linux-kernel - Re: [PATCH] cxl/region: Fix memregion leaks in devm_cxl_add

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: <0782e741-5478-4b62-866e-f966ab5cad91@fujitsu.com>
Date: Tue, 7 May 2024 02:42:37 +0000
From: "Zhijian Li (Fujitsu)" <lizhijian@...itsu.com>
To: Ira Weiny <ira.weiny@...el.com>, "dave@...olabs.net" <dave@...olabs.net>,
	"jonathan.cameron@...wei.com" <jonathan.cameron@...wei.com>,
	"dave.jiang@...el.com" <dave.jiang@...el.com>, "alison.schofield@...el.com"
	<alison.schofield@...el.com>, "vishal.l.verma@...el.com"
	<vishal.l.verma@...el.com>, "dan.j.williams@...el.com"
	<dan.j.williams@...el.com>, "linux-cxl@...r.kernel.org"
	<linux-cxl@...r.kernel.org>
CC: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] cxl/region: Fix memregion leaks in devm_cxl_add_region()



On 30/04/2024 00:23, Ira Weiny wrote:
> Li Zhijian wrote:
>> Move memregion_free(id) out of cxl_region_alloc() and
>> explicately free memregion in devm_cxl_add_region() error path.
>    ^^^^
>    explicitly
> 
> BTW you should run checkpatch.pl which will check spelling.


I remember I've done this check, but it seems the it doesn't work for me.
Did I miss something?

$ ./scripts/checkpatch.pl 0001-cxl-region-Fix-memregion-leaks-in-devm_cxl_add_regio.patch
total: 0 errors, 0 warnings, 23 lines checked

0001-cxl-region-Fix-memregion-leaks-in-devm_cxl_add_regio.patch has no obvious style problems and is ready for submission.



> 
> I'm not following what the problem is you are trying to fix.  This seems
> like it just moves the memregion_free() around a bit but the logic is the
> same.
> 

Sorry, I think my commit log may be misleading. In fact, this patch mainly fixes
the memregion leak in devm_cxl_add_region() when it gets an invalid mode.

>>   	default:
>>   		dev_err(&cxlrd->cxlsd.cxld.dev, "unsupported mode %d\n", mode);
>> +		memregion_free(id);
>>   		return ERR_PTR(-EINVAL);
>>   	}

Additionally, to maintain consistent error handling, I also moved memregion_free(id) from
cxl_region_alloc() to devm_cxl_add_region() so that devm_cxl_add_region() can
free memregion explicitly in error path.


Thanks
Zhijian


> Ira
> 
>>
>> After cxl_region_alloc() succeed, memregion will be freed by
>> cxl_region_type.release()
>>
>> Fixes: 6e099264185d ("cxl/region: Add volatile region creation support")
>> Signed-off-by: Li Zhijian <lizhijian@...itsu.com>
>> ---
>>   drivers/cxl/core/region.c | 9 +++++----
>>   1 file changed, 5 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
>> index 812b2948b6c6..8535718a20f0 100644
>> --- a/drivers/cxl/core/region.c
>> +++ b/drivers/cxl/core/region.c
>> @@ -2250,10 +2250,8 @@ static struct cxl_region *cxl_region_alloc(struct cxl_root_decoder *cxlrd, int i
>>   	struct device *dev;
>>   
>>   	cxlr = kzalloc(sizeof(*cxlr), GFP_KERNEL);
>> -	if (!cxlr) {
>> -		memregion_free(id);
>> +	if (!cxlr)
>>   		return ERR_PTR(-ENOMEM);
>> -	}
>>   
>>   	dev = &cxlr->dev;
>>   	device_initialize(dev);
>> @@ -2358,12 +2356,15 @@ static struct cxl_region *devm_cxl_add_region(struct cxl_root_decoder *cxlrd,
>>   		break;
>>   	default:
>>   		dev_err(&cxlrd->cxlsd.cxld.dev, "unsupported mode %d\n", mode);
>> +		memregion_free(id);
>>   		return ERR_PTR(-EINVAL);
>>   	}
>>   
>>   	cxlr = cxl_region_alloc(cxlrd, id);
>> -	if (IS_ERR(cxlr))
>> +	if (IS_ERR(cxlr)) {
>> +		memregion_free(id);
>>   		return cxlr;
>> +	}
>>   	cxlr->mode = mode;
>>   	cxlr->type = type;
>>   
>> -- 
>> 2.29.2
>>
> 
> 
>