Received-SPF: pass (google.com: domain of linux-kernel+bounces-13344-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Message-ID: <621c7483-9a8c-431c-86e0-85d23b983471@web.de>
Date: Sat, 30 Dec 2023 07:45:51 +0100
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: [PATCH 2/2] hugetlbfs: Improve exception handling in
 hugetlbfs_fill_super()
To: Matthew Wilcox <willy@infradead.org>, linux-mm@kvack.org
Cc: kernel-janitors@vger.kernel.org, Muchun Song <muchun.song@linux.dev>,
 LKML <linux-kernel@vger.kernel.org>
References: <9ce3f553-24bc-4ecd-ac5e-7ba27caeff57@web.de>
 <b109b7dc-3972-4b2e-ae4c-89bf8eecf8f2@web.de>
 <ZY7dEbRJb1dHkQPd@casper.infradead.org>
Content-Language: en-GB
From: Markus Elfring <Markus.Elfring@web.de>
In-Reply-To: <ZY7dEbRJb1dHkQPd@casper.infradead.org>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
UI-OutboundReport: notjunk:1;M01:P0:M7PlL5FeIXY=;P9CA05MuiTsu8Dx9xQU42UD9CtZ
 LDI4EywlodTtkCBe6x9Sz2G/e0wDw+0kn53GXNqM2o0sisVkvVUAFeQPUpxhC3VKhxjtLfRPW
 /egtBlhmnM6s6MWd2hitASrMibEmiZGchb6CDPKyasJ+azcGFfybofZEzTzAVncYkPqaNEutB
 hYRGUHc8IaWo7/XdbHjOU1skOELwJQ6rgJU2WQT52/VS4TkIl9fOnyJLHjRQVN6ZGhlI81lKD
 oP9HAXrGbSYGb/0+1zIL+CawJ8fRMa3IWu6Fn3Qg8kf6/tQyRQwl0rUhJTyeZlDvDtl/ftFqO
 1oPLgpe7o1+FebxrkOJgbi9iB3BzGAf0BlLiD/0UZLWHj1A+BeCRXH3iWyUSQDL1JeJbC9SbO
 dYDJJKt6heT3OXxlQgrqTI3B4+4JPjGdMbhchQ3MzdsylWZ/Z9pC+QD5QWoVtQdPnsYTiO/+c
 k1I80Eiqa8WfDJJoDZrQm0SN7jEdmgtlrFKn7DHjHFK21JxCMpeHC5aly7xajtYd5nr2A/RmC
 WJabT0ZKJou9U7sTkoUoCG/ek8usKP7FmY02aqWosl3dXWi5Ns9nm1cItEbSP3PXEqQ09TIMv
 x9uBABU6P/Gzs1w6H9kfJTUVfIYLjqP8jUqG2leSfM0oEJPk0awurtYRGjQXr+CeuSnwklxPW
 6YfDKwy9f1OElxm0AGvWpjo5NVic/Mdc8GF313axP9ICGODsYtpTqyrZIFvStTT2waUjvyqSJ
 a0AXAQB9HPHmNLrkmBnVWvIiZTNMTOIGo0KsLO0ck1xO968zG5VdiZNXqFsyOLDNCAJfkf32Y
 cWhdQ1++FjFeLhe/kIxPgV4xRgNB2KFU3inTJBXCemktUDTsWX/SHhl3upzmH+dlpnKqRW6Wy
 7ix7xwFhZvsg6UUc6vRqYgH+NUKvB3K4Q30tPs1BmB78/6UR+C3RIllUFGpUd5XllI5v0zK8S
 ci8axg==

>> +++ b/fs/hugetlbfs/inode.c
>> @@ -1483,7 +1483,7 @@ hugetlbfs_fill_super(struct super_block *sb, stru=
ct fs_context *fc)
>>  						     ctx->max_hpages,
>>  						     ctx->min_hpages);
>>  		if (!sbinfo->spool)
>> -			goto out_free;
>> +			goto free_sbinfo;
>>  	}
>>  	sb->s_maxbytes =3D MAX_LFS_FILESIZE;
>>  	sb->s_blocksize =3D huge_page_size(ctx->hstate);
>> @@ -1499,10 +1499,12 @@ hugetlbfs_fill_super(struct super_block *sb, st=
ruct fs_context *fc)
>>  	sb->s_stack_depth =3D FILESYSTEM_MAX_STACK_DEPTH;
>>  	sb->s_root =3D d_make_root(hugetlbfs_get_root(sb, ctx));
>>  	if (!sb->s_root)
>> -		goto out_free;
>> +		goto free_spool;
>>  	return 0;
>> -out_free:
>> +
>> +free_spool:
>>  	kfree(sbinfo->spool);
>> +free_sbinfo:
>>  	kfree(sbinfo);
>>  	return -ENOMEM;
>>  }
>
> This is more complex.  NACK.

I am curious how coding style preferences will evolve further.

See also:
https://wiki.sei.cmu.edu/confluence/display/c/MEM12-C.+Consider+using+a+go=
to+chain+when+leaving+a+function+on+error+when+using+and+releasing+resourc=
es

Regards,
Markus