From: Eugene Syromyatnikov <evgsyr-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org
Cc: linux-man-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: [PATCH 3/4] quotactl.2: Updated information regarding XFS-specific quotactl subcommands
Date: Wed, 21 Sep 2016 06:28:57 +0300 [thread overview]
Message-ID: <20160921032857.GA5550@obsidian> (raw)
* man2/quotactl.2: Added information regarding structure definitions
used for XFS-specific subcommands, updated flag constants, added
information regarding ignored syscall arguments, added notes on usage
of kernel UAPI header.
---
man2/quotactl.2 | 237 +++++++++++++++++++++++++++++++++++++++++++++++++------
1 file changed, 215 insertions(+), 22 deletions(-)
diff --git a/man2/quotactl.2 b/man2/quotactl.2
index 3910379..ea076f9 100644
--- a/man2/quotactl.2
+++ b/man2/quotactl.2
@@ -30,7 +30,7 @@ quotactl \- manipulate disk quotas
.SH SYNOPSIS
.nf
.B #include <sys/quota.h>
-.B #include <xfs/xqm.h>
+.B #include <xfs/xqm.h> /* for XFS quotas */
.LP
.BI "int quotactl(int " cmd ", const char *" special ", int " id \
", caddr_t " addr );
@@ -389,18 +389,25 @@ Therefore, XFS expects
.I addr
to be a pointer to an
.I "unsigned int"
-that contains either the flags
-.B XFS_QUOTA_UDQ_ACCT
-and/or
-.B XFS_QUOTA_UDQ_ENFD
-(for user quota), or
-.B XFS_QUOTA_GDQ_ACCT
-and/or
-.B XFS_QUOTA_GDQ_ENFD
-(for group quota), as defined in
-.IR <xfs/xqm.h> .
+that contains the combination of the following flags (defined in
+.IR <xfs/xqm.h> ):
+
+.nf
+.in +4n
+#define XFS_QUOTA_UDQ_ACCT (1<<0) /* user quota accounting */
+#define XFS_QUOTA_UDQ_ENFD (1<<1) /* user quota limits enforcement */
+#define XFS_QUOTA_GDQ_ACCT (1<<2) /* group quota accounting */
+#define XFS_QUOTA_GDQ_ENFD (1<<3) /* group quota limits enforcement */
+#define XFS_QUOTA_PDQ_ACCT (1<<4) /* project quota accounting */
+#define XFS_QUOTA_PDQ_ENFD (1<<5) /* project quota limits enforcement */
+.in
+.fi
+
This operation requires privilege
.RB ( CAP_SYS_ADMIN ).
+The
+.I id
+argument is ignored.
.TP
.B Q_XQUOTAOFF
Turn off quotas for an XFS filesystem.
@@ -409,9 +416,14 @@ As with
XFS filesystems expect a pointer to an
.I "unsigned int"
that specifies whether quota accounting and/or limit enforcement need
-to be turned off.
+to be turned off (using the same flags as for
+.RB Q_XQUOTAON
+subcommand).
This operation requires privilege
.RB ( CAP_SYS_ADMIN ).
+The
+.I id
+argument is ignored.
.TP
.B Q_XGETQUOTA
Get disk quota limits and current usage for user
@@ -420,8 +432,48 @@ The
.I addr
argument is a pointer to an
.I fs_disk_quota
-structure (defined in
-.IR <xfs/xqm.h> ).
+structure which is defined in
+.I <xfs/xqm.h>
+as follows:
+
+.nf
+.in +4n
+/* All the blk units are in BBs (Basic Blocks) of 512 bytes. */
+
+#define FS_DQUOT_VERSION 1 /* fs_disk_quota.d_version */
+
+#define XFS_USER_QUOTA (1<<0) /* user quota type */
+#define XFS_PROJ_QUOTA (1<<1) /* project quota type */
+#define XFS_GROUP_QUOTA (1<<2) /* group quota type */
+
+struct fs_disk_quota {
+ int8_t d_version; /* version of this structure */
+ int8_t d_flags; /* XFS_{USER,PROJ,GROUP}_QUOTA */
+ uint16_t d_fieldmask; /* field specifier */
+ uint32_t d_id; /* user, project, or group ID */
+ uint64_t d_blk_hardlimit; /* absolute limit on disk blks */
+ uint64_t d_blk_softlimit; /* preferred limit on disk blks */
+ uint64_t d_ino_hardlimit; /* maximum # allocated inodes */
+ uint64_t d_ino_softlimit; /* preferred inode limit */
+ uint64_t d_bcount; /* # disk blocks owned by the user */
+ uint64_t d_icount; /* # inodes owned by the user */
+ int32_t d_itimer; /* zero if within inode limits */
+ /* if not, we refuse service */
+ int32_t d_btimer; /* similar to above; for disk blocks */
+ uint16_t d_iwarns; /* # warnings issued wrt num inodes */
+ uint16_t d_bwarns; /* # warnings issued wrt disk blocks */
+ int32_t d_padding2; /* padding2 - for future use */
+ uint64_t d_rtb_hardlimit; /* absolute limit on realtime blks */
+ uint64_t d_rtb_softlimit; /* preferred limit on RT disk blks */
+ uint64_t d_rtbcount; /* # realtime blocks owned */
+ int32_t d_rtbtimer; /* similar to above; for RT disk blks */
+ uint16_t d_rtbwarns; /* # warnings issued wrt RT disk blks */
+ int16_t d_padding3; /* padding3 - for future use */
+ char d_padding4[8]; /* yet more padding */
+};
+.in
+.fi
+
Unprivileged users may retrieve only their own quotas;
a privileged user
.RB ( CAP_SYS_ADMIN )
@@ -431,9 +483,21 @@ may retrieve the quotas of any user.
.\" commit 8b37524962b9c54423374717786198f5c0820a28
This operation is the same as
.BR Q_XGETQUOTA ,
-but it returns quota information for the next ID greater than or equal to
+but it returns (in the
+.I fs_disk_quota
+structure pointed by
+.IR addr )
+quota information for the next ID greater than or equal to
.IR id
-that has a quota set.
+that has a quota set. Note that since
+.I fs_disk_quota
+already has
+.I q_id
+field, no separate structure type is needed (in contrast with
+.B Q_GETQUOTA
+and
+.B Q_GETNEXTQUOTA
+commands)
.TP
.B Q_XSETQLIM
Set disk quota limits for user
@@ -442,22 +506,123 @@ The
.I addr
argument is a pointer to an
.I fs_disk_quota
-structure (defined in
-.IR <xfs/xqm.h> ).
+structure.
This operation requires privilege
.RB ( CAP_SYS_ADMIN ).
.TP
.B Q_XGETQSTAT
-Returns an
+Returns (in a buffer pointed by
+.IR addr )
+an
.I fs_quota_stat
structure containing XFS filesystem-specific quota information.
This is useful for finding out how much space is used to store quota
information, and also to get quotaon/off status of a given local XFS
-filesystem.
+filesystem. The
+.I fs_quota_stat
+structure itself is defined as follows:
+
+.nf
+.in +4n
+#define FS_QSTAT_VERSION 1 /* fs_quota_stat.qs_version */
+
+struct fs_qfilestat {
+ uint64_t qfs_ino; /* inode number */
+ uint64_t qfs_nblks; /* number of BBs 512-byte-blks */
+ uint32_t qfs_nextents; /* number of extents */
+};
+
+struct fs_quota_stat {
+ int8_t qs_version; /* version number for future changes */
+ uint16_t qs_flags; /* XFS_QUOTA_{U,P,G}DQ_{ACCT,ENFD} */
+ int8_t qs_pad; /* unused */
+ struct fs_qfilestat qs_uquota; /* user quota storage information */
+ struct fs_qfilestat qs_gquota; /* group quota storage information */
+ uint32_t qs_incoredqs; /* number of dquots incore */
+ int32_t qs_btimelimit; /* limit for blks timer */
+ int32_t qs_itimelimit; /* limit for inodes timer */
+ int32_t qs_rtbtimelimit; /* limit for rt blks timer */
+ uint16_t qs_bwarnlimit; /* limit for num warnings */
+ uint16_t qs_iwarnlimit; /* limit for num warnings */
+};
+.in
+.fi
+
+The
+.I id
+argument is ignored.
+.TP
+.B Q_XGETQSTATV
+Returns (in a buffer pointed by
+.IR addr )
+an
+.I fs_quota_statv
+structure containing XFS filesystem-specific quota information. This version
+of the command use structure with proper versioning support along with
+appropriate layout (all fields are naturally aligned) and adding for avoiding
+special compat handling; it also provides ability to get statistics regarding
+project quota file. The
+.I fs_quota_statv
+structure itself is defined as follows:
+
+.nf
+.in +4n
+#define FS_QSTATV_VERSION1 1 /* fs_quota_statv.qs_version */
+
+struct fs_qfilestatv {
+ uint64_t qfs_ino; /* inode number */
+ uint64_t qfs_nblks; /* number of BBs 512-byte-blks */
+ uint32_t qfs_nextents; /* number of extents */
+ uint32_t qfs_pad; /* pad for 8-byte alignment */
+};
+
+struct fs_quota_statv {
+ int8_t qs_version; /* version for future changes */
+ uint8_t qs_pad1; /* pad for 16bit alignment */
+ uint16_t qs_flags; /* XFS_QUOTA_.* flags */
+ uint32_t qs_incoredqs; /* number of dquots incore */
+ struct fs_qfilestatv qs_uquota; /* user quota information */
+ struct fs_qfilestatv qs_gquota; /* group quota information */
+ struct fs_qfilestatv qs_pquota; /* project quota information */
+ int32_t qs_btimelimit; /* limit for blks timer */
+ int32_t qs_itimelimit; /* limit for inodes timer */
+ int32_t qs_rtbtimelimit; /* limit for rt blks timer */
+ uint16_t qs_bwarnlimit; /* limit for num warnings */
+ uint16_t qs_iwarnlimit; /* limit for num warnings */
+ uint64_t qs_pad2[8]; /* for future proofing */
+};
+.in
+.fi
+
+The
+.I qs_version
+field of the structure should be filled with the version of the structure
+supported by the callee (for now, only
+.I FS_QSTAT_VERSION1
+is supported). Kernel will fill the structure in accordance with
+version provided.
+The
+.I id
+argument is ignored.
.TP
.B Q_XQUOTARM
-Free the disk space taken by disk quotas.
-Quotas must have already been turned off.
+Free the disk space taken by disk quotas. The
+.I addr
+argument should be a pointer to an
+.I "unsigned int"
+value containing flags (the same as in
+.I d_flags
+field of
+.I fs_disk_quota
+structure) which identify what types of quota should be removed (note that quota
+type passed in
+.I cmd
+argument is ignored, but should remain valid in order to pass preliminary
+quotactl syscall handler checks).
+
+Quotas must have already been turned off. The
+.I id
+argument is ignored.
.PP
There is no command equivalent to
.B Q_SYNC
@@ -557,6 +722,34 @@ or
but there is no ID greater than or equal to
.IR id
that has an active quota.
+.SH NOTES
+Instead of
+.I <xfs/xqm.h>
+one can use
+.IR <linux/dqblk_xfs.h> ,
+taking into account that there are several naming discrepancies:
+.IP \(bu 3
+Quota enabling flags (of format
+.BR XFS_QUOTA_[UGP]DQ_{ACCT,ENFD} )
+are defined without leading "X", as
+.BR FS_QUOTA_[UGP]DQ_{ACCT,ENFD} .
+.IP \(bu
+The same is applied to
+.B XFS_{USER,GROUP,PROJ}_QUOTA
+quota type flags which are defined as
+.BR FS_{USER,GROUP,PROJ}_QUOTA .
+.IP \(bu
+The
+.I dqblk_xfs.h
+defines its own
+.BR XQM_USRQUOTA ,
+.BR XQM_GRPQUOTA ,
+and
+.B XQM_PRJQUOTA
+constants for the available quota types, but their values are the same as for
+constants without the
+.B XQM_
+prefix.
.SH SEE ALSO
.BR quota (1),
.BR getrlimit (2),
--
1.7.10.4
--
To unsubscribe from this list: send the line "unsubscribe linux-man" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
reply other threads:[~2016-09-21 3:28 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=20160921032857.GA5550@obsidian \
--to=evgsyr-re5jqeeqqe8avxtiumwx3w@public.gmane.org \
--cc=linux-man-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=mtk.manpages-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.