linux-kernel - Re: [PATCH 5.10 49/96] net/mlx5: Fix return value from tracer initialization

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:   Wed, 18 Aug 2021 07:44:03 +0000
From:   Saeed Mahameed <saeedm@...dia.com>
To:     "pavel@...x.de" <pavel@...x.de>,
        "gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>
CC:     Moshe Shemesh <moshe@...dia.com>, Tariq Toukan <tariqt@...dia.com>,
        "stable@...r.kernel.org" <stable@...r.kernel.org>,
        "sashal@...nel.org" <sashal@...nel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Aya Levin <ayal@...dia.com>
Subject: Re: [PATCH 5.10 49/96] net/mlx5: Fix return value from tracer
 initialization

On Tue, 2021-08-17 at 19:51 +0200, Pavel Machek wrote:
> Hi!
> 
> > [ Upstream commit bd37c2888ccaa5ceb9895718f6909b247cc372e0 ]
> > 
> > Check return value of mlx5_fw_tracer_start(), set error path and
> > fix
> > return value of mlx5_fw_tracer_init() accordingly.
> 
> This is actually two fixes in one: There's cancel_work_sync() added
> to

Yes, the reasoning was that the patch is fixing the whole error path of
the function in one-shot since we can blame it on a single commit.

> the error path, but there's additional error that needs fixing.

Yes.

> 
> Could someone familiar with the code verify it after me?
> 
> Best regards,
>                                                                 Pavel
> 
> Signed-off-by: Pavel Machek (CIP) <pavel@...x.de>
> 
> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
> b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
> index 3dfcb20e97c6..857be86b4a11 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
> @@ -1007,7 +1007,7 @@ int mlx5_fw_tracer_init(struct mlx5_fw_tracer
> *tracer)
>         err = mlx5_core_alloc_pd(dev, &tracer->buff.pdn);
>         if (err) {
>                 mlx5_core_warn(dev, "FWTracer: Failed to allocate PD
> %d\n", err);
> -               return err;
> +               goto err_cancel_work;
>         }
>  
>         err = mlx5_fw_tracer_create_mkey(tracer);
> @@ -1031,6 +1031,7 @@ int mlx5_fw_tracer_init(struct mlx5_fw_tracer
> *tracer)
>         mlx5_core_destroy_mkey(dev, &tracer->buff.mkey);
>  err_dealloc_pd:
>         mlx5_core_dealloc_pd(dev, tracer->buff.pdn);
> +err_cancel_work:
>         cancel_work_sync(&tracer->read_fw_strings_work);
>         return err;

this is correct, do you want to submit this patch or do you want us to
handle ?
maybe it is better if we delayed queue_work() to after all the fragile
code behind it, to reduce the error path handling .. 

thanks for pointing this out.