Re: [dpdk-dev] [PATCH 2/3] net/mlx5: convert return errno to negative ones

[Adrien Mazarguil <adrien.mazarguil@6wind.com>]

How about reusing the title/commit log of its mlx4 counterpart:

 net/mlx5: standardize on negative errno values

(see 9d14b27308a0 "net/mlx4: standardize on negative errno values")

More below.

On Thu, Feb 15, 2018 at 10:29:26AM +0100, Nelio Laranjeiro wrote:
> Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
> Acked-by: Yongseok Koh <yskoh@mellanox.com>
> ---
>  drivers/net/mlx5/mlx5.c         | 19 ++++-----
>  drivers/net/mlx5/mlx5_ethdev.c  | 30 ++++++++------
>  drivers/net/mlx5/mlx5_flow.c    | 92 +++++++++++++++++++++--------------------
>  drivers/net/mlx5/mlx5_mac.c     |  7 ++--
>  drivers/net/mlx5/mlx5_mr.c      |  4 +-
>  drivers/net/mlx5/mlx5_rss.c     | 16 +++----
>  drivers/net/mlx5/mlx5_rxq.c     | 20 ++++-----
>  drivers/net/mlx5/mlx5_socket.c  | 41 ++++++++++++------
>  drivers/net/mlx5/mlx5_trigger.c | 27 ++++++------
>  drivers/net/mlx5/mlx5_txq.c     | 14 +++----
>  drivers/net/mlx5/mlx5_vlan.c    |  2 +-
>  11 files changed, 145 insertions(+), 127 deletions(-)
> 
> diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c
> index f52edf74f..d24f2a37c 100644
> --- a/drivers/net/mlx5/mlx5.c
> +++ b/drivers/net/mlx5/mlx5.c
> @@ -413,7 +413,7 @@ mlx5_args_check(const char *key, const char *val, void *opaque)
>   *   Device arguments structure.
>   *
>   * @return
> - *   0 on success, errno value on failure.
> + *   0 on success, negative errno value on failure.

How about s/on \(failure\|error\)/otherwise/ here and everywhere else?
Documentation should be identical for all relevant functions.

<snip>

In mlx5_ethdev.c, priv_get_ifname() is still documented to return "0 on
success, -1 on failure and errno is set". You must get rid of the reliance
on external errno as part of this commit; you can optionally set rte_errno,
but for consistency all int-returning functions must return a valid negative
errno value, never -1.

The same applies to:

- priv_sysfs_read
- priv_sysfs_write
- priv_get_sysfs_ulong
- priv_set_sysfs_ulong
- priv_ifreq
- priv_get_num_vfs
- priv_get_mtu
- priv_get_cntr_sysfs
- priv_set_mtu
- priv_set_flags
- mlx5_link_update (lacks documentation)
- mlx5_ibv_device_to_pci_addr
- priv_dev_set_link
- mlx5_flow_item_validate
- mlx5_flow_create_* (unsure)
- priv_rx_intr_vec_enable ("negative" what?)
- mlx5_rx_intr_enable (ditto)
- mlx5_rx_intr_disable (ditto)
- check_cqe (returning a valid errno wouldn't impact performance)
- priv_read_dev_counters ("negative" what?)
- priv_ethtool_get_stats_n
- priv_xstats_get ("negative" what?)
- mlx5_stats_get (lacks documentation)
- mlx5_xstats_get ("negative" what?)
- mlx5_xstats_get_names (lacks documentation)
- priv_dev_traffic_disable (no error defined?)
- mlx5_priv_txq_ibv_releasable (lacks documentation)
- mlx5_priv_txq_ibv_verify (ditto)
- mlx5_vlan_offload_set (ditto, to be checked)

Also, some of them additionally set rte_errno while most of them do not. I'd
suggest to *always* set rte_errno in case of error then add the "...and
rte_errno is set" to documentation.

mlx4 approach to errors:

 if (boom) {
      rte_errno = ECRAP;
      return -rte_errno;
 }

Shorter, also valid but frowned upon:

 if (boom)
     return -(rte_errno = EBOOM);

Alternatively when calling another rte_errno-aware function:

 ret = boom();
 if (ret)
      return ret;

-- 
Adrien Mazarguil
6WIND