linux-kernel - Re: [PATCH v1 3/5] soc: qcom: memory

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1a5b2381-8987-2f92-d018-29fdbc23b826@quicinc.com>
Date:   Mon, 23 Oct 2023 20:19:14 +0800
From:   Zhenhua Huang <quic_zhenhuah@...cinc.com>
To:     Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>,
        <agross@...nel.org>, <andersson@...nel.org>,
        <konrad.dybcio@...aro.org>, <robh+dt@...nel.org>,
        <krzysztof.kozlowski+dt@...aro.org>, <conor+dt@...nel.org>
CC:     <linux-arm-msm@...r.kernel.org>, <devicetree@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>, <kernel@...cinc.com>,
        <quic_tingweiz@...cinc.com>
Subject: Re: [PATCH v1 3/5] soc: qcom: memory_dump: Add memory dump driver



On 2023/10/23 19:46, Krzysztof Kozlowski wrote:
> On 23/10/2023 13:43, Zhenhua Huang wrote:
>>>> +
>>>> +			mem_dump_apply_offset(&dump_vaddr, &phys_addr,
>>>> +					      size + QCOM_DUMP_DATA_SIZE);
>>>> +			if (phys_addr > phys_end_addr) {
>>>> +				dev_err_probe(dev, -ENOMEM, "Exceeding allocated region\n");
>>>
>>> ENOMEM? Does not look right then.
>>
>> ENOMEM means the memory allocated not enough? any suggestion? Thanks.
> 
> The error code is okay, but we rarely need to print error messages for
> memory allocation failures. Core prints it already.

It's not same as below case. Allocation is successful here, while the 
driver used more than allocated size.

> 
>>
>>>
>>>> +				return -ENOMEM;
>>>> +			}
>>>> +		} else {
>>>> +			continue;
>>>> +		}
>>>> +	}
>>>> +
>>>> +	return ret;
>>>> +}
>>>> +
>>>> +static int __init mem_dump_probe(struct platform_device *pdev)
>>>> +{
>>>> +	struct qcom_memory_dump *memdump;
>>>> +	struct device *dev = &pdev->dev;
>>>> +	struct page *page;
>>>> +	size_t total_size;
>>>> +	int ret = 0;
>>>> +
>>>> +	memdump = devm_kzalloc(dev, sizeof(struct qcom_memory_dump),
>>>> +			       GFP_KERNEL);
>>>> +	if (!memdump)
>>>> +		return -ENOMEM;
>>>> +
>>>> +	dev_set_drvdata(dev, memdump);
>>>> +
>>>> +	/* check and initiate CMA region */
>>>> +	ret = mem_dump_reserve_mem(dev);
>>>> +	if (ret)
>>>> +		return ret;
>>>> +
>>>> +	/* allocate and populate */
>>>> +	page = mem_dump_alloc_mem(dev, &total_size);
>>>> +	if (IS_ERR(page)) {
>>>> +		ret = PTR_ERR(page);
>>>> +		dev_err_probe(dev, ret, "mem dump alloc failed\n");
>>>
>>> No, the syntax is:
>>> ret = dev_err_probe
>>>
>>> But why do you print messgaes for memory allocations?
>>
>> Do you think it's useless because mm codes should print error as well if
>> failure ?
> 
> We fixed this in kernel long, long, long time ago so we have even
> coccicheck scripts to find misuses.
> 

Thanks, yeah.. I know mm codes already have for example warn_alloc() to 
print error messages. Thanks for pointing out.

> 
> 
> Best regards,
> Krzysztof
> 

Thanks,
Zhenhua