[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d8f346e6-9227-a6c0-5cdb-1db819b7d848@huawei.com>
Date: Mon, 15 May 2023 17:07:10 +0800
From: shaozhengchao <shaozhengchao@...wei.com>
To: Steen Hegelund <steen.hegelund@...rochip.com>,
<linux-wireless@...r.kernel.org>, <netdev@...r.kernel.org>,
<johannes@...solutions.net>, <kvalo@...nel.org>, <davem@...emloft.net>,
<edumazet@...gle.com>, <kuba@...nel.org>, <pabeni@...hat.com>
CC: <weiyongjun1@...wei.com>, <yuehaibing@...wei.com>,
<syzbot+904ce6fbb38532d9795c@...kaller.appspotmail.com>
Subject: Re: [PATCH net-next] mac80211_hwsim: fix memory leak in
hwsim_new_radio_nl
On 2023/5/15 16:30, Steen Hegelund wrote:
> Hi Shao,
>
> On Mon, 2023-05-15 at 11:47 +0800, Zhengchao Shao wrote:
>> [You don't often get email from shaozhengchao@...wei.com. Learn why this is
>> important at https://aka.ms/LearnAboutSenderIdentification ]
>>
>> EXTERNAL EMAIL: Do not click links or open attachments unless you know the
>> content is safe
>>
>> When parse_pmsr_capa failed in hwsim_new_radio_nl, the memory resources
>> applied for by pmsr_capa are not released. Add release processing to the
>> incorrect path.
>>
>> Fixes: 92d13386ec55 ("mac80211_hwsim: add PMSR capability support")
>> Reported-by: syzbot+904ce6fbb38532d9795c@...kaller.appspotmail.com
>> Signed-off-by: Zhengchao Shao <shaozhengchao@...wei.com>
>> ---
>> drivers/net/wireless/virtual/mac80211_hwsim.c | 4 +++-
>> 1 file changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/wireless/virtual/mac80211_hwsim.c
>> b/drivers/net/wireless/virtual/mac80211_hwsim.c
>> index 9a8faaf4c6b6..6a50858a5645 100644
>> --- a/drivers/net/wireless/virtual/mac80211_hwsim.c
>> +++ b/drivers/net/wireless/virtual/mac80211_hwsim.c
>> @@ -5965,8 +5965,10 @@ static int hwsim_new_radio_nl(struct sk_buff *msg,
>> struct genl_info *info)
>> goto out_free;
>> }
>> ret = parse_pmsr_capa(info->attrs[HWSIM_ATTR_PMSR_SUPPORT],
>> pmsr_capa, info);
>> - if (ret)
>> + if (ret) {
>> + kfree(pmsr_capa);
>
> This should not be needed, see below.
>
>> goto out_free;
>> + }
>> param.pmsr_capa = pmsr_capa;
>
>
> Why don't you just move this line up before the parse_pmsr_capa as there is
> already a kfree(param.pmsr_capa) under the out_free label?
>
Hi Steen:
Your suggestion looks good. I will send V2.
Zhengchao Shao
>> }
>>
>> --
>> 2.34.1
>>
>>
>
> BR
> Steen
Powered by blists - more mailing lists