From: Xueming Li <xuemingl@nvidia.com>
To: Dariusz Sosnowski <dsosnowski@nvidia.com>
Cc: Xueming Li <xuemingl@nvidia.com>, Bing Zhao <bingz@nvidia.com>,
"dpdk stable" <stable@dpdk.org>
Subject: patch 'net/mlx5: fix header modify action on group 0' has been queued to stable release 23.11.5
Date: Thu, 26 Jun 2025 20:00:41 +0800 [thread overview]
Message-ID: <20250626120145.27369-22-xuemingl@nvidia.com> (raw)
In-Reply-To: <20250626120145.27369-1-xuemingl@nvidia.com>
Hi,
FYI, your patch has been queued to stable release 23.11.5
Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
It will be pushed if I get no objections before 06/28/25. So please
shout if anyone has objections.
Also note that after the patch there's a diff of the upstream commit vs the
patch applied to the branch. This will indicate if there was any rebasing
needed to apply to the stable branch. If there were code changes for rebasing
(ie: not only metadata diffs), please double check that the rebase was
correctly done.
Queued patches are on a temporary branch at:
https://git.dpdk.org/dpdk-stable/log/?h=23.11-staging
This queued commit can be viewed at:
https://git.dpdk.org/dpdk-stable/commit/?h=23.11-staging&id=5789022ac7b77a88a8805fff678cc2f572492956
Thanks.
Xueming Li <xuemingl@nvidia.com>
---
From 5789022ac7b77a88a8805fff678cc2f572492956 Mon Sep 17 00:00:00 2001
From: Dariusz Sosnowski <dsosnowski@nvidia.com>
Date: Fri, 25 Apr 2025 21:32:57 +0200
Subject: [PATCH] net/mlx5: fix header modify action on group 0
Cc: Xueming Li <xuemingl@nvidia.com>
[ upstream commit f1ebb26d0d9388b6df65d493b06ce3b55adfb893 ]
HW modify header commands generated for multiple modify field
flow actions, which modify/access the same packet fields
do not have to be separated by NOPs when used on group 0.
This is because:
- On group > 0, HW uses Modify Header Pattern objects
which require NOP explicitly.
- On group 0, modify field action is implemented using
Modify Header Context object managed by FW.
FW inserts requires NOPs internally.
mlx5 PMD inserted NOP always, which caused flow/table creation
failures on group 0 flow rules.
This patch addresses that.
Fixes: 0f4aa72b99da ("net/mlx5: support flow modify field with HWS")
Signed-off-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
Acked-by: Bing Zhao <bingz@nvidia.com>
---
drivers/net/mlx5/mlx5_flow_hw.c | 21 ++++++++++++++++-----
1 file changed, 16 insertions(+), 5 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c
index c53d407746..fac6c51f90 100644
--- a/drivers/net/mlx5/mlx5_flow_hw.c
+++ b/drivers/net/mlx5/mlx5_flow_hw.c
@@ -1105,7 +1105,8 @@ flow_hw_action_modify_field_is_shared(const struct rte_flow_action *action,
static __rte_always_inline bool
flow_hw_should_insert_nop(const struct mlx5_hw_modify_header_action *mhdr,
- const struct mlx5_modification_cmd *cmd)
+ const struct mlx5_modification_cmd *cmd,
+ const struct rte_flow_attr *attr)
{
struct mlx5_modification_cmd last_cmd = { { 0 } };
struct mlx5_modification_cmd new_cmd = { { 0 } };
@@ -1113,6 +1114,15 @@ flow_hw_should_insert_nop(const struct mlx5_hw_modify_header_action *mhdr,
unsigned int last_type;
bool should_insert = false;
+ /*
+ * Modify header action list does not require NOPs in root table,
+ * because different type of underlying object is used:
+ * - in root table - MODIFY_HEADER_CONTEXT (does not require NOPs),
+ * - in non-root - either inline modify action or based on Modify Header Pattern
+ * (which requires NOPs).
+ */
+ if (attr->group == 0)
+ return false;
if (cmds_num == 0)
return false;
last_cmd = *(&mhdr->mhdr_cmds[cmds_num - 1]);
@@ -1188,7 +1198,8 @@ flow_hw_mhdr_cmd_append(struct mlx5_hw_modify_header_action *mhdr,
static __rte_always_inline int
flow_hw_converted_mhdr_cmds_append(struct mlx5_hw_modify_header_action *mhdr,
- struct mlx5_flow_dv_modify_hdr_resource *resource)
+ struct mlx5_flow_dv_modify_hdr_resource *resource,
+ const struct rte_flow_attr *attr)
{
uint32_t idx;
int ret;
@@ -1196,7 +1207,7 @@ flow_hw_converted_mhdr_cmds_append(struct mlx5_hw_modify_header_action *mhdr,
for (idx = 0; idx < resource->actions_num; ++idx) {
struct mlx5_modification_cmd *src = &resource->actions[idx];
- if (flow_hw_should_insert_nop(mhdr, src)) {
+ if (flow_hw_should_insert_nop(mhdr, src, attr)) {
ret = flow_hw_mhdr_cmd_nop_append(mhdr);
if (ret)
return ret;
@@ -1312,14 +1323,14 @@ flow_hw_modify_field_compile(struct rte_eth_dev *dev,
* This NOP command will not be a part of action's command range used to update commands
* on rule creation.
*/
- if (flow_hw_should_insert_nop(mhdr, &resource->actions[0])) {
+ if (flow_hw_should_insert_nop(mhdr, &resource->actions[0], attr)) {
ret = flow_hw_mhdr_cmd_nop_append(mhdr);
if (ret)
return rte_flow_error_set(error, ret, RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
NULL, "too many modify field operations specified");
}
cmds_start = mhdr->mhdr_cmds_num;
- ret = flow_hw_converted_mhdr_cmds_append(mhdr, resource);
+ ret = flow_hw_converted_mhdr_cmds_append(mhdr, resource, attr);
if (ret)
return rte_flow_error_set(error, ret, RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
NULL, "too many modify field operations specified");
--
2.34.1
---
Diff of the applied patch vs upstream commit (please double-check if non-empty:
---
--- - 2025-06-26 19:59:18.707953789 +0800
+++ 0021-net-mlx5-fix-header-modify-action-on-group-0.patch 2025-06-26 19:59:17.246418050 +0800
@@ -1 +1 @@
-From f1ebb26d0d9388b6df65d493b06ce3b55adfb893 Mon Sep 17 00:00:00 2001
+From 5789022ac7b77a88a8805fff678cc2f572492956 Mon Sep 17 00:00:00 2001
@@ -4,0 +5,3 @@
+Cc: Xueming Li <xuemingl@nvidia.com>
+
+[ upstream commit f1ebb26d0d9388b6df65d493b06ce3b55adfb893 ]
@@ -22 +24,0 @@
-Cc: stable@dpdk.org
@@ -31 +33 @@
-index 2f7b769aec..cbe6bd3ef9 100644
+index c53d407746..fac6c51f90 100644
@@ -34 +36 @@
-@@ -1422,7 +1422,8 @@ flow_hw_action_modify_field_is_shared(const struct rte_flow_action *action,
+@@ -1105,7 +1105,8 @@ flow_hw_action_modify_field_is_shared(const struct rte_flow_action *action,
@@ -44 +46 @@
-@@ -1430,6 +1431,15 @@ flow_hw_should_insert_nop(const struct mlx5_hw_modify_header_action *mhdr,
+@@ -1113,6 +1114,15 @@ flow_hw_should_insert_nop(const struct mlx5_hw_modify_header_action *mhdr,
@@ -60 +62 @@
-@@ -1508,7 +1518,8 @@ flow_hw_mhdr_cmd_append(struct mlx5_hw_modify_header_action *mhdr,
+@@ -1188,7 +1198,8 @@ flow_hw_mhdr_cmd_append(struct mlx5_hw_modify_header_action *mhdr,
@@ -70 +72 @@
-@@ -1516,7 +1527,7 @@ flow_hw_converted_mhdr_cmds_append(struct mlx5_hw_modify_header_action *mhdr,
+@@ -1196,7 +1207,7 @@ flow_hw_converted_mhdr_cmds_append(struct mlx5_hw_modify_header_action *mhdr,
@@ -79 +81 @@
-@@ -1639,14 +1650,14 @@ flow_hw_modify_field_compile(struct rte_eth_dev *dev,
+@@ -1312,14 +1323,14 @@ flow_hw_modify_field_compile(struct rte_eth_dev *dev,
next prev parent reply other threads:[~2025-06-26 12:04 UTC|newest]
Thread overview: 86+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-26 12:00 patch " Xueming Li
2025-06-26 12:00 ` patch 'ethdev: convert string initialization' " Xueming Li
2025-06-26 12:00 ` patch 'net/fm10k/base: fix compilation warnings' " Xueming Li
2025-06-26 12:00 ` patch 'net/ixgbe/base: correct definition of endianness macro' " Xueming Li
2025-06-26 12:00 ` patch 'net/ixgbe/base: fix compilation warnings' " Xueming Li
2025-06-26 12:00 ` patch 'net/i40e/base: fix unused value " Xueming Li
2025-06-26 12:00 ` patch 'net/i40e/base: fix compiler " Xueming Li
2025-06-26 12:00 ` patch 'acl: fix build with GCC 15 on aarch64' " Xueming Li
2025-06-26 12:00 ` patch 'eal/linux: improve ASLR check' " Xueming Li
2025-06-26 12:00 ` patch 'net/e1000: fix xstats name' " Xueming Li
2025-06-26 12:00 ` patch 'net/e1000: fix EEPROM dump' " Xueming Li
2025-06-26 12:00 ` patch 'net/ixgbe: enable ethertype filter for E610' " Xueming Li
2025-06-26 12:00 ` patch 'net/ixgbe: fix port mask default value in filter' " Xueming Li
2025-06-26 12:00 ` patch 'net/e1000: fix igb Tx queue offloads capability' " Xueming Li
2025-06-26 12:00 ` patch 'net/ice: fix flow creation failure' " Xueming Li
2025-06-26 12:00 ` patch 'vhost: fix wrapping on control virtqueue rings' " Xueming Li
2025-06-26 12:00 ` patch 'vhost/crypto: fix cipher data length' " Xueming Li
2025-06-26 12:00 ` patch 'crypto/virtio: fix cipher data source " Xueming Li
2025-06-26 12:00 ` patch 'app/crypto-perf: fix AAD offset alignment' " Xueming Li
2025-06-26 12:00 ` patch 'crypto/qat: fix out-of-place header bytes in AEAD raw API' " Xueming Li
2025-06-26 12:00 ` patch 'crypto/qat: fix out-of-place chain/cipher/auth headers' " Xueming Li
2025-06-26 12:00 ` Xueming Li [this message]
2025-06-26 12:00 ` patch 'net/mlx5: validate GTP PSC QFI width' " Xueming Li
2025-06-26 12:00 ` patch 'net/mlx5: fix counter service cleanup on init failure' " Xueming Li
2025-06-26 12:00 ` patch 'net/mlx5/hws: fix send queue drain on FW WQE destroy' " Xueming Li
2025-06-26 12:00 ` patch 'net/mlx5: remove unsupported flow meter action in HWS' " Xueming Li
2025-06-26 12:00 ` patch 'net/mlx5: fix maximal queue size query' " Xueming Li
2025-06-26 12:00 ` patch 'net/mlx5: fix mark action with shared Rx queue' " Xueming Li
2025-06-26 12:00 ` patch 'net/mlx5: align PF and VF/SF MAC address handling' " Xueming Li
2025-06-26 12:00 ` patch 'net/sfc: fix action order on start failure' " Xueming Li
2025-06-26 12:00 ` patch 'net/nfp: fix crash with null RSS hash key' " Xueming Li
2025-06-26 12:00 ` patch 'net/nfp: fix hash key length logic' " Xueming Li
2025-06-26 12:00 ` patch 'app/testpmd: fix RSS hash key update' " Xueming Li
2025-06-26 12:00 ` patch 'net/af_xdp: fix use after free in zero-copy Tx' " Xueming Li
2025-06-26 12:00 ` patch 'net/hns3: fix integer overflow in interrupt unmap' " Xueming Li
2025-06-26 12:00 ` patch 'net/hns3: fix memory leak on failure' " Xueming Li
2025-06-26 12:00 ` patch 'net/hns3: fix extra wait for link up' " Xueming Li
2025-06-26 12:00 ` patch 'net/hns3: fix memory leak for indirect flow action' " Xueming Li
2025-06-26 12:00 ` patch 'net/hns3: fix interrupt rollback' " Xueming Li
2025-06-26 12:00 ` patch 'net/hns3: fix divide by zero' " Xueming Li
2025-06-26 12:01 ` patch 'net/hns3: fix resources release on reset' " Xueming Li
2025-06-26 12:01 ` patch 'net/nfp: standardize NFD3 Tx descriptor endianness' " Xueming Li
2025-06-26 12:01 ` patch 'net/nfp: standardize NFDk " Xueming Li
2025-06-26 12:01 ` patch 'net/qede: fix use after free' " Xueming Li
2025-06-26 12:01 ` patch 'bus/fslmc: " Xueming Li
2025-06-26 12:01 ` patch 'net/null: fix packet copy' " Xueming Li
2025-06-26 12:01 ` patch 'bus/vmbus: align ring buffer data to page boundary' " Xueming Li
2025-06-26 12:01 ` patch 'bus/vmbus: use Hyper-V page size' " Xueming Li
2025-06-26 12:01 ` patch 'net/netvsc: " Xueming Li
2025-06-26 12:01 ` patch 'net/netvsc: add stats counters from VF' " Xueming Li
2025-06-26 12:01 ` patch 'app/testpmd: relax number of TCs in DCB command' " Xueming Li
2025-06-26 12:01 ` patch 'net/mana: check vendor ID when probing RDMA device' " Xueming Li
2025-06-26 12:01 ` patch 'net/hns3: fix CRC data segment' " Xueming Li
2025-06-26 12:01 ` patch 'net/tap: fix qdisc add failure handling' " Xueming Li
2025-06-26 12:01 ` patch 'net/mlx5: fix VLAN stripping on hairpin queue' " Xueming Li
2025-06-26 12:01 ` patch 'mem: fix lockup on address space shortage' " Xueming Li
2025-06-26 12:01 ` patch 'test/malloc: improve resiliency' " Xueming Li
2025-06-26 12:01 ` patch 'trace: fix overflow in per-lcore trace buffer' " Xueming Li
2025-06-26 12:01 ` patch 'common/cnxk: fix E-tag pattern parsing' " Xueming Li
2025-06-26 12:01 ` patch 'common/cnxk: fix CQ tail drop' " Xueming Li
2025-06-26 12:01 ` patch 'net/cnxk: fix descriptor count update on reconfig' " Xueming Li
2025-06-26 12:01 ` patch 'ethdev: fix error struct in flow configure' " Xueming Li
2025-06-26 12:01 ` patch 'net/ice/base: fix integer overflow' " Xueming Li
2025-06-26 12:01 ` patch 'net/ice/base: fix typo in device ID description' " Xueming Li
2025-06-26 12:01 ` patch 'common/dpaax: fix PDCP key command race condition' " Xueming Li
2025-06-26 12:01 ` patch 'common/dpaax: fix PDCP AES only 12-bit SN' " Xueming Li
2025-06-26 12:01 ` patch 'crypto/dpaa2_sec: fix uninitialized variable' " Xueming Li
2025-06-26 12:01 ` patch 'crypto/virtio: add request check on request side' " Xueming Li
2025-06-26 12:01 ` patch 'crypto/virtio: fix driver cleanup' " Xueming Li
2025-06-26 12:01 ` patch 'crypto/virtio: fix driver ID' " Xueming Li
2025-06-26 12:01 ` patch 'ethdev: keep promiscuous/allmulti value before disabling' " Xueming Li
2025-06-26 12:01 ` patch 'eal: fix return value of lcore role' " Xueming Li
2025-06-26 12:01 ` patch 'eal: warn if no lcore is available' " Xueming Li
2025-06-26 12:01 ` patch 'test/lcore: fix race in per-lcore test' " Xueming Li
2025-06-26 12:01 ` patch 'bus: cleanup device lists' " Xueming Li
2025-06-26 12:01 ` patch 'eal/linux: unregister alarm callback before free' " Xueming Li
2025-06-26 12:01 ` patch 'eal/freebsd: " Xueming Li
2025-06-26 12:01 ` patch 'bus/pci/bsd: fix device existence check' " Xueming Li
2025-06-26 12:01 ` patch 'power/intel_uncore: fix crash closing uninitialized driver' " Xueming Li
2025-06-26 12:01 ` patch 'crypto/qat: fix size calculation for memset' " Xueming Li
2025-06-26 12:01 ` patch 'net/mlx5: avoid setting kernel MTU if not needed' " Xueming Li
2025-06-26 12:01 ` patch 'doc: add kernel options required for mlx5' " Xueming Li
2025-06-26 12:01 ` patch 'net/mlx5: fix hypervisor detection in VLAN workaround' " Xueming Li
2025-06-26 12:01 ` patch 'net/hns3: check requirement for hardware GRO' " Xueming Li
2025-06-26 12:01 ` patch 'net/hns3: allow Tx vector when fast free not enabled' " Xueming Li
2025-06-26 12:01 ` patch 'net/hns3: allow Rx vector mode with VLAN filter' " Xueming Li
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250626120145.27369-22-xuemingl@nvidia.com \
--to=xuemingl@nvidia.com \
--cc=bingz@nvidia.com \
--cc=dsosnowski@nvidia.com \
--cc=stable@dpdk.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).