Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <02fc228b-7cc5-b470-9b5c-8ad726b18158@partition-saving.com>
Date:   Mon, 19 Sep 2022 20:40:34 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.11.0
Subject: Re: [PATCH -next 1/2] ext4: fix potential memory leak in
 ext4_fc_record_regions()
Content-Language: en-US
To:     Ye Bin <yebin10@huawei.com>, tytso@mit.edu,
        adilger.kernel@dilger.ca, linux-ext4@vger.kernel.org
Cc:     linux-kernel@vger.kernel.org, jack@suse.cz
References: <20220919144021.2162295-1-yebin10@huawei.com>
 <20220919144021.2162295-2-yebin10@huawei.com>
From:   Damien Guibouret <damien.guibouret@partition-saving.com>
In-Reply-To: <20220919144021.2162295-2-yebin10@huawei.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

Hello,

Le 19/09/2022 à 16:40, Ye Bin a écrit :
> As krealloc may return NULL, in this case 'state->fc_regions' may not be
> freed by krealloc, but 'state->fc_regions' already set NULL. Then will
> lead to 'state->fc_regions' memory leak.
> 
> Signed-off-by: Ye Bin <yebin10@huawei.com>
> ---
>   fs/ext4/fast_commit.c | 14 ++++++++------
>   1 file changed, 8 insertions(+), 6 deletions(-)
> 
> diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c
> index 9217a588afd1..cc8c8db075ba 100644
> --- a/fs/ext4/fast_commit.c
> +++ b/fs/ext4/fast_commit.c
> @@ -1677,15 +1677,17 @@ int ext4_fc_record_regions(struct super_block *sb, int ino,
>   	if (replay && state->fc_regions_used != state->fc_regions_valid)
>   		state->fc_regions_used = state->fc_regions_valid;
>   	if (state->fc_regions_used == state->fc_regions_size) {
> +		struct ext4_fc_alloc_region *fc_regions;
> +
>   		state->fc_regions_size +=
>   			EXT4_FC_REPLAY_REALLOC_INCREMENT;
> -		state->fc_regions = krealloc(
> -					state->fc_regions,
> -					state->fc_regions_size *
> -					sizeof(struct ext4_fc_alloc_region),
> -					GFP_KERNEL);
> -		if (!state->fc_regions)
> +		fc_regions = krealloc(state->fc_regions,
> +				      state->fc_regions_size *
> +				      sizeof(struct ext4_fc_alloc_region),
> +				      GFP_KERNEL);
> +		if (!fc_regions)

Would it not be safer to restore state->fc_regions_size to its previous 
value in that case to keep consistency between size value and allocated 
size (or to update state->fc_regions_size only after allocation as it is 
done in second part of this patch) ?

>   			return -ENOMEM;
> +		state->fc_regions = fc_regions;
>   	}
>   	region = &state->fc_regions[state->fc_regions_used++];
>   	region->ino = ino;

Regards,

Damien