[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <50ef36cd-d095-9abe-26ea-d363d11ce521@wanadoo.fr>
Date: Sat, 9 May 2020 18:47:08 +0200
From: Christophe JAILLET <christophe.jaillet@...adoo.fr>
To: Jakub Kicinski <kuba@...nel.org>
Cc: davem@...emloft.net, fthain@...egraphics.com.au,
netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
kernel-janitors@...r.kernel.org
Subject: Re: [PATCH] net/sonic: Fix some resource leaks in error handling
paths
Le 09/05/2020 à 03:54, Jakub Kicinski a écrit :
> On Fri, 8 May 2020 19:25:57 +0200 Christophe JAILLET wrote:
>> @@ -527,8 +531,9 @@ static int mac_sonic_platform_remove(struct platform_device *pdev)
>> struct sonic_local* lp = netdev_priv(dev);
>>
>> unregister_netdev(dev);
>> - dma_free_coherent(lp->device, SIZEOF_SONIC_DESC * SONIC_BUS_SCALE(lp->dma_bitmode),
>> - lp->descriptors, lp->descriptors_laddr);
>> + dma_free_coherent(lp->device,
>> + SIZEOF_SONIC_DESC * SONIC_BUS_SCALE(lp->dma_bitmode),
>> + lp->descriptors, lp->descriptors_laddr);
>> free_netdev(dev);
>>
>> return 0;
> This is a white-space only change, right? Since this is a fix we should
> avoid making cleanups which are not strictly necessary.
>
Right.
The reason of this clean-up is that I wanted to avoid a checkpatch
warning with the proposed patch and I felt that having the same layout
in the error handling path of the probe function and in the remove
function was clearer.
So I updated also the remove function.
Fell free to ignore this hunk if not desired. I will not sent a V2 only
for that.
CJ
Powered by blists - more mailing lists