aboutsummaryrefslogtreecommitdiffstats
path: root/fs/jfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/jfs')
-rw-r--r--fs/jfs/Makefile2
-rw-r--r--fs/jfs/acl.c4
-rw-r--r--fs/jfs/file.c10
-rw-r--r--fs/jfs/inode.c20
-rw-r--r--fs/jfs/ioctl.c43
-rw-r--r--fs/jfs/jfs_discard.c121
-rw-r--r--fs/jfs/jfs_discard.h26
-rw-r--r--fs/jfs/jfs_dmap.c126
-rw-r--r--fs/jfs/jfs_dmap.h2
-rw-r--r--fs/jfs/jfs_filsys.h3
-rw-r--r--fs/jfs/jfs_imap.c22
-rw-r--r--fs/jfs/jfs_incore.h9
-rw-r--r--fs/jfs/jfs_txnmgr.c9
-rw-r--r--fs/jfs/namei.c18
-rw-r--r--fs/jfs/super.c104
-rw-r--r--fs/jfs/xattr.c4
16 files changed, 452 insertions, 71 deletions
diff --git a/fs/jfs/Makefile b/fs/jfs/Makefile
index a58fa72d7e5..d20d4737b3e 100644
--- a/fs/jfs/Makefile
+++ b/fs/jfs/Makefile
@@ -6,7 +6,7 @@ obj-$(CONFIG_JFS_FS) += jfs.o
jfs-y := super.o file.o inode.o namei.o jfs_mount.o jfs_umount.o \
jfs_xtree.o jfs_imap.o jfs_debug.o jfs_dmap.o \
- jfs_unicode.o jfs_dtree.o jfs_inode.o \
+ jfs_unicode.o jfs_dtree.o jfs_inode.o jfs_discard.o \
jfs_extent.o symlink.o jfs_metapage.o \
jfs_logmgr.o jfs_txnmgr.o jfs_uniupr.o \
resize.o xattr.o ioctl.o
diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
index 45559dc3ea2..d254d6d3599 100644
--- a/fs/jfs/acl.c
+++ b/fs/jfs/acl.c
@@ -64,7 +64,7 @@ struct posix_acl *jfs_get_acl(struct inode *inode, int type)
else
acl = ERR_PTR(size);
} else {
- acl = posix_acl_from_xattr(value, size);
+ acl = posix_acl_from_xattr(&init_user_ns, value, size);
}
kfree(value);
if (!IS_ERR(acl))
@@ -100,7 +100,7 @@ static int jfs_set_acl(tid_t tid, struct inode *inode, int type,
value = kmalloc(size, GFP_KERNEL);
if (!value)
return -ENOMEM;
- rc = posix_acl_to_xattr(acl, value, size);
+ rc = posix_acl_to_xattr(&init_user_ns, acl, value, size);
if (rc < 0)
goto out;
}
diff --git a/fs/jfs/file.c b/fs/jfs/file.c
index 844f9460cb1..dd7442c5835 100644
--- a/fs/jfs/file.c
+++ b/fs/jfs/file.c
@@ -108,8 +108,8 @@ int jfs_setattr(struct dentry *dentry, struct iattr *iattr)
if (is_quota_modification(inode, iattr))
dquot_initialize(inode);
- if ((iattr->ia_valid & ATTR_UID && iattr->ia_uid != inode->i_uid) ||
- (iattr->ia_valid & ATTR_GID && iattr->ia_gid != inode->i_gid)) {
+ if ((iattr->ia_valid & ATTR_UID && !uid_eq(iattr->ia_uid, inode->i_uid)) ||
+ (iattr->ia_valid & ATTR_GID && !gid_eq(iattr->ia_gid, inode->i_gid))) {
rc = dquot_transfer(inode, iattr);
if (rc)
return rc;
@@ -119,9 +119,12 @@ int jfs_setattr(struct dentry *dentry, struct iattr *iattr)
iattr->ia_size != i_size_read(inode)) {
inode_dio_wait(inode);
- rc = vmtruncate(inode, iattr->ia_size);
+ rc = inode_newsize_ok(inode, iattr->ia_size);
if (rc)
return rc;
+
+ truncate_setsize(inode, iattr->ia_size);
+ jfs_truncate(inode);
}
setattr_copy(inode, iattr);
@@ -133,7 +136,6 @@ int jfs_setattr(struct dentry *dentry, struct iattr *iattr)
}
const struct inode_operations jfs_file_inode_operations = {
- .truncate = jfs_truncate,
.setxattr = jfs_setxattr,
.getxattr = jfs_getxattr,
.listxattr = jfs_listxattr,
diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
index 4692bf3ca8c..b7dc47ba675 100644
--- a/fs/jfs/inode.c
+++ b/fs/jfs/inode.c
@@ -300,6 +300,16 @@ static int jfs_readpages(struct file *file, struct address_space *mapping,
return mpage_readpages(mapping, pages, nr_pages, jfs_get_block);
}
+static void jfs_write_failed(struct address_space *mapping, loff_t to)
+{
+ struct inode *inode = mapping->host;
+
+ if (to > inode->i_size) {
+ truncate_pagecache(inode, to, inode->i_size);
+ jfs_truncate(inode);
+ }
+}
+
static int jfs_write_begin(struct file *file, struct address_space *mapping,
loff_t pos, unsigned len, unsigned flags,
struct page **pagep, void **fsdata)
@@ -308,11 +318,8 @@ static int jfs_write_begin(struct file *file, struct address_space *mapping,
ret = nobh_write_begin(mapping, pos, len, flags, pagep, fsdata,
jfs_get_block);
- if (unlikely(ret)) {
- loff_t isize = mapping->host->i_size;
- if (pos + len > isize)
- vmtruncate(mapping->host, isize);
- }
+ if (unlikely(ret))
+ jfs_write_failed(mapping, pos + len);
return ret;
}
@@ -326,6 +333,7 @@ static ssize_t jfs_direct_IO(int rw, struct kiocb *iocb,
const struct iovec *iov, loff_t offset, unsigned long nr_segs)
{
struct file *file = iocb->ki_filp;
+ struct address_space *mapping = file->f_mapping;
struct inode *inode = file->f_mapping->host;
ssize_t ret;
@@ -341,7 +349,7 @@ static ssize_t jfs_direct_IO(int rw, struct kiocb *iocb,
loff_t end = offset + iov_length(iov, nr_segs);
if (end > isize)
- vmtruncate(inode, isize);
+ jfs_write_failed(mapping, end);
}
return ret;
diff --git a/fs/jfs/ioctl.c b/fs/jfs/ioctl.c
index f19d1e04a37..bc555ff417e 100644
--- a/fs/jfs/ioctl.c
+++ b/fs/jfs/ioctl.c
@@ -11,13 +11,17 @@
#include <linux/mount.h>
#include <linux/time.h>
#include <linux/sched.h>
+#include <linux/blkdev.h>
#include <asm/current.h>
#include <asm/uaccess.h>
+#include "jfs_filsys.h"
+#include "jfs_debug.h"
#include "jfs_incore.h"
#include "jfs_dinode.h"
#include "jfs_inode.h"
-
+#include "jfs_dmap.h"
+#include "jfs_discard.h"
static struct {
long jfs_flag;
@@ -123,6 +127,40 @@ setflags_out:
mnt_drop_write_file(filp);
return err;
}
+
+ case FITRIM:
+ {
+ struct super_block *sb = inode->i_sb;
+ struct request_queue *q = bdev_get_queue(sb->s_bdev);
+ struct fstrim_range range;
+ s64 ret = 0;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+
+ if (!blk_queue_discard(q)) {
+ jfs_warn("FITRIM not supported on device");
+ return -EOPNOTSUPP;
+ }
+
+ if (copy_from_user(&range, (struct fstrim_range __user *)arg,
+ sizeof(range)))
+ return -EFAULT;
+
+ range.minlen = max_t(unsigned int, range.minlen,
+ q->limits.discard_granularity);
+
+ ret = jfs_ioc_trim(inode, &range);
+ if (ret < 0)
+ return ret;
+
+ if (copy_to_user((struct fstrim_range __user *)arg, &range,
+ sizeof(range)))
+ return -EFAULT;
+
+ return 0;
+ }
+
default:
return -ENOTTY;
}
@@ -142,6 +180,9 @@ long jfs_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
case JFS_IOC_SETFLAGS32:
cmd = JFS_IOC_SETFLAGS;
break;
+ case FITRIM:
+ cmd = FITRIM;
+ break;
}
return jfs_ioctl(filp, cmd, arg);
}
diff --git a/fs/jfs/jfs_discard.c b/fs/jfs/jfs_discard.c
new file mode 100644
index 00000000000..dfcd5030455
--- /dev/null
+++ b/fs/jfs/jfs_discard.c
@@ -0,0 +1,121 @@
+/*
+ * Copyright (C) Tino Reichardt, 2012
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/blkdev.h>
+
+#include "jfs_incore.h"
+#include "jfs_superblock.h"
+#include "jfs_discard.h"
+#include "jfs_dmap.h"
+#include "jfs_debug.h"
+
+
+/*
+ * NAME: jfs_issue_discard()
+ *
+ * FUNCTION: TRIM the specified block range on device, if supported
+ *
+ * PARAMETERS:
+ * ip - pointer to in-core inode
+ * blkno - starting block number to be trimmed (0..N)
+ * nblocks - number of blocks to be trimmed
+ *
+ * RETURN VALUES:
+ * none
+ *
+ * serialization: IREAD_LOCK(ipbmap) held on entry/exit;
+ */
+void jfs_issue_discard(struct inode *ip, u64 blkno, u64 nblocks)
+{
+ struct super_block *sb = ip->i_sb;
+ int r = 0;
+
+ r = sb_issue_discard(sb, blkno, nblocks, GFP_NOFS, 0);
+ if (unlikely(r != 0)) {
+ jfs_err("JFS: sb_issue_discard" \
+ "(%p, %llu, %llu, GFP_NOFS, 0) = %d => failed!\n",
+ sb, (unsigned long long)blkno,
+ (unsigned long long)nblocks, r);
+ }
+
+ jfs_info("JFS: sb_issue_discard" \
+ "(%p, %llu, %llu, GFP_NOFS, 0) = %d\n",
+ sb, (unsigned long long)blkno,
+ (unsigned long long)nblocks, r);
+
+ return;
+}
+
+/*
+ * NAME: jfs_ioc_trim()
+ *
+ * FUNCTION: attempt to discard (TRIM) all free blocks from the
+ * filesystem.
+ *
+ * PARAMETERS:
+ * ip - pointer to in-core inode;
+ * range - the range, given by user space
+ *
+ * RETURN VALUES:
+ * 0 - success
+ * -EIO - i/o error
+ */
+int jfs_ioc_trim(struct inode *ip, struct fstrim_range *range)
+{
+ struct inode *ipbmap = JFS_SBI(ip->i_sb)->ipbmap;
+ struct bmap *bmp = JFS_SBI(ip->i_sb)->bmap;
+ struct super_block *sb = ipbmap->i_sb;
+ int agno, agno_end;
+ u64 start, end, minlen;
+ u64 trimmed = 0;
+
+ /**
+ * convert byte values to block size of filesystem:
+ * start: First Byte to trim
+ * len: number of Bytes to trim from start
+ * minlen: minimum extent length in Bytes
+ */
+ start = range->start >> sb->s_blocksize_bits;
+ end = start + (range->len >> sb->s_blocksize_bits) - 1;
+ minlen = range->minlen >> sb->s_blocksize_bits;
+ if (minlen == 0)
+ minlen = 1;
+
+ if (minlen > bmp->db_agsize ||
+ start >= bmp->db_mapsize ||
+ range->len < sb->s_blocksize)
+ return -EINVAL;
+
+ if (end >= bmp->db_mapsize)
+ end = bmp->db_mapsize - 1;
+
+ /**
+ * we trim all ag's within the range
+ */
+ agno = BLKTOAG(start, JFS_SBI(ip->i_sb));
+ agno_end = BLKTOAG(end, JFS_SBI(ip->i_sb));
+ while (agno <= agno_end) {
+ trimmed += dbDiscardAG(ip, agno, minlen);
+ agno++;
+ }
+ range->len = trimmed << sb->s_blocksize_bits;
+
+ return 0;
+}
diff --git a/fs/jfs/jfs_discard.h b/fs/jfs/jfs_discard.h
new file mode 100644
index 00000000000..40d1ee6081a
--- /dev/null
+++ b/fs/jfs/jfs_discard.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (C) Tino Reichardt, 2012
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+#ifndef _H_JFS_DISCARD
+#define _H_JFS_DISCARD
+
+struct fstrim_range;
+
+extern void jfs_issue_discard(struct inode *ip, u64 blkno, u64 nblocks);
+extern int jfs_ioc_trim(struct inode *ip, struct fstrim_range *range);
+
+#endif /* _H_JFS_DISCARD */
diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c
index 9cbd11a3f80..9a55f53be5f 100644
--- a/fs/jfs/jfs_dmap.c
+++ b/fs/jfs/jfs_dmap.c
@@ -1,5 +1,6 @@
/*
* Copyright (C) International Business Machines Corp., 2000-2004
+ * Portions Copyright (C) Tino Reichardt, 2012
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -25,6 +26,7 @@
#include "jfs_lock.h"
#include "jfs_metapage.h"
#include "jfs_debug.h"
+#include "jfs_discard.h"
/*
* SERIALIZATION of the Block Allocation Map.
@@ -104,7 +106,6 @@ static int dbFreeBits(struct bmap * bmp, struct dmap * dp, s64 blkno,
static int dbFreeDmap(struct bmap * bmp, struct dmap * dp, s64 blkno,
int nblocks);
static int dbMaxBud(u8 * cp);
-s64 dbMapFileSizeToMapSize(struct inode *ipbmap);
static int blkstol2(s64 nb);
static int cntlz(u32 value);
@@ -145,7 +146,6 @@ static const s8 budtab[256] = {
2, 1, 1, 1, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, -1
};
-
/*
* NAME: dbMount()
*
@@ -310,7 +310,6 @@ int dbSync(struct inode *ipbmap)
return (0);
}
-
/*
* NAME: dbFree()
*
@@ -337,6 +336,7 @@ int dbFree(struct inode *ip, s64 blkno, s64 nblocks)
s64 lblkno, rem;
struct inode *ipbmap = JFS_SBI(ip->i_sb)->ipbmap;
struct bmap *bmp = JFS_SBI(ip->i_sb)->bmap;
+ struct super_block *sb = ipbmap->i_sb;
IREAD_LOCK(ipbmap, RDWRLOCK_DMAP);
@@ -351,6 +351,13 @@ int dbFree(struct inode *ip, s64 blkno, s64 nblocks)
return -EIO;
}
+ /**
+ * TRIM the blocks, when mounted with discard option
+ */
+ if (JFS_SBI(sb)->flag & JFS_DISCARD)
+ if (JFS_SBI(sb)->minblks_trim <= nblocks)
+ jfs_issue_discard(ipbmap, blkno, nblocks);
+
/*
* free the blocks a dmap at a time.
*/
@@ -1095,7 +1102,6 @@ static int dbExtend(struct inode *ip, s64 blkno, s64 nblocks, s64 addnblocks)
/* we were not successful */
release_metapage(mp);
-
return (rc);
}
@@ -1590,6 +1596,118 @@ static int dbAllocAny(struct bmap * bmp, s64 nblocks, int l2nb, s64 * results)
/*
+ * NAME: dbDiscardAG()
+ *
+ * FUNCTION: attempt to discard (TRIM) all free blocks of specific AG
+ *
+ * algorithm:
+ * 1) allocate blocks, as large as possible and save them
+ * while holding IWRITE_LOCK on ipbmap
+ * 2) trim all these saved block/length values
+ * 3) mark the blocks free again
+ *
+ * benefit:
+ * - we work only on one ag at some time, minimizing how long we
+ * need to lock ipbmap
+ * - reading / writing the fs is possible most time, even on
+ * trimming
+ *
+ * downside:
+ * - we write two times to the dmapctl and dmap pages
+ * - but for me, this seems the best way, better ideas?
+ * /TR 2012
+ *
+ * PARAMETERS:
+ * ip - pointer to in-core inode
+ * agno - ag to trim
+ * minlen - minimum value of contiguous blocks
+ *
+ * RETURN VALUES:
+ * s64 - actual number of blocks trimmed
+ */
+s64 dbDiscardAG(struct inode *ip, int agno, s64 minlen)
+{
+ struct inode *ipbmap = JFS_SBI(ip->i_sb)->ipbmap;
+ struct bmap *bmp = JFS_SBI(ip->i_sb)->bmap;
+ s64 nblocks, blkno;
+ u64 trimmed = 0;
+ int rc, l2nb;
+ struct super_block *sb = ipbmap->i_sb;
+
+ struct range2trim {
+ u64 blkno;
+ u64 nblocks;
+ } *totrim, *tt;
+
+ /* max blkno / nblocks pairs to trim */
+ int count = 0, range_cnt;
+ u64 max_ranges;
+
+ /* prevent others from writing new stuff here, while trimming */
+ IWRITE_LOCK(ipbmap, RDWRLOCK_DMAP);
+
+ nblocks = bmp->db_agfree[agno];
+ max_ranges = nblocks;
+ do_div(max_ranges, minlen);
+ range_cnt = min_t(u64, max_ranges + 1, 32 * 1024);
+ totrim = kmalloc(sizeof(struct range2trim) * range_cnt, GFP_NOFS);
+ if (totrim == NULL) {
+ jfs_error(bmp->db_ipbmap->i_sb,
+ "dbDiscardAG: no memory for trim array");
+ IWRITE_UNLOCK(ipbmap);
+ return 0;
+ }
+
+ tt = totrim;
+ while (nblocks >= minlen) {
+ l2nb = BLKSTOL2(nblocks);
+
+ /* 0 = okay, -EIO = fatal, -ENOSPC -> try smaller block */
+ rc = dbAllocAG(bmp, agno, nblocks, l2nb, &blkno);
+ if (rc == 0) {
+ tt->blkno = blkno;
+ tt->nblocks = nblocks;
+ tt++; count++;
+
+ /* the whole ag is free, trim now */
+ if (bmp->db_agfree[agno] == 0)
+ break;
+
+ /* give a hint for the next while */
+ nblocks = bmp->db_agfree[agno];
+ continue;
+ } else if (rc == -ENOSPC) {
+ /* search for next smaller log2 block */
+ l2nb = BLKSTOL2(nblocks) - 1;
+ nblocks = 1 << l2nb;
+ } else {
+ /* Trim any already allocated blocks */
+ jfs_error(bmp->db_ipbmap->i_sb,
+ "dbDiscardAG: -EIO");
+ break;
+ }
+
+ /* check, if our trim array is full */
+ if (unlikely(count >= range_cnt - 1))
+ break;
+ }
+ IWRITE_UNLOCK(ipbmap);
+
+ tt->nblocks = 0; /* mark the current end */
+ for (tt = totrim; tt->nblocks != 0; tt++) {
+ /* when mounted with online discard, dbFree() will
+ * call jfs_issue_discard() itself */
+ if (!(JFS_SBI(sb)->flag & JFS_DISCARD))
+ jfs_issue_discard(ip, tt->blkno, tt->nblocks);
+ dbFree(ip, tt->blkno, tt->nblocks);
+ trimmed += tt->nblocks;
+ }
+ kfree(totrim);
+
+ return trimmed;
+}
+
+/*
* NAME: dbFindCtl()
*
* FUNCTION: starting at a specified dmap control page level and block
diff --git a/fs/jfs/jfs_dmap.h b/fs/jfs/jfs_dmap.h
index 6dcb906c55d..562b9a7e431 100644
--- a/fs/jfs/jfs_dmap.h
+++ b/fs/jfs/jfs_dmap.h
@@ -311,4 +311,6 @@ extern int dbAllocBottomUp(struct inode *ip, s64 blkno, s64 nblocks);
extern int dbExtendFS(struct inode *ipbmap, s64 blkno, s64 nblocks);
extern void dbFinalizeBmap(struct inode *ipbmap);
extern s64 dbMapFileSizeToMapSize(struct inode *ipbmap);
+extern s64 dbDiscardAG(struct inode *ip, int agno, s64 minlen);
+
#endif /* _H_JFS_DMAP */
diff --git a/fs/jfs/jfs_filsys.h b/fs/jfs/jfs_filsys.h
index b3f5463fbe5..b67d64671bb 100644
--- a/fs/jfs/jfs_filsys.h
+++ b/fs/jfs/jfs_filsys.h
@@ -45,6 +45,9 @@
/* mount time flag to disable journaling to disk */
#define JFS_NOINTEGRITY 0x00000040
+/* mount time flag to enable TRIM to ssd disks */
+#define JFS_DISCARD 0x00000080
+
/* commit option */
#define JFS_COMMIT 0x00000f00 /* commit option mask */
#define JFS_GROUPCOMMIT 0x00000100 /* group (of 1) commit */
diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c
index 1b6f15f191b..6ba4006e011 100644
--- a/fs/jfs/jfs_imap.c
+++ b/fs/jfs/jfs_imap.c
@@ -3078,15 +3078,15 @@ static int copy_from_dinode(struct dinode * dip, struct inode *ip)
}
set_nlink(ip, le32_to_cpu(dip->di_nlink));
- jfs_ip->saved_uid = le32_to_cpu(dip->di_uid);
- if (sbi->uid == -1)
+ jfs_ip->saved_uid = make_kuid(&init_user_ns, le32_to_cpu(dip->di_uid));
+ if (!uid_valid(sbi->uid))
ip->i_uid = jfs_ip->saved_uid;
else {
ip->i_uid = sbi->uid;
}
- jfs_ip->saved_gid = le32_to_cpu(dip->di_gid);
- if (sbi->gid == -1)
+ jfs_ip->saved_gid = make_kgid(&init_user_ns, le32_to_cpu(dip->di_gid));
+ if (!gid_valid(sbi->gid))
ip->i_gid = jfs_ip->saved_gid;
else {
ip->i_gid = sbi->gid;
@@ -3150,14 +3150,16 @@ static void copy_to_dinode(struct dinode * dip, struct inode *ip)
dip->di_size = cpu_to_le64(ip->i_size);
dip->di_nblocks = cpu_to_le64(PBLK2LBLK(ip->i_sb, ip->i_blocks));
dip->di_nlink = cpu_to_le32(ip->i_nlink);
- if (sbi->uid == -1)
- dip->di_uid = cpu_to_le32(ip->i_uid);
+ if (!uid_valid(sbi->uid))
+ dip->di_uid = cpu_to_le32(i_uid_read(ip));
else
- dip->di_uid = cpu_to_le32(jfs_ip->saved_uid);
- if (sbi->gid == -1)
- dip->di_gid = cpu_to_le32(ip->i_gid);
+ dip->di_uid =cpu_to_le32(from_kuid(&init_user_ns,
+ jfs_ip->saved_uid));
+ if (!gid_valid(sbi->gid))
+ dip->di_gid = cpu_to_le32(i_gid_read(ip));
else
- dip->di_gid = cpu_to_le32(jfs_ip->saved_gid);
+ dip->di_gid = cpu_to_le32(from_kgid(&init_user_ns,
+ jfs_ip->saved_gid));
jfs_get_inode_flags(jfs_ip);
/*
* mode2 is only needed for storing the higher order bits.
diff --git a/fs/jfs/jfs_incore.h b/fs/jfs/jfs_incore.h
index 584a4a1a6e8..cf47f09e8ac 100644
--- a/fs/jfs/jfs_incore.h
+++ b/fs/jfs/jfs_incore.h
@@ -38,8 +38,8 @@
struct jfs_inode_info {
int fileset; /* fileset number (always 16)*/
uint mode2; /* jfs-specific mode */
- uint saved_uid; /* saved for uid mount option */
- uint saved_gid; /* saved for gid mount option */
+ kuid_t saved_uid; /* saved for uid mount option */
+ kgid_t saved_gid; /* saved for gid mount option */
pxd_t ixpxd; /* inode extent descriptor */
dxd_t acl; /* dxd describing acl */
dxd_t ea; /* dxd describing ea */
@@ -192,9 +192,10 @@ struct jfs_sb_info {
uint state; /* mount/recovery state */
unsigned long flag; /* mount time flags */
uint p_state; /* state prior to going no integrity */
- uint uid; /* uid to override on-disk uid */
- uint gid; /* gid to override on-disk gid */
+ kuid_t uid; /* uid to override on-disk uid */
+ kgid_t gid; /* gid to override on-disk gid */
uint umask; /* umask to override on-disk umask */
+ uint minblks_trim; /* minimum blocks, for online trim */
};
/* jfs_sb_info commit_state */
diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c
index bb8b661bcc5..5fcc02eaa64 100644
--- a/fs/jfs/jfs_txnmgr.c
+++ b/fs/jfs/jfs_txnmgr.c
@@ -2977,12 +2977,9 @@ int jfs_sync(void *arg)
* put back on the anon_list.
*/
- /* Take off anon_list */
- list_del(&jfs_ip->anon_inode_list);
-
- /* Put on anon_list2 */
- list_add(&jfs_ip->anon_inode_list,
- &TxAnchor.anon_list2);
+ /* Move from anon_list to anon_list2 */
+ list_move(&jfs_ip->anon_inode_list,
+ &TxAnchor.anon_list2);
TXN_UNLOCK();
iput(ip);
diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c
index 07c91ca6017..3b91a7ad608 100644
--- a/fs/jfs/namei.c
+++ b/fs/jfs/namei.c
@@ -73,7 +73,7 @@ static inline void free_ea_wmap(struct inode *inode)
*
*/
static int jfs_create(struct inode *dip, struct dentry *dentry, umode_t mode,
- struct nameidata *nd)
+ bool excl)
{
int rc = 0;
tid_t tid; /* transaction id */
@@ -176,8 +176,8 @@ static int jfs_create(struct inode *dip, struct dentry *dentry, umode_t mode,
unlock_new_inode(ip);
iput(ip);
} else {
- d_instantiate(dentry, ip);
unlock_new_inode(ip);
+ d_instantiate(dentry, ip);
}
out2:
@@ -309,8 +309,8 @@ static int jfs_mkdir(struct inode *dip, struct dentry *dentry, umode_t mode)
unlock_new_inode(ip);
iput(ip);
} else {
- d_instantiate(dentry, ip);
unlock_new_inode(ip);
+ d_instantiate(dentry, ip);
}
out2:
@@ -1043,8 +1043,8 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry,
unlock_new_inode(ip);
iput(ip);
} else {
- d_instantiate(dentry, ip);
unlock_new_inode(ip);
+ d_instantiate(dentry, ip);
}
out2:
@@ -1424,8 +1424,8 @@ static int jfs_mknod(struct inode *dir, struct dentry *dentry,
unlock_new_inode(ip);
iput(ip);
} else {
- d_instantiate(dentry, ip);
unlock_new_inode(ip);
+ d_instantiate(dentry, ip);
}
out1:
@@ -1436,7 +1436,7 @@ static int jfs_mknod(struct inode *dir, struct dentry *dentry,
return rc;
}
-static struct dentry *jfs_lookup(struct inode *dip, struct dentry *dentry, struct nameidata *nd)
+static struct dentry *jfs_lookup(struct inode *dip, struct dentry *dentry, unsigned int flags)
{
struct btstack btstack;
ino_t inum;
@@ -1570,7 +1570,7 @@ out:
return result;
}
-static int jfs_ci_revalidate(struct dentry *dentry, struct nameidata *nd)
+static int jfs_ci_revalidate(struct dentry *dentry, unsigned int flags)
{
/*
* This is not negative dentry. Always valid.
@@ -1589,7 +1589,7 @@ static int jfs_ci_revalidate(struct dentry *dentry, struct nameidata *nd)
* This may be nfsd (or something), anyway, we can't see the
* intent of this. So, since this can be for creation, drop it.
*/
- if (!nd)
+ if (!flags)
return 0;
/*
@@ -1597,7 +1597,7 @@ static int jfs_ci_revalidate(struct dentry *dentry, struct nameidata *nd)
* case sensitive name which is specified by user if this is
* for creation.
*/
- if (nd->flags & (LOOKUP_CREATE | LOOKUP_RENAME_TARGET))
+ if (flags & (LOOKUP_CREATE | LOOKUP_RENAME_TARGET))
return 0;
return 1;
}
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index 4a82950f412..1a543be09c7 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -33,6 +33,7 @@
#include <linux/slab.h>
#include <asm/uaccess.h>
#include <linux/seq_file.h>
+#include <linux/blkdev.h>
#include "jfs_incore.h"
#include "jfs_filsys.h"
@@ -100,7 +101,7 @@ void jfs_error(struct super_block *sb, const char * function, ...)
vsnprintf(error_buf, sizeof(error_buf), function, args);
va_end(args);
- printk(KERN_ERR "ERROR: (device %s): %s\n", sb->s_id, error_buf);
+ pr_err("ERROR: (device %s): %s\n", sb->s_id, error_buf);
jfs_handle_error(sb);
}
@@ -197,7 +198,8 @@ static void jfs_put_super(struct super_block *sb)
enum {
Opt_integrity, Opt_nointegrity, Opt_iocharset, Opt_resize,
Opt_resize_nosize, Opt_errors, Opt_ignore, Opt_err, Opt_quota,
- Opt_usrquota, Opt_grpquota, Opt_uid, Opt_gid, Opt_umask
+ Opt_usrquota, Opt_grpquota, Opt_uid, Opt_gid, Opt_umask,
+ Opt_discard, Opt_nodiscard, Opt_discard_minblk
};
static const match_table_t tokens = {
@@ -214,6 +216,9 @@ static const match_table_t tokens = {
{Opt_uid, "uid=%u"},
{Opt_gid, "gid=%u"},
{Opt_umask, "umask=%u"},
+ {Opt_discard, "discard"},
+ {Opt_nodiscard, "nodiscard"},
+ {Opt_discard_minblk, "discard=%u"},
{Opt_err, NULL}
};
@@ -255,8 +260,7 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
else {
nls_map = load_nls(args[0].from);
if (!nls_map) {
- printk(KERN_ERR
- "JFS: charset not found\n");
+ pr_err("JFS: charset not found\n");
goto cleanup;
}
}
@@ -272,8 +276,7 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
*newLVSize = sb->s_bdev->bd_inode->i_size >>
sb->s_blocksize_bits;
if (*newLVSize == 0)
- printk(KERN_ERR
- "JFS: Cannot determine volume size\n");
+ pr_err("JFS: Cannot determine volume size\n");
break;
}
case Opt_errors:
@@ -294,8 +297,7 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
*flag &= ~JFS_ERR_REMOUNT_RO;
*flag |= JFS_ERR_PANIC;
} else {
- printk(KERN_ERR
- "JFS: %s is an invalid error handler\n",
+ pr_err("JFS: %s is an invalid error handler\n",
errors);
goto cleanup;
}
@@ -314,33 +316,76 @@ static int parse_options(char *options, struct super_block *sb, s64 *newLVSize,
case Opt_usrquota:
case Opt_grpquota:
case Opt_quota:
- printk(KERN_ERR
- "JFS: quota operations not supported\n");
+ pr_err("JFS: quota operations not supported\n");
break;
#endif
case Opt_uid:
{
char *uid = args[0].from;
- sbi->uid = simple_strtoul(uid, &uid, 0);
+ uid_t val = simple_strtoul(uid, &uid, 0);
+ sbi->uid = make_kuid(current_user_ns(), val);
+ if (!uid_valid(sbi->uid))
+ goto cleanup;
break;
}
+
case Opt_gid:
{
char *gid = args[0].from;
- sbi->gid = simple_strtoul(gid, &gid, 0);
+ gid_t val = simple_strtoul(gid, &gid, 0);
+ sbi->gid = make_kgid(current_user_ns(), val);
+ if (!gid_valid(sbi->gid))
+ goto cleanup;
break;
}
+
case Opt_umask:
{
char *umask = args[0].from;
sbi->umask = simple_strtoul(umask, &umask, 8);
if (sbi->umask & ~0777) {
- printk(KERN_ERR
- "JFS: Invalid value of umask\n");
+ pr_err("JFS: Invalid value of umask\n");
goto cleanup;
}
break;
}
+
+ case Opt_discard:
+ {
+ struct request_queue *q = bdev_get_queue(sb->s_bdev);
+ /* if set to 1, even copying files will cause
+ * trimming :O
+ * -> user has more control over the online trimming
+ */
+ sbi->minblks_trim = 64;
+ if (blk_queue_discard(q)) {
+ *flag |= JFS_DISCARD;
+ } else {
+ pr_err("JFS: discard option " \
+ "not supported on device\n");
+ }
+ break;
+ }
+
+ case Opt_nodiscard:
+ *flag &= ~JFS_DISCARD;
+ break;
+
+ case Opt_discard_minblk:
+ {
+ struct request_queue *q = bdev_get_queue(sb->s_bdev);
+ char *minblks_trim = args[0].from;
+ if (blk_queue_discard(q)) {
+ *flag |= JFS_DISCARD;
+ sbi->minblks_trim = simple_strtoull(
+ minblks_trim, &minblks_trim, 0);
+ } else {
+ pr_err("JFS: discard option " \
+ "not supported on device\n");
+ }
+ break;
+ }
+
default:
printk("jfs: Unrecognized mount option \"%s\" "
" or missing value\n", p);
@@ -374,8 +419,8 @@ static int jfs_remount(struct super_block *sb, int *flags, char *data)
if (newLVSize) {
if (sb->s_flags & MS_RDONLY) {
- printk(KERN_ERR
- "JFS: resize requires volume to be mounted read-write\n");
+ pr_err("JFS: resize requires volume" \
+ " to be mounted read-write\n");
return -EROFS;
}
rc = jfs_extendfs(sb, newLVSize, 0);
@@ -443,7 +488,9 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
sb->s_fs_info = sbi;
sb->s_max_links = JFS_LINK_MAX;
sbi->sb = sb;
- sbi->uid = sbi->gid = sbi->umask = -1;
+ sbi->uid = INVALID_UID;
+ sbi->gid = INVALID_GID;
+ sbi->umask = -1;
/* initialize the mount flag and determine the default error handler */
flag = JFS_ERR_REMOUNT_RO;
@@ -457,7 +504,7 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent)
#endif
if (newLVSize) {
- printk(KERN_ERR "resize option for remount only\n");
+ pr_err("resize option for remount only\n");
goto out_kfree;
}
@@ -601,6 +648,11 @@ static int jfs_sync_fs(struct super_block *sb, int wait)
/* log == NULL indicates read-only mount */
if (log) {
+ /*
+ * Write quota structures to quota file, sync_blockdev() will
+ * write them to disk later
+ */
+ dquot_writeback_dquots(sb, -1);
jfs_flush_journal(log, wait);
jfs_syncpt(log, 0);
}
@@ -612,14 +664,16 @@ static int jfs_show_options(struct seq_file *seq, struct dentry *root)
{
struct jfs_sb_info *sbi = JFS_SBI(root->d_sb);
- if (sbi->uid != -1)
- seq_printf(seq, ",uid=%d", sbi->uid);
- if (sbi->gid != -1)
- seq_printf(seq, ",gid=%d", sbi->gid);
+ if (uid_valid(sbi->uid))
+ seq_printf(seq, ",uid=%d", from_kuid(&init_user_ns, sbi->uid));
+ if (gid_valid(sbi->gid))
+ seq_printf(seq, ",gid=%d", from_kgid(&init_user_ns, sbi->gid));
if (sbi->umask != -1)
seq_printf(seq, ",umask=%03o", sbi->umask);
if (sbi->flag & JFS_NOINTEGRITY)
seq_puts(seq, ",nointegrity");
+ if (sbi->flag & JFS_DISCARD)
+ seq_printf(seq, ",discard=%u", sbi->minblks_trim);
if (sbi->nls_tab)
seq_printf(seq, ",iocharset=%s", sbi->nls_tab->charset);
if (sbi->flag & JFS_ERR_CONTINUE)
@@ -898,6 +952,12 @@ static void __exit exit_jfs_fs(void)
jfs_proc_clean();
#endif
unregister_filesystem(&jfs_fs_type);
+
+ /*
+ * Make sure all delayed rcu free inodes are flushed before we
+ * destroy cache.
+ */
+ rcu_barrier();
kmem_cache_destroy(jfs_inode_cachep);
}
diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c
index 26683e15b3a..42d67f9757b 100644
--- a/fs/jfs/xattr.c
+++ b/fs/jfs/xattr.c
@@ -685,7 +685,7 @@ static int can_set_system_xattr(struct inode *inode, const char *name,
* POSIX_ACL_XATTR_ACCESS is tied to i_mode
*/
if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0) {
- acl = posix_acl_from_xattr(value, value_len);
+ acl = posix_acl_from_xattr(&init_user_ns, value, value_len);
if (IS_ERR(acl)) {
rc = PTR_ERR(acl);
printk(KERN_ERR "posix_acl_from_xattr returned %d\n",
@@ -710,7 +710,7 @@ static int can_set_system_xattr(struct inode *inode, const char *name,
return 0;
} else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0) {
- acl = posix_acl_from_xattr(value, value_len);
+ acl = posix_acl_from_xattr(&init_user_ns, value, value_len);
if (IS_ERR(acl)) {
rc = PTR_ERR(acl);
printk(KERN_ERR "posix_acl_from_xattr returned %d\n",