[v3] net/mlx5: fix init with zero Rx queues
Checks
Commit Message
Recent patch [1] added, at the end of mlx5_dev_configure(), a call to
mlx5_proc_priv_init(), initializing process_private data of eth_dev.
This call is not reached if PMD is started with zero Rx queues.
In this case mlx5_dev_configure() returns earlier due to the check:
if (rxqs_n == priv->rxqs_n)
return 0;
In such a scenario, later references to uninitialized process_private
data will result in segmentation fault.
For example see in function txq_uar_init().
This patch changes the check logic. The following code is executed
if (rxqs_n != priv->rxqs_n), and skipped otherwise.
Function mlx5_proc_priv_init() is always invoked, to ensure
process_private data is initialized.
[1] http://patches.dpdk.org/patch/52629/
Fixes: 120dc4a7dcd3 ("net/mlx5: remove device register remap")
Cc: stable@dpdk.org
Signed-off-by: Dekel Peled <dekelp@mellanox.com>
Acked-by: Yongseok Koh <yskoh@mellanox.com>
---
v2: Update logic after code review.
v3: Remove unrelated code that was wrongly added in v2.
---
---
drivers/net/mlx5/mlx5_ethdev.c | 46 +++++++++++++++++++++++-------------------
1 file changed, 25 insertions(+), 21 deletions(-)
Comments
Sunday, May 5, 2019 2:44 PM, Dekel Peled:
> Subject: [dpdk-dev] [PATCH v3] net/mlx5: fix init with zero Rx queues
>
> Recent patch [1] added, at the end of mlx5_dev_configure(), a call to
> mlx5_proc_priv_init(), initializing process_private data of eth_dev.
> This call is not reached if PMD is started with zero Rx queues.
> In this case mlx5_dev_configure() returns earlier due to the check:
> if (rxqs_n == priv->rxqs_n)
> return 0;
> In such a scenario, later references to uninitialized process_private data will
> result in segmentation fault.
> For example see in function txq_uar_init().
>
> This patch changes the check logic. The following code is executed if (rxqs_n
> != priv->rxqs_n), and skipped otherwise.
> Function mlx5_proc_priv_init() is always invoked, to ensure process_private
> data is initialized.
>
> [1]
> https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fpatch
> es.dpdk.org%2Fpatch%2F52629%2F&data=02%7C01%7Cshahafs%40mel
> lanox.com%7C18b3d5bb065348a57ef408d6d14f6b26%7Ca652971c7d2e4d9ba
> 6a4d149256f461b%7C0%7C0%7C636926536374830910&sdata=7KXyV1U7
> 9P%2Bhc4HtZ8W05m95ZYyhY0m0Yc3A9NECvwo%3D&reserved=0
>
> Fixes: 120dc4a7dcd3 ("net/mlx5: remove device register remap")
> Cc: stable@dpdk.org
>
> Signed-off-by: Dekel Peled <dekelp@mellanox.com>
> Acked-by: Yongseok Koh <yskoh@mellanox.com>
Applied to next-net-mlx, thanks.
@@ -428,27 +428,31 @@ struct ethtool_link_settings {
rte_errno = EINVAL;
return -rte_errno;
}
- if (rxqs_n == priv->rxqs_n)
- return 0;
- DRV_LOG(INFO, "port %u Rx queues number update: %u -> %u",
- dev->data->port_id, priv->rxqs_n, rxqs_n);
- priv->rxqs_n = rxqs_n;
- /* If the requested number of RX queues is not a power of two, use the
- * maximum indirection table size for better balancing.
- * The result is always rounded to the next power of two. */
- reta_idx_n = (1 << log2above((rxqs_n & (rxqs_n - 1)) ?
- priv->config.ind_table_max_size :
- rxqs_n));
- ret = mlx5_rss_reta_index_resize(dev, reta_idx_n);
- if (ret)
- return ret;
- /* When the number of RX queues is not a power of two, the remaining
- * table entries are padded with reused WQs and hashes are not spread
- * uniformly. */
- for (i = 0, j = 0; (i != reta_idx_n); ++i) {
- (*priv->reta_idx)[i] = j;
- if (++j == rxqs_n)
- j = 0;
+ if (rxqs_n != priv->rxqs_n) {
+ DRV_LOG(INFO, "port %u Rx queues number update: %u -> %u",
+ dev->data->port_id, priv->rxqs_n, rxqs_n);
+ priv->rxqs_n = rxqs_n;
+ /*
+ * If the requested number of RX queues is not a power of two,
+ * use the maximum indirection table size for better balancing.
+ * The result is always rounded to the next power of two.
+ */
+ reta_idx_n = (1 << log2above((rxqs_n & (rxqs_n - 1)) ?
+ priv->config.ind_table_max_size :
+ rxqs_n));
+ ret = mlx5_rss_reta_index_resize(dev, reta_idx_n);
+ if (ret)
+ return ret;
+ /*
+ * When the number of RX queues is not a power of two,
+ * the remaining table entries are padded with reused WQs
+ * and hashes are not spread uniformly.
+ */
+ for (i = 0, j = 0; (i != reta_idx_n); ++i) {
+ (*priv->reta_idx)[i] = j;
+ if (++j == rxqs_n)
+ j = 0;
+ }
}
ret = mlx5_proc_priv_init(dev);
if (ret)