Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <e8d9d010-7fa1-da61-feeb-43f0a101a323@quicinc.com>
Date:   Thu, 2 Jun 2022 13:32:48 -0700
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101
 Thunderbird/91.10.0
Subject: Re: [PATCH v4 1/2] devcoredump: remove the useless gfp_t parameter in
 dev_coredumpv
Content-Language: en-US
To:     Duoming Zhou <duoming@zju.edu.cn>,
        <linux-wireless@vger.kernel.org>, <linux-kernel@vger.kernel.org>
CC:     <amitkarwar@gmail.com>, <ganapathi017@gmail.com>,
        <sharvari.harisangam@nxp.com>, <huxinming820@gmail.com>,
        <kvalo@kernel.org>, <davem@davemloft.net>, <edumazet@google.com>,
        <kuba@kernel.org>, <pabeni@redhat.com>, <netdev@vger.kernel.org>,
        <gregkh@linuxfoundation.org>, <johannes@sipsolutions.net>,
        <rafael@kernel.org>
References: <cover.1654175941.git.duoming@zju.edu.cn>
 <338a65fe8f30d23339cfc09fe1fb7be751ad655b.1654175941.git.duoming@zju.edu.cn>
From:   Jeff Johnson <quic_jjohnson@quicinc.com>
In-Reply-To: <338a65fe8f30d23339cfc09fe1fb7be751ad655b.1654175941.git.duoming@zju.edu.cn>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 6/2/2022 6:33 AM, Duoming Zhou wrote:
> The dev_coredumpv() could not be used in atomic context, because
> it calls kvasprintf_const() and kstrdup() with GFP_KERNEL parameter.
> The process is shown below:
> 
> dev_coredumpv(..., gfp_t gfp)
>    dev_coredumpm
>      dev_set_name
>        kobject_set_name_vargs
>          kvasprintf_const(GFP_KERNEL, ...); //may sleep
>            kstrdup(s, GFP_KERNEL); //may sleep
> 
> This patch removes gfp_t parameter of dev_coredumpv() and changes the
> gfp_t parameter of dev_coredumpm() to GFP_KERNEL in order to show
> dev_coredumpv() could not be used in atomic context.

shouldn't you remove the gfp parameter to dev_coredumpm() as well since 
it is actually within that function where dev_set_name() is called which 
cannot be done in atomic context?