxe2/sxe2_drv_cmd.h
index 3fabf351af..03ef3b315d 100644
--- a/drivers/net/sxe2/sxe2_drv_cmd.h
+++ b/drivers/net/sxe2/sxe2_drv_cmd.h
@@ -651,6 +651,23 @@ struct __rte_aligned(4) __rte_packed_begin sxe2_drv_sfp_resp {
uint8_t data[];
} __rte_packed_end;
+enum sxe2_fc_type {
+ SXE2_FC_T_DIS = 0,
+ SXE2_FC_T_LFC,
+ SXE2_FC_T_PFC,
+ SXE2_FC_T_UNKNOWN = 255,
+};
+
+struct __rte_aligned(4) __rte_packed_begin sxe2_drv_vsi_fc_get_req {
+ uint16_t vsi_id;
+ uint8_t rsv[2];
+} __rte_packed_end;
+
+struct __rte_aligned(4) __rte_packed_begin sxe2_drv_vsi_fc_get_resp {
+ uint8_t fc_enable;
+ uint8_t rsv[3];
+} __rte_packed_end;
+
enum sxe2_drv_cmd_module {
SXE2_DRV_CMD_MODULE_HANDSHAKE = 0,
SXE2_DRV_CMD_MODULE_DEV = 1,
diff --git a/drivers/net/sxe2/sxe2_dump.c b/drivers/net/sxe2/sxe2_dump.c
index d43473e083..c7cdde5f4f 100644
--- a/drivers/net/sxe2/sxe2_dump.c
+++ b/drivers/net/sxe2/sxe2_dump.c
@@ -186,6 +186,20 @@ static void sxe2_dump_filter_info(FILE *file, struct rte_eth_dev *dev)
return;
}
+static void sxe2_dump_fc_state(FILE *file, struct rte_eth_dev *dev)
+{
+ struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+
+ if (!(adapter->cap_flags & SXE2_DEV_CAPS_OFFLOAD_FC_STATE))
+ goto l_end;
+
+ fprintf(file, " -- fc state:\n"
+ "\t -- curr_state: %u\n",
+ adapter->fc_state_ctx.curr_state);
+l_end:
+ return;
+}
+
static const char *sxe2_vsi_id_str(uint16_t vsi_id, char *buf, size_t len)
{
if (vsi_id == SXE2_INVALID_VSI_ID)
@@ -272,6 +286,7 @@ int32_t sxe2_eth_dev_priv_dump(struct rte_eth_dev *dev, FILE *file)
sxe2_dump_dev_args_info(str, dev);
sxe2_dump_filter_info(str, dev);
sxe2_dump_switchdev_info(str, dev);
+ sxe2_dump_fc_state(str, dev);
(void)fflush(str);
diff --git a/drivers/net/sxe2/sxe2_ethdev.c b/drivers/net/sxe2/sxe2_ethdev.c
index 6b60b6d75f..8a129d87ad 100644
--- a/drivers/net/sxe2/sxe2_ethdev.c
+++ b/drivers/net/sxe2/sxe2_ethdev.c
@@ -69,6 +69,15 @@ static const struct rte_pci_id pci_id_sxe2_tbl[] = {
{ RTE_PCI_DEVICE(SXE2_PCI_VENDOR_ID_206F, SXE2_PCI_DEVICE_ID_VF_1)},
{ .vendor_id = 0, },
};
+#define SXE2_TXSCH_NODE_ADJ_LVL_MAX 3
+#define SXE2_DEVARG_FLOW_DULP_PATTERN_MODE "flow-duplicate-pattern"
+#define SXE2_DEVARG_FUNC_FLOW_DIRCT "function-flow-direct"
+#define SXE2_DEVARG_FNAV_STAT_TYPE "fnav-stat-type"
+#define SXE2_DEVARG_NO_SCHED_MODE "no-sched-mode"
+#define SXE2_DEVARG_SCHED_LAYER_MODE "sched-layer-mode"
+#define SXE2_DEVARG_RX_LOW_LATENCY "rx-low-latency"
+
+#define SXE2_FLOW_DUP_PATTERN_DEFAULT 1
static struct sxe2_pci_map_addr_info sxe2_net_map_addr_info_pf[SXE2_PCI_MAP_RES_MAX_COUNT] = {
[SXE2_PCI_MAP_RES_INVALID] = {.addr_base = 0,
@@ -965,6 +974,149 @@ sxe2_buffer_split_supported_hdr_ptypes_get(struct rte_eth_dev *dev __rte_unused,
return ptypes;
}
+static int32_t sxe2_parse_fnav_stat_type(const char *key, const char *value, void *args)
+{
+ int32_t ret = -EINVAL;
+ uint8_t *num = (uint8_t *)args;
+ unsigned long fnav_stat_type;
+ char *endptr = NULL;
+
+ if (value == NULL || args == NULL) {
+ ret = 0;
+ goto l_end;
+ }
+ errno = 0;
+ fnav_stat_type = strtoul(value, &endptr, 10);
+ if (errno != 0 || endptr == value || *endptr != '\0') {
+ PMD_LOG_WARN(INIT, "%s: \"%s\" is not a valid int value.",
+ key, value);
+ goto l_end;
+ }
+ if (fnav_stat_type > SXE2_FNAV_STAT_ENA_ALL ||
+ fnav_stat_type == SXE2_FNAV_STAT_ENA_NONE) {
+ PMD_LOG_ERR(INIT, "%s: \"%s\" out of range [1-3].",
+ key, value);
+ goto l_end;
+ }
+ *num = (uint8_t)fnav_stat_type;
+ ret = 0;
+l_end:
+ return ret;
+}
+static int32_t sxe2_parse_sched_layer_mode(const char *key, const char *value, void *args)
+{
+ int32_t ret = -EINVAL;
+ uint8_t *num = (uint8_t *)args;
+ unsigned long sched_layer_mode;
+ char *endptr = NULL;
+
+ if (value == NULL || args == NULL) {
+ ret = 0;
+ goto l_end;
+ }
+ errno = 0;
+ sched_layer_mode = strtoul(value, &endptr, 10);
+ if (errno != 0 || endptr == value || *endptr != '\0') {
+ PMD_LOG_WARN(INIT, "%s: \"%s\" is not a valid int value.",
+ key, value);
+ goto l_end;
+ }
+ if (sched_layer_mode > SXE2_TXSCH_NODE_ADJ_LVL_MAX) {
+ PMD_LOG_ERR(INIT, "%s: \"%s\" > 3.",
+ key, value);
+ goto l_end;
+ }
+ *num = (uint8_t)sched_layer_mode;
+ ret = 0;
+l_end:
+ return ret;
+}
+static int32_t sxe2_parse_no_sched_mode(const char *key, const char *value, void *args)
+{
+ int32_t ret = -EINVAL;
+ uint8_t *num = (uint8_t *)args;
+ unsigned long high_performance_mode;
+ char *endptr = NULL;
+
+ if (value == NULL || args == NULL) {
+ ret = 0;
+ goto l_end;
+ }
+ errno = 0;
+ high_performance_mode = strtoul(value, &endptr, 10);
+ if (errno != 0 || endptr == value || *endptr != '\0') {
+ PMD_LOG_WARN(INIT, "%s: \"%s\" is not a valid int value.",
+ key, value);
+ goto l_end;
+ }
+ if (high_performance_mode != 1) {
+ PMD_LOG_ERR(INIT, "%s: \"%s\" != 1.",
+ key, value);
+ goto l_end;
+ }
+ *num = (uint8_t)high_performance_mode;
+ ret = 0;
+l_end:
+ return ret;
+}
+static int32_t sxe2_parse_u8(const char *key, const char *value, void *args)
+{
+ uint8_t *num = (uint8_t *)args;
+ char *end;
+ unsigned long val;
+ int32_t ret = -EINVAL;
+
+ if (value == NULL || args == NULL) {
+ ret = 0;
+ goto l_end;
+ }
+ errno = 0;
+ val = strtoul(value, &end, 10);
+ if (errno != 0 || end == value || *end != '\0') {
+ PMD_LOG_ERR(INIT, "Invalid 8-bit integer value for key %s: %s", key, value);
+ return -EINVAL;
+ }
+
+ if (val > UINT8_MAX) {
+ PMD_LOG_ERR(INIT, "%s: \"%s\" out of range [0-255].",
+ key, value);
+ return -ERANGE;
+ }
+
+ *num = (uint8_t)val;
+ ret = 0;
+l_end:
+ return ret;
+}
+static int32_t sxe2_parse_bool(const char *key, const char *value, void *args)
+{
+ int32_t ret = -EINVAL;
+ uint8_t *num = (uint8_t *)args;
+ unsigned long bool_val;
+ char *endptr = NULL;
+
+ if (value == NULL || args == NULL) {
+ ret = 0;
+ goto l_end;
+ }
+ errno = 0;
+ bool_val = strtoul(value, &endptr, 10);
+ if (errno != 0 || endptr == value || *endptr != '\0') {
+ PMD_LOG_WARN(INIT, "%s: \"%s\" is not a valid int value.",
+ key, value);
+ goto l_end;
+ }
+ if (bool_val != 0 && bool_val != 1) {
+ PMD_LOG_ERR(INIT, "%s: \"%s\" out of range [0|1].",
+ key, value);
+ goto l_end;
+ }
+ *num = (uint8_t)bool_val;
+ ret = 0;
+l_end:
+ return ret;
+}
+
struct sxe2_pci_map_bar_info *sxe2_dev_get_bar_info(struct sxe2_adapter *adapter,
enum sxe2_pci_map_resource res_type)
{
@@ -1032,6 +1184,65 @@ void *sxe2_pci_map_addr_get(struct sxe2_adapter *adapter,
return addr;
}
+static int32_t sxe2_args_parse(struct rte_eth_dev *dev, struct sxe2_dev_kvargs_info *kvargs)
+{
+ struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+ int32_t ret = 0;
+ PMD_INIT_FUNC_TRACE();
+
+ adapter->devargs.flow_dup_pattern_mode = SXE2_FLOW_DUP_PATTERN_DEFAULT;
+
+ if (kvargs == NULL)
+ goto l_end;
+ ret = sxe2_kvargs_process(kvargs, SXE2_DEVARG_FNAV_STAT_TYPE,
+ &sxe2_parse_fnav_stat_type,
+ &adapter->devargs.fnav_stat_type);
+ if (ret) {
+ PMD_DEV_LOG_ERR(adapter, INIT, "Failed to parse fnav stat type, ret:%d", ret);
+ goto l_end;
+ }
+
+ ret = sxe2_kvargs_process(kvargs, SXE2_DEVARG_NO_SCHED_MODE,
+ &sxe2_parse_no_sched_mode,
+ &adapter->devargs.no_sched_mode);
+ if (ret) {
+ PMD_DEV_LOG_ERR(adapter, INIT, "Failed to parse no sched mode, ret:%d", ret);
+ goto l_end;
+ }
+ ret = sxe2_kvargs_process(kvargs, SXE2_DEVARG_SCHED_LAYER_MODE,
+ &sxe2_parse_sched_layer_mode,
+ &adapter->devargs.sched_layer_mode);
+ if (ret) {
+ PMD_DEV_LOG_ERR(adapter, INIT, "Failed to parse sched layer mode, ret:%d", ret);
+ goto l_end;
+ }
+ ret = sxe2_kvargs_process(kvargs, SXE2_DEVARG_FLOW_DULP_PATTERN_MODE,
+ &sxe2_parse_u8,
+ &adapter->devargs.flow_dup_pattern_mode);
+ if (ret) {
+ PMD_DEV_LOG_ERR(adapter, INIT, "Failed to parse switch dulpliate flow pattern mode,"
+ "ret:%d", ret);
+ goto l_end;
+ }
+ ret = sxe2_kvargs_process(kvargs, SXE2_DEVARG_FUNC_FLOW_DIRCT,
+ &sxe2_parse_bool,
+ &adapter->devargs.func_flow_direct_en);
+ if (ret) {
+ PMD_DEV_LOG_ERR(adapter, INIT, "Failed to parse function flow rule enable,"
+ "ret:%d", ret);
+ goto l_end;
+ }
+ ret = sxe2_kvargs_process(kvargs, SXE2_DEVARG_RX_LOW_LATENCY,
+ &sxe2_parse_bool,
+ &adapter->devargs.rx_low_latency);
+ if (ret) {
+ PMD_DEV_LOG_ERR(adapter, INIT, "Failed to parse rx low latency, ret:%d", ret);
+ goto l_end;
+ }
+l_end:
+ return ret;
+}
+
static int32_t sxe2_eth_init(struct rte_eth_dev *dev)
{
int32_t ret = 0;
@@ -1584,6 +1795,37 @@ void sxe2_dev_pci_map_uinit(struct rte_eth_dev *dev)
adapter->dev_info.dev_data = NULL;
}
+static int32_t sxe2_fc_state_init(struct rte_eth_dev *dev)
+{
+ struct sxe2_adapter *adapter =
+ SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+ struct sxe2_drv_vsi_fc_get_resp fc_resp = {0};
+ int32_t ret;
+
+ if (!(adapter->cap_flags & SXE2_DEV_CAPS_OFFLOAD_FC_STATE)) {
+ adapter->fc_state_ctx.cfg_state = 0;
+ adapter->fc_state_ctx.curr_state = 0;
+ ret = 0;
+ goto l_end;
+ }
+ ret = sxe2_drv_fc_state_get(adapter, &fc_resp);
+ if (ret) {
+ PMD_LOG_ERR(INIT, "Failed to get fc state, ret=[%d]", ret);
+ goto l_end;
+ }
+ adapter->fc_state_ctx.cfg_state = fc_resp.fc_enable;
+ adapter->fc_state_ctx.curr_state = 0;
+l_end:
+ return ret;
+}
+static void sxe2_fc_state_uinit(struct rte_eth_dev *dev)
+{
+ struct sxe2_adapter *adapter =
+ SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+ adapter->fc_state_ctx.cfg_state = 0;
+ adapter->fc_state_ctx.curr_state = 0;
+}
+
uint32_t sxe2_sched_mode_get(struct sxe2_adapter *adapter)
{
uint32_t ret_mode = SXE2_SCHED_MODE_INVALID;
@@ -1646,18 +1888,6 @@ static int32_t sxe2_sched_uinit(struct rte_eth_dev *dev)
return ret;
}
-static int32_t sxe2_args_parse(struct rte_eth_dev *dev,
- __rte_unused struct sxe2_dev_kvargs_info *kvargs)
-{
- struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
- int32_t ret = 0;
- PMD_INIT_FUNC_TRACE();
-
- adapter->devargs.flow_dup_pattern_mode = SXE2_FLOW_DUP_PATTERN_DEFAULT;
-
- return ret;
-}
-
static int32_t sxe2_dev_init(struct rte_eth_dev *dev,
struct sxe2_dev_kvargs_info *kvargs)
{
@@ -1682,7 +1912,7 @@ static int32_t sxe2_dev_init(struct rte_eth_dev *dev,
ret = sxe2_args_parse(dev, kvargs);
if (ret) {
- PMD_LOG_ERR(INIT, "Failed to parse args, ret=[%d]", ret);
+ PMD_LOG_ERR(INIT, "Failed to parse devargs, ret=%d", ret);
goto l_end;
}
@@ -1752,6 +1982,12 @@ static int32_t sxe2_dev_init(struct rte_eth_dev *dev,
goto init_flow_err;
}
+ ret = sxe2_fc_state_init(dev);
+ if (ret) {
+ PMD_LOG_ERR(INIT, "Failed to init fc state, ret=%d", ret);
+ goto init_fc_state_err;
+ }
+
ret = sxe2_sched_init(dev);
if (ret) {
PMD_LOG_ERR(INIT, "Failed to init sched, ret=%d", ret);
@@ -1775,6 +2011,8 @@ static int32_t sxe2_dev_init(struct rte_eth_dev *dev,
init_xstats_err:
(void)sxe2_sched_uinit(dev);
init_sched_err:
+ sxe2_fc_state_uinit(dev);
+init_fc_state_err:
(void)sxe2_flow_uninit(dev);
init_flow_err:
init_rss_err:
@@ -1820,6 +2058,7 @@ static int32_t sxe2_dev_close(struct rte_eth_dev *dev)
sxe2_eth_uinit(dev);
sxe2_dev_pci_map_uinit(dev);
sxe2_free_repr_info(dev);
+ sxe2_fc_state_uinit(dev);
l_end:
return 0;
@@ -2125,6 +2364,13 @@ RTE_INIT(rte_sxe2_pmd_init)
RTE_PMD_EXPORT_NAME(net_sxe2);
RTE_PMD_REGISTER_PCI_TABLE(net_sxe2, pci_id_sxe2_tbl);
RTE_PMD_REGISTER_KMOD_DEP(net_sxe2, "* sxe2");
+RTE_PMD_REGISTER_PARAM_STRING(net_sxe2,
+ "flow-duplicate-pattern=<0|1> "
+ "function-flow-direct=<0|1> "
+ "fnav-stat-type=<1|2|3> "
+ "no-sched-mode=<0|1> "
+ "sched-layer-mode=<0-3> "
+ "rx-low-latency=<0|1>");
RTE_LOG_REGISTER_SUFFIX(sxe2_log_init, init, NOTICE);
RTE_LOG_REGISTER_SUFFIX(sxe2_log_driver, driver, NOTICE);
diff --git a/drivers/net/sxe2/sxe2_ethdev.h b/drivers/net/sxe2/sxe2_ethdev.h
index b63238f423..f00a69112a 100644
--- a/drivers/net/sxe2/sxe2_ethdev.h
+++ b/drivers/net/sxe2/sxe2_ethdev.h
@@ -310,6 +310,11 @@ struct sxe2_filter_context {
bool cur_l2_config;
};
+struct sxe2_fc_state_ctxt {
+ uint8_t curr_state;
+ uint8_t cfg_state;
+};
+
struct sxe2_adapter {
struct sxe2_common_device *cdev;
struct sxe2_dev_info dev_info;
@@ -331,6 +336,7 @@ struct sxe2_adapter {
struct sxe2_security_ctx security_ctx;
struct sxe2_repr_context repr_ctxt;
struct sxe2_switchdev_info switchdev_info;
+ struct sxe2_fc_state_ctxt fc_state_ctx;
bool rule_started;
bool flow_isolated;
bool flow_isolate_cfg;
diff --git a/drivers/net/sxe2/sxe2_flow.c b/drivers/net/sxe2/sxe2_flow.c
index 63cfc36968..1aa5813ee4 100644
--- a/drivers/net/sxe2/sxe2_flow.c
+++ b/drivers/net/sxe2/sxe2_flow.c
@@ -762,6 +762,7 @@ static int32_t sxe2_flow_validate_with_flow(struct rte_eth_dev *dev,
const struct rte_flow_action actions[],
struct rte_flow_error *error)
{
+ struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
int32_t ret = 0;
struct sxe2_flow *flow = NULL;
@@ -804,9 +805,11 @@ static int32_t sxe2_flow_validate_with_flow(struct rte_eth_dev *dev,
ret = sxe2_flow_check_flow_list_duplicate(dev, flow_list);
if (ret != 0) {
- rte_flow_error_set(error, EEXIST, RTE_FLOW_ERROR_TYPE_ITEM,
- NULL, "Duplicate flow.");
- PMD_LOG_ERR(DRV, "Duplicate flow.");
+ rte_flow_error_set(error, EEXIST, RTE_FLOW_ERROR_TYPE_ITEM, NULL,
+ adapter->devargs.flow_dup_pattern_mode ?
+ "Duplicate flow pattern." :
+ "Duplicate flow pattern is not allowed.");
+ PMD_LOG_ERR(DRV, "Duplicate flow pattern.");
goto l_end;
}
l_end:
diff --git a/drivers/net/sxe2/sxe2_irq.c b/drivers/net/sxe2/sxe2_irq.c
index d8e0b19463..3306504761 100644
--- a/drivers/net/sxe2/sxe2_irq.c
+++ b/drivers/net/sxe2/sxe2_irq.c
@@ -10,6 +10,7 @@
#include <rte_alarm.h>
#include <fcntl.h>
#include <rte_stdatomic.h>
+#include <rte_common.h>
#include "sxe2_ethdev.h"
#include "sxe2_irq.h"
@@ -47,6 +48,31 @@ static struct sxe2_event_handler event_handler = {
static RTE_ATOMIC(uint32_t)event_thread_run;
+static int32_t sxe2_fc_state_callback(struct rte_eth_dev *dev)
+{
+ struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+ struct sxe2_drv_vsi_fc_get_resp fc_resp = {0};
+ int32_t ret;
+
+ if (!(adapter->cap_flags & SXE2_DEV_CAPS_OFFLOAD_FC_STATE)) {
+ ret = 0;
+ goto l_end;
+ }
+ ret = sxe2_drv_fc_state_get(adapter, &fc_resp);
+ if (ret) {
+ PMD_LOG_ERR(INIT, "Failed to get fc state, ret=[%d]", ret);
+ goto l_end;
+ }
+ adapter->fc_state_ctx.cfg_state = fc_resp.fc_enable;
+ if (dev->data->dev_started) {
+ PMD_LOG_NOTICE(DRV, "Interrupt event: FC status changed."
+ "cfg_state:%u curr_state:%u",
+ adapter->fc_state_ctx.cfg_state,
+ adapter->fc_state_ctx.curr_state);
+ }
+l_end:
+ return ret;
+}
static void sxe2_event_irq_common_handler(struct sxe2_adapter *adapter, uint64_t oicr)
{
@@ -68,6 +94,10 @@ static void sxe2_event_irq_common_handler(struct sxe2_adapter *adapter, uint64_t
PMD_DEV_LOG_INFO(adapter, DRV, "event notify legacy");
(void)sxe2_switchdev_notify_callback(adapter, false);
}
+ if (oicr & RTE_BIT32(SXE2_COM_FC_ST_CHANGE)) {
+ PMD_DEV_LOG_INFO(adapter, DRV, "fc event notify legacy");
+ (void)sxe2_fc_state_callback(dev);
+ }
}
static uint32_t sxe2_event_intr_handle(void *param __rte_unused)
diff --git a/drivers/net/sxe2/sxe2_rx.c b/drivers/net/sxe2/sxe2_rx.c
index 820d4f0620..d700c60083 100644
--- a/drivers/net/sxe2/sxe2_rx.c
+++ b/drivers/net/sxe2/sxe2_rx.c
@@ -467,12 +467,24 @@ int32_t __rte_cold sxe2_rx_queue_start(struct rte_eth_dev *dev, uint16_t rx_queu
int32_t __rte_cold sxe2_rxqs_all_start(struct rte_eth_dev *dev)
{
struct rte_eth_dev_data *data = dev->data;
+ struct sxe2_adapter *adapter = SXE2_DEV_PRIVATE_TO_ADAPTER(dev);
+ struct sxe2_drv_vsi_fc_get_resp fc_resp = {0};
struct sxe2_rx_queue *rxq;
uint16_t nb_rxq;
uint16_t nb_started_rxq;
int32_t ret;
PMD_INIT_FUNC_TRACE();
+ if (adapter->cap_flags & SXE2_DEV_CAPS_OFFLOAD_FC_STATE) {
+ ret = sxe2_drv_fc_state_get(adapter, &fc_resp);
+ if (ret) {
+ PMD_LOG_ERR(RX, "Failed to get fc state, ret=[%d]", ret);
+ goto l_end;
+ }
+ adapter->fc_state_ctx.cfg_state = fc_resp.fc_enable;
+ adapter->fc_state_ctx.curr_state = adapter->fc_state_ctx.cfg_state;
+ }
+
for (nb_rxq = 0; nb_rxq < data->nb_rx_queues; nb_rxq++) {
rxq = dev->data->rx_queues[nb_rxq];
if (!rxq || rxq->rx_deferred_start)
--
2.52.0
^ permalink raw reply related
* [PATCH v4 23/23] net/sxe2: update sxe2 feature matrix docs
From: liujie5 @ 2026-06-19 8:12 UTC (permalink / raw)
To: stephen; +Cc: dev, Jie Liu
In-Reply-To: <20260618082723.571054-21-liujie5@linkdatatechnology.com>
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=a, Size: 10288 bytes --]
From: Jie Liu <liujie5@linkdatatechnology.com>
Update the sxe2.ini feature sheet to accurately reflect the recently
implemented hardware capabilities in the sxe2 PMD.
Signed-off-by: Jie Liu <liujie5@linkdatatechnology.com>
---
doc/guides/nics/features/sxe2.ini | 56 ++++++++++
doc/guides/nics/sxe2.rst | 164 ++++++++++++++++++++++++++++++
2 files changed, 220 insertions(+)
diff --git a/doc/guides/nics/features/sxe2.ini b/doc/guides/nics/features/sxe2.ini
index 09ba2f558c..3c1e6a8a39 100644
--- a/doc/guides/nics/features/sxe2.ini
+++ b/doc/guides/nics/features/sxe2.ini
@@ -7,17 +7,73 @@
; is selected.
;
[Features]
+Speed capabilities = Y
+Link status = Y
+Link status event = Y
+Rx interrupt = Y
Fast mbuf free = P
Free Tx mbuf on demand = Y
Burst mode info = Y
Queue start/stop = Y
+Power mgmt address monitor = Y
Buffer split on Rx = P
Scattered Rx = Y
+Traffic manager = Y
CRC offload = Y
+VLAN offload = Y
+QinQ offload = P
L3 checksum offload = Y
L4 checksum offload = Y
+Timestamp offload = P
+Inner L3 checksum = P
+Inner L4 checksum = P
Rx descriptor status = Y
Tx descriptor status = Y
+MTU update = Y
+TSO = P
+Promiscuous mode = Y
+Allmulticast mode = Y
+Unicast MAC filter = Y
+RSS hash = Y
+RSS key update = Y
+RSS reta update = Y
+VLAN filter = Y
+Inline crypto = Y
+Packet type parsing = Y
+Timesync = Y
+Basic stats = Y
+Extended stats = Y
+FW version = Y
+Module EEPROM dump = Y
+Multiprocess aware = Y
Linux = Y
x86-32 = Y
x86-64 = Y
+
+[rte_flow items]
+eth = P
+geneve = Y
+gre = Y
+gtpu = Y
+ipv4 = Y
+ipv6 = Y
+ipv6_frag_ext = Y
+nvgre = Y
+sctp = Y
+tcp = Y
+udp = Y
+vlan = P
+vxlan = Y
+vxlan_gpe = Y
+
+[rte_flow actions]
+count = Y
+drop = Y
+mark = Y
+passthru = Y
+port_representor = Y
+queue = Y
+represented_port = Y
+rss = Y
+send_to_kernel = Y
+port_id = Y
diff --git a/doc/guides/nics/sxe2.rst b/doc/guides/nics/sxe2.rst
index 539072b076..06541bf66e 100644
--- a/doc/guides/nics/sxe2.rst
+++ b/doc/guides/nics/sxe2.rst
@@ -35,3 +35,167 @@ preventing unauthorized access to random physical memory.
This capability allows the PMD to coexist with kernel network interfaces
which remain functional, although they stop receiving unicast packets
as long as they share the same MAC address.
+
+Configuration
+-------------
+
+Runtime Configuration
+~~~~~~~~~~~~~~~~~~~~~
+
+- ``Traffic Management Scheduling Levels``
+
+ The DPDK Traffic Management (rte_tm) APIs can be used to configure the Tx scheduler on the NIC.
+ The ``sched-layer-mode`` parameter can be used to set the number of scheduling levels
+ in the transmit scheduling hierarchy.
+ The provided value must be between 0 and 3.
+ If the value provided is greater than the number of levels supported by the HW,
+ the driver will use the hardware maximum value.
+
+- ``flow-duplicate-pattern`` parameter [int]
+
+ There are two options to choose:
+
+ - 0. Prevent insertion of flow rules with the same pattern items.
+ In this case, duplicate rules are rejected and error code EEXIST is returned.
+
+ - 1. Allow insertion of flow rules with the same pattern items.
+ In this case, the per-rule metadata flag ``switch_pattern_dup_allow`` is set
+ to instruct the hardware switch engine that rules with identical pattern
+ items are permitted.
+
+ This option only applies to the switch engine flow type.
+ For the Fnav flow engine type, duplicate rules are always rejected.
+
+ By default, the PMD will set this value to 1.
+
+- ``fnav-stat-type`` parameter [int]
+
+ This parameter controls the Fnav flow engine statistics type used
+ for flow rule hit counting (via ``rte_flow_query``).
+
+ - 1: Only count the number of packets.
+ - 2: Only count the number of bytes.
+ - 3: Count both packets and bytes (default).
+
+ Default value is 3 (count both packets and bytes).
+
+- ``drv-sw-stats`` parameter [int]
+
+ This parameter controls whether per-packet software statistics
+ (SW stats) are collected in the Rx data path.
+
+ Hardware packet statistic counters may be inaccurate for certain
+ packet types due to hardware design limitations.
+ When accuracy of Rx packet classification statistics is critical,
+ enabling this parameter allows the driver to accumulate statistics
+ in software as packets are received, providing an alternative
+ statistical path that bypasses hardware counter inaccuracies.
+
+ - 0: Disable software statistics collection (default).
+ The basic port statistics (``ipackets``, ``ibytes``) are reported
+ from the hardware counters.
+ - 1: Enable software statistics collection.
+ Per-packet software statistics are accumulated for unicast,
+ multicast, broadcast, and dropped packets in the Rx data path.
+
+ When enabled, the following extended statistics (xstats) are available:
+ ``rx_sw_unicast_packets``, ``rx_sw_multicast_packets``,
+ ``rx_sw_broadcast_packets``, ``rx_sw_drop_packets``,
+ and ``rx_sw_drop_bytes``.
+
+- ``no-sched-mode`` parameter [int]
+
+ This parameter enables non-scheduling mode (no-sched mode).
+ When enabled, the transmit path bypasses the hardware scheduling module
+ and packets are sent directly out through the port.
+ This results in lower transmit latency and higher throughput,
+ but Traffic Management (rte_tm) APIs are not supported in this mode.
+
+ - 0: Disable non-scheduling mode (default).
+ The transmit path goes through the hardware scheduling hierarchy.
+ Traffic Management (rte_tm) APIs can be used to configure the Tx scheduler.
+ - 1: Enable non-scheduling mode.
+ The transmit path bypasses the hardware scheduling module.
+ Packets are sent directly from the port at full speed without scheduling.
+ Traffic Management (rte_tm) APIs are not available in this mode.
+
+- ``rx-low-latency`` parameter [int]
+
+ This parameter controls the interrupt throttling (ITR) interval
+ for Rx queue interrupts.
+
+ When enabled, the driver sets a shorter interrupt coalescing timeout
+ (``SXE2_ITR_INTERVAL_LOW``, approximately 1 μs),
+ reducing the time between packet arrival and interrupt delivery to the CPU.
+ This lowers receive latency at the cost of increased CPU interrupt rate.
+
+ When disabled (default), the driver uses the normal interrupt throttling
+ interval (``SXE2_ITR_INTERVAL_NORMAL``, approximately 20 μs),
+ which reduces the CPU interrupt rate at the expense of higher receive latency.
+
+ - 0: Disable Rx low latency (default).
+ Normal interrupt throttling interval (~20 μs) is used.
+ - 1: Enable Rx low latency.
+ Low interrupt throttling interval (~1 μs) is used
+ for reduced receive latency.
+
+- ``function-flow-direct`` parameter [int]
+
+ This parameter controls whether flow rules from different functional units
+ (DPDK vs kernel driver) are isolated or combined when both drivers
+ control the same physical port.
+
+ When the DPDK PMD and the kernel network driver coexist on the same port,
+ flow rules may originate from either driver.
+ This parameter determines how the source VSI (Virtual Switch Interface)
+ of each flow rule is handled during hardware programming.
+
+ - 0 (default): Isolate flow rules between DPDK and kernel.
+ When ``flow_isolated`` is enabled (``rte_flow_isolate()`` called),
+ kernel-side flow rules take priority and DPDK-side flow rules are suppressed.
+ When ``flow_isolated`` is disabled, DPDK-side flow rules take priority
+ and kernel-side flow rules are suppressed.
+ Only one functional unit's flows are active at a time.
+
+ - 1: Allow direct flow rules from both DPDK and kernel simultaneously.
+ Both DPDK and kernel source VSIs are preserved in the hardware flow table.
+ Flow rules from both sides are programmed without isolation.
+
+ This option only applies to FNAV and ACL flow engine types
+ and does not apply to PF bond devices.
+
+Extended Statistics (xstats)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The PMD provides the following extended statistics (xstats) for detailed
+monitoring of receive-side packet classification and software-level accounting.
+The software statistics path is provided as a workaround for hardware
+counter inaccuracies on certain packet types --- it accumulates per-packet
+statistics directly in the Rx data path, ensuring that unicast, multicast,
+broadcast, and drop counts reflect the actual packets processed by the driver.
+
+Receive Software Statistics
+ These counters are collected in the Rx data path when ``drv-sw-stats=1``
+ is configured (see the ``drv-sw-stats`` devarg above).
+ When ``drv-sw-stats`` is disabled (default), these xstats report zero.
+
+ - ``rx_sw_unicast_packets``: Number of unicast packets received.
+ - ``rx_sw_multicast_packets``: Number of multicast packets received.
+ - ``rx_sw_broadcast_packets``: Number of broadcast packets received.
+ - ``rx_sw_drop_packets``: Number of packets dropped in the Rx data path.
+ - ``rx_sw_drop_bytes``: Number of bytes dropped in the Rx data path.
+
+ When ``drv-sw-stats`` is enabled, the basic counters ``ipackets`` and
+ ``ibytes`` (from ``rte_eth_stats``) also reflect the software-accumulated
+ packet and byte counts. Otherwise, they are reported from hardware counters.
+
+Fnav Flow Engine Statistics
+ The Fnav flow engine statistics type is controlled by the ``fnav-stat-type``
+ devarg (see above). Depending on the configuration:
+
+ - ``fnav-stat-type=1``: Only packet count is available.
+ - ``fnav-stat-type=2``: Only byte count is available.
+ - ``fnav-stat-type=3`` (default): Both packet and byte counts are available.
+
+ Flow query results (via ``rte_flow_query``) expose these per-flow counters
+ through the query API, not via xstats.
--
2.52.0
^ permalink raw reply related
* Re: [PATCH v1 1/1] net/i40e: do not reject RSS types parameter
From: Bruce Richardson @ 2026-06-19 8:44 UTC (permalink / raw)
To: Anatoly Burakov; +Cc: dev
In-Reply-To: <7f6f11fb07bc550784335123af6a270cd514f7d7.1781692817.git.anatoly.burakov@intel.com>
On Wed, Jun 17, 2026 at 11:40:26AM +0100, Anatoly Burakov wrote:
> After the recent refactor, global RSS configuration started rejecting the
> RSS types parameter because it was not used for anything.
>
> However, because testpmd will specify RSS types by default if omitted from
> the flow command (i.e. `actions rss queues 0 1 end` vs `actions rss queues
> 0 1 end types end`), RSS action based flows that are created without
> mentioning RSS types will still have RSS types as non-zero, causing flow
> creation failures not directly related to the pattern.
>
> Fix it by printing a warning but allowing spurious RSS types as opposed to
> rejecting it outright.
>
> Fixes: 0185303c2e24 ("net/i40e: refactor RSS flow parameter checks")
>
> Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
> ---
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Applied to dpdk-next-net-intel.
Thanks,
/Bruce
> drivers/net/intel/i40e/i40e_hash.c | 18 ++++++++++++------
> 1 file changed, 12 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/net/intel/i40e/i40e_hash.c b/drivers/net/intel/i40e/i40e_hash.c
> index 8b80d0a91c..5a6543a9ec 100644
> --- a/drivers/net/intel/i40e/i40e_hash.c
> +++ b/drivers/net/intel/i40e/i40e_hash.c
> @@ -1089,12 +1089,18 @@ i40e_hash_validate_rss_common(const struct rte_flow_action_rss *rss_act,
> "Symmetric hash function not supported without specific patterns");
> }
>
> - /* hash types are not supported for global RSS configuration */
> - if (rss_act->types != 0) {
> - return rte_flow_error_set(error, EINVAL,
> - RTE_FLOW_ERROR_TYPE_ACTION_CONF, rss_act,
> - "RSS types not supported without a pattern");
> - }
> + /*
> + * When RSS types is not specified in testpmd, it will set up a default
> + * RSS types value for the flow. Even though no hash engine part calling
> + * this particular function will use RSS types parameter for anything,
> + * we cannot reject having it because it is extra effort for testpmd
> + * user to avoid specifying it.
> + *
> + * So, instead, accept types value even though we are not using it for
> + * anything, but produce a warning for the user.
> + */
> + if (rss_act->types != 0)
> + PMD_DRV_LOG(WARNING, "RSS types specified but will not be used");
>
> /* check RSS key length if it is specified */
> if (rss_act->key_len != 0 && rss_act->key_len != I40E_RSS_KEY_LEN) {
> --
> 2.47.3
>
^ permalink raw reply
* Re: [PATCH v2] net/iavf: report selected burst mode when no-poll active
From: Bruce Richardson @ 2026-06-19 9:27 UTC (permalink / raw)
To: Ciara Loftus; +Cc: dev, stable
In-Reply-To: <20260618121711.371057-1-ciara.loftus@intel.com>
On Thu, Jun 18, 2026 at 12:17:11PM +0000, Ciara Loftus wrote:
> When the no-poll feature is enabled (it is enabled by default), the
> device burst functions point at the no-poll wrapper for the lifetime of
> the port. As the wrapper occupies the "Disabled" slot in the burst mode
> path-info tables, the Rx/Tx burst mode was always reported as "Disabled"
> regardless of link state, even though the wrapper only drops traffic
> while the link is down and otherwise dispatches to the selected path.
>
> Report the burst mode of the selected path directly by indexing the
> path-info tables with the selected path type. This fixes the misreport
> while the no-poll wrapper is active and also simplifies the burst mode
> lookup: the previous pointer comparison and table search loop are no
> longer needed.
>
> Fixes: 0d5a856f5be9 ("net/iavf: support Rx/Tx burst mode info")
> Cc: stable@dpdk.org
>
> Signed-off-by: Ciara Loftus <ciara.loftus@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Applied to dpdk-next-net-intel.
Thanks,
/Bruce
^ permalink raw reply
* Re: [PATCH v2 1/4] net/bonding: make 8023ad enable function void
From: Bruce Richardson @ 2026-06-19 9:48 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: dev, Chas Williams, Min Hu (Connor)
In-Reply-To: <20260529000157.235931-2-stephen@networkplumber.org>
On Thu, May 28, 2026 at 04:59:13PM -0700, Stephen Hemminger wrote:
> The function never returns an error. Cleanup the call sites.
>
> The 8023ad disable function was never implemented,
> remove prototype.
>
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
> ---
While I'd actually suggest this would be better as two patches - one for
enable, one for disable, it's probably ok to keep them combined. If
splitting in a new version, please keep my ack.
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
> drivers/net/bonding/eth_bond_8023ad_private.h | 17 +----------------
> drivers/net/bonding/rte_eth_bond_8023ad.c | 4 +---
> drivers/net/bonding/rte_eth_bond_pmd.c | 7 +++----
> 3 files changed, 5 insertions(+), 23 deletions(-)
>
> diff --git a/drivers/net/bonding/eth_bond_8023ad_private.h b/drivers/net/bonding/eth_bond_8023ad_private.h
> index ab7d15f81a..bd7a5848de 100644
> --- a/drivers/net/bonding/eth_bond_8023ad_private.h
> +++ b/drivers/net/bonding/eth_bond_8023ad_private.h
> @@ -209,25 +209,10 @@ bond_mode_8023ad_setup(struct rte_eth_dev *dev,
> * @internal
> *
> * Enables 802.1AX mode and all active members on bonding interface.
> - *
> - * @param dev Bonding interface
> - * @return
> - * 0 on success, negative value otherwise.
> */
> -int
> +void
> bond_mode_8023ad_enable(struct rte_eth_dev *dev);
>
> -/**
> - * @internal
> - *
> - * Disables 802.1AX mode of the bonding interface and members.
> - *
> - * @param dev Bonding interface
> - * @return
> - * 0 on success, negative value otherwise.
> - */
> -int bond_mode_8023ad_disable(struct rte_eth_dev *dev);
> -
> /**
> * @internal
> *
> diff --git a/drivers/net/bonding/rte_eth_bond_8023ad.c b/drivers/net/bonding/rte_eth_bond_8023ad.c
> index ba88f6d261..eba713e381 100644
> --- a/drivers/net/bonding/rte_eth_bond_8023ad.c
> +++ b/drivers/net/bonding/rte_eth_bond_8023ad.c
> @@ -1287,7 +1287,7 @@ bond_mode_8023ad_setup(struct rte_eth_dev *dev,
> bond_mode_8023ad_start(dev);
> }
>
> -int
> +void
> bond_mode_8023ad_enable(struct rte_eth_dev *bond_dev)
> {
> struct bond_dev_private *internals = bond_dev->data->dev_private;
> @@ -1296,8 +1296,6 @@ bond_mode_8023ad_enable(struct rte_eth_dev *bond_dev)
> for (i = 0; i < internals->active_member_count; i++)
> bond_mode_8023ad_activate_member(bond_dev,
> internals->active_members[i]);
> -
> - return 0;
> }
>
> int
> diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c b/drivers/net/bonding/rte_eth_bond_pmd.c
> index 96725071da..7fcb3ec7d7 100644
> --- a/drivers/net/bonding/rte_eth_bond_pmd.c
> +++ b/drivers/net/bonding/rte_eth_bond_pmd.c
> @@ -1619,8 +1619,7 @@ bond_ethdev_mode_set(struct rte_eth_dev *eth_dev, uint8_t mode)
> eth_dev->rx_pkt_burst = bond_ethdev_rx_burst;
> break;
> case BONDING_MODE_8023AD:
> - if (bond_mode_8023ad_enable(eth_dev) != 0)
> - return -1;
> + bond_mode_8023ad_enable(eth_dev);
>
> if (internals->mode4.dedicated_queues.enabled == 0) {
> eth_dev->rx_pkt_burst = bond_ethdev_rx_burst_8023ad;
> @@ -1641,13 +1640,13 @@ bond_ethdev_mode_set(struct rte_eth_dev *eth_dev, uint8_t mode)
> eth_dev->rx_pkt_burst = bond_ethdev_rx_burst_active_backup;
> break;
> case BONDING_MODE_ALB:
> - if (bond_mode_alb_enable(eth_dev) != 0)
> - return -1;
> + bond_mode_alb_enable(eth_dev);
>
> eth_dev->tx_pkt_burst = bond_ethdev_tx_burst_alb;
> eth_dev->rx_pkt_burst = bond_ethdev_rx_burst_alb;
> break;
> default:
> + RTE_BOND_LOG(ERR, "Invalid mode %#x", mode);
> return -1;
> }
>
> --
> 2.53.0
>
^ permalink raw reply
* Re: [PATCH v2 2/4] net/bonding: check mode before setting dedicated queues
From: Bruce Richardson @ 2026-06-19 9:52 UTC (permalink / raw)
To: Stephen Hemminger
Cc: dev, stable, Chas Williams, Min Hu (Connor), Declan Doherty,
Tomasz Kulasek
In-Reply-To: <20260529000157.235931-3-stephen@networkplumber.org>
On Thu, May 28, 2026 at 04:59:14PM -0700, Stephen Hemminger wrote:
> The calls to enable and disable dedicated queues are missing
> check for mode.
>
> Fixes: 112891cd27e5 ("net/bonding: add dedicated HW queues for LACP control")
> Cc: stable@dpdk.org
>
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
> ---
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
^ permalink raw reply
* Re: [PATCH v2 3/4] net/bonding: prevent crash on Rx/Tx from secondary process
From: Bruce Richardson @ 2026-06-19 9:54 UTC (permalink / raw)
To: Stephen Hemminger
Cc: dev, stable, Chas Williams, Min Hu (Connor), Anatoly Burakov,
Qi Zhang
In-Reply-To: <20260529000157.235931-4-stephen@networkplumber.org>
On Thu, May 28, 2026 at 04:59:15PM -0700, Stephen Hemminger wrote:
> The bonding PMD's secondary process attach path registered the
> ethdev but never installed rx_pkt_burst or tx_pkt_burst, leaving
> both as NULL. Any rx_burst or tx_burst call from a secondary
> process therefore crashed with a NULL pointer dereference.
>
> Fully sharing bonding state across processes would be
> overly complex. Instead, install blackhole burst functions
> in the secondary so the data path is safe by default. Rx returns 0
> and Tx frees the mbufs and reports them as transmitted,
> matching /dev/null semantics so applications do not spin
> retrying. Each stub logs once at NOTICE level on first use.
>
> Also reject bond mode changes from a secondary process.
> rte_eth_bond_mode_set() is callable from secondary, and
> without this guard it would overwrite the secondary's safe
> burst stubs with real per-mode functions whose internal
> state is not valid outside the primary, reintroducing the
> crash by another path.
>
> This keeps secondary support available for the more
> common use cases of procinfo and packet capture.
>
> Bugzilla ID: 1698
> Fixes: 4852aa8f6e21 ("drivers/net: enable hotplug on secondary process")
> Cc: stable@dpdk.org
>
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
> ---
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
^ permalink raw reply
* Re: [PATCH v2 4/4] net/bonding: remove redundant function names from log
From: Bruce Richardson @ 2026-06-19 9:55 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: dev, Chas Williams, Min Hu (Connor)
In-Reply-To: <20260529000157.235931-5-stephen@networkplumber.org>
On Thu, May 28, 2026 at 04:59:16PM -0700, Stephen Hemminger wrote:
> The function name is already printed as part of RTE_BOND_LOG().
> Fix grammar in log messages.
>
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
> ---
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
^ permalink raw reply
* [PATCH v3 00/18] net/dpaa: bug fixes for bus, net and fmlib drivers
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev
In-Reply-To: <20260619060916.485258-1-hemant.agrawal@nxp.com>
This series contains bug fixes for the DPAA PMD (bus/dpaa, net/dpaa,
net/dpaa/fmlib and dma/dpaa).
v3 changes (AI code review feedback):
- P05: Clarify commit message: p_dev == NULL is equivalent to h_scheme == NULL
since p_dev = (t_device *)h_scheme; consistent with all sibling functions
- P16: Add comment explaining the intentional loop continuation; clarify
commit message about the loop design
- P17: Add DPAA_DP_LOG(WARNING) before silent return on l3_len == 0 to
aid debugging of corrupt/uninitialized mbufs
v2 changes:
- P05: Fix commit message API name
- P08: Guard DPAA_PUSH_QUEUES_NUMBER env-var for LS1043A (errata)
- P09: Document dpaa_finish() removal
- P10: Fix wrong Fixes: tag
- P11: Split into two patches with correct Fixes: tags
- P13: Also fix rx_buf_diallocate -> rx_buf_deallocate
All patches are bug fixes tagged with Fixes: and Cc: stable@dpdk.org.
Gagandeep Singh (3):
bus/dpaa: fix device probe issue
net/dpaa: fix device remove
net/dpaa: fix invalid check on interrupt unregister
Hemant Agrawal (11):
bus/dpaa: fix error handling of qman_create_fq
bus/dpaa: fix fqid endianness
bus/dpaa: fix error handling in qman_query
net/dpaa: fix modify cgr to use index
bus/dpaa: fix fd leak for ccsr mmap
net/dpaa: fix xstat name for tx undersized counter
net/dpaa: fix xstat string typos in BMI stats table
net/dpaa: remove duplicate ptype entries
net/dpaa: fix wrong buffer in xstats get by id
net/dpaa: fix null l3_len check in checksum offload
net/dpaa: fix mbuf leak in SG fd creation
Jun Yang (1):
bus/dpaa: fix BMI RX stats register offset
Prashant Gupta (1):
net/dpaa/fmlib: add null check in scheme delete
Vanshika Shukla (2):
net/dpaa: fix port_handle leak in fm_prev_cleanup
dma/dpaa: fix out-of-bounds access in SG descriptor enqueue
drivers/bus/dpaa/base/qbman/bman_driver.c | 3 ++-
drivers/bus/dpaa/base/qbman/qman.c | 11 ++++++---
drivers/bus/dpaa/base/qbman/qman_driver.c | 6 ++---
drivers/bus/dpaa/dpaa_bus.c | 17 ++++++-------
drivers/bus/dpaa/include/fman.h | 6 ++---
drivers/dma/dpaa/dpaa_qdma.c | 7 +++++-
drivers/net/dpaa/dpaa_ethdev.c | 30 +++++++++++------------
drivers/net/dpaa/dpaa_flow.c | 4 +++
drivers/net/dpaa/dpaa_rxtx.c | 5 ++++
drivers/net/dpaa/fmlib/fm_lib.c | 3 +++
10 files changed, 56 insertions(+), 36 deletions(-)
--
2.43.0
^ permalink raw reply
* [PATCH v3 01/18] bus/dpaa: fix error handling of qman_create_fq
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
Fix the error handling path in qman_create_fq() to properly
return error codes instead of silently ignoring failures.
Fixes: c47ff048b99a ("bus/dpaa: add QMAN driver core routines")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
---
drivers/bus/dpaa/base/qbman/qman.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/bus/dpaa/base/qbman/qman.c b/drivers/bus/dpaa/base/qbman/qman.c
index 5534e1846c..9a99eb9785 100644
--- a/drivers/bus/dpaa/base/qbman/qman.c
+++ b/drivers/bus/dpaa/base/qbman/qman.c
@@ -1579,6 +1579,9 @@ int qman_create_fq(u32 fqid, u32 flags, struct qman_fq *fq)
err:
if (flags & QMAN_FQ_FLAG_DYNAMIC_FQID)
qman_release_fqid(fqid);
+#ifdef CONFIG_FSL_QMAN_FQ_LOOKUP
+ clear_fq_table_entry(fq->key);
+#endif
return -EIO;
}
--
2.43.0
^ permalink raw reply related
* [PATCH v3 02/18] bus/dpaa: fix fqid endianness
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
In qman_fq_flow_control(), the fqid field in the management
command was set using the host-endian fqid instead of the
pre-converted big-endian fqid_be. Fix it to use fqid_be
consistent with all other enqueue paths.
Fixes: c47ff048b99a ("bus/dpaa: add QMAN driver core routines")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
---
drivers/bus/dpaa/base/qbman/qman.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/bus/dpaa/base/qbman/qman.c b/drivers/bus/dpaa/base/qbman/qman.c
index 9a99eb9785..2da1b3e3f7 100644
--- a/drivers/bus/dpaa/base/qbman/qman.c
+++ b/drivers/bus/dpaa/base/qbman/qman.c
@@ -1921,7 +1921,7 @@ int qman_fq_flow_control(struct qman_fq *fq, int xon)
goto out;
}
mcc = qm_mc_start(&p->p);
- mcc->alterfq.fqid = fq->fqid;
+ mcc->alterfq.fqid = fq->fqid_be;
mcc->alterfq.count = 0;
myverb = xon ? QM_MCC_VERB_ALTER_FQXON : QM_MCC_VERB_ALTER_FQXOFF;
--
2.43.0
^ permalink raw reply related
* [PATCH v3 03/18] bus/dpaa: fix error handling in qman_query
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
Optimize error handling in qman_query() to avoid redundant
checks and properly propagate error codes.
Fixes: 06268e2cb175 ("bus/dpaa: query queue frame count support")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
---
drivers/bus/dpaa/base/qbman/qman.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/bus/dpaa/base/qbman/qman.c b/drivers/bus/dpaa/base/qbman/qman.c
index 2da1b3e3f7..d289df2d33 100644
--- a/drivers/bus/dpaa/base/qbman/qman.c
+++ b/drivers/bus/dpaa/base/qbman/qman.c
@@ -1955,11 +1955,11 @@ int qman_query_fq(struct qman_fq *fq, struct qm_fqd *fqd)
cpu_relax();
DPAA_ASSERT((mcr->verb & QM_MCR_VERB_MASK) == QM_MCR_VERB_QUERYFQ);
res = mcr->result;
- if (res == QM_MCR_RESULT_OK)
- *fqd = mcr->queryfq.fqd;
- hw_fqd_to_cpu(fqd);
if (res != QM_MCR_RESULT_OK)
return -EIO;
+
+ *fqd = mcr->queryfq.fqd;
+ hw_fqd_to_cpu(fqd);
return 0;
}
--
2.43.0
^ permalink raw reply related
* [PATCH v3 04/18] net/dpaa: fix modify cgr to use index
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
In dpaa_modify_cgr(), the code was always using the pointer to
the first CGR element instead of indexing by the queue index.
Fix it to use the correct CGR entry by index.
Fixes: 62f53995caaf ("net/dpaa: add frame count based tail drop with CGR")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
---
drivers/net/dpaa/dpaa_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 9f976d179b..424458857e 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -1304,7 +1304,7 @@ int dpaa_eth_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
rxq->nb_desc = nb_desc;
/* Enable tail drop with cgr on this queue */
qm_cgr_cs_thres_set64(&cgr_opts.cgr.cs_thres, nb_desc, 0);
- ret = qman_modify_cgr(dpaa_intf->cgr_rx, 0, &cgr_opts);
+ ret = qman_modify_cgr(&dpaa_intf->cgr_rx[queue_idx], 0, &cgr_opts);
if (ret) {
DPAA_PMD_WARN(
"rx taildrop modify fail on fqid %d (ret=%d)",
--
2.43.0
^ permalink raw reply related
* [PATCH v3 05/18] net/dpaa/fmlib: add null check in scheme delete
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable, Prashant Gupta
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
From: Prashant Gupta <prashant.gupta_3@nxp.com>
Add a null pointer check at the entry of fm_pcd_kg_scheme_delete().
Since p_dev is assigned directly from h_scheme via a cast
(t_device *)h_scheme, checking p_dev == NULL is equivalent to
checking h_scheme == NULL. This matches the defensive pattern used
in all sibling functions in fm_lib.c and returns E_NO_DEVICE on a
null handle.
Fixes: 663ff698e38f ("net/dpaa: support VSP in fmlib")
Cc: stable@dpdk.org
Signed-off-by: Prashant Gupta <prashant.gupta_3@nxp.com>
---
drivers/net/dpaa/fmlib/fm_lib.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/dpaa/fmlib/fm_lib.c b/drivers/net/dpaa/fmlib/fm_lib.c
index b35feba004..65a818372e 100644
--- a/drivers/net/dpaa/fmlib/fm_lib.c
+++ b/drivers/net/dpaa/fmlib/fm_lib.c
@@ -305,6 +305,9 @@ fm_pcd_kg_scheme_delete(t_handle h_scheme)
_fml_dbg("Calling...");
+ if (p_dev == NULL)
+ return E_NO_DEVICE;
+
p_pcd_dev = (t_device *)p_dev->h_user_priv;
id.obj = UINT_TO_PTR(p_dev->id);
--
2.43.0
^ permalink raw reply related
* [PATCH v3 06/18] bus/dpaa: fix BMI RX stats register offset
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable, Jun Yang
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
From: Jun Yang <jun.yang@nxp.com>
Fix incorrect register offset for BMI RX statistics counters
in the fman.h header. The wrong offset caused incorrect stats
values to be reported.
Fixes: 0095306cdbda ("bus/dpaa: add FMan node")
Cc: stable@dpdk.org
Signed-off-by: Jun Yang <jun.yang@nxp.com>
---
drivers/bus/dpaa/include/fman.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/bus/dpaa/include/fman.h b/drivers/bus/dpaa/include/fman.h
index c33fe81516..6e3abf1b50 100644
--- a/drivers/bus/dpaa/include/fman.h
+++ b/drivers/bus/dpaa/include/fman.h
@@ -2,7 +2,7 @@
*
* Copyright 2010-2012 Freescale Semiconductor, Inc.
* All rights reserved.
- * Copyright 2019-2024 NXP
+ * Copyright 2019-2026 NXP
*
*/
@@ -263,8 +263,8 @@ struct rx_bmi_regs {
/**< Buffer Manager pool Information-*/
uint32_t fmbm_acnt[FMAN_PORT_MAX_EXT_POOLS_NUM];
/**< Allocate Counter-*/
- uint32_t reserved0120[16];
- /**< 0x130/0x140 - 0x15F reserved -*/
+ uint32_t reserved0140[8];
+ /**< 0x140 - 0x15F reserved -*/
uint32_t fmbm_rcgm[FMAN_PORT_CG_MAP_NUM];
/**< Congestion Group Map*/
uint32_t fmbm_mpd; /**< BM Pool Depletion */
--
2.43.0
^ permalink raw reply related
* [PATCH v3 07/18] bus/dpaa: fix fd leak for ccsr mmap
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable, Jun Yang
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
The CCSR file descriptor was kept open after mmap() was done.
Close the fd immediately after mmap() as it is no longer needed,
preventing a file descriptor leak.
Fixes: 8e253882cd31 ("bus/dpaa: support interrupt portal based fd")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Signed-off-by: Jun Yang <jun.yang@nxp.com>
---
drivers/bus/dpaa/base/qbman/bman_driver.c | 3 ++-
drivers/bus/dpaa/base/qbman/qman_driver.c | 6 +++---
2 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/drivers/bus/dpaa/base/qbman/bman_driver.c b/drivers/bus/dpaa/base/qbman/bman_driver.c
index 23e44ac10b..71a2028383 100644
--- a/drivers/bus/dpaa/base/qbman/bman_driver.c
+++ b/drivers/bus/dpaa/base/qbman/bman_driver.c
@@ -145,7 +145,7 @@ void bman_thread_irq(void)
int bman_init_ccsr(const struct device_node *node)
{
- static int ccsr_map_fd;
+ int ccsr_map_fd;
uint64_t phys_addr;
const uint32_t *bman_addr;
uint64_t regs_size;
@@ -169,6 +169,7 @@ int bman_init_ccsr(const struct device_node *node)
bman_ccsr_map = mmap(NULL, regs_size, PROT_READ |
PROT_WRITE, MAP_SHARED, ccsr_map_fd, phys_addr);
+ close(ccsr_map_fd);
if (bman_ccsr_map == MAP_FAILED) {
pr_err("Can not map BMan CCSR base Bman: "
"0x%x Phys: 0x%" PRIx64 " size 0x%" PRIu64,
diff --git a/drivers/bus/dpaa/base/qbman/qman_driver.c b/drivers/bus/dpaa/base/qbman/qman_driver.c
index 3bab8b8337..45b094e0c6 100644
--- a/drivers/bus/dpaa/base/qbman/qman_driver.c
+++ b/drivers/bus/dpaa/base/qbman/qman_driver.c
@@ -1,7 +1,7 @@
/* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0
*
* Copyright 2008-2016 Freescale Semiconductor Inc.
- * Copyright 2017-2022, 2025 NXP
+ * Copyright 2017-2022, 2025-2026 NXP
*
*/
@@ -270,7 +270,7 @@ int qman_global_init(void)
const struct device_node *dt_node;
size_t lenp;
const u32 *chanid;
- static int ccsr_map_fd;
+ int ccsr_map_fd;
const uint32_t *qman_addr;
uint64_t phys_addr;
uint64_t regs_size;
@@ -358,9 +358,9 @@ int qman_global_init(void)
pr_err("Can not open /dev/mem for qman ccsr map\n");
return ccsr_map_fd;
}
-
qman_ccsr_map = mmap(NULL, regs_size, PROT_READ | PROT_WRITE,
MAP_SHARED, ccsr_map_fd, phys_addr);
+ close(ccsr_map_fd);
if (qman_ccsr_map == MAP_FAILED) {
pr_err("Can not map qman ccsr base\n");
return -EINVAL;
--
2.43.0
^ permalink raw reply related
* [PATCH v3 08/18] bus/dpaa: fix device probe issue
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable, Gagandeep Singh
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
From: Gagandeep Singh <g.singh@nxp.com>
Remove an unintended early return in the LS1043 SoC version check
that was preventing device probing from completing successfully on
LS1043A platforms.
The early return did two things: set max_push_rxq_num = 0 and skip
the DPAA_PUSH_QUEUES_NUMBER env-var override. With the return gone,
the env-var could inadvertently re-enable push mode on LS1043A, which
must remain disabled due to the FMAN push-mode errata handled in
dpaa_rxtx.c. Guard the env-var override so it only applies to
non-LS1043A SoCs.
Fixes: 164e9e13e50f ("bus/dpaa: enhance SoC version")
Cc: stable@dpdk.org
Signed-off-by: Gagandeep Singh <g.singh@nxp.com>
---
drivers/bus/dpaa/dpaa_bus.c | 17 ++++++++---------
1 file changed, 8 insertions(+), 9 deletions(-)
diff --git a/drivers/bus/dpaa/dpaa_bus.c b/drivers/bus/dpaa/dpaa_bus.c
index ee467b94d5..02a8c5882e 100644
--- a/drivers/bus/dpaa/dpaa_bus.c
+++ b/drivers/bus/dpaa/dpaa_bus.c
@@ -1,6 +1,6 @@
/* SPDX-License-Identifier: BSD-3-Clause
*
- * Copyright 2017-2025 NXP
+ * Copyright 2017-2026 NXP
*
*/
/* System headers */
@@ -724,18 +724,17 @@ rte_dpaa_bus_scan(void)
dpaa_bus.svr_ver);
}
- /* Disabling the default push mode for LS1043A */
+ /* Disabling the default push mode for LS1043A due to errata */
if (dpaa_bus.svr_ver == SVR_LS1043A_FAMILY) {
dpaa_bus.max_push_rxq_num = 0;
- return 0;
+ } else {
+ penv = getenv("DPAA_PUSH_QUEUES_NUMBER");
+ if (penv)
+ dpaa_bus.max_push_rxq_num = atoi(penv);
+ if (dpaa_bus.max_push_rxq_num > DPAA_MAX_PUSH_MODE_QUEUE)
+ dpaa_bus.max_push_rxq_num = DPAA_MAX_PUSH_MODE_QUEUE;
}
- penv = getenv("DPAA_PUSH_QUEUES_NUMBER");
- if (penv)
- dpaa_bus.max_push_rxq_num = atoi(penv);
- if (dpaa_bus.max_push_rxq_num > DPAA_MAX_PUSH_MODE_QUEUE)
- dpaa_bus.max_push_rxq_num = DPAA_MAX_PUSH_MODE_QUEUE;
-
/* Device list creation is only done once */
if (!process_once) {
rte_dpaa_bus_dev_build();
--
2.43.0
^ permalink raw reply related
* [PATCH v3 09/18] net/dpaa: fix device remove
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable, Gagandeep Singh
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
From: Gagandeep Singh <g.singh@nxp.com>
Add a check to avoid closing a device that is already closed,
preventing a double-close condition during device removal.
Note: this also removes the explicit dpaa_finish() call that was
made at last-device remove time (!dpaa_valid_dev). dpaa_finish() is
registered as RTE_FINI_PRIO(dpaa_finish, 103) and will still run at
process exit, so for the normal run-then-exit path behaviour is
unchanged. For a remove-all-then-re-probe scenario, is_global_init
will remain set until exit; re-probe in a running process is not a
supported use case for this driver.
Fixes: 78ea4b4fcb52 ("bus/dpaa: improve cleanup")
Cc: stable@dpdk.org
Signed-off-by: Gagandeep Singh <g.singh@nxp.com>
---
drivers/net/dpaa/dpaa_ethdev.c | 13 +++++++------
1 file changed, 7 insertions(+), 6 deletions(-)
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 424458857e..9a9c5ee817 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -2674,18 +2674,19 @@ static int
rte_dpaa_remove(struct rte_dpaa_device *dpaa_dev)
{
struct rte_eth_dev *eth_dev;
- int ret;
+ int ret = 0;
PMD_INIT_FUNC_TRACE();
eth_dev = dpaa_dev->eth_dev;
- dpaa_eth_dev_close(eth_dev);
- ret = rte_eth_dev_release_port(eth_dev);
+ if (eth_dev->state != RTE_ETH_DEV_UNUSED) {
+ dpaa_eth_dev_close(eth_dev);
+ ret = rte_eth_dev_release_port(eth_dev);
+ }
dpaa_valid_dev--;
- if (!dpaa_valid_dev) {
+ if (!dpaa_valid_dev)
rte_mempool_free(dpaa_tx_sg_pool);
- dpaa_finish();
- }
+
return ret;
}
--
2.43.0
^ permalink raw reply related
* [PATCH v3 10/18] net/dpaa: fix invalid check on interrupt unregister
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable, Gagandeep Singh
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
From: Gagandeep Singh <g.singh@nxp.com>
rte_intr_callback_unregister() returns the number of callbacks
removed (>= 1) on success and a negative value on failure. The
previous check 'if (ret)' logged a spurious warning on every
successful unregister. Fix it to 'if (ret < 0)'.
Fixes: 2aa10990a8dd ("bus/dpaa: enable link state interrupt")
Cc: stable@dpdk.org
Signed-off-by: Gagandeep Singh <g.singh@nxp.com>
---
drivers/net/dpaa/dpaa_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index 9a9c5ee817..c143e66f77 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -559,7 +559,7 @@ static int dpaa_eth_dev_close(struct rte_eth_dev *dev)
}
ret = rte_intr_callback_unregister(intr_handle,
dpaa_interrupt_handler, (void *)dev);
- if (ret) {
+ if (ret < 0) {
DPAA_PMD_WARN("%s: unregister interrupt failed(%d)",
dev->data->name, ret);
}
--
2.43.0
^ permalink raw reply related
* [PATCH v3 11/18] net/dpaa: fix port_handle leak in fm_prev_cleanup
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable, Vanshika Shukla
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
From: Vanshika Shukla <vanshika.shukla@nxp.com>
In fm_prev_cleanup(), the port_handle was not closed before being
overwritten on each iteration, causing a resource leak. Add a null
check and close the existing handle before opening a new one.
Fixes: e498f3b51f38 ("net/dpaa: improve port cleanup")
Cc: stable@dpdk.org
Signed-off-by: Vanshika Shukla <vanshika.shukla@nxp.com>
---
drivers/net/dpaa/dpaa_flow.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/dpaa/dpaa_flow.c b/drivers/net/dpaa/dpaa_flow.c
index 417b9b6fbb..f21950f64d 100644
--- a/drivers/net/dpaa/dpaa_flow.c
+++ b/drivers/net/dpaa/dpaa_flow.c
@@ -81,6 +81,10 @@ static void fm_prev_cleanup(void)
devid = fm_model.device_order[i];
/* FM Port Open */
fm_model.fm_port_params[devid].h_fm = fm_info.fman_handle;
+ if (dpaa_intf.port_handle) {
+ fm_port_close(dpaa_intf.port_handle);
+ dpaa_intf.port_handle = NULL;
+ }
dpaa_intf.port_handle =
fm_port_open(&fm_model.fm_port_params[devid]);
dpaa_intf.scheme_handle[0] = create_device(fm_info.pcd_handle,
--
2.43.0
^ permalink raw reply related
* [PATCH v3 12/18] dma/dpaa: fix out-of-bounds access in SG descriptor enqueue
From: Hemant Agrawal @ 2026-06-19 10:38 UTC (permalink / raw)
To: stephen, david.marchand, dev; +Cc: stable, Vanshika Shukla
In-Reply-To: <20260619103901.2274740-1-hemant.agrawal@nxp.com>
From: Vanshika Shukla <vanshika.shukla@nxp.com>
In fsl_qdma_enqueue_desc_sg(), the code accesses desc_ssge[num - 1]
without validating num first. If pending_num is 0, num will be 0 and
the access underflows. Add a bounds check to return -EINVAL when num
is 0 or exceeds FSL_QDMA_SG_MAX_ENTRY.
Fixes: a77261f61245 ("dma/dpaa: support scatter-gather")
Cc: stable@dpdk.org
Signed-off-by: Vanshika Shukla <vanshika.shukla@nxp.com>
---
drivers/dma/dpaa/dpaa_qdma.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/drivers/dma/dpaa/dpaa_qdma.c b/drivers/dma/dpaa/dpaa_qdma.c
index 74e23d2ee5..b20ff24ab6 100644
--- a/drivers/dma/dpaa/dpaa_qdma.c
+++ b/drivers/dma/dpaa/dpaa_qdma.c
@@ -1,5 +1,5 @@
/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright 2021-2024 NXP
+ * Copyright 2021-2026 NXP
*/
#include <bus_dpaa_driver.h>
@@ -827,6 +827,11 @@ fsl_qdma_enqueue_desc_sg(struct fsl_qdma_queue *fsl_queue)
}
}
+ if (num == 0 || num > FSL_QDMA_SG_MAX_ENTRY) {
+ DPAA_QDMA_ERR("Invalid scatter-gather entry count: num=%u", num);
+ return -EINVAL;
+ }
+
ft->desc_ssge[num - 1].final = 1;
ft->desc_dsge[num - 1].final = 1;
csgf_src->length = total_len;
--
2.43.0
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox