* [PATCH v7 03/10] app/testpmd: support selective Rx
From: Thomas Monjalon @ 2026-06-03 18:23 UTC (permalink / raw)
To: dev; +Cc: Stephen Hemminger, Gregory Etelson, Aman Singh
In-Reply-To: <20260603184503.652030-1-thomas@monjalon.net>
From: Gregory Etelson <getelson@nvidia.com>
Add support for selective Rx using existing rxoffs and rxpkts
command line parameters.
When both rxoffs and rxpkts are specified
on PMDs supporting selective Rx, testpmd automatically:
1. Inserts segments with NULL mempool
for gaps between configured segments to discard unwanted data.
2. Adds a trailing segment with NULL mempool
to cover any remaining data up to the max packet length.
Example usage to receive only Ethernet header and a segment at offset 128:
--rxoffs=0,128 --rxpkts=14,64
This creates segments:
- [0-13]: 14 bytes with mempool (received)
- [14-127]: 114 bytes with NULL mempool (discarded)
- [128-191]: 64 bytes with mempool (received)
- [192-max]: remaining bytes with NULL mempool (discarded)
Note: RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT is required for this feature
and is checked at ethdev API level.
This check is removed from testpmd to allow negative testing of the API.
Signed-off-by: Gregory Etelson <getelson@nvidia.com>
---
app/test-pmd/testpmd.c | 71 ++++++++++++++++++++++-----
doc/guides/testpmd_app_ug/run_app.rst | 20 ++++++++
2 files changed, 80 insertions(+), 11 deletions(-)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index a9b35f530a..d14341d3ff 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -2731,6 +2731,16 @@ port_is_started(portid_t port_id)
return 1;
}
+static struct rte_eth_rxseg_split *
+next_rx_seg(union rte_eth_rxseg *segs, uint16_t *idx)
+{
+ if (*idx >= MAX_SEGS_BUFFER_SPLIT) {
+ fprintf(stderr, "Too many segments (max %u)\n", MAX_SEGS_BUFFER_SPLIT);
+ return NULL;
+ }
+ return &segs[(*idx)++].split;
+}
+
/* Configure the Rx with optional split. */
int
rx_queue_setup(uint16_t port_id, uint16_t rx_queue_id,
@@ -2744,31 +2754,70 @@ rx_queue_setup(uint16_t port_id, uint16_t rx_queue_id,
uint32_t prev_hdrs = 0;
int ret;
- if ((rx_pkt_nb_segs > 1) &&
- (rx_conf->offloads & RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT)) {
+ if (rx_pkt_nb_segs > 1 || rx_pkt_nb_offs > 0) {
+ struct rte_eth_dev_info dev_info;
+ uint16_t seg_idx = 0;
+ uint16_t next_offset = 0;
+
+ ret = rte_eth_dev_info_get(port_id, &dev_info);
+ if (ret != 0)
+ return ret;
+
/* multi-segment configuration */
for (i = 0; i < rx_pkt_nb_segs; i++) {
- struct rte_eth_rxseg_split *rx_seg = &rx_useg[i].split;
- /*
- * Use last valid pool for the segments with number
- * exceeding the pool index.
- */
+ struct rte_eth_rxseg_split *rx_seg;
+ uint16_t seg_offset;
+
+ if (i < rx_pkt_nb_offs)
+ seg_offset = rx_pkt_seg_offsets[i];
+ else
+ seg_offset = rx_pkt_nb_offs > 0 ? next_offset : 0;
+
+ /* Insert selective Rx discard segment if there's a gap */
+ if (seg_offset > next_offset) {
+ rx_seg = next_rx_seg(rx_useg, &seg_idx);
+ if (rx_seg == NULL)
+ return -EINVAL;
+ rx_seg->offset = next_offset;
+ rx_seg->length = seg_offset - next_offset;
+ rx_seg->mp = NULL;
+ next_offset = seg_offset;
+ }
+
+ rx_seg = next_rx_seg(rx_useg, &seg_idx);
+ if (rx_seg == NULL)
+ return -EINVAL;
mp_n = (i >= mbuf_data_size_n) ? mbuf_data_size_n - 1 : i;
mpx = mbuf_pool_find(socket_id, mp_n);
- /* Handle zero as mbuf data buffer size. */
- rx_seg->offset = i < rx_pkt_nb_offs ?
- rx_pkt_seg_offsets[i] : 0;
+ rx_seg->offset = seg_offset;
rx_seg->mp = mpx ? mpx : mp;
if (rx_pkt_hdr_protos[i] != 0 && rx_pkt_seg_lengths[i] == 0) {
rx_seg->proto_hdr = rx_pkt_hdr_protos[i] & ~prev_hdrs;
prev_hdrs |= rx_seg->proto_hdr;
+ } else if (rx_pkt_nb_offs > 0 && rx_pkt_seg_lengths[i] == 0) {
+ /* Insert fake discard segment if explicitly requested */
+ rx_seg->mp = NULL;
+ rx_seg->length = 0;
} else {
rx_seg->length = rx_pkt_seg_lengths[i] ?
rx_pkt_seg_lengths[i] :
mbuf_data_size[mp_n];
}
+
+ next_offset = seg_offset + rx_seg->length;
}
- rx_conf->rx_nseg = rx_pkt_nb_segs;
+
+ /* Add trailing selective Rx discard segment up to max packet length */
+ if (rx_pkt_nb_offs > 0 && next_offset < dev_info.max_rx_pktlen) {
+ struct rte_eth_rxseg_split *rx_seg = next_rx_seg(rx_useg, &seg_idx);
+ if (rx_seg == NULL)
+ return -EINVAL;
+ rx_seg->offset = next_offset;
+ rx_seg->length = dev_info.max_rx_pktlen - next_offset;
+ rx_seg->mp = NULL;
+ }
+
+ rx_conf->rx_nseg = seg_idx;
rx_conf->rx_seg = rx_useg;
rx_conf->rx_mempools = NULL;
rx_conf->rx_nmempool = 0;
diff --git a/doc/guides/testpmd_app_ug/run_app.rst b/doc/guides/testpmd_app_ug/run_app.rst
index 1a4a4b6c12..b59991ed89 100644
--- a/doc/guides/testpmd_app_ug/run_app.rst
+++ b/doc/guides/testpmd_app_ug/run_app.rst
@@ -364,6 +364,11 @@ The command line options are:
feature is engaged. Affects only the queues configured
with split offloads (currently BUFFER_SPLIT is supported only).
+ When used with ``--rxpkts`` on PMDs supporting selective Rx,
+ enables receiving only specific packet segments and discarding the rest.
+ Gaps between configured segments and any trailing data up to the max packet length
+ are automatically filled with NULL mempool segments (data is discarded).
+
* ``--rxpkts=X[,Y]``
Set the length of segments to scatter packets on receiving if split
@@ -373,6 +378,21 @@ The command line options are:
command line parameter and the mbufs to receive will be allocated
sequentially from these extra memory pools.
+ Note: ``--rxoffs`` is required to enable selective Rx in testpmd.
+ To receive only the first N bytes, use ``--rxoffs=0 --rxpkts=N``.
+
+ To receive only the Ethernet header (14 bytes at offset 0) and
+ a 64-byte segment starting at offset 128, while discarding the rest::
+
+ --rxoffs=0,128 --rxpkts=14,64
+
+ This configuration will:
+
+ * Receive 14 bytes at offset 0 (Ethernet header)
+ * Discard bytes 14-127 (inserted NULL mempool segment)
+ * Receive 64 bytes at offset 128
+ * Discard remaining bytes (inserted NULL mempool segment)
+
* ``--txpkts=X[,Y]``
Set TX segment sizes or total packet length. Valid for ``tx-only``
--
2.54.0
^ permalink raw reply related
* [PATCH v7 02/10] ethdev: introduce selective Rx
From: Thomas Monjalon @ 2026-06-03 18:23 UTC (permalink / raw)
To: dev; +Cc: Stephen Hemminger, Gregory Etelson, Andrew Rybchenko, Aman Singh
In-Reply-To: <20260603184503.652030-1-thomas@monjalon.net>
From: Gregory Etelson <getelson@nvidia.com>
Receiving an entire packet is not always needed.
The Rx performance can be improved by receiving only partial data
and safely discard the rest of the packet data,
because it reduces the PCI bandwidth and the memory consumption.
Selective Rx allows an application to receive
only pre-configured packet segments and discard the rest.
For example:
- Deliver the first N bytes only.
- Deliver the last N bytes only.
- Deliver N1 bytes from offset Off1 and N2 bytes from offset Off2.
Selective Rx is implemented on top of the Rx buffer split API:
- rte_eth_rxseg_split uses the null mempool for segments
that should be discarded.
- the PMD does not create mbuf segments if no data read.
For example: Deliver Ethernet header only
Rx queue segments configuration:
struct rte_eth_rxseg_split split[2] = {
{
.mp = <some mempool>,
.length = sizeof(struct rte_ether_hdr)
},
{
.mp = NULL, /* discard data */
.length = 0 /* default to buffer size */
}
};
Received mbuf:
pkt_len = sizeof(struct rte_ether_hdr);
data_len = sizeof(struct rte_ether_hdr);
next = NULL; /* The next segment did not deliver data */
After selective Rx, the mbuf packet length reflects only the data
that was actually received,
and can be less than the original wire packet length.
A PMD activates the selective Rx capability by setting
the rte_eth_rxseg_capa.selective_rx bit.
This new capability bit is inserted in a bitmap hole
of the struct rte_eth_rxseg_capa,
but it needs to be ignored in the ABI check as libabigail sees a change.
Signed-off-by: Gregory Etelson <getelson@nvidia.com>
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
---
app/test-pmd/config.c | 1 +
devtools/libabigail.abignore | 7 +++++++
doc/guides/nics/features.rst | 14 ++++++++++++++
doc/guides/nics/features/default.ini | 1 +
doc/guides/rel_notes/release_26_07.rst | 7 +++++++
lib/ethdev/rte_ethdev.c | 24 ++++++++++++++++--------
lib/ethdev/rte_ethdev.h | 14 +++++++++++++-
7 files changed, 59 insertions(+), 9 deletions(-)
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index 55d1c6d696..9d457ca88e 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -925,6 +925,7 @@ port_infos_display(portid_t port_id)
print_bool_capa("\tBuffer offset", dev_info.rx_seg_capa.offset_allowed);
printf("\tOffset alignment: %u\n",
RTE_BIT32(dev_info.rx_seg_capa.offset_align_log2));
+ print_bool_capa("\tSelective Rx", dev_info.rx_seg_capa.selective_rx);
}
if (dev_info.max_vfs)
diff --git a/devtools/libabigail.abignore b/devtools/libabigail.abignore
index 21b8cd6113..2a0efd718e 100644
--- a/devtools/libabigail.abignore
+++ b/devtools/libabigail.abignore
@@ -33,3 +33,10 @@
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
; Temporary exceptions till next major ABI version ;
;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
+
+; Ignore new bit selective_rx in rte_eth_rxseg_capa bitmap hole
+[suppress_type]
+ name = rte_eth_rxseg_capa
+ type_kind = struct
+ has_size_change = no
+ has_data_member_inserted_at = 6
diff --git a/doc/guides/nics/features.rst b/doc/guides/nics/features.rst
index a075c057ec..26357036ca 100644
--- a/doc/guides/nics/features.rst
+++ b/doc/guides/nics/features.rst
@@ -199,6 +199,20 @@ Scatters the packets being received on specified boundaries to segmented mbufs.
* **[related] API**: ``rte_eth_rx_queue_setup()``, ``rte_eth_buffer_split_get_supported_hdr_ptypes()``.
+.. _nic_features_selective_rx:
+
+Selective Rx
+------------
+
+Discards some segments of buffer split on Rx.
+
+* **[uses] rte_eth_rxconf,rte_eth_rxmode**: ``offloads:RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT``.
+* **[uses] rte_eth_rxconf**: ``rx_seg.mp = NULL`` to discard segments.
+* **[provides] rte_eth_dev_info**: ``rx_offload_capa:RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT``.
+* **[provides] rte_eth_dev_info**: ``rx_seg_capa.selective_rx``.
+* **[related] API**: ``rte_eth_rx_queue_setup()``.
+
+
.. _nic_features_lro:
LRO
diff --git a/doc/guides/nics/features/default.ini b/doc/guides/nics/features/default.ini
index e50514d750..8303a530c1 100644
--- a/doc/guides/nics/features/default.ini
+++ b/doc/guides/nics/features/default.ini
@@ -25,6 +25,7 @@ Burst mode info =
Power mgmt address monitor =
MTU update =
Buffer split on Rx =
+Selective Rx =
Scattered Rx =
LRO =
TSO =
diff --git a/doc/guides/rel_notes/release_26_07.rst b/doc/guides/rel_notes/release_26_07.rst
index b8a3e2ced9..90a7948c1d 100644
--- a/doc/guides/rel_notes/release_26_07.rst
+++ b/doc/guides/rel_notes/release_26_07.rst
@@ -81,6 +81,13 @@ New Features
Added no-IOMMU mode for devices without or not enabling IOMMU/SVA.
+* **Added selective Rx in ethdev API.**
+
+ Some parts of packets may be discarded in Rx
+ by configuring a split of packets received in a queue,
+ and assigning no mempool to some configuration segments.
+ This is a driver capability advertised in the ``selective_rx`` bit.
+
* **Added LinkData sxe2 ethernet driver.**
Added network driver for the LinkData network adapters.
diff --git a/lib/ethdev/rte_ethdev.c b/lib/ethdev/rte_ethdev.c
index d0273e3f7b..67d609820c 100644
--- a/lib/ethdev/rte_ethdev.c
+++ b/lib/ethdev/rte_ethdev.c
@@ -2129,7 +2129,7 @@ rte_eth_rx_queue_check_split(uint16_t port_id,
const struct rte_eth_dev_info *dev_info)
{
const struct rte_eth_rxseg_capa *seg_capa = &dev_info->rx_seg_capa;
- struct rte_mempool *mp_first;
+ struct rte_mempool *mp_first = NULL;
uint32_t offset_mask;
uint16_t seg_idx;
int ret = 0;
@@ -2148,7 +2148,6 @@ rte_eth_rx_queue_check_split(uint16_t port_id,
* Check the sizes and offsets against buffer sizes
* for each segment specified in extended configuration.
*/
- mp_first = rx_seg[0].mp;
offset_mask = RTE_BIT32(seg_capa->offset_align_log2) - 1;
ptypes = NULL;
@@ -2160,13 +2159,17 @@ rte_eth_rx_queue_check_split(uint16_t port_id,
uint32_t offset = rx_seg[seg_idx].offset;
uint32_t proto_hdr = rx_seg[seg_idx].proto_hdr;
- if (mpl == NULL) {
- RTE_ETHDEV_LOG_LINE(ERR, "null mempool pointer");
- ret = -EINVAL;
- goto out;
+ if (mpl == NULL) { /* discarded segment */
+ if (seg_capa->selective_rx == 0) { /* not supported */
+ RTE_ETHDEV_LOG_LINE(ERR, "null mempool pointer");
+ ret = -EINVAL;
+ goto out;
+ }
+ continue; /* next checks are not relevant if no mempool */
}
- if (seg_idx != 0 && mp_first != mpl &&
- seg_capa->multi_pools == 0) {
+ if (mp_first == NULL)
+ mp_first = mpl;
+ if (mp_first != mpl && seg_capa->multi_pools == 0) {
RTE_ETHDEV_LOG_LINE(ERR, "Receiving to multiple pools is not supported");
ret = -ENOTSUP;
goto out;
@@ -2233,6 +2236,11 @@ rte_eth_rx_queue_check_split(uint16_t port_id,
if (ret != 0)
goto out;
}
+ if (mp_first == NULL) {
+ RTE_ETHDEV_LOG_LINE(ERR, "At least one Rx segment must have a mempool");
+ ret = -EINVAL;
+ goto out;
+ }
out:
free(ptypes);
return ret;
diff --git a/lib/ethdev/rte_ethdev.h b/lib/ethdev/rte_ethdev.h
index dedbc05554..d7e1560828 100644
--- a/lib/ethdev/rte_ethdev.h
+++ b/lib/ethdev/rte_ethdev.h
@@ -1073,6 +1073,7 @@ struct rte_eth_txmode {
* - The first network buffer will be allocated from the memory pool,
* specified in the first array element, the second buffer, from the
* pool in the second element, and so on.
+ * If the pool is NULL, the segment will be discarded, i.e. not received.
*
* - The proto_hdrs in the elements define the split position of
* received packets.
@@ -1121,7 +1122,15 @@ struct rte_eth_txmode {
* The rest will be put into the last valid pool.
*/
struct rte_eth_rxseg_split {
- struct rte_mempool *mp; /**< Memory pool to allocate segment from. */
+ /**
+ * Memory pool to allocate segment from.
+ *
+ * NULL means discarded segment.
+ * Length of discarded segment is not reflected in mbuf packet length
+ * and not accounted in ibytes statistics.
+ * @see rte_eth_rxseg_capa::selective_rx
+ */
+ struct rte_mempool *mp;
uint16_t length; /**< Segment data length, configures split point. */
uint16_t offset; /**< Data offset from beginning of mbuf data buffer. */
/**
@@ -1752,12 +1761,15 @@ struct rte_eth_switch_info {
* @b EXPERIMENTAL: this structure may change without prior notice.
*
* Ethernet device Rx buffer segmentation capabilities.
+ *
+ * @see rte_eth_rxseg_split
*/
struct rte_eth_rxseg_capa {
__extension__
uint32_t multi_pools:1; /**< Supports receiving to multiple pools.*/
uint32_t offset_allowed:1; /**< Supports buffer offsets. */
uint32_t offset_align_log2:4; /**< Required offset alignment. */
+ uint32_t selective_rx:1; /**< Supports discarding segment. */
uint16_t max_nseg; /**< Maximum amount of segments to split. */
uint16_t reserved; /**< Reserved field. */
};
--
2.54.0
^ permalink raw reply related
* [PATCH v7 01/10] app/testpmd: print Rx split capabilities
From: Thomas Monjalon @ 2026-06-03 18:23 UTC (permalink / raw)
To: dev; +Cc: Stephen Hemminger, Aman Singh
In-Reply-To: <20260603184503.652030-1-thomas@monjalon.net>
The capabilities from rte_eth_rxseg_capa are added
to the command "show port info".
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
---
app/test-pmd/config.c | 16 ++++++++++++++++
1 file changed, 16 insertions(+)
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index c950793aaf..55d1c6d696 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -790,6 +790,12 @@ rss_offload_types_display(uint64_t offload_types, uint16_t char_num_per_line)
printf("\n");
}
+static void
+print_bool_capa(const char *label, int value)
+{
+ printf("%s: %s\n", label, value ? "supported" : "not supported");
+}
+
void
port_infos_display(portid_t port_id)
{
@@ -911,6 +917,16 @@ port_infos_display(portid_t port_id)
dev_info.max_rx_pktlen);
printf("Maximum configurable size of LRO aggregated packet: %u\n",
dev_info.max_lro_pkt_size);
+
+ printf("Rx split:\n");
+ printf("\tMax segments: %hu\n", dev_info.rx_seg_capa.max_nseg);
+ if (dev_info.rx_seg_capa.max_nseg > 0) {
+ print_bool_capa("\tMulti-pool", dev_info.rx_seg_capa.multi_pools);
+ print_bool_capa("\tBuffer offset", dev_info.rx_seg_capa.offset_allowed);
+ printf("\tOffset alignment: %u\n",
+ RTE_BIT32(dev_info.rx_seg_capa.offset_align_log2));
+ }
+
if (dev_info.max_vfs)
printf("Maximum number of VFs: %u\n", dev_info.max_vfs);
if (dev_info.max_vmdq_pools)
--
2.54.0
^ permalink raw reply related
* [PATCH v7 00/10] selective Rx
From: Thomas Monjalon @ 2026-06-03 18:23 UTC (permalink / raw)
To: dev; +Cc: Stephen Hemminger
In-Reply-To: <20260202160903.254621-1-getelson@nvidia.com>
This is a new feature in ethdev with tests and mlx5 implementation.
Selective Rx allows to receive partial data,
saving some hardware bandwidth.
Note 1: mlx5 support patch is not correctly indented
to make review easier. An indent patch follows to be squashed.
Note 2: DTS patch is an attempt to test the feature on day 1,
it is not mandatory if it is blocking the merge.
v2: rework after Gregory
v3: fix bugs found with AI by Stephen
v4: fix packet type in DTS test
v5: fix mlx5 Rx to handle discarding first segment
v6: fix reindent patch
v7: fix mlx5 CQE error handling + outdated mcqe + redundant assignment
Gregory Etelson (4):
ethdev: introduce selective Rx
app/testpmd: support selective Rx
common/mlx5: add null MR functions
net/mlx5: support selective Rx
Thomas Monjalon (6):
app/testpmd: print Rx split capabilities
net/mlx5: fix Rx split segment counter type
net/mlx5: reindent previous changes
common/mlx5: remove callbacks for MR registration
dts: fix topology capability comparison
dts: add selective Rx tests
app/test-pmd/config.c | 17 ++
app/test-pmd/testpmd.c | 71 ++++-
devtools/libabigail.abignore | 7 +
doc/guides/nics/features.rst | 14 +
doc/guides/nics/features/default.ini | 1 +
doc/guides/nics/features/mlx5.ini | 1 +
doc/guides/nics/mlx5.rst | 86 ++++--
doc/guides/rel_notes/release_26_07.rst | 11 +
doc/guides/testpmd_app_ug/run_app.rst | 20 ++
drivers/common/mlx5/linux/mlx5_common_verbs.c | 53 ++--
drivers/common/mlx5/mlx5_common.c | 6 +-
drivers/common/mlx5/mlx5_common_mr.c | 37 ++-
drivers/common/mlx5/mlx5_common_mr.h | 29 +-
drivers/common/mlx5/windows/mlx5_common_os.c | 31 ++-
drivers/compress/mlx5/mlx5_compress.c | 4 +-
drivers/crypto/mlx5/mlx5_crypto.h | 2 -
drivers/crypto/mlx5/mlx5_crypto_gcm.c | 6 +-
drivers/net/mlx5/mlx5.c | 7 +
drivers/net/mlx5/mlx5.h | 4 +-
drivers/net/mlx5/mlx5_ethdev.c | 25 ++
drivers/net/mlx5/mlx5_flow_aso.c | 21 +-
drivers/net/mlx5/mlx5_flow_hw.c | 11 +-
drivers/net/mlx5/mlx5_flow_quota.c | 6 +-
drivers/net/mlx5/mlx5_hws_cnt.c | 19 +-
drivers/net/mlx5/mlx5_rx.c | 187 ++++++++-----
drivers/net/mlx5/mlx5_rx.h | 5 +-
drivers/net/mlx5/mlx5_rxq.c | 75 +++--
drivers/net/mlx5/mlx5_trigger.c | 64 ++++-
dts/api/capabilities.py | 2 +
dts/api/testpmd/__init__.py | 17 ++
dts/api/testpmd/types.py | 6 +
dts/framework/testbed_model/capability.py | 10 +-
dts/tests/TestSuite_rx_split.py | 262 ++++++++++++++++++
lib/ethdev/rte_ethdev.c | 24 +-
lib/ethdev/rte_ethdev.h | 14 +-
35 files changed, 881 insertions(+), 274 deletions(-)
create mode 100644 dts/tests/TestSuite_rx_split.py
--
2.54.0
^ permalink raw reply
* [PATCH v2] net/bnxt: fix RSS hash mode configuration for VFs
From: Mohammad Shuab Siddique @ 2026-06-03 18:36 UTC (permalink / raw)
To: dev; +Cc: kishore.padmanabha, stable, Mohammad Shuab Siddique
In-Reply-To: <20260603175137.1990204-1-Mohammad-Shuab.Siddique@broadcom.com>
From: Mohammad Shuab Siddique <mohammad-shuab.siddique@broadcom.com>
Fixed VFs attempting global RSS configuration which is not
permitted by firmware. VFs (including trusted VFs) must use
per-VNIC RSS configuration with actual vnic_id and rss_ctx_idx
values.
Fixes: 8b9adaf0da6b ("net/bnxt: support RSS on ESP and AH headers")
Cc: stable@dpdk.org
Signed-off-by: Mohammad Shuab Siddique <mohammad-shuab.siddique@broadcom.com>
---
drivers/net/bnxt/bnxt_hwrm.c | 18 ++++++++++++++++--
1 file changed, 16 insertions(+), 2 deletions(-)
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 0c82935de9..afc948ac29 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -2970,8 +2970,22 @@ bnxt_hwrm_vnic_rss_cfg_hash_mode_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic
req.hash_mode_flags = BNXT_HASH_MODE_INNERMOST;
else
req.hash_mode_flags = vnic->hash_mode;
- req.vnic_id = rte_cpu_to_le_16(BNXT_DFLT_VNIC_ID_INVALID);
- req.rss_ctx_idx = rte_cpu_to_le_16(BNXT_RSS_CTX_IDX_INVALID);
+
+ /* VFs must use actual vnic_id for per-VNIC configuration.
+ * PFs can use INVALID vnic_id for global configuration.
+ * This is because VFs don't have permission to configure
+ * global hash mode, even if they're trusted.
+ */
+ if (BNXT_VF(bp)) {
+ req.vnic_id = rte_cpu_to_le_16(vnic->fw_vnic_id);
+ req.rss_ctx_idx = rte_cpu_to_le_16(vnic->fw_grp_ids[0]);
+ PMD_DRV_LOG_LINE(DEBUG, "VF using per-VNIC RSS config (vnic_id=%u)",
+ vnic->fw_vnic_id);
+ } else {
+ req.vnic_id = rte_cpu_to_le_16(BNXT_DFLT_VNIC_ID_INVALID);
+ req.rss_ctx_idx = rte_cpu_to_le_16(BNXT_RSS_CTX_IDX_INVALID);
+ PMD_DRV_LOG_LINE(DEBUG, "PF using global RSS config");
+ }
PMD_DRV_LOG_LINE(DEBUG, "RSS CFG: Hash level %d", req.hash_mode_flags);
rc = bnxt_hwrm_send_message(bp, &req, sizeof(req),
--
2.47.3
^ permalink raw reply related
* Re: [PATCH v8 19/20] drivers: add testpmd commands for private features
From: Stephen Hemminger @ 2026-06-03 18:23 UTC (permalink / raw)
To: liujie5; +Cc: dev
In-Reply-To: <20260603062945.1253672-20-liujie5@linkdatatechnology.com>
On Wed, 3 Jun 2026 14:29:44 +0800
liujie5@linkdatatechnology.com wrote:
> From: Jie Liu <liujie5@linkdatatechnology.com>
>
> Introduce private testpmd commands and implementation files to enable
> debugging and testing of sxe2-specific hardware features (such as
> packet scheduling reset, UDP tunnel configuration, and IPsec ingress/
> egress offloads) directly within the testpmd application.
>
> The parameters are parsed using the standard 'rte_kvargs' library during
> the PCI/vdev probing phase. Documentation for these parameters is also
> updated.
>
> 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.
>
> This commit introduces sxe2_memseg_walk_cb() as a helper that walks
> memory segments using the thread-unsafe variant
> rte_memseg_walk_thread_unsafe(), which is safe to call from
> memory-related callbacks [citation:1][citation:3][citation:5].
>
> 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>
> ---
WARNING: [TYPO_SPELLING] 'thw' may be misspelled - perhaps 'the'?
# drivers/net/sxe2/sxe2_testpmd_lib.c:90:
+ cmdline_printf(cl, "\thw flow id: %d\n", hw_flow->flow_id);
total: 0 errors, 1 warnings, 0 checks, 2371 lines checked
18/20 valid patches
^ permalink raw reply
* Re: [PATCH v8 04/20] net/sxe2: support L2 filtering and MAC config
From: Stephen Hemminger @ 2026-06-03 18:21 UTC (permalink / raw)
To: liujie5; +Cc: dev
In-Reply-To: <20260603062945.1253672-5-liujie5@linkdatatechnology.com>
On Wed, 3 Jun 2026 14:29:29 +0800
liujie5@linkdatatechnology.com wrote:
> From: Jie Liu <liujie5@linkdatatechnology.com>
>
> - Support primary/secondary MAC address setup.
> - Enable L2 broadcast/multicast filter bits.
> - Add multicast address update logic.
>
> Signed-off-by: Jie Liu <liujie5@linkdatatechnology.com>
> ---
As part of "inclusive naming" DPDK does not allow use of variables name master or slave.
### [PATCH] net/sxe2: support L2 filtering and MAC config
WARNING:TYPO_SPELLING: 'master' may be misspelled - perhaps 'primary'?
#264: FILE: drivers/net/sxe2/sxe2_drv_cmd.h:238:
+ uint8_t master;
^^^^^^
^ permalink raw reply
* Re: [PATCH v8 07/20] net/sxe2: support IPsec inline protocol offload
From: Stephen Hemminger @ 2026-06-03 18:17 UTC (permalink / raw)
To: liujie5; +Cc: dev
In-Reply-To: <20260603062945.1253672-8-liujie5@linkdatatechnology.com>
On Wed, 3 Jun 2026 14:29:32 +0800
liujie5@linkdatatechnology.com wrote:
> From: Jie Liu <liujie5@linkdatatechnology.com>
>
> This patch adds support for IPsec inline protocol offload for both
> inbound and outbound traffic.
>
> - Implement rte_security_ops: session_create, session_destroy.
> - Add hardware SA table management.
> - Update Rx/Tx data path to handle security offload flags.
>
> The hardware offloads the ESP encapsulation/decapsulation and
> cryptographic processing.
>
> Signed-off-by: Jie Liu <liujie5@linkdatatechnology.com>
Git complains about blank line:
Applying: net/sxe2: support IPsec inline protocol offload
/home/shemminger/DPDK/main/.git/worktrees/sxe2/rebase-apply/patch:236: new blank line at EOF.
+
warning: 1 line adds whitespace errors.
^ permalink raw reply
* Re: [PATCH v8 11/20] drivers: add support for VF representors
From: Stephen Hemminger @ 2026-06-03 18:22 UTC (permalink / raw)
To: liujie5; +Cc: dev
In-Reply-To: <20260603062945.1253672-12-liujie5@linkdatatechnology.com>
On Wed, 3 Jun 2026 14:29:36 +0800
liujie5@linkdatatechnology.com wrote:
> From: Jie Liu <liujie5@linkdatatechnology.com>
>
> Add support for VF representors in sxe2 PMD. This allows the host
> application (e.g., OVS-DPDK) to control and monitor virtual functions
> through a dedicated ethdev on the PF (Physical Function) side.
>
> Key changes include:
> - Added representor enumeration and identification logic.
> - Implemented representor-specific dev_ops (link update, stats, etc.).
> - Configured back-channel communication between PF and VF for control
> messages.
> - Supported the "-a <DBDF>,representor=[0-N]" EAL parameter to
> instantiate representor ports.
>
> Signed-off-by: Jie Liu <liujie5@linkdatatechnology.com>
> ---
### [PATCH] drivers: add support for VF representors
WARNING: [TYPO_SPELLING] 'macth' may be misspelled - perhaps 'match'?
# drivers/net/sxe2/sxe2_flow_parse_pattern.c:1786:
+ "Invalid pattern spec miss macth mask for rule.");
WARNING: [TYPO_SPELLING] 'macth' may be misspelled - perhaps 'match'?
# drivers/net/sxe2/sxe2_flow_parse_pattern.c:1787:
+ PMD_LOG_ERR(DRV, "Invalid pattern spec miss macth mask for rule.");
WARNING: [TYPO_SPELLING] 'macth' may be misspelled - perhaps 'match'?
# drivers/net/sxe2/sxe2_switchdev.c:197:
+ PMD_DEV_LOG_INFO(adapter, DRV, "current switchdev mode miss macth");
total: 0 errors, 3 warnings, 0 checks, 6571 lines checked
^ permalink raw reply
* Re: [PATCH v8 20/20] net/sxe2: update sxe2 feature matrix docs
From: Stephen Hemminger @ 2026-06-03 18:19 UTC (permalink / raw)
To: liujie5; +Cc: dev
In-Reply-To: <20260603062945.1253672-21-liujie5@linkdatatechnology.com>
On Wed, 3 Jun 2026 14:29:45 +0800
liujie5@linkdatatechnology.com wrote:
> 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>
> ---
The flow support documentation does not match code.
$ ./devtools/check-doc-vs-code.sh
rte_flow doc out of sync for sxe2
item ah
item any
item ecpri
item esp
item icmp
item mpls
item pfcp
item raw
^ permalink raw reply
* [PATCH] net/af_packet: fix parsing of numeric device args
From: Stephen Hemminger @ 2026-06-03 18:13 UTC (permalink / raw)
To: dev; +Cc: Stephen Hemminger, Denis Sergeev, stable, John W. Linville
In-Reply-To: <20260603170812.212262-1-denserg.edu@gmail.com>
This driver has several numeric arguments but it was using
atoi() which allows garbage and negative values.
Convert to a helper using strtoul() with upper bound.
First found by Linux Verification Center (linuxtesting.org) with SVACE.
Reported-by: Denis Sergeev <denserg.edu@gmail.com>
Fixes: 364e08f2bbc0 ("af_packet: add PMD for AF_PACKET-based virtual devices")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
---
drivers/net/af_packet/rte_eth_af_packet.c | 58 +++++++++++++++++++----
1 file changed, 48 insertions(+), 10 deletions(-)
diff --git a/drivers/net/af_packet/rte_eth_af_packet.c b/drivers/net/af_packet/rte_eth_af_packet.c
index 8303ff5ca9..b0ff22ea55 100644
--- a/drivers/net/af_packet/rte_eth_af_packet.c
+++ b/drivers/net/af_packet/rte_eth_af_packet.c
@@ -15,7 +15,9 @@
#include <rte_kvargs.h>
#include <bus_vdev_driver.h>
+#include <ctype.h>
#include <errno.h>
+#include <limits.h>
#include <linux/if_ether.h>
#include <linux/if_packet.h>
#include <arpa/inet.h>
@@ -1138,6 +1140,42 @@ rte_pmd_init_internals(struct rte_vdev_device *dev,
return -1;
}
+/* Parse an unsigned integer device argument. */
+static int
+parse_uint(const char *key, const char *value,
+ unsigned int *out, unsigned long limit)
+{
+ unsigned long val;
+ char *endptr;
+
+ if (value == NULL) {
+ PMD_LOG(ERR, "no value for argument \"%s\"", key);
+ return -1;
+ }
+
+ /* Skip leading whitespace so a leading sign can be detected. */
+ while (isspace((unsigned char)*value))
+ value++;
+
+ /* strtoul() silently accepts and negates a leading '-'. */
+ if (*value == '\0' || *value == '-') {
+ PMD_LOG(ERR, "invalid value \"%s\" for argument \"%s\"",
+ value, key);
+ return -1;
+ }
+
+ errno = 0;
+ val = strtoul(value, &endptr, 10);
+ if (errno != 0 || *endptr != '\0' || val > limit) {
+ PMD_LOG(ERR, "invalid value \"%s\" for argument \"%s\"",
+ value, key);
+ return -1;
+ }
+
+ *out = (unsigned int)val;
+ return 0;
+}
+
static int
rte_eth_from_packet(struct rte_vdev_device *dev,
int const *sockfd,
@@ -1168,7 +1206,9 @@ rte_eth_from_packet(struct rte_vdev_device *dev,
for (k_idx = 0; k_idx < kvlist->count; k_idx++) {
pair = &kvlist->pairs[k_idx];
if (strstr(pair->key, ETH_AF_PACKET_NUM_Q_ARG) != NULL) {
- qpairs = atoi(pair->value);
+ if (parse_uint(pair->key, pair->value,
+ &qpairs, RTE_MAX_QUEUES_PER_PORT) < 0)
+ return -1;
if (qpairs < 1) {
PMD_LOG(ERR,
"%s: invalid qpairs value",
@@ -1178,7 +1218,8 @@ rte_eth_from_packet(struct rte_vdev_device *dev,
continue;
}
if (strstr(pair->key, ETH_AF_PACKET_BLOCKSIZE_ARG) != NULL) {
- blocksize = atoi(pair->value);
+ if (parse_uint(pair->key, pair->value, &blocksize, UINT_MAX) < 0)
+ return -1;
if (!blocksize) {
PMD_LOG(ERR,
"%s: invalid blocksize value",
@@ -1188,7 +1229,8 @@ rte_eth_from_packet(struct rte_vdev_device *dev,
continue;
}
if (strstr(pair->key, ETH_AF_PACKET_FRAMESIZE_ARG) != NULL) {
- framesize = atoi(pair->value);
+ if (parse_uint(pair->key, pair->value, &framesize, UINT_MAX) < 0)
+ return -1;
if (!framesize) {
PMD_LOG(ERR,
"%s: invalid framesize value",
@@ -1198,7 +1240,8 @@ rte_eth_from_packet(struct rte_vdev_device *dev,
continue;
}
if (strstr(pair->key, ETH_AF_PACKET_FRAMECOUNT_ARG) != NULL) {
- framecount = atoi(pair->value);
+ if (parse_uint(pair->key, pair->value, &framecount, UINT_MAX) < 0)
+ return -1;
if (!framecount) {
PMD_LOG(ERR,
"%s: invalid framecount value",
@@ -1208,13 +1251,8 @@ rte_eth_from_packet(struct rte_vdev_device *dev,
continue;
}
if (strstr(pair->key, ETH_AF_PACKET_QDISC_BYPASS_ARG) != NULL) {
- qdisc_bypass = atoi(pair->value);
- if (qdisc_bypass > 1) {
- PMD_LOG(ERR,
- "%s: invalid bypass value",
- name);
+ if (parse_uint(pair->key, pair->value, &qdisc_bypass, 1) < 0)
return -1;
- }
continue;
}
if (strstr(pair->key, ETH_AF_PACKET_FANOUT_MODE_ARG) != NULL) {
--
2.53.0
^ permalink raw reply related
* [PATCH 4/4] net/bnxt: fix RSS hash mode configuration for VFs
From: Mohammad Shuab Siddique @ 2026-06-03 17:51 UTC (permalink / raw)
To: dev; +Cc: kishore.padmanabha, stable, Mohammad Shuab Siddique
In-Reply-To: <20260603175137.1990204-1-Mohammad-Shuab.Siddique@broadcom.com>
From: Mohammad Shuab Siddique <mohammad-shuab.siddique@broadcom.com>
Fixed VFs attempting global RSS configuration which is not
permitted by firmware. VFs (including trusted VFs) must use
per-VNIC RSS configuration with actual vnic_id and rss_ctx_idx
values.
Cc: stable@dpdk.org
Signed-off-by: Mohammad Shuab Siddique <mohammad-shuab.siddique@broadcom.com>
---
drivers/net/bnxt/bnxt_hwrm.c | 18 ++++++++++++++++--
1 file changed, 16 insertions(+), 2 deletions(-)
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index 0c82935de9..afc948ac29 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -2970,8 +2970,22 @@ bnxt_hwrm_vnic_rss_cfg_hash_mode_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic
req.hash_mode_flags = BNXT_HASH_MODE_INNERMOST;
else
req.hash_mode_flags = vnic->hash_mode;
- req.vnic_id = rte_cpu_to_le_16(BNXT_DFLT_VNIC_ID_INVALID);
- req.rss_ctx_idx = rte_cpu_to_le_16(BNXT_RSS_CTX_IDX_INVALID);
+
+ /* VFs must use actual vnic_id for per-VNIC configuration.
+ * PFs can use INVALID vnic_id for global configuration.
+ * This is because VFs don't have permission to configure
+ * global hash mode, even if they're trusted.
+ */
+ if (BNXT_VF(bp)) {
+ req.vnic_id = rte_cpu_to_le_16(vnic->fw_vnic_id);
+ req.rss_ctx_idx = rte_cpu_to_le_16(vnic->fw_grp_ids[0]);
+ PMD_DRV_LOG_LINE(DEBUG, "VF using per-VNIC RSS config (vnic_id=%u)",
+ vnic->fw_vnic_id);
+ } else {
+ req.vnic_id = rte_cpu_to_le_16(BNXT_DFLT_VNIC_ID_INVALID);
+ req.rss_ctx_idx = rte_cpu_to_le_16(BNXT_RSS_CTX_IDX_INVALID);
+ PMD_DRV_LOG_LINE(DEBUG, "PF using global RSS config");
+ }
PMD_DRV_LOG_LINE(DEBUG, "RSS CFG: Hash level %d", req.hash_mode_flags);
rc = bnxt_hwrm_send_message(bp, &req, sizeof(req),
--
2.47.3
^ permalink raw reply related
* [PATCH 3/4] net/bnxt: remove implicit integer sign-extension
From: Mohammad Shuab Siddique @ 2026-06-03 17:51 UTC (permalink / raw)
To: dev; +Cc: kishore.padmanabha, stable, Zoe Cheimets, Mohammad Shuab Siddique
In-Reply-To: <20260603175137.1990204-1-Mohammad-Shuab.Siddique@broadcom.com>
From: Zoe Cheimets <zoe.cheimets@broadcom.com>
In bnxt_ring.c, the result on line 389 was auto-sign extended by
the compiler because the arithmetic result is an int, but the
dpi_offset is uint64_t. Fix by casting the result to uint64_t
before the multiplication forces extension. To ensure that a
negative integer is not being cast to uint64_t, add a check in
the if-statement.
Fixes: 7a1f9c782b50 ("net/bnxt: add multi-doorbell support")
Cc: stable@dpdk.org
Signed-off-by: Zoe Cheimets <zoe.cheimets@broadcom.com>
Signed-off-by: Mohammad Shuab Siddique <mohammad-shuab.siddique@broadcom.com>
---
drivers/net/bnxt/bnxt_ring.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/net/bnxt/bnxt_ring.c b/drivers/net/bnxt/bnxt_ring.c
index ccca779b97..579b73d2ce 100644
--- a/drivers/net/bnxt/bnxt_ring.c
+++ b/drivers/net/bnxt/bnxt_ring.c
@@ -385,9 +385,10 @@ void bnxt_set_db(struct bnxt *bp,
db->doorbell = (char *)bp->doorbell_base + db_offset;
if (bp->fw_cap & BNXT_FW_CAP_MULTI_DB &&
- dpi != BNXT_PRIVILEGED_DPI) {
- dpi_offset = (dpi - bp->nq_dpi_start) *
- bp->db_page_size;
+ dpi != BNXT_PRIVILEGED_DPI &&
+ dpi >= bp->nq_dpi_start) {
+ dpi_offset = (uint64_t)(dpi - bp->nq_dpi_start) *
+ bp->db_page_size;
db->doorbell = (char *)db->doorbell + dpi_offset;
}
db->db_key64 |= (uint64_t)fid << DBR_XID_SFT;
--
2.47.3
^ permalink raw reply related
* [PATCH 2/4] net/bnxt: fix QP resource count in backing store config
From: Mohammad Shuab Siddique @ 2026-06-03 17:51 UTC (permalink / raw)
To: dev; +Cc: kishore.padmanabha, stable, Ajit Khaparde,
Mohammad Shuab Siddique
In-Reply-To: <20260603175137.1990204-1-Mohammad-Shuab.Siddique@broadcom.com>
From: Ajit Khaparde <ajit.khaparde@broadcom.com>
The driver is not passing the QP1 count while
configuring backing store for QP type.
This can result in a smaller set of entries allocated by the
driver with the firmware.
The number of entries is provided by the firmware as a
part of backing store qcaps v2 HWRM command.
Fixes: 5b5d398434f9 ("net/bnxt: add support for backing store v2")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: Mohammad Shuab Siddique <mohammad-shuab.siddique@broadcom.com>
---
drivers/net/bnxt/bnxt_ethdev.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index b677f9491d..7d70d0f3ec 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -5404,11 +5404,11 @@ int bnxt_alloc_ctx_pg_tbls(struct bnxt *bp)
if (ctxm->type == HWRM_FUNC_BACKING_STORE_CFG_V2_INPUT_TYPE_CQ)
entries = ctxm->cq_l2_entries;
else if (ctxm->type == HWRM_FUNC_BACKING_STORE_CFG_V2_INPUT_TYPE_QP)
- entries = ctxm->qp_l2_entries;
+ entries = ctxm->qp_l2_entries + ctxm->qp_qp1_entries;
else if (ctxm->type == HWRM_FUNC_BACKING_STORE_CFG_V2_INPUT_TYPE_MRAV)
entries = ctxm->mrav_av_entries;
else if (ctxm->type == HWRM_FUNC_BACKING_STORE_CFG_V2_INPUT_TYPE_TIM)
- entries = ctx2->qp_l2_entries;
+ entries = ctx2->qp_l2_entries + ctx2->qp_qp1_entries;
entries = clamp_t(uint32_t, entries, ctxm->min_entries,
ctxm->max_entries);
ctx_pg[i].entries = entries;
--
2.47.3
^ permalink raw reply related
* [PATCH 1/4] net/bnxt: modify check for short Tx BDs
From: Mohammad Shuab Siddique @ 2026-06-03 17:51 UTC (permalink / raw)
To: dev; +Cc: kishore.padmanabha, stable, Ajit Khaparde,
Mohammad Shuab Siddique
In-Reply-To: <20260603175137.1990204-1-Mohammad-Shuab.Siddique@broadcom.com>
From: Ajit Khaparde <ajit.khaparde@broadcom.com>
There is no need to use the long BDs for transmits
where only checksum offload is needed.
Modify the check for long BD and use long BDs only in cases
where TSO and other offloads are requested.
Fixes: 527b10089cc5 ("net/bnxt: optimize Tx completion handling")
Cc: stable@dpdk.org
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Signed-off-by: Mohammad Shuab Siddique <mohammad-shuab.siddique@broadcom.com>
---
drivers/net/bnxt/bnxt_txr.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/net/bnxt/bnxt_txr.c b/drivers/net/bnxt/bnxt_txr.c
index 27758898b0..7ef5b15ae8 100644
--- a/drivers/net/bnxt/bnxt_txr.c
+++ b/drivers/net/bnxt/bnxt_txr.c
@@ -111,8 +111,7 @@ int bnxt_init_tx_ring_struct(struct bnxt_tx_queue *txq, unsigned int socket_id)
static bool
bnxt_xmit_need_long_bd(struct rte_mbuf *tx_pkt, struct bnxt_tx_queue *txq)
{
- if (tx_pkt->ol_flags & (RTE_MBUF_F_TX_TCP_SEG | RTE_MBUF_F_TX_TCP_CKSUM |
- RTE_MBUF_F_TX_UDP_CKSUM | RTE_MBUF_F_TX_IP_CKSUM |
+ if (tx_pkt->ol_flags & (RTE_MBUF_F_TX_TCP_SEG |
RTE_MBUF_F_TX_VLAN | RTE_MBUF_F_TX_OUTER_IP_CKSUM |
RTE_MBUF_F_TX_TUNNEL_GRE | RTE_MBUF_F_TX_TUNNEL_VXLAN |
RTE_MBUF_F_TX_TUNNEL_GENEVE | RTE_MBUF_F_TX_IEEE1588_TMST |
--
2.47.3
^ permalink raw reply related
* [PATCH 0/4] net/bnxt: miscellaneous bug fixes
From: Mohammad Shuab Siddique @ 2026-06-03 17:51 UTC (permalink / raw)
To: dev; +Cc: kishore.padmanabha, stable, Mohammad Shuab Siddique
From: Mohammad Shuab Siddique <mohammad-shuab.siddique@broadcom.com>
This series collects four independent bug fixes for the bnxt PMD:
- Eliminate unnecessary long TX BDs when only checksum offload is needed
- Pass QP1 resource count correctly when configuring backing store
- Fix implicit integer sign-extension in the doorbell calculation
- Prevent VFs from attempting global RSS configuration
All patches carry Fixes: tags and Cc: stable@dpdk.org.
Ajit Khaparde (2):
net/bnxt: modify check for short Tx BDs
net/bnxt: fix QP resource count in backing store config
Mohammad Shuab Siddique (1):
net/bnxt: fix RSS hash mode configuration for VFs
Zoe Cheimets (1):
net/bnxt: remove implicit integer sign-extension
drivers/net/bnxt/bnxt_ethdev.c | 4 ++--
drivers/net/bnxt/bnxt_hwrm.c | 18 ++++++++++++++++--
drivers/net/bnxt/bnxt_ring.c | 7 ++++---
drivers/net/bnxt/bnxt_txr.c | 3 +--
4 files changed, 23 insertions(+), 9 deletions(-)
--
2.47.3
^ permalink raw reply
* Re: [PATCH v6 00/10] selective Rx
From: Stephen Hemminger @ 2026-06-03 17:31 UTC (permalink / raw)
To: Thomas Monjalon; +Cc: dev
In-Reply-To: <20260602215022.3698662-1-thomas@monjalon.net>
On Tue, 2 Jun 2026 23:49:12 +0200
Thomas Monjalon <thomas@monjalon.net> wrote:
> This is a new feature in ethdev with tests and mlx5 implementation.
> Selective Rx allows to receive partial data,
> saving some hardware bandwidth.
>
> Note 1: mlx5 support patch is not correctly indented
> to make review easier. An indent patch follows to be squashed.
>
> Note 2: DTS patch is an attempt to test the feature on day 1,
> it is not mandatory if it is blocking the merge.
>
> v2: rework after Gregory
> v3: fix bugs found with AI by Stephen
> v4: fix packet type in DTS test
> v5: fix mlx5 Rx to handle discarding first segment
> v6: fix reindent patch
>
>
> Gregory Etelson (4):
> ethdev: introduce selective Rx
> app/testpmd: support selective Rx
> common/mlx5: add null MR functions
> net/mlx5: support selective Rx
>
> Thomas Monjalon (6):
> app/testpmd: print Rx split capabilities
> net/mlx5: fix Rx split segment counter type
> net/mlx5: reindent previous changes
> common/mlx5: remove callbacks for MR registration
> dts: fix topology capability comparison
> dts: add selective Rx tests
>
> app/test-pmd/config.c | 17 ++
> app/test-pmd/testpmd.c | 71 ++++-
> devtools/libabigail.abignore | 7 +
> doc/guides/nics/features.rst | 14 +
> doc/guides/nics/features/default.ini | 1 +
> doc/guides/nics/features/mlx5.ini | 1 +
> doc/guides/nics/mlx5.rst | 86 ++++--
> doc/guides/rel_notes/release_26_07.rst | 11 +
> doc/guides/testpmd_app_ug/run_app.rst | 20 ++
> drivers/common/mlx5/linux/mlx5_common_verbs.c | 53 ++--
> drivers/common/mlx5/mlx5_common.c | 6 +-
> drivers/common/mlx5/mlx5_common_mr.c | 37 ++-
> drivers/common/mlx5/mlx5_common_mr.h | 29 +-
> drivers/common/mlx5/windows/mlx5_common_os.c | 31 ++-
> drivers/compress/mlx5/mlx5_compress.c | 4 +-
> drivers/crypto/mlx5/mlx5_crypto.h | 2 -
> drivers/crypto/mlx5/mlx5_crypto_gcm.c | 6 +-
> drivers/net/mlx5/mlx5.c | 7 +
> drivers/net/mlx5/mlx5.h | 4 +-
> drivers/net/mlx5/mlx5_ethdev.c | 25 ++
> drivers/net/mlx5/mlx5_flow_aso.c | 21 +-
> drivers/net/mlx5/mlx5_flow_hw.c | 11 +-
> drivers/net/mlx5/mlx5_flow_quota.c | 6 +-
> drivers/net/mlx5/mlx5_hws_cnt.c | 19 +-
> drivers/net/mlx5/mlx5_rx.c | 186 ++++++++-----
> drivers/net/mlx5/mlx5_rx.h | 5 +-
> drivers/net/mlx5/mlx5_rxq.c | 75 +++--
> drivers/net/mlx5/mlx5_trigger.c | 64 ++++-
> dts/api/capabilities.py | 2 +
> dts/api/testpmd/__init__.py | 17 ++
> dts/api/testpmd/types.py | 6 +
> dts/framework/testbed_model/capability.py | 10 +-
> dts/tests/TestSuite_rx_split.py | 262 ++++++++++++++++++
> lib/ethdev/rte_ethdev.c | 24 +-
> lib/ethdev/rte_ethdev.h | 14 +-
> 35 files changed, 880 insertions(+), 274 deletions(-)
> create mode 100644 dts/tests/TestSuite_rx_split.py
>
Still has some issues:
Patch 6: net/mlx5: support selective Rx
Error: after a non-critical Rx error CQE, the next packet is processed with
a stale length and stale CQE; the queue effectively wedges.
To avoid re-polling the CQE on each leading discard segment of one packet,
v6 wraps the poll in "if (len == 0)" and resets len to 0 at the points where
a packet finishes. Two of the three exits reset it (normal completion and
the "no real segment found" skip both do "len = 0;"), but the non-critical
error-CQE path does not:
if (unlikely(len & MLX5_ERROR_CQE_MASK)) {
if (seg->pool)
rte_mbuf_raw_free(rep);
if (len == MLX5_CRITICAL_ERROR_CQE_RET) {
rq_ci = rxq->rq_ci << sges_n;
break;
}
rq_ci >>= sges_n;
rq_ci += skip_cnt;
rq_ci <<= sges_n;
MLX5_ASSERT(!pkt);
continue; /* len still == MLX5_ERROR_CQE_MASK (0x40000000) */
}
MLX5_ERROR_CQE_MASK is 0x40000000, so len is non-zero on this continue. The
next iteration hits "if (len == 0)" == false and skips mlx5_rx_poll_len()
entirely. The following real segment then sets pkt with PKT_LEN(pkt) ==
0x40000000 and rxq_cq_to_mbuf() reads the stale cqe/mcqe. Because
0x40000000 > DATA_LEN(seg), the "more data" branch keeps consuming
descriptors as one bogus giant packet, walking the whole ring and emitting
nothing. Pre-v6 this worked because the poll was unconditional in the !pkt
block.
Suggested fix: reset len before the error-skip continue, matching the other
two exits:
rq_ci >>= sges_n;
rq_ci += skip_cnt;
rq_ci <<= sges_n;
MLX5_ASSERT(!pkt);
len = 0;
continue;
Minor: "tail = seg;" is now set in both the "first real segment" block and
the "real segment: replenish WQE" block; the first is redundant since the
second always runs for the same segment. Harmless, but the duplicate can be
dropped.
^ permalink raw reply
* Re: [PATCH] net/ark: validate IPv4 octets in address parser
From: Stephen Hemminger @ 2026-06-03 17:29 UTC (permalink / raw)
To: Denis Sergeev; +Cc: dev, shepard.siegel, ed.czeck, john.miller, sdl.dpdk
In-Reply-To: <20260603054742.120101-1-denserg.edu@gmail.com>
On Wed, 3 Jun 2026 08:47:38 +0300
Denis Sergeev <denserg.edu@gmail.com> wrote:
> The IPv4 parsing helper used by pktgen and pktchkr reads each octet
> with "%u". This allows values above 255 to be accepted from the
> configuration file and encoded into unintended device register values.
>
> Reject parsed octets outside the IPv4 byte range before assembling
> the 32-bit address.
>
> Found by Linux Verification Center (linuxtesting.org) with SVACE.
>
> Fixes: 9c7188a68d7b ("net/ark: provide API for hardware modules pktchkr and pktgen")
>
> Signed-off-by: Denis Sergeev <denserg.edu@gmail.com>
> ---
Have to ask, why a driver is rolling it own IP address parser, that is a bad idea.
But then the whole builtin vendor pktgen in a driver is bad idea.
Claude scan of the driver found lots more issues.
Hate to think what Mythos would find here...
^ permalink raw reply
* [DPDK/ethdev Bug 1950] Lots of problems found from review of ARK PMD
From: bugzilla @ 2026-06-03 17:24 UTC (permalink / raw)
To: dev
http://bugs.dpdk.org/show_bug.cgi?id=1950
Bug ID: 1950
Summary: Lots of problems found from review of ARK PMD
Product: DPDK
Version: 22.03
Hardware: All
OS: All
Status: UNCONFIRMED
Severity: major
Priority: Normal
Component: ethdev
Assignee: dev@dpdk.org
Reporter: stephen@networkplumber.org
Target Milestone: ---
In addition to the problems identified by LF scan, using Claude Opus found a
lot more in this driver:
Component: drivers/net/ark
Version: main
Manual review of the full ark driver tree (ark_ethdev.c, ark_ethdev_rx.c,
ark_ethdev_tx.c, ark_pktgen.{c,h}, ark_pktchkr.{c,h}, ark_mpu.c, ark_pktdir.c)
turned up the issues below, in addition to existing scanner reports (strncat
length-arg misuse and non-NUL-terminating strncpy in process_file_args;
unchecked rte_zmalloc_socket before memcpy in ark_dev_init).
== SECURITY (suggest separate tracking item) ==
S1. Unvalidated dlopen() driver overload in check_for_ext() (ark_ethdev.c)
Loads an arbitrary .so from the ARK_EXT_PATH environment variable
(dlopen(path, RTLD_LOCAL | RTLD_LAZY)) and resolves ~12 rte_pmd_ark_*
symbols that are cast to function pointers and called on the control and
data paths. No path validation, allow-list, ownership/permission check, or
integrity check. DPDK apps commonly run privileged, so any process able to
influence the environment gets arbitrary in-process code execution -- a
local code-injection / privesc vector. RTLD_LAZY defeats up-front symbol
validation; unchecked dlsym casts are UB on ABI mismatch.
Recommendation: gate behind a build option off by default; refuse to load
when privileged or when the file is not root-owned / is world-writable;
document as debug-only.
== HIGH - crashes / memory safety ==
1. NULL deref in ark_pktgen_setup() (ark_pktgen.c): pktgen toptions[] has no
"port" entry (only pktchkr's does), but the run branch calls
options("port")->v.INT; options() returns NULL on miss. Crashes whenever
Pkt_gen config sets run=1.
2. ark_tx_queue_release() missing NULL guard (ark_ethdev_tx.c): ark_dev_close()
calls it for every tx_queues[] slot and ark_dev_stop() loops over
ark_tx_queue_stop() without checks; both deref queue->ddm / cons_index. RX
side guards (queue == 0); TX side does not -> NULL deref on close/stop with
any un-setup TX slot.
3. ark_dev_rx_queue_count() missing NULL guard (ark_ethdev_rx.c): dereferences
queue with no queue == 0 check, unlike every other entry point in the file.
4. mac_addrs alloc failure not handled (ark_ethdev.c ark_dev_init, ~L381):
failure is logged but execution falls through with mac_addrs == NULL; the
secondary-port path (~L461) uses goto error, the primary does not.
5. error: path leaks secondary ports (ark_ethdev.c ark_dev_init, ~L480): only
the primary mac_addrs is freed; already-probed secondary eth_devs and their
dev_private/mac_addrs leak.
== HIGH - silent data corruption ==
6. OTLONG written through wrong union member (ark_pktgen.c pmd_set_arg,
ark_pktchkr.c set_arg): "case OTLONG: o->v.INT = atoll(val);" truncates to
32 bits and leaves stale upper bits in .LONG. Affects num_pkts,
src_mac_addr, dst_mac_addr. Should be o->v.LONG.
7. src_mac_addr / num_pkts read via .v.INT in setup (both *_setup()): these are
OTLONG; reading .INT drops upper MAC bytes (default 0xdC3cF6425060).
Adjacent
dst_mac_addr correctly uses .LONG.
== MEDIUM ==
8. parse_ipv4_string() return type (ark_pktgen.c, ark_pktchkr.c): declared
int32_t but builds values up to 0xFFFFFFFF via unsigned math; default
169.254.10.240 goes negative on conversion. Should be uint32_t. "return 0"
on parse failure also aliases a valid 0.0.0.0.
9. Divide-by-zero risk in ark_api_num_queues_per_port() (ark_mpu.c):
num_queues / ark_ports; ark_ports derives from untrusted
user_ext.dev_get_port_count(), unchecked for 0.
10. pkt_size_min > pkt_size_max defaults: 2006/2005 vs 1514 in both option
tables; likely a typo.
11. dst_port / src_port default 65536: out of range for a 16-bit port.
== LOW / cleanup ==
12. Misplaced log in ark_pktchkr_wait_done(): "pktgen done" prints every loop
iteration instead of once after completion.
13. ark_pktchkr_get_pkts_sent() returns int for a uint32_t register
(signedness); pktgen version returns uint32_t.
14. *_parse() tokenizer/doc mismatch: comments document comma-separated args;
toks[] omits ','.
15. inst->dev_info never set/used in ark_pkt_gen_inst / ark_pkt_chkr_inst
(allocated via rte_malloc, not zmalloc).
16. check_for_ext() "found" is dead: never reassigned; only the -1
dlopen-failure path is meaningful.
17. ark_mpu_configure() logs ring_size (uint32_t) with %d.
== PROCESS / TOOLING ==
P1. Non-inclusive naming not caught by review tooling: en_slaved_start
(ark_pktgen.c toptions[] and the ark_pktgen_set_pkt_ctrl() parameter) uses
terminology DPDK has been retiring. The substantive point is that
checkpatches.sh / the checkpatch master|slave test should have flagged this
and did not -- indicating the file was either grandfathered in or the check
is not being run against this tree. Coverage gap in the tooling, not just a
naming nit. Suggest renaming (e.g. en_gated_start) and reviewing why the
automated check missed it.
--
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply
* [PATCH v2] net/af_packet: fix qpairs argument upper bound check
From: Denis Sergeev @ 2026-06-03 17:08 UTC (permalink / raw)
To: dev; +Cc: stephen, Denis Sergeev
In-Reply-To: <20260603044228.117357-1-denserg.edu@gmail.com>
The qpairs vdev argument was parsed with atoi(), which does no
validation: trailing garbage is ignored and a negative input such
as "-1" wraps to UINT_MAX when stored in the unsigned qpairs field,
passing the existing "< 1" check and reaching rte_pmd_init_internals()
as nb_queues. This causes excessive socket and memory allocation in
the per-queue loop.
Parse the value with strtoul() and reject non-numeric input, trailing
characters, negative values and values outside the
[1, RTE_MAX_QUEUES_PER_PORT] range.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: ccd37d341e8d ("net/af_packet: remove queue number limitation")
Signed-off-by: Denis Sergeev <denserg.edu@gmail.com>
---
v2:
* Replace atoi() with strtoul() and validate the parsed value
drivers/net/af_packet/rte_eth_af_packet.c | 11 +++++++++--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/drivers/net/af_packet/rte_eth_af_packet.c b/drivers/net/af_packet/rte_eth_af_packet.c
index 8303ff5ca9..b7758a5c75 100644
--- a/drivers/net/af_packet/rte_eth_af_packet.c
+++ b/drivers/net/af_packet/rte_eth_af_packet.c
@@ -1168,13 +1168,20 @@ rte_eth_from_packet(struct rte_vdev_device *dev,
for (k_idx = 0; k_idx < kvlist->count; k_idx++) {
pair = &kvlist->pairs[k_idx];
if (strstr(pair->key, ETH_AF_PACKET_NUM_Q_ARG) != NULL) {
- qpairs = atoi(pair->value);
- if (qpairs < 1) {
+ char *endptr;
+ unsigned long num;
+
+ errno = 0;
+ num = strtoul(pair->value, &endptr, 10);
+ if (errno != 0 || endptr == pair->value ||
+ *endptr != '\0' || pair->value[0] == '-' ||
+ num < 1 || num > RTE_MAX_QUEUES_PER_PORT) {
PMD_LOG(ERR,
"%s: invalid qpairs value",
name);
return -1;
}
+ qpairs = num;
continue;
}
if (strstr(pair->key, ETH_AF_PACKET_BLOCKSIZE_ARG) != NULL) {
--
2.50.1
^ permalink raw reply related
* [PATCH v3] net/ark: fix null dereference on allocation failure
From: Denis Sergeev @ 2026-06-03 16:32 UTC (permalink / raw)
To: dev; +Cc: shepard.siegel, ed.czeck, john.miller, stephen, stable,
Denis Sergeev
In-Reply-To: <20260603053214.119296-1-denserg.edu@gmail.com>
rte_zmalloc_socket() can return NULL when memory is exhausted.
The result is passed directly to memcpy() without a NULL check,
leading to undefined behavior. Add a NULL check consistent with
the existing check for mac_addrs allocation in the same function.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: bf73ee28f47b ("net/ark: support single function with multiple port")
Cc: stable@dpdk.org
Signed-off-by: Denis Sergeev <denserg.edu@gmail.com>
---
v3:
* Use a single-line error message without "!" and "Exiting"
v2:
* Fix Fixes: tag to use 12-char sha1 (checkpatch BAD_FIXES_TAG)
drivers/net/ark/ark_ethdev.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ark/ark_ethdev.c b/drivers/net/ark/ark_ethdev.c
index 8b25ed948f..546a44704f 100644
--- a/drivers/net/ark/ark_ethdev.c
+++ b/drivers/net/ark/ark_ethdev.c
@@ -445,6 +445,10 @@ ark_dev_init(struct rte_eth_dev *dev)
sizeof(struct ark_adapter),
RTE_CACHE_LINE_SIZE,
rte_socket_id());
+ if (eth_dev->data->dev_private == NULL) {
+ ARK_PMD_LOG(ERR, "Memory allocation for dev_private failed\n");
+ goto error;
+ }
memcpy(eth_dev->data->dev_private, ark,
sizeof(struct ark_adapter));
--
2.50.1
^ permalink raw reply related
* Re: [PATCH v2] eal: fix data race in hugepage prefault
From: Thomas Monjalon @ 2026-06-03 15:57 UTC (permalink / raw)
To: Stephen Hemminger
Cc: dev, stable, Michal Sieron, Anatoly Burakov, Bruce Richardson
In-Reply-To: <20260601160019.116067-1-stephen@networkplumber.org>
01/06/2026 18:00, Stephen Hemminger:
> The prefault step in alloc_seg() reads a value from the hugepage and
> writes it back unchanged to force the kernel to commit the backing
> page. The read and write were not atomic, which races with concurrent
> access to the same physical page from a secondary process attaching
> to the hugetlbfs-backed mapping during rte_eal_init().
>
> Replace the non-atomic load+store with a single atomic fetch-or of
> zero. This touches the page with an atomic read-modify-write without
> changing its contents, eliminating the race while preserving the
> original intent of forcing a write fault.
>
> Fixes: 0f1631be24bd ("mem: fix page fault trigger")
> Cc: stable@dpdk.org
>
> Reported-by: Michal Sieron <michal.sieron@nokia.com>
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Applied, thanks.
^ permalink raw reply
* Re: [PATCH] eal: fix function versioning with LTO
From: Stephen Hemminger @ 2026-06-03 15:53 UTC (permalink / raw)
To: David Marchand; +Cc: dev, stable, Thomas Monjalon
In-Reply-To: <CAJFAV8w+bUJGTBSOgqp8xOHXm5av2xdFMo8TNwr=kP87R30AYg@mail.gmail.com>
On Wed, 3 Jun 2026 12:01:48 +0200
David Marchand <david.marchand@redhat.com> wrote:
> Hello,
>
> On Wed, 3 Jun 2026 at 00:57, Stephen Hemminger
> <stephen@networkplumber.org> wrote:
> >
> > When using function versioning and building with Link Time Optimization,
> > the compiler does not see the __asm__ annotation of symbols and
> > therefore thinks there are two versions of the same symbol.
> >
> > The fix is to use compiler symver attribute on the function which
> > was added in GCC 10. Keep the older method for backward compatibility
> > with older compilers.
> >
> > Bugzilla ID: 1949
> > Fixes: e30e194c4d06 ("eal: rework function versioning macros")
>
> We never used the symver stuff, so it seems unlikely the issue was
> introduced with this rework.
>
> The fact that clang does not support this attribute is a concern.
Clang doesn't have this problem. It works as is.
> > Cc: stable@dpdk.org
>
> Why do we need to backport?
Well LTO has worked for a long time, it is not experimental just
not commonly done since it takes so long to build.
We were doing it years ago at MSFT.
>
> LTO is kind of experimental, so it seems good enough to reply "not
> expected to work in older LTS" if someone reported an issue.
>
> And in practice, no LTS release call the versioning macros, since a
> LTS drops all compatibility.
>
> > Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
>
> I would like to reproduce, but I can't build main with LTO.
> What patches did you apply locally to avoid warnings on the hash library?
I use Debian testing and GCC 15 but shows up on older versions as well
I get no warnings building main
$ git am ~/Downloads/v2-ethdev-add-buffer-size-parameter-to-rte_eth_dev_get_name_by_port.patch
$ meson setup build-lto -Db_lto=true
$ ninja -C build-lto
ninja: Entering directory `build-lto'
[620/3781] Linking target lib/librte_ethdev.so.26.2
FAILED: [code=1] lib/librte_ethdev.so.26.2
cc -o lib/librte_ethdev.so.26.2 lib/librte_ethdev.so.26.2.p/ethdev_ethdev_driver.c.o lib/librte_ethdev.so.26.2.p/ethdev_ethdev_private.c.o lib/librte_ethdev.so.26.2.p/ethdev_ethdev_profile.c.o lib/librte_ethdev.so.26.2.p/ethdev_ethdev_trace_points.c.o lib/librte_ethdev.so.26.2.p/ethdev_rte_class_eth.c.o lib/librte_ethdev.so.26.2.p/ethdev_rte_ethdev.c.o lib/librte_ethdev.so.26.2.p/ethdev_rte_ethdev_cman.c.o lib/librte_ethdev.so.26.2.p/ethdev_rte_ethdev_telemetry.c.o lib/librte_ethdev.so.26.2.p/ethdev_rte_flow.c.o lib/librte_ethdev.so.26.2.p/ethdev_rte_mtr.c.o lib/librte_ethdev.so.26.2.p/ethdev_rte_tm.c.o lib/librte_ethdev.so.26.2.p/ethdev_sff_telemetry.c.o lib/librte_ethdev.so.26.2.p/ethdev_sff_common.c.o lib/librte_ethdev.so.26.2.p/ethdev_sff_8079.c.o lib/librte_ethdev.so.26.2.p/ethdev_sff_8472.c.o lib/librte_ethdev.so.26.2.p/ethdev_sff_8636.c.o lib/librte_ethdev.so.26.2.p/ethdev_ethdev_linux_ethtool.c.o -flto=auto -Wl,--as-needed -Wl,--no-undefined -Wl,-O1 -shared -fPIC -Wl,-soname,librte_ethdev.so.26 -Wl,--no-as-needed -Wl,--undefined-version -pthread -Wl,--start-group -lm -ldl -lnuma -lfdt '-Wl,-rpath,$ORIGIN/' lib/librte_eal.so.26.2 lib/librte_kvargs.so.26.2 lib/librte_log.so.26.2 lib/librte_telemetry.so.26.2 lib/librte_argparse.so.26.2 lib/librte_net.so.26.2 lib/librte_mbuf.so.26.2 lib/librte_mempool.so.26.2 lib/librte_ring.so.26.2 lib/librte_meter.so.26.2 -Wl,--version-script=/home/shemminger/DPDK/lto/build-lto/lib/ethdev_exports.map /usr/lib/x86_64-linux-gnu/libbsd.so /usr/lib/x86_64-linux-gnu/libarchive.so -Wl,--end-group
/tmp/cc3RQyqL.s: Assembler messages:
/tmp/cc3RQyqL.s: Error: invalid attempt to declare external version name as default in symbol `rte_eth_dev_get_name_by_port@@DPDK_27'
make: *** [/tmp/ccVzgiZ2.mk:2: /tmp/ccTlGfA9.ltrans0.ltrans.o] Error 1
make: *** Waiting for unfinished jobs....
lto-wrapper: fatal error: make returned 2 exit status
compilation terminated.
^ permalink raw reply
* Re: [PATCH v4 0/2] net/intel: optimize for fast-free hint
From: Bruce Richardson @ 2026-06-03 15:56 UTC (permalink / raw)
To: Morten Brørup; +Cc: dev, ciara.loftus
In-Reply-To: <98CBD80474FA8B44BF855DF32C47DC35F658DD@smartserver.smartshare.dk>
On Tue, Jun 02, 2026 at 06:26:13PM +0200, Morten Brørup wrote:
> > From: Bruce Richardson [mailto:bruce.richardson@intel.com]
> > Sent: Tuesday, 2 June 2026 17.45
> >
> > When the fast-free hint is provided to the driver we know that the
> > mbufs
> > have refcnt of 1 and are from the same mempool. Therefore, we can
> > optimize a bit for this case even in the scalar path of our drivers.
> >
> > ---
> > v4:
> > * add precursor patch to adjust mbuf pointers so that the DD bit
> > is written to a descriptor with a valid mbuf pointer associated
> > with it.
> >
> > v3:
> > * used mbuf_raw_free_bulk rather than mempool function directly
> > * check for fast_free via mp pointer rather than flags
> > * remove unnecessary prefetches
> >
> > V2: Fix issues with original submission:
> > * missed check for NULL mbufs
> > * fixed issue with freeing directly from sw_ring in scalar path which
> > doesn't work as thats not a flag array of pointers
> > * fixed missing null assignment in case of large segments for TSO
> >
> >
> > Bruce Richardson (2):
> > net/intel: write mbuf for last Tx desc of segment
> > net/intel: optimize for fast-free hint
> >
> > drivers/net/intel/common/tx.h | 21 ++++--
> > drivers/net/intel/common/tx_scalar.h | 98 +++++++++++++++++++++-------
> > 2 files changed, 90 insertions(+), 29 deletions(-)
> >
> > --
> > 2.53.0
>
> Good catch by Ciara, and good solution to it.
> Series-Acked-by: Morten Brørup <mb@smartsharesystems.com>
>
Applied to dpdk-next-net-intel.
Thanks for all the reviews folks. Quick tests on next-net-intel tree with
scalar driver show a nice perf bump from this change.
/Bruce
^ permalink raw reply
* Re: [PATCH] net/ark: fix unsafe env variable in extension loading
From: Stephen Hemminger @ 2026-06-03 15:30 UTC (permalink / raw)
To: Denis Sergeev
Cc: dev, shepard.siegel, ed.czeck, john.miller, stable, sdl.dpdk
In-Reply-To: <20260603052604.118850-1-denserg.edu@gmail.com>
On Wed, 3 Jun 2026 08:26:00 +0300
Denis Sergeev <denserg.edu@gmail.com> wrote:
> diff --git a/drivers/net/ark/ark_ethdev.c b/drivers/net/ark/ark_ethdev.c
> index 8b25ed948f..e25478103b 100644
> --- a/drivers/net/ark/ark_ethdev.c
> +++ b/drivers/net/ark/ark_ethdev.c
> @@ -211,9 +211,19 @@ static int
> check_for_ext(struct ark_adapter *ark)
> {
> int found = 0;
> + const char *dllpath;
> +
> + /*
> + * A basic security check is necessary before trusting
> + * ARK_EXT_PATH environment variable.
> + */
> + if (geteuid() != getuid() || getegid() != getgid()) {
> + ARK_PMD_LOG(DEBUG, "EXT ignoring ARK_EXT_PATH under setuid/setgid\n");
> + return 0;
> + }
>
DPDK may be run in containers. This would break that.
The whole dlopen extension stuff in this driver is rubbish and should not have been allowed in.
It creates testing and security nightmares.
^ permalink raw reply
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