[PATCH] net/mlx5: optimize counters ID greneration logic

Enqueue generated counter IDs on a ring in bulk. Generate them and store in an array before putting them on a ring all at once. That bring better cache access and speeds up the mlx5_hws_cnt_pool_create() function. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_hws_cnt.c | 19

Re: [PATCH v3] net/mlx5: fix RSS hash for non-RSS CQE zipping

Raslan, please revert this patch. I rejected it last week. This fix is incorrect without the FW changes. Regards, Alex From: Raslan Darawsheh Sent: Sunday, January 19, 2025 6:47:48 a.m. To: Alexander Kozyrev ; dev@dpdk.org Cc: sta...@dpdk.org ; Slava Ovsiienko

[PATCH] doc: document hash RSS limitations with CQE zipping

RSS hash is only fully supported when the Hash RSS format is selected. Update MLX5 documentation about lack of RTE_MBUF_F_RX_RSS_HASH flag for some packets in case of any other CQE compression format is configured. Signed-off-by: Alexander Kozyrev --- doc/guides/nics/mlx5.rst | 7 +-- 1

[PATCH v2] net/mlx5/hws: fix fragmented ptype match

matching in case of L2/L3 mask is provided as well. Mask out L2/L3 types when setting L4_FRAG. Fixes: 761439a20f net/mlx5/hws: support fragmented packet type matching Cc: sta...@dpdk.org Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 4 ++-- 1 file changed, 2

[PATCH] net/mlx5: fix fragmented ptype match

matching in case of L2/L3 mask is provided as well. Mask out L2/L3 types when setting L4_FRAG. Fixes: 761439a20f net/mlx5/hws: support fragmented packet type matching Cc: sta...@dpdk.org Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 4 ++-- 1 file changed, 2

[PATCH v3] net/mlx5: fix RSS hash for non-RSS CQE zipping

.@dpdk.org Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_rxtx_vec_altivec.h | 32 +--- drivers/net/mlx5/mlx5_rxtx_vec_neon.h| 18 ++--- drivers/net/mlx5/mlx5_rxtx_vec_sse.h | 18 ++--- 3 files changed, 35 insertions(+), 33 deletions(-) di

[PATCH v2] net/mlx5: fix RSS hash for non-RSS CQE zipping

Take the RSS hash and flow tag values from the title packet before they get overwritten by the decompressing routine. Set the RSS hash flag in the packet mbuf if RSS is enabled in case of non-RSS CQE zipping format. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_rxtx_vec_altivec.h

[PATCH] net/mlx5: fix RSS hash for non-RSS CQE zipping

Take the RSS hash value for the title packet for flow tag and packet header CQE zipping formats. Fixes: 54c2d46b16 ("net/mlx5: support flow tag and packet header miniCQEs") Cc: sta...@dpdk.org Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_rxtx_vec_altivec.h | 9 +---

[PATCH v3] net/mlx5: fix miniCQEs number calculation

: add vectorized Rx/Tx burst for x86") Signed-off-by: Alexander Kozyrev Acked-by: Viacheslav Ovsiienko --- drivers/net/mlx5/mlx5_rxtx_vec_altivec.h | 3 +-- drivers/net/mlx5/mlx5_rxtx_vec_neon.h| 3 +-- drivers/net/mlx5/mlx5_rxtx_vec_sse.h | 3 +-- 3 files changed, 3 insertions(+), 6 deletion

[PATCH] net/mlx5: fix shared Rx queue port number in data path

Wrong CQE is used to get the shared Rx queue port number in vectorized Rx burst routine. Fix the CQE indexing. Fixes: 25ed2ebff1 ("net/mlx5: support shared Rx queue port data path") Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_rxtx_vec_altivec.h | 12 ++-- d

[PATCH v3] drivers: fix error CQE handling for 128 bytes CQE

bytes and doesn't take the padding into consideration on platforms with 128-byte cache lines. Fix the error CQE size in all error handling routines in mlx5. Fixes: 957e45fb7b ("net/mlx5: handle Tx completion with error") Cc: sta...@dpdk.org Signed-off-by: Alexander Kozyrev Acked

[PATCH v2] net/mlx5: fix miniCQEs number calculation

: add vectorized Rx/Tx burst for x86") Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_rxtx_vec_altivec.h | 3 +-- drivers/net/mlx5/mlx5_rxtx_vec_neon.h| 3 +-- drivers/net/mlx5/mlx5_rxtx_vec_sse.h | 3 +-- 3 files changed, 3 insertions(+), 6 deletions(-) diff --git a/driver

[PATCH] net/mlx5: fix miniCQEs number calculation

: add vectorized Rx/Tx burst for x86") Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_rxtx_vec_altivec.h | 3 +-- drivers/net/mlx5/mlx5_rxtx_vec_neon.h| 3 +-- drivers/net/mlx5/mlx5_rxtx_vec_sse.h | 3 +-- 3 files changed, 3 insertions(+), 6 deletions(-) diff --git a/driver

[PATCH v3 0/5] jump to table index support in mlx5

Support the new Flow API JUMP_TO_TABLE_INDEX action in mlx5. Seried-acked-by: Dariusz Sosnowski Alexander Kozyrev (3): net/mlx5: create array ste matcher net/mlx5: add flow rule insertion by index with pattern net/mlx5: implement jump to table index action Hamdan Igbaria (2): net/mlx5

[PATCH v3 5/5] net/mlx5: implement jump to table index action

insertion type is by index with pattern. Signed-off-by: Alexander Kozyrev --- doc/guides/nics/features/mlx5.ini | 1 + doc/guides/prog_guide/ethdev/flow_offload.rst | 24 +++ doc/guides/rel_notes/release_24_11.rst| 1 + drivers/net/mlx5/mlx5_flow.h | 8

[PATCH v4 3/5] net/mlx5: create array ste matcher

oceed with the default miss instead. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_flow_hw.c | 11 +-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c index c21eb1eaed..c236831e21 100644 --- a/dr

[PATCH v3 3/5] net/mlx5: create array ste matcher

oceed with the default miss instead. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_flow_hw.c | 11 +-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c index c21eb1eaed..c236831e21 100644 --- a/dr

[PATCH v4 1/5] net/mlx5/hws: introduce new matcher type

From: Hamdan Igbaria introduce STE array matcher, where this matcher can only be isolated under a parent table and not chained to the table matchers chain. Signed-off-by: Hamdan Igbaria --- drivers/net/mlx5/hws/mlx5dr.h | 13 +- drivers/net/mlx5/hws/mlx5dr_debug.c | 12 +- dr

Re: [PATCH] mbuf: add transport mode ESP packet type

>>And we definitely need RTE_PTYPE_INNER_L4_ESP for ESP over UDP support. >Isn't this already taken care when mbuf->packet_type = >(RTE_PTYPE_L4_UDP | RTE_PTYPE_TUNNEL_ESP) ? This is ambigous. And both UDP and ESP are L4 headers, which can lead to the undefined behavior when we specify both of the

[PATCH v4 5/5] net/mlx5: implement jump to table index action

insertion type is by index with pattern. Signed-off-by: Alexander Kozyrev --- doc/guides/nics/features/default.ini | 1 + doc/guides/nics/features/mlx5.ini | 1 + doc/guides/prog_guide/ethdev/flow_offload.rst | 24 +++ doc/guides/rel_notes/release_24_11.rst| 1

[PATCH v4 2/5] net/mlx5/hws: introduce jump to matcher action

From: Hamdan Igbaria Introduce jump to matcher action, this action will allow jumping to another matcher. For now this jump restricted to STE array matchers and matchers of size 1. Signed-off-by: Hamdan Igbaria --- drivers/net/mlx5/hws/mlx5dr.h| 29 ++ drivers/net/mlx5/hws/mlx5

[PATCH v4 4/5] net/mlx5: add flow rule insertion by index with pattern

Implement rte_flow_async_create_by_index_with_pattern() function. Rework the driver implementation to reduce code duplication by providing a single flow insertion routine, that can be called with different parameters depending on the insertion type. Signed-off-by: Alexander Kozyrev --- doc

[PATCH v4 0/5] jump to table index support in mlx5

Support the new Flow API JUMP_TO_TABLE_INDEX action in mlx5. Seried-acked-by: Dariusz Sosnowski Alexander Kozyrev (3): net/mlx5: create array ste matcher net/mlx5: add flow rule insertion by index with pattern net/mlx5: implement jump to table index action Hamdan Igbaria (2): net/mlx5

[PATCH v3 1/5] net/mlx5/hws: introduce new matcher type

From: Hamdan Igbaria introduce STE array matcher, where this matcher can only be isolated under a parent table and not chained to the table matchers chain. Signed-off-by: Hamdan Igbaria --- drivers/net/mlx5/hws/mlx5dr.h | 13 +- drivers/net/mlx5/hws/mlx5dr_debug.c | 12 +- dr

[PATCH v3 4/5] net/mlx5: add flow rule insertion by index with pattern

Implement rte_flow_async_create_by_index_with_pattern() function. Rework the driver implementation to reduce code duplication by providing a single flow insertion routine, that can be called with different parameters depending on the insertion type. Signed-off-by: Alexander Kozyrev --- doc

[PATCH v3 2/5] net/mlx5/hws: introduce jump to matcher action

From: Hamdan Igbaria Introduce jump to matcher action, this action will allow jumping to another matcher. For now this jump restricted to STE array matchers and matchers of size 1. Signed-off-by: Hamdan Igbaria --- drivers/net/mlx5/hws/mlx5dr.h| 29 ++ drivers/net/mlx5/hws/mlx5

[PATCH] net/mlx5: add support for transport mode ESP ptype

Implement matching on RTE_PTYPE_L4_ESP and RTE_PTYPE_INNER_L4_ESP for IP Encapsulating Security Payload (ESP) in transport mode. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 33 +++ drivers/net/mlx5/hws/mlx5dr_definer.h | 2 ++ 2 files

Re: [PATCH] mbuf: add transport mode ESP packet type

> I am curious, where is the driver that implements this? I'll send MLX5 implementation shortly.

[PATCH v2 4/5] net/mlx5: add flow rule insertion by index with pattern

Implement rte_flow_async_create_by_index_with_pattern() function. Rework the driver implementation to reduce code duplication by providing a single flow insertion routine, that can be called with different parameters depending on the insertion type. Signed-off-by: Alexander Kozyrev --- drivers

[PATCH v2 5/5] net/mlx5: implement jump to table index action

insertion type is by index with pattern. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_flow.h| 8 +- drivers/net/mlx5/mlx5_flow_hw.c | 145 2 files changed, 151 insertions(+), 2 deletions(-) diff --git a/drivers/net/mlx5/mlx5_flow.h b/drivers/net

[PATCH v2 2/5] net/mlx5/hws: introduce jump to matcher action

From: Hamdan Igbaria Introduce jump to matcher action, this action will allow jumping to another matcher. For now this jump restricted to STE array matchers and matchers of size 1. Signed-off-by: Hamdan Igbaria --- drivers/net/mlx5/hws/mlx5dr.h| 29 ++ drivers/net/mlx5/hws/mlx5

[PATCH v2 3/5] net/mlx5: create array ste matcher

oceed with the default miss instead. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_flow_hw.c | 11 +-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c index b9807f347d..6434937562 100644 --- a/dr

[PATCH v2 1/5] net/mlx5/hws: introduce new matcher type

From: Hamdan Igbaria introduce STE array matcher, where this matcher can only be isolated under a parent table and not chained to the table matchers chain. Signed-off-by: Hamdan Igbaria --- drivers/net/mlx5/hws/mlx5dr.h | 13 +- drivers/net/mlx5/hws/mlx5dr_debug.c | 12 +- dr

[PATCH 5/5] net/mlx5: implement jump to table index action

insertion type is by index with pattern. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_flow.h| 8 +- drivers/net/mlx5/mlx5_flow_hw.c | 145 2 files changed, 151 insertions(+), 2 deletions(-) diff --git a/drivers/net/mlx5/mlx5_flow.h b/drivers/net/mlx5

[PATCH 3/5] net/mlx5: create array ste matcher

oceed with the default miss instead. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_flow_hw.c | 11 +-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c index b9807f347d..6434937562 100644 --- a/dr

[PATCH 4/5] net/mlx5: add flow rule insertion by index with pattern

Implement rte_flow_async_create_by_index_with_pattern() function. Rework the driver implementaion to reduce the code duplication by providing a signle flow insertion routine, that can be called with different parameters depending on the insertion type. Signed-off-by: Alexander Kozyrev

[PATCH 1/5] net/mlx5/hws: introduce new matcher type

From: Hamdan Igbaria introduce STE array matcher, where this matcher can only be isolated under a parent table and not chained to the table matchers chain. Signed-off-by: Hamdan Igbaria --- drivers/net/mlx5/hws/mlx5dr.h | 13 +- drivers/net/mlx5/hws/mlx5dr_debug.c | 12 +- dr

[PATCH 2/5] net/mlx5/hws: introduce jump to matcher action

From: Hamdan Igbaria Introduce jump to matcher action, this action will allow jumping to another matcher. For now this jump restricted to STE array matchers and matchers of size 1. Signed-off-by: Hamdan Igbaria --- drivers/net/mlx5/hws/mlx5dr.h| 29 ++ drivers/net/mlx5/hws/mlx5

[PATCH v2] drivers: ignore non-critical syndromes for Tx queues

ned-off-by: Alexander Kozyrev --- drivers/common/mlx5/mlx5_prm.h | 17 + drivers/net/mlx5/mlx5_rx.c | 4 +--- drivers/net/mlx5/mlx5_tx.c | 4 ++-- 3 files changed, 20 insertions(+), 5 deletions(-) diff --git a/drivers/common/mlx5/mlx5_prm.h b/drivers/common/mlx5/mlx5_p

[PATCH v2] drivers: fix error CQE handling

bytes and doesn't take the padding into consideration on platforms with 128-byte cache lines. Fix the error CQE size in all error handling routines in mlx5. Fixes: 957e45fb7b ("net/mlx5: handle Tx completion with error") Cc: sta...@dpdk.org Signed-off-by: Alexander Kozyrev ---

[PATCH v2 6/7] app/testpmd: add jump to table index action

Signed-off-by: Alexander Kozyrev --- app/test-pmd/cmdline_flow.c | 34 ++ 1 file changed, 34 insertions(+) diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c index 65030936d2..d150f6cca8 100644 --- a/app/test-pmd/cmdline_flow.c +++ b/app/test-pmd

[PATCH v2 7/7] ethdev: add trace points to flow insertion by index

Adds trace points for rte_flow rule insertion by index functions: rte_flow_async_create_by_index and rte_flow_async_create_by_index_with_pattern. Signed-off-by: Alexander Kozyrev --- lib/ethdev/ethdev_trace.h| 44 lib/ethdev/ethdev_trace_points.c | 6

[PATCH v2 3/7] ethdev: add flow rule insertion by index with pattern

. Signed-off-by: Alexander Kozyrev --- doc/guides/prog_guide/rte_flow.rst | 20 ++ doc/guides/rel_notes/release_24_11.rst | 5 +++ lib/ethdev/rte_flow.c | 55 ++ lib/ethdev/rte_flow.h | 54 + lib/ethdev

[PATCH v2 5/7] ethdev: add jump to table index action

Introduce the RTE_FLOW_ACTION_TYPE_JUMP_TO_TABLE_INDEX action. It redirects packets to a particular index in a flow table. Signed-off-by: Alexander Kozyrev --- doc/guides/rel_notes/release_24_11.rst | 4 lib/ethdev/rte_flow.c | 1 + lib/ethdev/rte_flow.h

[PATCH v2 4/7] app/testpmd: add insertion by index with pattern option

actions count / queue index 1 / end Signed-off-by: Alexander Kozyrev --- app/test-pmd/cmdline_flow.c | 8 +++- app/test-pmd/config.c | 22 -- app/test-pmd/testpmd.h | 2 +- 3 files changed, 24 insertions(+), 8 deletions(-) diff --git a/app/test-pmd

[PATCH v2 2/7] app/testpmd: add index with pattern insertion type

Provide index_with_pattern command line option for the template table insertion type. flow template_table 0 create table_id 2 group 13 priority 0 insertion_type index_with_pattern ingress rules_number 64 pattern_template 2 actions_template 2 Signed-off-by: Alexander Kozyrev --- app/test

[PATCH v2 0/7] ethdev: jump to table support

X can redirect a packet to another matching criteria at the specified index in this case. RFC: https://patchwork.dpdk.org/project/dpdk/patch/20240822202753.3856703-1-akozy...@nvidia.com/ v2: added trace point to flow insertion by index functions. Alexander Kozyrev (7): ethdev: add insertion by

[PATCH v2 1/7] ethdev: add insertion by index with pattern

: Alexander Kozyrev --- doc/guides/rel_notes/release_24_11.rst | 4 lib/ethdev/rte_flow.h | 4 2 files changed, 8 insertions(+) diff --git a/doc/guides/rel_notes/release_24_11.rst b/doc/guides/rel_notes/release_24_11.rst index ef0124a9e6..8d311aead2 100644 --- a/doc/guides

[PATCH 6/6] app/testpmd: add jump to table index action

Signed-off-by: Alexander Kozyrev --- app/test-pmd/cmdline_flow.c | 34 ++ 1 file changed, 34 insertions(+) diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c index 855273365e..b7bcf18311 100644 --- a/app/test-pmd/cmdline_flow.c +++ b/app/test-pmd

[PATCH 5/6] ethdev: add jump to table index action

Introduce the RTE_FLOW_ACTION_TYPE_JUMP_TO_TABLE_INDEX action. It redirects packets to a particular index in a flow table. Signed-off-by: Alexander Kozyrev --- doc/guides/rel_notes/release_24_11.rst | 4 lib/ethdev/rte_flow.c | 1 + lib/ethdev/rte_flow.h

[PATCH 4/6] app/testpmd: add insertion by index with pattern option

actions count / queue index 1 / end Signed-off-by: Alexander Kozyrev --- app/test-pmd/cmdline_flow.c | 8 +++- app/test-pmd/config.c | 22 -- app/test-pmd/testpmd.h | 2 +- 3 files changed, 24 insertions(+), 8 deletions(-) diff --git a/app/test-pmd

[PATCH 2/6] app/testpmd: add index with pattern insertion type

Provide index_with_pattern command line option for the template table insertion type. flow template_table 0 create table_id 2 group 13 priority 0 insertion_type index_with_pattern ingress rules_number 64 pattern_template 2 actions_template 2 Signed-off-by: Alexander Kozyrev --- app/test

[PATCH 3/6] ethdev: add flow rule insertion by index with pattern

. Signed-off-by: Alexander Kozyrev --- doc/guides/prog_guide/rte_flow.rst | 20 ++ doc/guides/rel_notes/release_24_11.rst | 5 +++ lib/ethdev/ethdev_trace.h | 44 + lib/ethdev/ethdev_trace_points.c | 6 +++ lib/ethdev/rte_flow.c

[PATCH 0/6] ethdev: jump to table support

X can redirect a packet to another matching criteria at the specified index in this case. RFC: https://patchwork.dpdk.org/project/dpdk/patch/20240822202753.3856703-1-akozy...@nvidia.com/ Alexander Kozyrev (6): ethdev: add insertion by index with pattern app/testpmd: add index with pattern inse

[PATCH 1/6] ethdev: add insertion by index with pattern

: Alexander Kozyrev --- doc/guides/rel_notes/release_24_11.rst | 4 lib/ethdev/rte_flow.h | 4 2 files changed, 8 insertions(+) diff --git a/doc/guides/rel_notes/release_24_11.rst b/doc/guides/rel_notes/release_24_11.rst index 0ff70d9057..7056f17f3c 100644 --- a/doc/guides

Re: [PATCH] mbuf: add transport mode ESP packet type

> I think we already discussed this same patch in previous emails > (Aug-Oct 2023) at > https://mails.dpdk.org/archives/dev/2023-October/279390.html and > concluded that it is not needed ? > Did anything change from then ? Yes, Nithin, we found a way to distinguish the modes by looking into the ne

Re: [RFC] ethdev: jump to table support

>> +__rte_experimental >> +struct rte_flow * >> +rte_flow_async_create_by_index_with_pattern(uint16_t port_id, >> + uint32_t queue_id, >> + const struct rte_flow_op_attr >> *op_attr, >> +

[RFC] ethdev: jump to table support

n the packet and executes the actions immediately. Add a new index-based flow table with pattern matching. The JUMP_TO_TABLE_INDEX can redirect a packet to another matching criteria at the specified index in this case. Signed-off-by: Alexander Kozyrev --- lib/ethdev/rte_flow.c

[PATCH] mbuf: add transport mode ESP packet type

YPE_L4_ESP for the regular transport mode and RTE_PTYPE_INNER_L4_ESP for the ESP over UDP packets. Signed-off-by: Alexander Kozyrev --- lib/mbuf/rte_mbuf_ptype.c | 2 ++ lib/mbuf/rte_mbuf_ptype.h | 36 ++-- 2 files changed, 32 insertions(+), 6 deletions(-) diff -

[PATCH] app/testpmd: add postpone option to async flow destroy

ned-off-by: Alexander Kozyrev --- app/test-pmd/cmdline_flow.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c index a76b44bf39..fb6a552863 100644 --- a/app/test-pmd/cmdline_flow.c +++ b/app/test-pmd/cmdline_flow.c @@ -3703

[PATCH] net/mlx5: break resource release forever loop

exit. Don't wait for the tables release and try them to free only once and proceed with the exit. Fixes: d1559d66ed ("net/mlx5: add table management") Signed-off-by: Alexander Kozyrev Acked-by: Dariusz Sosnowski --- drivers/net/mlx5/mlx5_flow_hw.c | 62 +++---

[PATCH v2] net/mlx5: replenish MPRQ buffers for miniCQEs

a possible title CQE. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_rxtx_vec.c | 56 ++-- drivers/net/mlx5/mlx5_rxtx_vec_altivec.h | 6 ++- drivers/net/mlx5/mlx5_rxtx_vec_neon.h| 6 ++- drivers/net/mlx5/mlx5_rxtx_vec_sse.h | 6 ++- 4 files

[PATCH] net/mlx5: replenish MPRQ buffers for miniCQEs

a possible title CQE. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5_rxtx_vec.c | 46 drivers/net/mlx5/mlx5_rxtx_vec_altivec.h | 6 ++-- drivers/net/mlx5/mlx5_rxtx_vec_neon.h| 6 ++-- drivers/net/mlx5/mlx5_rxtx_vec_sse.h | 6 ++-- 4 files

[PATCH v4 4/4] doc: add packet type matching item to release notes

Document new RTE_FLOW_ITEM_TYPE_PTYPE in the release notes. Signed-off-by: Alexander Kozyrev --- doc/guides/rel_notes/release_23_11.rst | 5 + 1 file changed, 5 insertions(+) diff --git a/doc/guides/rel_notes/release_23_11.rst b/doc/guides/rel_notes/release_23_11.rst index 0a6fc76a9d

[PATCH v4 3/4] doc: add PMD ptype item limitations

From: Michael Baum Add limitations for ptype item support in "mlx5.rst" file. Signed-off-by: Michael Baum --- doc/guides/nics/features/mlx5.ini | 1 + doc/guides/nics/mlx5.rst | 15 +++ 2 files changed, 16 insertions(+) diff --git a/doc/guides/nics/features/mlx5.ini b/d

[PATCH v4 2/4] net/mlx5/hws: add support for fragmented ptype match

Expand packet type matching with support of the Fragmented IP (Internet Protocol) packet type. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 54 ++- drivers/net/mlx5/hws/mlx5dr_definer.h | 2 + 2 files changed, 46 insertions(+), 10

[PATCH v4 0/4] ptype matching support in mlx5

Add support for RTE_FLOW_ITEM_TYPE_PTYPE in mlx5 PMD. Alexander Kozyrev (3): net/mlx5: add support for ptype match in hardware steering net/mlx5/hws: add support for fragmented ptype match doc: add packet type matching item to release notes Michael Baum (1): doc: add PMD ptype item

[PATCH v4 1/4] net/mlx5: add support for ptype match in hardware steering

The packet type matching provides quick way of finding out L2/L3/L4 protocols in a given packet. That helps with optimized flow rules matching, eliminating the need of stacking all the packet headers in the matching criteria. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws

[PATCH] net/mlx5/hws: fix integrity bits level

ned-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 6 ++ 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/mlx5/hws/mlx5dr_definer.c b/drivers/net/mlx5/hws/mlx5dr_definer.c index 95b5d4b70e..600544c044 100644 --- a/drivers/net/mlx5/hws/mlx5dr

[PATCH] net/mlx5/hws: remove csum check from L3 ok check

From: Michael Baum This patch changes the integrity item behavior for HW steering. Old behavior: the "ipv4_csum_ok" checks only IPv4 checksum and "l3_ok" checks everything is ok including IPv4 checksum. New behavior: the "l3_ok" checks everything is ok excluding IPv4 checksum. This change enab

[PATCH] ethdev: fix ESP packet type description

The correct protocol number for ESP (IP Encapsulating Security Payload) packet type is 50. 51 is IPSec AH (Authentication Header). Fixes: 1e84afd3906b ("mbuf: add security crypto flags and fields") Signed-off-by: Alexander Kozyrev --- lib/mbuf/rte_mbuf_ptype.h | 4 ++-- 1 file

[PATCH v3 7/7] net/mlx5/hws: fix integrity bits level

ned-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 6 ++ 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/mlx5/hws/mlx5dr_definer.c b/drivers/net/mlx5/hws/mlx5dr_definer.c index c752896ca7..f1f9235956 100644 --- a/drivers/net/mlx5/hws/mlx5dr

[PATCH v3 4/7] doc: add PMD ptype item limitations

From: Michael Baum Add limitations for ptype item support in "mlx5.rst" file. Signed-off-by: Michael Baum --- doc/guides/nics/features/mlx5.ini | 1 + doc/guides/nics/mlx5.rst | 15 +++ 2 files changed, 16 insertions(+) diff --git a/doc/guides/nics/features/mlx5.ini b/d

[PATCH v3 6/7] net/mlx5/hws: remove csum check from L3 ok check

From: Michael Baum This patch changes the integrity item behavior for HW steering. Old behavior: the "ipv4_csum_ok" checks only IPv4 checksum and "l3_ok" checks everything is ok including IPv4 checksum. New behavior: the "l3_ok" checks everything is ok excluding IPv4 checksum. This change enab

[PATCH v3 5/7] doc: add packet type matching item to release notes

Document new RTE_FLOW_ITEM_TYPE_PTYPE in the release notes. Signed-off-by: Alexander Kozyrev --- doc/guides/rel_notes/release_23_11.rst | 4 1 file changed, 4 insertions(+) diff --git a/doc/guides/rel_notes/release_23_11.rst b/doc/guides/rel_notes/release_23_11.rst index 0a6fc76a9d

[PATCH v3 2/7] net/mlx5: add support for ptype match in hardware steering

The packet type matching provides quick way of finding out L2/L3/L4 protocols in a given packet. That helps with optimized flow rules matching, eliminating the need of stacking all the packet headers in the matching criteria. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws

[PATCH v3 3/7] net/mlx5/hws: add support for fragmented ptype match

Expand packet type matching with support of the Fragmented IP (Internet Protocol) packet type. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 54 ++- drivers/net/mlx5/hws/mlx5dr_definer.h | 2 + 2 files changed, 46 insertions(+), 10

[PATCH v3 1/7] ethdev: fix ESP packet type description

The correct protocol number for ESP (IP Encapsulating Security Payload) packet type is 50. 51 is IPSec AH (Authentication Header). Fixes: 1e84afd3906b ("mbuf: add security crypto flags and fields") Signed-off-by: Alexander Kozyrev --- lib/mbuf/rte_mbuf_ptype.h | 4 ++-- 1 file

[PATCH v3 0/7] ptype matching support in mlx5

Add support for RTE_FLOW_ITEM_TYPE_PTYPE in mlx5 PMD. Alexander Kozyrev (5): ethdev: fix ESP packet type description net/mlx5: add support for ptype match in hardware steering net/mlx5/hws: add support for fragmented ptype match doc: add packet type matching item to release notes net

[PATCH v2 3/7] net/mlx5/hws: add support for fragmented ptype match

Expand packet type matching with support of the Fragmented IP (Internet Protocol) packet type. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 54 ++- drivers/net/mlx5/hws/mlx5dr_definer.h | 2 + 2 files changed, 46 insertions(+), 10

[PATCH v2 7/7] net/mlx5/hws: fix integrity bits level

ned-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 6 ++ 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/mlx5/hws/mlx5dr_definer.c b/drivers/net/mlx5/hws/mlx5dr_definer.c index c752896ca7..f1f9235956 100644 --- a/drivers/net/mlx5/hws/mlx5dr

[PATCH v2 6/7] net/mlx5/hws: remove csum check from L3 ok check

From: Michael Baum This patch changes the integrity item behavior for HW steering. Old behavior: the "ipv4_csum_ok" checks only IPv4 checksum and "l3_ok" checks everything is ok including IPv4 checksum. New behavior: the "l3_ok" checks everything is ok excluding IPv4 checksum. This change enab

[PATCH v2 5/7] doc: add packet type matching item to release notes

Document new RTE_FLOW_ITEM_TYPE_PTYPE in the release notes. Signed-off-by: Alexander Kozyrev --- doc/guides/rel_notes/release_23_11.rst | 4 1 file changed, 4 insertions(+) diff --git a/doc/guides/rel_notes/release_23_11.rst b/doc/guides/rel_notes/release_23_11.rst index 0a6fc76a9d

[PATCH v2 4/7] doc: add PMD ptype item limitations

From: Michael Baum Add limitations for ptype item support in "mlx5.rst" file. Signed-off-by: Michael Baum --- doc/guides/nics/features/mlx5.ini | 1 + doc/guides/nics/mlx5.rst | 15 +++ 2 files changed, 16 insertions(+) diff --git a/doc/guides/nics/features/mlx5.ini b/d

[PATCH v2 2/7] net/mlx5: add support for ptype match in hardware steering

The packet type matching provides quick way of finding out L2/L3/L4 protocols in a given packet. That helps with optimized flow rules matching, eliminating the need of stacking all the packet headers in the matching criteria. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws

[PATCH v2 1/7] ethdev: fix ESP packet type description

The correct protocol number for ESP (IP Encapsulating Security Payload) packet type is 50. 51 is IPSec AH (Authentication Header). Fixes: 1e84afd3906b ("mbuf: add security crypto flags and fields") Signed-off-by: Alexander Kozyrev --- lib/mbuf/rte_mbuf_ptype.h | 4 ++-- 1 file

[PATCH v2 0/7] ptype matching support in mlx5

Add support for RTE_FLOW_ITEM_TYPE_PTYPE in mlx5 PMD. Alexander Kozyrev (5): ethdev: fix ESP packet type description net/mlx5: add support for ptype match in hardware steering net/mlx5/hws: add support for fragmented ptype match doc: add packet type matching item to release notes net

RE: [PATCH v3] mbuf: add ESP packet type

> As per IPSEC ESP RFC 4303, for both tunnel mode or transport mode, > next proto 50, so we cannot identify a packet is for tunnel mode or > transport mode by just packet parsing. > Am I missing something ? You are absolutely correct, the only way to tell the difference is to parse the next_proto f

[PATCH] net/mlx5: fix MPRQ stride size check

("net/mlx5: fix MPRQ stride size to accommodate the headroom") Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/mlx5.c | 2 +- drivers/net/mlx5/mlx5_rxq.c | 25 ++--- 2 files changed, 15 insertions(+), 12 deletions(-) diff --git a/drivers/net/mlx5/mlx5.c

[PATCH 5/5] net/mlx5/hws: fix integrity bits level

ned-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 6 ++ 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/mlx5/hws/mlx5dr_definer.c b/drivers/net/mlx5/hws/mlx5dr_definer.c index 84d15a41df..b092249c0a 100644 --- a/drivers/net/mlx5/hws/mlx5dr

[PATCH 4/5] net/mlx5/hws: remove csum check from L3 ok check

From: Michael Baum This patch changes the integrity item behavior for HW steering. Old behavior: the "ipv4_csum_ok" checks only IPv4 checksum and "l3_ok" checks everything is ok including IPv4 checksum. New behavior: the "l3_ok" checks everything is ok excluding IPv4 checksum. This change enab

[PATCH 0/5] ptype matching support in mlx5

Add support for RTE_FLOW_ITEM_TYPE_PTYPE in mlx5 PMD. Alexander Kozyrev (3): net/mlx5: add support for ptype match in hardware steering net/mlx5/hws: add support for fragmented ptype match net/mlx5/hws: fix integrity bits level Michael Baum (2): doc: add PMD ptype item limitations net

[PATCH 3/5] doc: add PMD ptype item limitations

From: Michael Baum Add limitations for ptype item support in "mlx5.rst" file. Signed-off-by: Michael Baum --- doc/guides/nics/features/mlx5.ini | 1 + doc/guides/nics/mlx5.rst | 19 +++ 2 files changed, 20 insertions(+) diff --git a/doc/guides/nics/features/mlx5.ini

[PATCH 2/5] net/mlx5/hws: add support for fragmented ptype match

Expand packet type matching with support of the Fragmented IP (Internet Protocol) packet type. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws/mlx5dr_definer.c | 74 +++ drivers/net/mlx5/hws/mlx5dr_definer.h | 2 + 2 files changed, 56 insertions(+), 20

[PATCH 1/5] net/mlx5: add support for ptype match in hardware steering

The packet type matching provides quick way of finding out L2/L3/L4 protocols in a given packet. That helps with optimized flow rules matching, eliminating the need of stacking all the packet headers in the matching criteria. Signed-off-by: Alexander Kozyrev --- drivers/net/mlx5/hws

[PATCH v3] ethdev: add packet type matching item

0 template_table 1 pattern_template 0 actions_template 0 pattern ptype packet_type is 0x02000100 / end actions queue index 1 / end Signed-off-by: Alexander Kozyrev Acked-by: Ori Kam --- app/test-pmd/cmdline_flow.c | 28 + doc/guides/nics/features

[PATCH v2] ethdev: add packet type matching item

0 template_table 1 pattern_template 0 actions_template 0 pattern ptype packet_type is 0x02000100 / end actions queue index 1 / end Signed-off-by: Alexander Kozyrev Acked-by: Ori Kam --- app/test-pmd/cmdline_flow.c | 27 + doc/guides/nics/features

[PATCH v3] mbuf: add ESP packet type

Support the IP Encapsulating Security Payload (ESP) in transport mode. Signed-off-by: Alexander Kozyrev Acked-by: Morten Brørup --- lib/mbuf/rte_mbuf_ptype.h | 36 ++-- 1 file changed, 30 insertions(+), 6 deletions(-) diff --git a/lib/mbuf/rte_mbuf_ptype.h b

[PATCH v2] mbuf: add ESP packet type

Support the IP Encapsulating Security Payload (ESP) in transport mode. Signed-off-by: Alexander Kozyrev --- lib/mbuf/rte_mbuf_ptype.h | 31 +++ 1 file changed, 27 insertions(+), 4 deletions(-) diff --git a/lib/mbuf/rte_mbuf_ptype.h b/lib/mbuf/rte_mbuf_ptype.h index

RE: [PATCH] mbuf: add ESP packet type

> From: Morten Brørup > Non-fragment criteria seems to be missing: > > * | 'version'=4, 'protocol'=50, 'MF'=0, 'frag_offset'=0> > > Also missing here: > * | 'version'=4, 'protocol'=50, 'MF'=0, 'frag_offset'=0> > You are right, thank you, sending v2 with updated description.

  1   2   3   4   5   6   >