From: vanshika.shukla@nxp.com
To: dev@dpdk.org, Hemant Agrawal <hemant.agrawal@nxp.com>,
Sachin Saxena <sachin.saxena@nxp.com>,
Gagandeep Singh <g.singh@nxp.com>
Cc: stable@dpdk.org
Subject: [v1 35/43] net/dpaa2: fix memory corruption in TM
Date: Fri, 13 Sep 2024 11:29:51 +0530 [thread overview]
Message-ID: <20240913055959.3246917-36-vanshika.shukla@nxp.com> (raw)
In-Reply-To: <20240913055959.3246917-1-vanshika.shukla@nxp.com>
From: Gagandeep Singh <g.singh@nxp.com>
driver was reserving memory in an array for 8 queues only,
but it can support many more queues configuration.
This patch fixes the memory corruption issue by defining the
queue array with correct size.
Fixes: 72100f0dee21 ("net/dpaa2: support level 2 in traffic management")
Cc: g.singh@nxp.com
Cc: stable@dpdk.org
Signed-off-by: Gagandeep Singh <g.singh@nxp.com>
---
drivers/net/dpaa2/dpaa2_tm.c | 29 ++++++++++++++++++-----------
1 file changed, 18 insertions(+), 11 deletions(-)
diff --git a/drivers/net/dpaa2/dpaa2_tm.c b/drivers/net/dpaa2/dpaa2_tm.c
index cb854964b4..83d0d669ce 100644
--- a/drivers/net/dpaa2/dpaa2_tm.c
+++ b/drivers/net/dpaa2/dpaa2_tm.c
@@ -684,6 +684,7 @@ dpaa2_hierarchy_commit(struct rte_eth_dev *dev, int clear_on_fail,
struct dpaa2_tm_node *leaf_node, *temp_leaf_node, *channel_node;
struct fsl_mc_io *dpni = (struct fsl_mc_io *)dev->process_private;
int ret, t;
+ bool conf_schedule = false;
/* Populate TCs */
LIST_FOREACH(channel_node, &priv->nodes, next) {
@@ -757,7 +758,7 @@ dpaa2_hierarchy_commit(struct rte_eth_dev *dev, int clear_on_fail,
}
LIST_FOREACH(channel_node, &priv->nodes, next) {
- int wfq_grp = 0, is_wfq_grp = 0, conf[DPNI_MAX_TC];
+ int wfq_grp = 0, is_wfq_grp = 0, conf[priv->nb_tx_queues];
struct dpni_tx_priorities_cfg prio_cfg;
memset(&prio_cfg, 0, sizeof(prio_cfg));
@@ -767,6 +768,7 @@ dpaa2_hierarchy_commit(struct rte_eth_dev *dev, int clear_on_fail,
if (channel_node->level_id != CHANNEL_LEVEL)
continue;
+ conf_schedule = false;
LIST_FOREACH(leaf_node, &priv->nodes, next) {
struct dpaa2_queue *leaf_dpaa2_q;
uint8_t leaf_tc_id;
@@ -789,6 +791,7 @@ dpaa2_hierarchy_commit(struct rte_eth_dev *dev, int clear_on_fail,
if (leaf_node->parent != channel_node)
continue;
+ conf_schedule = true;
leaf_dpaa2_q = (struct dpaa2_queue *)dev->data->tx_queues[leaf_node->id];
leaf_tc_id = leaf_dpaa2_q->tc_index;
/* Process sibling leaf nodes */
@@ -829,8 +832,8 @@ dpaa2_hierarchy_commit(struct rte_eth_dev *dev, int clear_on_fail,
goto out;
}
is_wfq_grp = 1;
- conf[temp_leaf_node->id] = 1;
}
+ conf[temp_leaf_node->id] = 1;
}
if (is_wfq_grp) {
if (wfq_grp == 0) {
@@ -851,6 +854,9 @@ dpaa2_hierarchy_commit(struct rte_eth_dev *dev, int clear_on_fail,
}
conf[leaf_node->id] = 1;
}
+ if (!conf_schedule)
+ continue;
+
if (wfq_grp > 1) {
prio_cfg.separate_groups = 1;
if (prio_cfg.prio_group_B < prio_cfg.prio_group_A) {
@@ -864,6 +870,16 @@ dpaa2_hierarchy_commit(struct rte_eth_dev *dev, int clear_on_fail,
prio_cfg.prio_group_A = 1;
prio_cfg.channel_idx = channel_node->channel_id;
+ DPAA2_PMD_DEBUG("########################################\n");
+ DPAA2_PMD_DEBUG("Channel idx = %d\n", prio_cfg.channel_idx);
+ for (t = 0; t < DPNI_MAX_TC; t++)
+ DPAA2_PMD_DEBUG("tc = %d mode = %d, delta = %d\n", t,
+ prio_cfg.tc_sched[t].mode,
+ prio_cfg.tc_sched[t].delta_bandwidth);
+
+ DPAA2_PMD_DEBUG("prioritya = %d, priorityb = %d, separate grps"
+ " = %d\n\n", prio_cfg.prio_group_A,
+ prio_cfg.prio_group_B, prio_cfg.separate_groups);
ret = dpni_set_tx_priorities(dpni, 0, priv->token, &prio_cfg);
if (ret) {
ret = -rte_tm_error_set(error, EINVAL,
@@ -871,15 +887,6 @@ dpaa2_hierarchy_commit(struct rte_eth_dev *dev, int clear_on_fail,
"Scheduling Failed\n");
goto out;
}
- DPAA2_PMD_DEBUG("########################################\n");
- DPAA2_PMD_DEBUG("Channel idx = %d\n", prio_cfg.channel_idx);
- for (t = 0; t < DPNI_MAX_TC; t++) {
- DPAA2_PMD_DEBUG("tc = %d mode = %d ", t, prio_cfg.tc_sched[t].mode);
- DPAA2_PMD_DEBUG("delta = %d\n", prio_cfg.tc_sched[t].delta_bandwidth);
- }
- DPAA2_PMD_DEBUG("prioritya = %d\n", prio_cfg.prio_group_A);
- DPAA2_PMD_DEBUG("priorityb = %d\n", prio_cfg.prio_group_B);
- DPAA2_PMD_DEBUG("separate grps = %d\n\n", prio_cfg.separate_groups);
}
return 0;
--
2.25.1
next prev parent reply other threads:[~2024-09-13 6:04 UTC|newest]
Thread overview: 88+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-09-13 5:59 [v1 00/43] DPAA2 specific patches vanshika.shukla
2024-09-13 5:59 ` [v1 01/43] net/dpaa2: enhance Tx scatter-gather mempool vanshika.shukla
2024-09-13 5:59 ` [v1 02/43] net/dpaa2: support PTP packet one-step timestamp vanshika.shukla
2024-09-13 5:59 ` [v1 03/43] net/dpaa2: add proper MTU debugging print vanshika.shukla
2024-09-13 5:59 ` [v1 04/43] net/dpaa2: add support to dump dpdmux counters vanshika.shukla
2024-09-13 5:59 ` [v1 05/43] bus/fslmc: change dpcon close as internal symbol vanshika.shukla
2024-09-13 5:59 ` [v1 06/43] bus/fslmc: add close API to close DPAA2 device vanshika.shukla
2024-09-13 5:59 ` [v1 07/43] net/dpaa2: dpdmux: add support for CVLAN vanshika.shukla
2024-09-13 5:59 ` [v1 08/43] bus/fslmc: upgrade with MC version 10.37 vanshika.shukla
2024-09-13 5:59 ` [v1 09/43] net/dpaa2: support link state for eth interfaces vanshika.shukla
2024-09-13 5:59 ` [v1 10/43] net/dpaa2: update DPNI link status method vanshika.shukla
2024-09-13 5:59 ` [v1 11/43] net/dpaa2: add new PMD API to check dpaa platform version vanshika.shukla
2024-09-13 5:59 ` [v1 12/43] bus/fslmc: improve BMAN buffer acquire vanshika.shukla
2024-09-13 5:59 ` [v1 13/43] bus/fslmc: get MC VFIO group FD directly vanshika.shukla
2024-09-13 5:59 ` [v1 14/43] bus/fslmc: enhance MC VFIO multiprocess support vanshika.shukla
2024-09-13 5:59 ` [v1 15/43] bus/fslmc: free VFIO group FD in case of add group failure vanshika.shukla
2024-09-13 5:59 ` [v1 16/43] bus/fslmc: dynamic IOVA mode configuration vanshika.shukla
2024-09-13 5:59 ` [v1 17/43] bus/fslmc: remove VFIO IRQ mapping vanshika.shukla
2024-09-13 5:59 ` [v1 18/43] bus/fslmc: create dpaa2 device with it's object vanshika.shukla
2024-09-13 5:59 ` [v1 19/43] bus/fslmc: fix coverity issue vanshika.shukla
2024-09-13 5:59 ` [v1 20/43] bus/fslmc: fix invalid error FD code vanshika.shukla
2024-09-13 5:59 ` [v1 21/43] bus/fslmc: change qbman eq desc from d to desc vanshika.shukla
2024-09-13 5:59 ` [v1 22/43] bus/fslmc: introduce VFIO DMA mapping API for fslmc vanshika.shukla
2024-09-13 5:59 ` [v1 23/43] net/dpaa2: change miss flow ID macro name vanshika.shukla
2024-09-13 5:59 ` [v1 24/43] net/dpaa2: flow API refactor vanshika.shukla
2024-09-13 5:59 ` [v1 25/43] net/dpaa2: dump Rx parser result vanshika.shukla
2024-09-13 5:59 ` [v1 26/43] net/dpaa2: enhancement of raw flow extract vanshika.shukla
2024-09-13 5:59 ` [v1 27/43] net/dpaa2: frame attribute flags parser vanshika.shukla
2024-09-13 5:59 ` [v1 28/43] net/dpaa2: add VXLAN distribution support vanshika.shukla
2024-09-13 5:59 ` [v1 29/43] net/dpaa2: protocol inside tunnel distribution vanshika.shukla
2024-09-13 5:59 ` [v1 30/43] net/dpaa2: eCPRI support by parser result vanshika.shukla
2024-09-13 5:59 ` [v1 31/43] net/dpaa2: add GTP flow support vanshika.shukla
2024-09-13 5:59 ` [v1 32/43] net/dpaa2: check if Soft parser is loaded vanshika.shukla
2024-09-13 5:59 ` [v1 33/43] net/dpaa2: soft parser flow verification vanshika.shukla
2024-09-13 5:59 ` [v1 34/43] net/dpaa2: add flow support for IPsec AH and ESP vanshika.shukla
2024-09-13 5:59 ` vanshika.shukla [this message]
2024-09-13 5:59 ` [v1 36/43] net/dpaa2: support software taildrop vanshika.shukla
2024-09-13 5:59 ` [v1 37/43] net/dpaa2: check IOVA before sending MC command vanshika.shukla
2024-09-13 5:59 ` [v1 38/43] net/dpaa2: improve DPDMUX error behavior settings vanshika.shukla
2024-09-13 5:59 ` [v1 39/43] net/dpaa2: store drop priority in mbuf vanshika.shukla
2024-09-13 5:59 ` [v1 40/43] net/dpaa2: add API to get endpoint name vanshika.shukla
2024-09-13 5:59 ` [v1 41/43] net/dpaa2: support VLAN traffic splitting vanshika.shukla
2024-09-13 5:59 ` [v1 42/43] net/dpaa2: add support for C-VLAN and MAC vanshika.shukla
2024-09-13 5:59 ` [v1 43/43] net/dpaa2: dpdmux single flow/multiple rules support vanshika.shukla
2024-09-18 7:50 ` [v2 00/43] DPAA2 specific patches vanshika.shukla
2024-09-18 7:50 ` [v2 01/43] net/dpaa2: enhance Tx scatter-gather mempool vanshika.shukla
2024-09-18 7:50 ` [v2 02/43] net/dpaa2: support PTP packet one-step timestamp vanshika.shukla
2024-09-18 7:50 ` [v2 03/43] net/dpaa2: add proper MTU debugging print vanshika.shukla
2024-09-18 7:50 ` [v2 04/43] net/dpaa2: add support to dump dpdmux counters vanshika.shukla
2024-09-18 7:50 ` [v2 05/43] bus/fslmc: change dpcon close as internal symbol vanshika.shukla
2024-09-18 7:50 ` [v2 06/43] bus/fslmc: add close API to close DPAA2 device vanshika.shukla
2024-09-18 7:50 ` [v2 07/43] net/dpaa2: dpdmux: add support for CVLAN vanshika.shukla
2024-09-18 7:50 ` [v2 08/43] bus/fslmc: upgrade with MC version 10.37 vanshika.shukla
2024-09-18 7:50 ` [v2 09/43] net/dpaa2: support link state for eth interfaces vanshika.shukla
2024-09-18 7:50 ` [v2 10/43] net/dpaa2: update DPNI link status method vanshika.shukla
2024-09-18 7:50 ` [v2 11/43] net/dpaa2: add new PMD API to check dpaa platform version vanshika.shukla
2024-09-18 7:50 ` [v2 12/43] bus/fslmc: improve BMAN buffer acquire vanshika.shukla
2024-09-18 7:50 ` [v2 13/43] bus/fslmc: get MC VFIO group FD directly vanshika.shukla
2024-09-18 7:50 ` [v2 14/43] bus/fslmc: enhance MC VFIO multiprocess support vanshika.shukla
2024-09-18 7:50 ` [v2 15/43] bus/fslmc: free VFIO group FD in case of add group failure vanshika.shukla
2024-09-18 7:50 ` [v2 16/43] bus/fslmc: dynamic IOVA mode configuration vanshika.shukla
2024-09-18 7:50 ` [v2 17/43] bus/fslmc: remove VFIO IRQ mapping vanshika.shukla
2024-09-18 7:50 ` [v2 18/43] bus/fslmc: create dpaa2 device with it's object vanshika.shukla
2024-09-18 7:50 ` [v2 19/43] bus/fslmc: fix coverity issue vanshika.shukla
2024-09-18 7:50 ` [v2 20/43] bus/fslmc: fix invalid error FD code vanshika.shukla
2024-09-18 7:50 ` [v2 21/43] bus/fslmc: change qbman eq desc from d to desc vanshika.shukla
2024-09-18 7:50 ` [v2 22/43] bus/fslmc: introduce VFIO DMA mapping API for fslmc vanshika.shukla
2024-09-18 7:50 ` [v2 23/43] net/dpaa2: change miss flow ID macro name vanshika.shukla
2024-09-18 7:50 ` [v2 24/43] net/dpaa2: flow API refactor vanshika.shukla
2024-09-18 7:50 ` [v2 25/43] net/dpaa2: dump Rx parser result vanshika.shukla
2024-09-18 7:50 ` [v2 26/43] net/dpaa2: enhancement of raw flow extract vanshika.shukla
2024-09-18 7:50 ` [v2 27/43] net/dpaa2: frame attribute flags parser vanshika.shukla
2024-09-18 7:50 ` [v2 28/43] net/dpaa2: add VXLAN distribution support vanshika.shukla
2024-09-18 7:50 ` [v2 29/43] net/dpaa2: protocol inside tunnel distribution vanshika.shukla
2024-09-18 7:50 ` [v2 30/43] net/dpaa2: eCPRI support by parser result vanshika.shukla
2024-09-18 7:50 ` [v2 31/43] net/dpaa2: add GTP flow support vanshika.shukla
2024-09-18 7:50 ` [v2 32/43] net/dpaa2: check if Soft parser is loaded vanshika.shukla
2024-09-18 7:50 ` [v2 33/43] net/dpaa2: soft parser flow verification vanshika.shukla
2024-09-18 7:50 ` [v2 34/43] net/dpaa2: add flow support for IPsec AH and ESP vanshika.shukla
2024-09-18 7:50 ` [v2 35/43] net/dpaa2: fix memory corruption in TM vanshika.shukla
2024-09-18 7:50 ` [v2 36/43] net/dpaa2: support software taildrop vanshika.shukla
2024-09-18 7:50 ` [v2 37/43] net/dpaa2: check IOVA before sending MC command vanshika.shukla
2024-09-18 7:50 ` [v2 38/43] net/dpaa2: improve DPDMUX error behavior settings vanshika.shukla
2024-09-18 7:50 ` [v2 39/43] net/dpaa2: store drop priority in mbuf vanshika.shukla
2024-09-18 7:50 ` [v2 40/43] net/dpaa2: add API to get endpoint name vanshika.shukla
2024-09-18 7:50 ` [v2 41/43] net/dpaa2: support VLAN traffic splitting vanshika.shukla
2024-09-18 7:50 ` [v2 42/43] net/dpaa2: add support for C-VLAN and MAC vanshika.shukla
2024-09-18 7:50 ` [v2 43/43] net/dpaa2: dpdmux single flow/multiple rules support vanshika.shukla
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=20240913055959.3246917-36-vanshika.shukla@nxp.com \
--to=vanshika.shukla@nxp.com \
--cc=dev@dpdk.org \
--cc=g.singh@nxp.com \
--cc=hemant.agrawal@nxp.com \
--cc=sachin.saxena@nxp.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).