[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5a7cb5c7c0bf2f9f9540616a2a0e70d36a166a9f.camel@redhat.com>
Date: Thu, 26 May 2022 13:14:20 +0200
From: Paolo Abeni <pabeni@...hat.com>
To: Jianglei Nie <niejianglei2021@....com>, saeedm@...dia.com,
leon@...nel.org, davem@...emloft.net, edumazet@...gle.com,
kuba@...nel.org, parav@...dia.com, vuhuong@...dia.com,
shayd@...dia.com
Cc: netdev@...r.kernel.org, linux-rdma@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH] net/mlx5: Fix memory leak in mlx5_sf_dev_add()
On Thu, 2022-05-26 at 16:44 +0800, Jianglei Nie wrote:
> The variable id is allocated by mlx5_adev_idx_alloc(). When some error
> happens, the id should be freed by mlx5_adev_idx_free().But when
> auxiliary_device_add() and xa_insert() fail, the id is not freed,which
> will lead to a potential memory leak.
>
> We can fix it by calling mlx5_sf_dev_add() when auxiliary_device_add()
> and xa_insert() fail.
>
> Signed-off-by: Jianglei Nie <niejianglei2021@....com>
> ---
> drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c | 5 ++++-
> 1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
> index 7da012ff0d41..9f222061a1c0 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
> @@ -125,13 +125,16 @@ static void mlx5_sf_dev_add(struct mlx5_core_dev *dev, u16 sf_index, u16 fn_id,
>
> err = auxiliary_device_add(&sf_dev->adev);
> if (err) {
> + mlx5_adev_idx_free(id);
> put_device(&sf_dev->adev.dev);
This looks not correct. put_device() -> mlx5_sf_dev_release() -> should
already take care of that.
> goto add_err;
> }
>
> err = xa_insert(&table->devices, sf_index, sf_dev, GFP_KERNEL);
> - if (err)
> + if (err) {
> + mlx5_adev_idx_free(id);
> goto xa_err;
> + }
> return;
>
> xa_err:
Powered by blists - more mailing lists