Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:3 as permitted sender) client-ip=2620:137:e000::3:3;
CC:     <james.clark@arm.com>, <suzuki.poulose@arm.com>,
        <yangyicong@hisilicon.com>, <coresight@lists.linaro.org>,
        <linux-arm-kernel@lists.infradead.org>,
        <linux-kernel@vger.kernel.org>, <linuxarm@huawei.com>,
        <jonathan.cameron@huawei.com>, <prime.zeng@hisilicon.com>
Subject: Re: [PATCH 3/3] coresight: ultrasoc-smb: fix uninitialized before use
 buf_hw_base
To:     Junhao He <hejunhao3@huawei.com>
References: <20231012094706.21565-1-hejunhao3@huawei.com>
 <20231012094706.21565-4-hejunhao3@huawei.com>
From:   Yicong Yang <yangyicong@huawei.com>
Message-ID: <fe5af067-d2f4-7967-e75b-272b00e6275f@huawei.com>
Date:   Thu, 19 Oct 2023 10:34:55 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:78.0) Gecko/20100101
 Thunderbird/78.5.1
MIME-Version: 1.0
In-Reply-To: <20231012094706.21565-4-hejunhao3@huawei.com>
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 2023/10/12 17:47, Junhao He wrote:
> In smb_reset_buffer, the sdb->buf_hw_base variable is uninitialized
> before use, which initializes it in smb_init_data_buffer. And the SMB
> regiester are set in smb_config_inport.
> So move the call after smb_config_inport.
> 
> Fixes: 06f5c2926aaa ("drivers/coresight: Add UltraSoc System Memory Buffer driver")
> 
> Signed-off-by: Junhao He <hejunhao3@huawei.com>
> ---
>  drivers/hwtracing/coresight/ultrasoc-smb.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/hwtracing/coresight/ultrasoc-smb.c b/drivers/hwtracing/coresight/ultrasoc-smb.c
> index e78edc3480ce..21ba473786e5 100644
> --- a/drivers/hwtracing/coresight/ultrasoc-smb.c
> +++ b/drivers/hwtracing/coresight/ultrasoc-smb.c
> @@ -475,7 +475,6 @@ static int smb_init_data_buffer(struct platform_device *pdev,
>  static void smb_init_hw(struct smb_drv_data *drvdata)
>  {
>  	smb_disable_hw(drvdata);
> -	smb_reset_buffer(drvdata);
>  
>  	writel(SMB_LB_CFG_LO_DEFAULT, drvdata->base + SMB_LB_CFG_LO_REG);
>  	writel(SMB_LB_CFG_HI_DEFAULT, drvdata->base + SMB_LB_CFG_HI_REG);
> @@ -597,6 +596,7 @@ static int smb_probe(struct platform_device *pdev)
>  	}
>  
>  	platform_set_drvdata(pdev, drvdata);
> +	smb_reset_buffer(drvdata);

Shouldn't we reset the buffer before registering the sink? Otherwise it'll
be possible for user to access an unreset one.

>  
>  	return 0;
>  }
>