public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: ebiederm@xmission.com (Eric W. Biederman)
To: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Cc: <linux-fsdevel@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
	Serge Hallyn <serge@hallyn.com>,
	Linux Containers <containers@lists.linux-foundation.org>
Subject: Re: [PATCH 01/25] userns: Convert fat to use kuid/kgid where appropriate
Date: Thu, 20 Sep 2012 05:04:02 -0700	[thread overview]
Message-ID: <87vcf8vr2l.fsf@xmission.com> (raw)
In-Reply-To: <87pq5gx60i.fsf@devron.myhome.or.jp> (OGAWA Hirofumi's message of "Thu, 20 Sep 2012 20:55:57 +0900")

OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> writes:

> "Eric W. Biederman" <ebiederm@xmission.com> writes:
>
>>  struct fat_mount_options {
>> -	uid_t fs_uid;
>> -	gid_t fs_gid;
>> +	kuid_t fs_uid;
>> +	kgid_t fs_gid;
>>  	unsigned short fs_fmask;
>>  	unsigned short fs_dmask;
>>  	unsigned short codepage;  /* Codepage for shortname conversions */
>
> I'm not reading this thread, so I may be wrong though. Doesn't this need
> to initialize opts->fs_uid/fs_gid by GLOBAL_ROOT_UID/GID?

Good question.

At the start of parse_options there is:

static int parse_options(struct super_block *sb, char *options, int is_vfat,
			 int silent, int *debug, struct fat_mount_options *opts)
{
	char *p;
	substring_t args[MAX_OPT_ARGS];
	int option;
	char *iocharset;

	opts->isvfat = is_vfat;

	opts->fs_uid = current_uid();
	opts->fs_gid = current_gid();

So there should be no problems with initialization.

Eric

>> diff --git a/fs/fat/file.c b/fs/fat/file.c
>> index e007b8b..a62e0ec 100644
>> --- a/fs/fat/file.c
>> +++ b/fs/fat/file.c
>> @@ -352,7 +352,7 @@ static int fat_allow_set_time(struct msdos_sb_info *sbi, struct inode *inode)
>>  {
>>  	umode_t allow_utime = sbi->options.allow_utime;
>>  
>> -	if (current_fsuid() != inode->i_uid) {
>> +	if (!uid_eq(current_fsuid(), inode->i_uid)) {
>>  		if (in_group_p(inode->i_gid))
>>  			allow_utime >>= 3;
>>  		if (allow_utime & MAY_WRITE)
>> @@ -407,9 +407,9 @@ int fat_setattr(struct dentry *dentry, struct iattr *attr)
>>  	}
>>  
>>  	if (((attr->ia_valid & ATTR_UID) &&
>> -	     (attr->ia_uid != sbi->options.fs_uid)) ||
>> +	     (!uid_eq(attr->ia_uid, sbi->options.fs_uid))) ||
>>  	    ((attr->ia_valid & ATTR_GID) &&
>> -	     (attr->ia_gid != sbi->options.fs_gid)) ||
>> +	     (!gid_eq(attr->ia_gid, sbi->options.fs_gid))) ||
>>  	    ((attr->ia_valid & ATTR_MODE) &&
>>  	     (attr->ia_mode & ~FAT_VALID_MODE)))
>>  		error = -EPERM;
>> diff --git a/fs/fat/inode.c b/fs/fat/inode.c
>> index 05e897f..47d9eb0 100644
>> --- a/fs/fat/inode.c
>> +++ b/fs/fat/inode.c
>> @@ -791,10 +791,12 @@ static int fat_show_options(struct seq_file *m, struct dentry *root)
>>  	struct fat_mount_options *opts = &sbi->options;
>>  	int isvfat = opts->isvfat;
>>  
>> -	if (opts->fs_uid != 0)
>> -		seq_printf(m, ",uid=%u", opts->fs_uid);
>> -	if (opts->fs_gid != 0)
>> -		seq_printf(m, ",gid=%u", opts->fs_gid);
>> +	if (!uid_eq(opts->fs_uid, GLOBAL_ROOT_UID))
>> +		seq_printf(m, ",uid=%u",
>> +				from_kuid_munged(&init_user_ns, opts->fs_uid));
>> +	if (!gid_eq(opts->fs_gid, GLOBAL_ROOT_GID))
>> +		seq_printf(m, ",gid=%u",
>> +				from_kgid_munged(&init_user_ns, opts->fs_gid));
>>  	seq_printf(m, ",fmask=%04o", opts->fs_fmask);
>>  	seq_printf(m, ",dmask=%04o", opts->fs_dmask);
>>  	if (opts->allow_utime)
>> @@ -1037,12 +1039,16 @@ static int parse_options(struct super_block *sb, char *options, int is_vfat,
>>  		case Opt_uid:
>>  			if (match_int(&args[0], &option))
>>  				return 0;
>> -			opts->fs_uid = option;
>> +			opts->fs_uid = make_kuid(current_user_ns(), option);
>> +			if (!uid_valid(opts->fs_uid))
>> +				return 0;
>>  			break;
>>  		case Opt_gid:
>>  			if (match_int(&args[0], &option))
>>  				return 0;
>> -			opts->fs_gid = option;
>> +			opts->fs_gid = make_kgid(current_user_ns(), option);
>> +			if (!gid_valid(opts->fs_gid))
>> +				return 0;
>>  			break;
>>  		case Opt_umask:
>>  			if (match_octal(&args[0], &option))
>> diff --git a/init/Kconfig b/init/Kconfig
>> index 296d48b..60bdff2 100644
>> --- a/init/Kconfig
>> +++ b/init/Kconfig
>> @@ -952,7 +952,6 @@ config UIDGID_CONVERTED
>>  	depends on ECRYPT_FS = n
>>  	depends on EFS_FS = n
>>  	depends on EXOFS_FS = n
>> -	depends on FAT_FS = n
>>  	depends on FUSE_FS = n
>>  	depends on GFS2_FS = n
>>  	depends on HFS_FS = n

  reply	other threads:[~2012-09-20 12:04 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-09-20 11:40 [REVIEW][PATCH 00/25] userns: Trivial filesystem conversions Eric W. Biederman
2012-09-20 11:41 ` [PATCH 01/25] userns: Convert fat to use kuid/kgid where appropriate Eric W. Biederman
2012-09-20 11:41   ` [PATCH 02/25] userns: Convert gadgetfs to use kuid and kgid " Eric W. Biederman
2012-09-20 11:58     ` Greg Kroah-Hartman
2012-09-21  9:33     ` Felipe Balbi
2012-09-20 11:41   ` [PATCH 03/25] userns: Convert usb functionfs to use kuid/kgid " Eric W. Biederman
2012-09-20 11:58     ` Greg Kroah-Hartman
2012-09-21  9:32     ` Felipe Balbi
2012-09-20 11:41   ` [PATCH 04/25] userns: Convert devtmpfs to use GLOBAL_ROOT_UID and GLOBAL_ROOT_GID Eric W. Biederman
2012-09-20 11:58     ` Greg Kroah-Hartman
2012-09-20 11:41   ` [PATCH 05/25] userns: Convert hugetlbfs to use kuid/kgid where appropriate Eric W. Biederman
2012-09-20 11:41   ` [PATCH 06/25] userns: Convert xenfs to use kuid and kgid " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 07/25] userns: Convert adfs " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 08/25] userns: Convert befs to use kuid/kgid " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 09/25] userns: Convert cramfs " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 10/25] userns: Convert ecryptfs " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 11/25] userns: Convert efs " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 12/25] userns: Convert exofs " Eric W. Biederman
2012-09-20 12:20     ` Boaz Harrosh
2012-09-20 11:41   ` [PATCH 13/25] userns: Convert hfs to use kuid and kgid " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 14/25] userns: Convert hfsplus " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 15/25] userns: Convert isofs to use kuid/kgid " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 16/25] userns: Convert logfs " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 17/25] userns: Convert minix " Eric W. Biederman
2012-09-20 11:41   ` [PATCH 18/25] userns: Convert nillfs2 " Eric W. Biederman
2012-09-20 14:34     ` Ryusuke Konishi
2012-09-20 11:42   ` [PATCH 19/25] userns: Convert ntfs to use kuid and kgid " Eric W. Biederman
2012-09-20 11:42   ` [PATCH 20/25] userns: Convert omfs " Eric W. Biederman
2012-09-20 12:50     ` Bob Copeland
2012-09-20 13:13       ` Eric W. Biederman
2012-09-20 13:19         ` Bob Copeland
2012-09-20 11:42   ` [PATCH 21/25] userns: Convert the qnx4 filesystem to use kuid/kgid " Eric W. Biederman
2012-09-20 12:34     ` Anders Larsen
2012-09-20 11:42   ` [PATCH 22/25] userns: Convert the qnx6 " Eric W. Biederman
2012-09-20 11:42   ` [PATCH 25/25] userns: Convert ipathfs to use GLOBAL_ROOT_UID and GLOBAL_ROOT_GID Eric W. Biederman
2012-09-20 16:47     ` Marciniszyn, Mike
2012-09-20 11:55   ` [PATCH 01/25] userns: Convert fat to use kuid/kgid where appropriate OGAWA Hirofumi
2012-09-20 12:04     ` Eric W. Biederman [this message]
2012-09-20 12:28       ` OGAWA Hirofumi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87vcf8vr2l.fsf@xmission.com \
    --to=ebiederm@xmission.com \
    --cc=containers@lists.linux-foundation.org \
    --cc=hirofumi@mail.parknet.co.jp \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=serge@hallyn.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox