netdev - Re: [PATCH net-next V2] net: ena: Add a driver for Amazon Elastic Network Adapters (ENA)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Thu, 07 Jul 2016 11:56:43 +0200
From:	Ben Hutchings <ben@...adent.org.uk>
To:	Netanel Belgazal <netanel@...apurnalabs.com>,
	netdev@...r.kernel.org, davem@...emloft.net,
	linux-kernel@...r.kernel.org
Cc:	zorik@...apurnalabs.com, saeed@...apurnalabs.com,
	alex@...apurnalabs.com, msw@...zon.com, aliguori@...zon.com,
	benjamin.poirier@...il.com, romieu@...zoreil.com,
	rami.rosen@...el.com, antoine.tenart@...e-electrons.com
Subject: Re: [PATCH net-next V2] net: ena: Add a driver for Amazon Elastic
 Network Adapters (ENA)

On Thu, 2016-07-07 at 08:58 +0300, Netanel Belgazal wrote:
[...]
> +int ena_get_sset_count(struct net_device *netdev, int sset)
> +{
> +       if (sset != ETH_SS_STATS)
> +               return -EOPNOTSUPP;
> +
> +       return  netdev->num_tx_queues *
> +               (ENA_STATS_ARRAY_TX + ENA_STATS_ARRAY_RX) +
> +               ENA_STATS_ARRAY_GLOBAL + ENA_STATS_ARRAY_ENA_COM;

num_tx_queues is the number of software queues originally allocated,
which may be larger than the number in use (real_num_tx_queues).

And when actually generating the strings:

[...]
> +static void ena_queue_strings(struct ena_adapter *adapter, u8 **data)
> +{
> +       const struct ena_stats *ena_stats;
> +       int i, j;
> +
> +       for (i = 0; i < adapter->num_queues; i++) {

you use adapter->num_queues.

[...]
> +static int ena_nway_reset(struct net_device *netdev)
> +{
> +       return -ENODEV;
> +}

That doesn't make sense.  The device is there but presumably doesn't
support autonegotiation.  So don't implement the operation.

[...]
> +static int ena_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
> +                       u8 *hfunc)
> +{
> +       struct ena_adapter *adapter = netdev_priv(netdev);
> +       enum ena_admin_hash_functions ena_func;
> +       u8 func;
> +       int rc;
> +
> +       rc = ena_com_indirect_table_get(adapter->ena_dev, indir);
> +       if (rc)
> +               return rc;
> +
> +       rc = ena_com_get_hash_function(adapter->ena_dev, &ena_func, key);
> +       if (rc)
> +               return rc;
> +
> +       switch (ena_func) {
> +       case ENA_ADMIN_TOEPLITZ:
> +               func = ETH_RSS_HASH_TOP;
> +       case ENA_ADMIN_CRC32:
> +               func = ETH_RSS_HASH_XOR;

Missing break statements.

> +       default:
> +               netif_err(adapter, drv, netdev,
> +                         "Command parameter is not supported\n");
> +               return -EOPNOTSUPP;
> +       }
[...]
> +static const struct ethtool_ops ena_ethtool_ops = {
> +       .get_settings           = ena_get_settings,
[...]

get_settings is now deprecated in favour of get_link_ksettings.

Ben.

Ben.

-- 

Ben Hutchings
For every action, there is an equal and opposite criticism. - Harrison

Download attachment "signature.asc" of type "application/pgp-signature" (820 bytes)