/*
 * linux/fs/jfs/ioctl.c
 *
 * Copyright (C) 2006 Herbert Poetzl
 * adapted from Remy Card's ext2/ioctl.c
 */

#include <linux/fs.h>
#include <linux/ctype.h>
#include <linux/capability.h>
#include <linux/mount.h>
#include <linux/time.h>
#include <linux/sched.h>
#include <linux/blkdev.h>
#include <asm/current.h>
#include <linux/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;
	long ext2_flag;
} jfs_map[] = {
	{JFS_NOATIME_FL,	FS_NOATIME_FL},
	{JFS_DIRSYNC_FL,	FS_DIRSYNC_FL},
	{JFS_SYNC_FL,		FS_SYNC_FL},
	{JFS_SECRM_FL,		FS_SECRM_FL},
	{JFS_UNRM_FL,		FS_UNRM_FL},
	{JFS_APPEND_FL,		FS_APPEND_FL},
	{JFS_IMMUTABLE_FL,	FS_IMMUTABLE_FL},
	{0, 0},
};

static long jfs_map_ext2(unsigned long flags, int from)
{
	int index=0;
	long mapped=0;

	while (jfs_map[index].jfs_flag) {
		if (from) {
			if (jfs_map[index].ext2_flag & flags)
				mapped |= jfs_map[index].jfs_flag;
		} else {
			if (jfs_map[index].jfs_flag & flags)
				mapped |= jfs_map[index].ext2_flag;
		}
		index++;
	}
	return mapped;
}


long jfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
	struct inode *inode = file_inode(filp);
	struct jfs_inode_info *jfs_inode = JFS_IP(inode);
	unsigned int flags;

	switch (cmd) {
	case JFS_IOC_GETFLAGS:
		jfs_get_inode_flags(jfs_inode);
		flags = jfs_inode->mode2 & JFS_FL_USER_VISIBLE;
		flags = jfs_map_ext2(flags, 0);
		return put_user(flags, (int __user *) arg);
	case JFS_IOC_SETFLAGS: {
		unsigned int oldflags;
		int err;

		err = mnt_want_write_file(filp);
		if (err)
			return err;

		if (!inode_owner_or_capable(inode)) {
			err = -EACCES;
			goto setflags_out;
		}
		if (get_user(flags, (int __user *) arg)) {
			err = -EFAULT;
			goto setflags_out;
		}

		flags = jfs_map_ext2(flags, 1);
		if (!S_ISDIR(inode->i_mode))
			flags &= ~JFS_DIRSYNC_FL;

		/* Is it quota file? Do not allow user to mess with it */
		if (IS_NOQUOTA(inode)) {
			err = -EPERM;
			goto setflags_out;
		}

		/* Lock against other parallel changes of flags */
		inode_lock(inode);

		jfs_get_inode_flags(jfs_inode);
		oldflags = jfs_inode->mode2;

		/*
		 * The IMMUTABLE and APPEND_ONLY flags can only be changed by
		 * the relevant capability.
		 */
		if ((oldflags & JFS_IMMUTABLE_FL) ||
			((flags ^ oldflags) &
			(JFS_APPEND_FL | JFS_IMMUTABLE_FL))) {
			if (!capable(CAP_LINUX_IMMUTABLE)) {
				inode_unlock(inode);
				err = -EPERM;
				goto setflags_out;
			}
		}

		flags = flags & JFS_FL_USER_MODIFIABLE;
		flags |= oldflags & ~JFS_FL_USER_MODIFIABLE;
		jfs_inode->mode2 = flags;

		jfs_set_inode_flags(inode);
		inode_unlock(inode);
		inode->i_ctime = current_time(inode);
		mark_inode_dirty(inode);
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;
	}
}

#ifdef CONFIG_COMPAT
long jfs_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
{
	/* While these ioctl numbers defined with 'long' and have different
	 * numbers than the 64bit ABI,
	 * the actual implementation only deals with ints and is compatible.
	 */
	switch (cmd) {
	case JFS_IOC_GETFLAGS32:
		cmd = JFS_IOC_GETFLAGS;
		break;
	case JFS_IOC_SETFLAGS32:
		cmd = JFS_IOC_SETFLAGS;
		break;
	}
	return jfs_ioctl(filp, cmd, arg);
}
#endif
c040319aaebfec09e0144dc3341'/><table><tr><td colspan='2'/></tr><tr><td class='label'>context:</td><td class='ctrl'><select name='context' onchange='this.form.submit();'><option value='1'>1</option><option value='2'>2</option><option value='3' selected='selected'>3</option><option value='4'>4</option><option value='5'>5</option><option value='6'>6</option><option value='7'>7</option><option value='8'>8</option><option value='9'>9</option><option value='10'>10</option><option value='15'>15</option><option value='20'>20</option><option value='25'>25</option><option value='30'>30</option><option value='35'>35</option><option value='40'>40</option></select></td></tr><tr><td class='label'>space:</td><td class='ctrl'><select name='ignorews' onchange='this.form.submit();'><option value='0' selected='selected'>include</option><option value='1'>ignore</option></select></td></tr><tr><td class='label'>mode:</td><td class='ctrl'><select name='dt' onchange='this.form.submit();'><option value='0' selected='selected'>unified</option><option value='1'>ssdiff</option><option value='2'>stat only</option></select></td></tr><tr><td/><td class='ctrl'><noscript><input type='submit' value='reload'/></noscript></td></tr></table></form></div><table summary='commit info' class='commit-info'>
<tr><th>author</th><td>Douglas Miller &lt;dougmill@linux.vnet.ibm.com&gt;</td><td class='right'>2017-01-28 06:42:20 -0600</td></tr>
<tr><th>committer</th><td>Tejun Heo &lt;tj@kernel.org&gt;</td><td class='right'>2017-01-28 07:49:42 -0500</td></tr>
<tr><th>commit</th><td colspan='2' class='oid'><a href='/cgit.cgi/linux/net-next.git/commit/tools/testing/selftests/powerpc/alignment/copy_first_unaligned.c?id=966d2b04e070bc040319aaebfec09e0144dc3341'>966d2b04e070bc040319aaebfec09e0144dc3341</a> (<a href='/cgit.cgi/linux/net-next.git/patch/tools/testing/selftests/powerpc/alignment/copy_first_unaligned.c?id=966d2b04e070bc040319aaebfec09e0144dc3341'>patch</a>)</td></tr>
<tr><th>tree</th><td colspan='2' class='oid'><a href='/cgit.cgi/linux/net-next.git/tree/?id=966d2b04e070bc040319aaebfec09e0144dc3341'>4b96156e3d1dd4dfd6039b7c219c9dc4616da52d</a> /<a href='/cgit.cgi/linux/net-next.git/tree/tools/testing/selftests/powerpc/alignment/copy_first_unaligned.c?id=966d2b04e070bc040319aaebfec09e0144dc3341'>tools/testing/selftests/powerpc/alignment/copy_first_unaligned.c</a></td></tr>
<tr><th>parent</th><td colspan='2' class='oid'><a href='/cgit.cgi/linux/net-next.git/commit/tools/testing/selftests/powerpc/alignment/copy_first_unaligned.c?id=1b1bc42c1692e9b62756323c675a44cb1a1f9dbd'>1b1bc42c1692e9b62756323c675a44cb1a1f9dbd</a> (<a href='/cgit.cgi/linux/net-next.git/diff/tools/testing/selftests/powerpc/alignment/copy_first_unaligned.c?id=966d2b04e070bc040319aaebfec09e0144dc3341&amp;id2=1b1bc42c1692e9b62756323c675a44cb1a1f9dbd'>diff</a>)</td></tr></table>
<div class='commit-subject'>percpu-refcount: fix reference leak during percpu-atomic transition</div><div class='commit-msg'>percpu_ref_tryget() and percpu_ref_tryget_live() should return
"true" IFF they acquire a reference. But the return value from
atomic_long_inc_not_zero() is a long and may have high bits set,
e.g. PERCPU_COUNT_BIAS, and the return value of the tryget routines
is bool so the reference may actually be acquired but the routines
return "false" which results in a reference leak since the caller
assumes it does not need to do a corresponding percpu_ref_put().

This was seen when performing CPU hotplug during I/O, as hangs in
blk_mq_freeze_queue_wait where percpu_ref_kill (blk_mq_freeze_queue_start)
raced with percpu_ref_tryget (blk_mq_timeout_work).
Sample stack trace:

__switch_to+0x2c0/0x450
__schedule+0x2f8/0x970
schedule+0x48/0xc0
blk_mq_freeze_queue_wait+0x94/0x120
blk_mq_queue_reinit_work+0xb8/0x180
blk_mq_queue_reinit_prepare+0x84/0xa0
cpuhp_invoke_callback+0x17c/0x600
cpuhp_up_callbacks+0x58/0x150
_cpu_up+0xf0/0x1c0
do_cpu_up+0x120/0x150
cpu_subsys_online+0x64/0xe0
device_online+0xb4/0x120
online_store+0xb4/0xc0
dev_attr_store+0x68/0xa0
sysfs_kf_write+0x80/0xb0
kernfs_fop_write+0x17c/0x250
__vfs_write+0x6c/0x1e0
vfs_write+0xd0/0x270
SyS_write+0x6c/0x110
system_call+0x38/0xe0

Examination of the queue showed a single reference (no PERCPU_COUNT_BIAS,
and __PERCPU_REF_DEAD, __PERCPU_REF_ATOMIC set) and no requests.
However, conditions at the time of the race are count of PERCPU_COUNT_BIAS + 0
and __PERCPU_REF_DEAD and __PERCPU_REF_ATOMIC set.

The fix is to make the tryget routines use an actual boolean internally instead
of the atomic long result truncated to a int.

Fixes: e625305b3907 percpu-refcount: make percpu_ref based on longs instead of ints
Link: https://bugzilla.kernel.org/show_bug.cgi?id=190751
Signed-off-by: Douglas Miller &lt;dougmill@linux.vnet.ibm.com&gt;
Reviewed-by: Jens Axboe &lt;axboe@fb.com&gt;
Signed-off-by: Tejun Heo &lt;tj@kernel.org&gt;
Fixes: e625305b3907 ("percpu-refcount: make percpu_ref based on longs instead of ints")
Cc: stable@vger.kernel.org # v3.18+
</div><div class='diffstat-header'><a href='/cgit.cgi/linux/net-next.git/diff/?id=966d2b04e070bc040319aaebfec09e0144dc3341'>Diffstat</a> (limited to 'tools/testing/selftests/powerpc/alignment/copy_first_unaligned.c')</div><table summary='diffstat' class='diffstat'>