Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH] scsi: ufs: ufshpb: Fix possible memory leak
To:     keosung.park@samsung.com, ALIM AKHTAR <alim.akhtar@samsung.com>,
        "avri.altman@wdc.com" <avri.altman@wdc.com>,
        "jejb@linux.ibm.com" <jejb@linux.ibm.com>,
        "martin.petersen@oracle.com" <martin.petersen@oracle.com>,
        Daejun Park <daejun7.park@samsung.com>,
        "beanhuo@micron.com" <beanhuo@micron.com>,
        "gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>,
        "linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
References: <CGME20210820014624epcms2p6724e146ca1f93ba6eac5e7cf95d4cfd2@epcms2p6>
 <1891546521.01629427801384.JavaMail.epsvc@epcpadp3>
From:   Bart Van Assche <bvanassche@acm.org>
Message-ID: <ca6e8507-743b-005a-faec-375d88d8aaf6@acm.org>
Date:   Fri, 20 Aug 2021 13:28:53 -0700
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.12.0
MIME-Version: 1.0
In-Reply-To: <1891546521.01629427801384.JavaMail.epsvc@epcpadp3>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 8/19/21 6:46 PM, Keoseong Park wrote:
> When HPB pinned region exists and mctx allocation for this region fails,
> memory leak is possible because memory is not released for the subregion
> table of the current region.
> 
> So, change to free memory for the subregion table of the current region.
> 
> Signed-off-by: Keoseong Park <keosung.park@samsung.com>
> ---
>   drivers/scsi/ufs/ufshpb.c | 5 ++++-
>   1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/scsi/ufs/ufshpb.c b/drivers/scsi/ufs/ufshpb.c
> index 9acce92a356b..052f584c789a 100644
> --- a/drivers/scsi/ufs/ufshpb.c
> +++ b/drivers/scsi/ufs/ufshpb.c
> @@ -1933,7 +1933,7 @@ static int ufshpb_alloc_region_tbl(struct ufs_hba *hba, struct ufshpb_lu *hpb)
>   		if (ufshpb_is_pinned_region(hpb, rgn_idx)) {
>   			ret = ufshpb_init_pinned_active_region(hba, hpb, rgn);
>   			if (ret)
> -				goto release_srgn_table;
> +				goto release_current_srgn_table;
>   		} else {
>   			rgn->rgn_state = HPB_RGN_INACTIVE;
>   		}
> @@ -1944,6 +1944,9 @@ static int ufshpb_alloc_region_tbl(struct ufs_hba *hba, struct ufshpb_lu *hpb)
>   
>   	return 0;
>   
> +release_current_srgn_table:
> +	kvfree(rgn_table[rgn_idx].srgn_tbl);
> +
>   release_srgn_table:
>   	for (i = 0; i < rgn_idx; i++)
>   		kvfree(rgn_table[i].srgn_tbl);

'rgn_table' is allocated with kvcalloc() so please merge the new kvfree() statement
with the for-loop below it.

There is another improvement that can be made in this function: hpb->rgn_tbl
is not cleared in the error path. I propose to move the "hpb->rgn_tbl = rgn_table"
assignment from the start of the function to just above the "return 0" statement.

Thanks,

Bart.