patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Kevin Traynor <ktraynor@redhat.com>
To: Junfeng Guo <junfengg@nvidia.com>
Cc: Dariusz Sosnowski <dsosnowski@nvidia.com>, dpdk stable <stable@dpdk.org>
Subject: patch 'net/mlx5: fix actions translation error overwrite' has been queued to stable release 24.11.2
Date: Fri,  7 Mar 2025 12:47:02 +0000	[thread overview]
Message-ID: <20250307124726.475001-29-ktraynor@redhat.com> (raw)
In-Reply-To: <20250307124726.475001-1-ktraynor@redhat.com>

Hi,

FYI, your patch has been queued to stable release 24.11.2

Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
It will be pushed if I get no objections before 03/12/25. 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://github.com/kevintraynor/dpdk-stable

This queued commit can be viewed at:
https://github.com/kevintraynor/dpdk-stable/commit/f43c794f603e818502096ccf3bfda3160cbfd018

Thanks.

Kevin

---
From f43c794f603e818502096ccf3bfda3160cbfd018 Mon Sep 17 00:00:00 2001
From: Junfeng Guo <junfengg@nvidia.com>
Date: Thu, 20 Feb 2025 09:08:50 +0200
Subject: [PATCH] net/mlx5: fix actions translation error overwrite

[ upstream commit 494da70e289c6a603185c890111f95568eb1fd63 ]

Function __flow_hw_translate_actions_template contains several
encapsulated functions that already have internal error handling
process via rte_flow_error_set for each case.

Thus the one (rte_flow_error_set) within the goto statement `err`
at the end of __flow_hw_translate_actions_template function may be
redundant for those failed cases. As a result, the error messages
would all be overwritten as "fail to create rte table", making it
displayed at quite large granularity.

To prevent above error messages overwrite, this patch add a local
variable `struct rte_flow_error sub_error` to the function and pass
this `sub_error` instead of `error` to each sub-function. Under error
handling process (`err` label), if `sub_error` was updated, copy its
contents to `error` and return. If it was not updated, return default
error message (`fail to create rte table`).

Also refactor the logic for SEND_TO_KERNEL, COUNT and AGE actions in
above function to align the error handling process.

Fixes: f13fab23922b ("net/mlx5: add flow jump action")

Signed-off-by: Junfeng Guo <junfengg@nvidia.com>
Acked-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
---
 .mailmap                        |  2 +-
 drivers/net/mlx5/mlx5_flow_hw.c | 65 ++++++++++++++++++++-------------
 2 files changed, 40 insertions(+), 27 deletions(-)

diff --git a/.mailmap b/.mailmap
index d9423aa250..0ee46fdb0f 100644
--- a/.mailmap
+++ b/.mailmap
@@ -758,5 +758,5 @@ Julien Massonneau <julien.massonneau@6wind.com>
 Julien Meunier <julien.meunier@nokia.com> <julien.meunier@6wind.com>
 Július Milan <jmilan.dev@gmail.com>
-Junfeng Guo <junfeng.guo@intel.com>
+Junfeng Guo <junfengg@nvidia.com> <junfeng.guo@intel.com>
 Junjie Chen <junjie.j.chen@intel.com>
 Junjie Wan <wanjunjie@bytedance.com>
diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c
index 501bf33f94..e72b87d70f 100644
--- a/drivers/net/mlx5/mlx5_flow_hw.c
+++ b/drivers/net/mlx5/mlx5_flow_hw.c
@@ -2544,4 +2544,9 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 	size_t data_size = 0, push_size = 0;
 	struct mlx5_hw_modify_header_action mhdr = { 0 };
+	struct rte_flow_error sub_error = {
+		.type = RTE_FLOW_ERROR_TYPE_NONE,
+		.cause = NULL,
+		.message = NULL,
+	};
 	bool actions_end = false;
 	uint32_t type;
@@ -2663,5 +2668,5 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 					actions->conf)->group;
 				acts->jump = flow_hw_jump_action_register
-						(dev, cfg, jump_group, error);
+						(dev, cfg, jump_group, &sub_error);
 				if (!acts->jump)
 					goto err;
@@ -2794,13 +2799,14 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 		case RTE_FLOW_ACTION_TYPE_SEND_TO_KERNEL:
 			ret = flow_hw_translate_group(dev, cfg, attr->group,
-						&target_grp, error);
+						&target_grp, &sub_error);
 			if (ret)
-				return ret;
+				goto err;
 			if (target_grp == 0) {
 				__flow_hw_action_template_destroy(dev, acts);
-				return rte_flow_error_set(error, ENOTSUP,
-						RTE_FLOW_ERROR_TYPE_ACTION,
-						NULL,
-						"Send to kernel action on root table is not supported in HW steering mode");
+				rte_flow_error_set(&sub_error, ENOTSUP,
+					RTE_FLOW_ERROR_TYPE_ACTION,
+					NULL,
+					"Send to kernel action on root table is not supported in HW steering mode");
+				goto err;
 			}
 			table_type = attr->ingress ? MLX5DR_TABLE_TYPE_NIC_RX :
@@ -2812,5 +2818,5 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 			err = flow_hw_modify_field_compile(dev, attr, actions,
 							   masks, acts, &mhdr,
-							   src_pos, error);
+							   src_pos, &sub_error);
 			if (err)
 				goto err;
@@ -2819,5 +2825,5 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 			if (flow_hw_represented_port_compile
 					(dev, attr, actions,
-					 masks, acts, src_pos, dr_pos, error))
+					 masks, acts, src_pos, dr_pos, &sub_error))
 				goto err;
 			break;
@@ -2833,5 +2839,6 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 			     masks->conf)->mtr_id) {
 				err = flow_hw_meter_compile(dev, cfg,
-							    dr_pos, jump_pos, actions, acts, error);
+							    dr_pos, jump_pos, actions, acts,
+							    &sub_error);
 				if (err)
 					goto err;
@@ -2844,13 +2851,14 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 		case RTE_FLOW_ACTION_TYPE_AGE:
 			ret = flow_hw_translate_group(dev, cfg, attr->group,
-						&target_grp, error);
+						&target_grp, &sub_error);
 			if (ret)
-				return ret;
+				goto err;
 			if (target_grp == 0) {
 				__flow_hw_action_template_destroy(dev, acts);
-				return rte_flow_error_set(error, ENOTSUP,
-						RTE_FLOW_ERROR_TYPE_ACTION,
-						NULL,
-						"Age action on root table is not supported in HW steering mode");
+				rte_flow_error_set(&sub_error, ENOTSUP,
+					RTE_FLOW_ERROR_TYPE_ACTION,
+					NULL,
+					"Age action on root table is not supported in HW steering mode");
+				goto err;
 			}
 			if (__flow_hw_act_data_general_append(priv, acts,
@@ -2862,13 +2870,14 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 		case RTE_FLOW_ACTION_TYPE_COUNT:
 			ret = flow_hw_translate_group(dev, cfg, attr->group,
-						&target_grp, error);
+						&target_grp, &sub_error);
 			if (ret)
-				return ret;
+				goto err;
 			if (target_grp == 0) {
 				__flow_hw_action_template_destroy(dev, acts);
-				return rte_flow_error_set(error, ENOTSUP,
-						RTE_FLOW_ERROR_TYPE_ACTION,
-						NULL,
-						"Counter action on root table is not supported in HW steering mode");
+				rte_flow_error_set(&sub_error, ENOTSUP,
+					RTE_FLOW_ERROR_TYPE_ACTION,
+					NULL,
+					"Counter action on root table is not supported in HW steering mode");
+				goto err;
 			}
 			if ((at->action_flags & MLX5_FLOW_ACTION_AGE) ||
@@ -2913,5 +2922,5 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 								 &acts->mtr_id,
 								 MLX5_HW_INV_QUEUE,
-								 error);
+								 &sub_error);
 				if (err)
 					goto err;
@@ -2980,9 +2989,9 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 	}
 	if (mhdr.pos != UINT16_MAX) {
-		ret = mlx5_tbl_translate_modify_header(dev, cfg, acts, mp_ctx, &mhdr, error);
+		ret = mlx5_tbl_translate_modify_header(dev, cfg, acts, mp_ctx, &mhdr, &sub_error);
 		if (ret)
 			goto err;
 		if (!nt_mode && mhdr.shared) {
-			ret = mlx5_tbl_ensure_shared_modify_header(dev, cfg, acts, error);
+			ret = mlx5_tbl_ensure_shared_modify_header(dev, cfg, acts, &sub_error);
 			if (ret)
 				goto err;
@@ -2995,5 +3004,5 @@ __flow_hw_translate_actions_template(struct rte_eth_dev *dev,
 						  mp_ctx, data_size,
 						  reformat_src,
-						  refmt_type, error);
+						  refmt_type, &sub_error);
 		if (ret)
 			goto err;
@@ -3021,4 +3030,8 @@ err:
 	err = rte_errno;
 	__flow_hw_action_template_destroy(dev, acts);
+	if (error != NULL && sub_error.type != RTE_FLOW_ERROR_TYPE_NONE) {
+		rte_memcpy(error, &sub_error, sizeof(sub_error));
+		return -EINVAL;
+	}
 	return rte_flow_error_set(error, err,
 				  RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL,
-- 
2.48.1

---
  Diff of the applied patch vs upstream commit (please double-check if non-empty:
---
--- -	2025-03-07 11:02:57.977737092 +0000
+++ 0029-net-mlx5-fix-actions-translation-error-overwrite.patch	2025-03-07 11:02:56.897335815 +0000
@@ -1 +1 @@
-From 494da70e289c6a603185c890111f95568eb1fd63 Mon Sep 17 00:00:00 2001
+From f43c794f603e818502096ccf3bfda3160cbfd018 Mon Sep 17 00:00:00 2001
@@ -5,0 +6,2 @@
+[ upstream commit 494da70e289c6a603185c890111f95568eb1fd63 ]
+
@@ -27 +28,0 @@
-Cc: stable@dpdk.org
@@ -37 +38 @@
-index 2a3f1b1bd4..e9896a2e5b 100644
+index d9423aa250..0ee46fdb0f 100644
@@ -40 +41 @@
-@@ -761,5 +761,5 @@ Julien Massonneau <julien.massonneau@6wind.com>
+@@ -758,5 +758,5 @@ Julien Massonneau <julien.massonneau@6wind.com>


  parent reply	other threads:[~2025-03-07 12:48 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-03-07 12:46 patch 'ci: build with MSVC in GHA' " Kevin Traynor
2025-03-07 12:46 ` patch 'net/netvsc: scan all net devices under the PCI device' " Kevin Traynor
2025-03-07 12:46 ` patch 'net/netvsc: remove device if its net devices removed' " Kevin Traynor
2025-03-07 12:46 ` patch 'doc: fix feature flags for queue start/stop' " Kevin Traynor
2025-03-07 12:46 ` patch 'doc: add VXLAN matching requirement in mlx5 guide' " Kevin Traynor
2025-03-07 12:46 ` patch 'app/testpmd: use VXLAN flow flags from user if set' " Kevin Traynor
2025-03-07 12:46 ` patch 'ethdev: fix registers info request' " Kevin Traynor
2025-03-07 12:46 ` patch 'app/testpmd: show all DCB priority TC map' " Kevin Traynor
2025-03-07 12:46 ` patch 'app/testpmd: avoid crash in DCB config' " Kevin Traynor
2025-03-07 12:46 ` patch 'net/mana: fix multi-process tracking' " Kevin Traynor
2025-03-07 12:46 ` patch 'app/testpmd: fix out-of-bound reference in offload config' " Kevin Traynor
2025-03-07 12:46 ` patch 'net/ngbe: fix WOL and NCSI capabilities' " Kevin Traynor
2025-03-07 12:46 ` patch 'net/txgbe: remove useless condition for SW-FW sync' " Kevin Traynor
2025-03-07 12:46 ` patch 'net/nfp: fix representor port statistics' " Kevin Traynor
2025-03-07 12:46 ` patch 'bus/pci: fix registered device name' " Kevin Traynor
2025-03-07 12:46 ` patch 'net/nfp: fix init failure handling' " Kevin Traynor
2025-03-07 12:46 ` patch 'vhost/crypto: skip fetch before vring init' " Kevin Traynor
2025-03-07 12:46 ` patch 'examples/vhost_crypto: fix user callbacks' " Kevin Traynor
2025-03-07 12:46 ` patch 'vhost: check descriptor chains length' " Kevin Traynor
2025-03-07 12:46 ` patch 'test/bbdev: update FFT test vectors' " Kevin Traynor
2025-03-07 12:46 ` patch 'event/dlb2: fix event weight handling in SSE code path' " Kevin Traynor
2025-03-07 12:46 ` patch 'test/event: fix number of queues in eventdev conf' " Kevin Traynor
2025-03-07 12:46 ` patch 'net/e1000: fix crashes in secondary processes' " Kevin Traynor
2025-03-07 12:46 ` patch 'net/ixgbe: " Kevin Traynor
2025-03-07 12:46 ` patch 'net/ixgbe: fix minimum Rx/Tx descriptors' " Kevin Traynor
2025-03-07 12:46 ` patch 'net/mlx5/hws: fix fragmented packet type matching' " Kevin Traynor
2025-03-07 12:47 ` patch 'net/mlx5: fix leak in HWS flow counter action' " Kevin Traynor
2025-03-07 12:47 ` patch 'net/mlx5: fix crash with null flow list creation' " Kevin Traynor
2025-03-07 12:47 ` Kevin Traynor [this message]
2025-03-07 12:47 ` patch 'net/mlx5: fix hardware packet type translation' " Kevin Traynor
2025-03-07 12:47 ` patch 'common/cnxk: fix inbound IPsec SA setup' " Kevin Traynor
2025-03-07 12:47 ` patch 'stack: fix pop in C11 implementation' " Kevin Traynor
2025-03-07 12:47 ` patch 'test/crypto: fix AES-ECB test lengths' " Kevin Traynor
2025-03-07 12:47 ` patch 'examples/ipsec-secgw: fix IV length in CTR 192/256' " Kevin Traynor
2025-03-07 12:47 ` patch 'crypto/cnxk: fix asymmetric operation status code' " Kevin Traynor
2025-03-07 12:47 ` patch 'test/crypto: remove unused variable' " Kevin Traynor
2025-03-07 12:47 ` patch 'crypto/openssl: validate incorrect RSA signature' " Kevin Traynor
2025-03-07 12:47 ` patch 'test/crypto: fix check for OOP header data' " Kevin Traynor
2025-03-07 12:47 ` patch 'crypto/qat: fix SM3 state size' " Kevin Traynor
2025-03-07 12:47 ` patch 'test/dma: fix pointers in IOVA as PA mode' " Kevin Traynor
2025-03-07 12:47 ` patch 'doc: update ionic driver guide' " Kevin Traynor
2025-03-07 12:47 ` patch 'ci: point at GitHub mirror' " Kevin Traynor
2025-03-07 12:47 ` patch 'ci: fix ccache for Ubuntu 22.04' " Kevin Traynor

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=20250307124726.475001-29-ktraynor@redhat.com \
    --to=ktraynor@redhat.com \
    --cc=dsosnowski@nvidia.com \
    --cc=junfengg@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).