[mirror_ubuntu-bionic-kernel.git] / fs / xfs / xfs_discard.c

/*
 * Copyright (C) 2010 Red Hat, Inc.
 * All Rights Reserved.
 *
 * 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.
 *
 * This program is distributed in the hope that it would 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 the Free Software Foundation,
 * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
 */
#include "xfs.h"
#include "xfs_format.h"
#include "xfs_log.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
#include "xfs_ag.h"
#include "xfs_mount.h"
#include "xfs_quota.h"
#include "xfs_alloc_btree.h"
#include "xfs_bmap_btree.h"
#include "xfs_ialloc_btree.h"
#include "xfs_btree.h"
#include "xfs_inode.h"
#include "xfs_alloc.h"
#include "xfs_error.h"
#include "xfs_extent_busy.h"
#include "xfs_discard.h"
#include "xfs_trace.h"

STATIC int
xfs_trim_extents(
	struct xfs_mount	*mp,
	xfs_agnumber_t		agno,
	xfs_daddr_t		start,
	xfs_daddr_t		end,
	xfs_daddr_t		minlen,
	__uint64_t		*blocks_trimmed)
{
	struct block_device	*bdev = mp->m_ddev_targp->bt_bdev;
	struct xfs_btree_cur	*cur;
	struct xfs_buf		*agbp;
	struct xfs_perag	*pag;
	int			error;
	int			i;

	pag = xfs_perag_get(mp, agno);

	error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp);
	if (error || !agbp)
		goto out_put_perag;

	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT);

	/*
	 * Force out the log.  This means any transactions that might have freed
	 * space before we took the AGF buffer lock are now on disk, and the
	 * volatile disk cache is flushed.
	 */
	xfs_log_force(mp, XFS_LOG_SYNC);

	/*
	 * Look up the longest btree in the AGF and start with it.
	 */
	error = xfs_alloc_lookup_ge(cur, 0,
			    be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest), &i);
	if (error)
		goto out_del_cursor;

	/*
	 * Loop until we are done with all extents that are large
	 * enough to be worth discarding.
	 */
	while (i) {
		xfs_agblock_t	fbno;
		xfs_extlen_t	flen;
		xfs_daddr_t	dbno;
		xfs_extlen_t	dlen;

		error = xfs_alloc_get_rec(cur, &fbno, &flen, &i);
		if (error)
			goto out_del_cursor;
		XFS_WANT_CORRUPTED_GOTO(i == 1, out_del_cursor);
		ASSERT(flen <= be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest));

		/*
		 * use daddr format for all range/len calculations as that is
		 * the format the range/len variables are supplied in by
		 * userspace.
		 */
		dbno = XFS_AGB_TO_DADDR(mp, agno, fbno);
		dlen = XFS_FSB_TO_BB(mp, flen);

		/*
		 * Too small?  Give up.
		 */
		if (dlen < minlen) {
			trace_xfs_discard_toosmall(mp, agno, fbno, flen);
			goto out_del_cursor;
		}

		/*
		 * If the extent is entirely outside of the range we are
		 * supposed to discard skip it.  Do not bother to trim
		 * down partially overlapping ranges for now.
		 */
		if (dbno + dlen < start || dbno > end) {
			trace_xfs_discard_exclude(mp, agno, fbno, flen);
			goto next_extent;
		}

		/*
		 * If any blocks in the range are still busy, skip the
		 * discard and try again the next time.
		 */
		if (xfs_extent_busy_search(mp, agno, fbno, flen)) {
			trace_xfs_discard_busy(mp, agno, fbno, flen);
			goto next_extent;
		}

		trace_xfs_discard_extent(mp, agno, fbno, flen);
		error = -blkdev_issue_discard(bdev, dbno, dlen, GFP_NOFS, 0);
		if (error)
			goto out_del_cursor;
		*blocks_trimmed += flen;

next_extent:
		error = xfs_btree_decrement(cur, 0, &i);
		if (error)
			goto out_del_cursor;
	}

out_del_cursor:
	xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
	xfs_buf_relse(agbp);
out_put_perag:
	xfs_perag_put(pag);
	return error;
}

/*
 * trim a range of the filesystem.
 *
 * Note: the parameters passed from userspace are byte ranges into the
 * filesystem which does not match to the format we use for filesystem block
 * addressing. FSB addressing is sparse (AGNO|AGBNO), while the incoming format
 * is a linear address range. Hence we need to use DADDR based conversions and
 * comparisons for determining the correct offset and regions to trim.
 */
int
xfs_ioc_trim(
	struct xfs_mount		*mp,
	struct fstrim_range __user	*urange)
{
	struct request_queue	*q = mp->m_ddev_targp->bt_bdev->bd_disk->queue;
	unsigned int		granularity = q->limits.discard_granularity;
	struct fstrim_range	range;
	xfs_daddr_t		start, end, minlen;
	xfs_agnumber_t		start_agno, end_agno, agno;
	__uint64_t		blocks_trimmed = 0;
	int			error, last_error = 0;

	if (!capable(CAP_SYS_ADMIN))
		return -XFS_ERROR(EPERM);
	if (!blk_queue_discard(q))
		return -XFS_ERROR(EOPNOTSUPP);
	if (copy_from_user(&range, urange, sizeof(range)))
		return -XFS_ERROR(EFAULT);

	/*
	 * Truncating down the len isn't actually quite correct, but using
	 * BBTOB would mean we trivially get overflows for values
	 * of ULLONG_MAX or slightly lower.  And ULLONG_MAX is the default
	 * used by the fstrim application.  In the end it really doesn't
	 * matter as trimming blocks is an advisory interface.
	 */
	if (range.start >= XFS_FSB_TO_B(mp, mp->m_sb.sb_dblocks) ||
	    range.minlen > XFS_FSB_TO_B(mp, XFS_ALLOC_AG_MAX_USABLE(mp)))
		return -XFS_ERROR(EINVAL);

	start = BTOBB(range.start);
	end = start + BTOBBT(range.len) - 1;
	minlen = BTOBB(max_t(u64, granularity, range.minlen));

	if (end > XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks) - 1)
		end = XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks)- 1;

	start_agno = xfs_daddr_to_agno(mp, start);
	end_agno = xfs_daddr_to_agno(mp, end);

	for (agno = start_agno; agno <= end_agno; agno++) {
		error = -xfs_trim_extents(mp, agno, start, end, minlen,
					  &blocks_trimmed);
		if (error)
			last_error = error;
	}

	if (last_error)
		return last_error;

	range.len = XFS_FSB_TO_B(mp, blocks_trimmed);
	if (copy_to_user(urange, &range, sizeof(range)))
		return -XFS_ERROR(EFAULT);
	return 0;
}

int
xfs_discard_extents(
	struct xfs_mount	*mp,
	struct list_head	*list)
{
	struct xfs_extent_busy	*busyp;
	int			error = 0;

	list_for_each_entry(busyp, list, list) {
		trace_xfs_discard_extent(mp, busyp->agno, busyp->bno,
					 busyp->length);

		error = -blkdev_issue_discard(mp->m_ddev_targp->bt_bdev,
				XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno),
				XFS_FSB_TO_BB(mp, busyp->length),
				GFP_NOFS, 0);
		if (error && error != EOPNOTSUPP) {
			xfs_info(mp,
	 "discard failed for extent [0x%llu,%u], error %d",
				 (unsigned long long)busyp->bno,
				 busyp->length,
				 error);
			return error;
		}
	}

	return 0;
}
Commit	Line	Data
a46db608 CH	1	/*
	2	* Copyright (C) 2010 Red Hat, Inc.
	3	* All Rights Reserved.
	4	*
	5	* This program is free software; you can redistribute it and/or
	6	* modify it under the terms of the GNU General Public License as
	7	* published by the Free Software Foundation.
	8	*
	9	* This program is distributed in the hope that it would be useful,
	10	* but WITHOUT ANY WARRANTY; without even the implied warranty of
	11	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
	12	* GNU General Public License for more details.
	13	*
	14	* You should have received a copy of the GNU General Public License
	15	* along with this program; if not, write the Free Software Foundation,
	16	* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
	17	*/
	18	#include "xfs.h"
6ca1c906	19	#include "xfs_format.h"
a46db608	20	#include "xfs_log.h"
7fd36c44 DC	21	#include "xfs_trans.h"
7fd36c44 DC	22	#include "xfs_sb.h"
a46db608 CH	23	#include "xfs_ag.h"
	24	#include "xfs_mount.h"
	25	#include "xfs_quota.h"
a46db608 CH	26	#include "xfs_alloc_btree.h"
	27	#include "xfs_bmap_btree.h"
	28	#include "xfs_ialloc_btree.h"
	29	#include "xfs_btree.h"
	30	#include "xfs_inode.h"
	31	#include "xfs_alloc.h"
	32	#include "xfs_error.h"
efc27b52	33	#include "xfs_extent_busy.h"
a46db608 CH	34	#include "xfs_discard.h"
	35	#include "xfs_trace.h"
	36
	37	STATIC int
	38	xfs_trim_extents(
	39	struct xfs_mount *mp,
	40	xfs_agnumber_t agno,
a66d6363 DC	41	xfs_daddr_t start,
	42	xfs_daddr_t end,
	43	xfs_daddr_t minlen,
a46db608 CH	44	__uint64_t *blocks_trimmed)
	45	{
	46	struct block_device *bdev = mp->m_ddev_targp->bt_bdev;
	47	struct xfs_btree_cur *cur;
	48	struct xfs_buf *agbp;
	49	struct xfs_perag *pag;
	50	int error;
	51	int i;
	52
	53	pag = xfs_perag_get(mp, agno);
	54
	55	error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp);
	56	if (error \|\| !agbp)
	57	goto out_put_perag;
	58
	59	cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT);
	60
	61	/*
	62	* Force out the log. This means any transactions that might have freed
	63	* space before we took the AGF buffer lock are now on disk, and the
	64	* volatile disk cache is flushed.
	65	*/
	66	xfs_log_force(mp, XFS_LOG_SYNC);
	67
	68	/*
	69	* Look up the longest btree in the AGF and start with it.
	70	*/
a66d6363	71	error = xfs_alloc_lookup_ge(cur, 0,
b1c770c2	72	be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest), &i);
a46db608 CH	73	if (error)
	74	goto out_del_cursor;
	75
	76	/*
	77	* Loop until we are done with all extents that are large
	78	* enough to be worth discarding.
	79	*/
	80	while (i) {
a66d6363 DC	81	xfs_agblock_t fbno;
	82	xfs_extlen_t flen;
	83	xfs_daddr_t dbno;
	84	xfs_extlen_t dlen;
a46db608 CH	85
	86	error = xfs_alloc_get_rec(cur, &fbno, &flen, &i);
	87	if (error)
	88	goto out_del_cursor;
	89	XFS_WANT_CORRUPTED_GOTO(i == 1, out_del_cursor);
b1c770c2	90	ASSERT(flen <= be32_to_cpu(XFS_BUF_TO_AGF(agbp)->agf_longest));
a46db608	91
a66d6363 DC	92	/*
	93	* use daddr format for all range/len calculations as that is
	94	* the format the range/len variables are supplied in by
	95	* userspace.
	96	*/
	97	dbno = XFS_AGB_TO_DADDR(mp, agno, fbno);
	98	dlen = XFS_FSB_TO_BB(mp, flen);
	99
a46db608 CH	100	/*
	101	* Too small? Give up.
	102	*/
a66d6363	103	if (dlen < minlen) {
a46db608 CH	104	trace_xfs_discard_toosmall(mp, agno, fbno, flen);
	105	goto out_del_cursor;
	106	}
	107
	108	/*
	109	* If the extent is entirely outside of the range we are
	110	* supposed to discard skip it. Do not bother to trim
	111	* down partially overlapping ranges for now.
	112	*/
a66d6363	113	if (dbno + dlen < start \|\| dbno > end) {
a46db608 CH	114	trace_xfs_discard_exclude(mp, agno, fbno, flen);
	115	goto next_extent;
	116	}
	117
	118	/*
	119	* If any blocks in the range are still busy, skip the
	120	* discard and try again the next time.
	121	*/
4ecbfe63	122	if (xfs_extent_busy_search(mp, agno, fbno, flen)) {
a46db608 CH	123	trace_xfs_discard_busy(mp, agno, fbno, flen);
	124	goto next_extent;
	125	}
	126
	127	trace_xfs_discard_extent(mp, agno, fbno, flen);
a66d6363	128	error = -blkdev_issue_discard(bdev, dbno, dlen, GFP_NOFS, 0);
a46db608 CH	129	if (error)
	130	goto out_del_cursor;
	131	*blocks_trimmed += flen;
	132
	133	next_extent:
	134	error = xfs_btree_decrement(cur, 0, &i);
	135	if (error)
	136	goto out_del_cursor;
	137	}
	138
	139	out_del_cursor:
	140	xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
	141	xfs_buf_relse(agbp);
	142	out_put_perag:
	143	xfs_perag_put(pag);
	144	return error;
	145	}
	146
a66d6363 DC	147	/*
	148	* trim a range of the filesystem.
	149	*
	150	* Note: the parameters passed from userspace are byte ranges into the
	151	* filesystem which does not match to the format we use for filesystem block
	152	* addressing. FSB addressing is sparse (AGNO\|AGBNO), while the incoming format
	153	* is a linear address range. Hence we need to use DADDR based conversions and
	154	* comparisons for determining the correct offset and regions to trim.
	155	*/
a46db608 CH	156	int
	157	xfs_ioc_trim(
	158	struct xfs_mount *mp,
	159	struct fstrim_range __user *urange)
	160	{
	161	struct request_queue *q = mp->m_ddev_targp->bt_bdev->bd_disk->queue;
	162	unsigned int granularity = q->limits.discard_granularity;
	163	struct fstrim_range range;
a66d6363	164	xfs_daddr_t start, end, minlen;
a46db608 CH	165	xfs_agnumber_t start_agno, end_agno, agno;
	166	__uint64_t blocks_trimmed = 0;
	167	int error, last_error = 0;
	168
	169	if (!capable(CAP_SYS_ADMIN))
	170	return -XFS_ERROR(EPERM);
be715140 LC	171	if (!blk_queue_discard(q))
be715140 LC	172	return -XFS_ERROR(EOPNOTSUPP);
a46db608 CH	173	if (copy_from_user(&range, urange, sizeof(range)))
	174	return -XFS_ERROR(EFAULT);
	175
	176	/*
	177	* Truncating down the len isn't actually quite correct, but using
a66d6363	178	* BBTOB would mean we trivially get overflows for values
a46db608 CH	179	* of ULLONG_MAX or slightly lower. And ULLONG_MAX is the default
	180	* used by the fstrim application. In the end it really doesn't
	181	* matter as trimming blocks is an advisory interface.
	182	*/
a672e1be TR	183	if (range.start >= XFS_FSB_TO_B(mp, mp->m_sb.sb_dblocks) \|\|
	184	range.minlen > XFS_FSB_TO_B(mp, XFS_ALLOC_AG_MAX_USABLE(mp)))
	185	return -XFS_ERROR(EINVAL);
	186
a66d6363 DC	187	start = BTOBB(range.start);
	188	end = start + BTOBBT(range.len) - 1;
	189	minlen = BTOBB(max_t(u64, granularity, range.minlen));
a46db608	190
a66d6363 DC	191	if (end > XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks) - 1)
a66d6363 DC	192	end = XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks)- 1;
a46db608	193
a66d6363 DC	194	start_agno = xfs_daddr_to_agno(mp, start);
a66d6363 DC	195	end_agno = xfs_daddr_to_agno(mp, end);
a46db608 CH	196
a46db608 CH	197	for (agno = start_agno; agno <= end_agno; agno++) {
c029a50d	198	error = -xfs_trim_extents(mp, agno, start, end, minlen,
a46db608 CH	199	&blocks_trimmed);
	200	if (error)
	201	last_error = error;
	202	}
	203
	204	if (last_error)
	205	return last_error;
	206
	207	range.len = XFS_FSB_TO_B(mp, blocks_trimmed);
	208	if (copy_to_user(urange, &range, sizeof(range)))
	209	return -XFS_ERROR(EFAULT);
	210	return 0;
	211	}
e84661aa CH	212
	213	int
	214	xfs_discard_extents(
	215	struct xfs_mount *mp,
	216	struct list_head *list)
	217	{
4ecbfe63	218	struct xfs_extent_busy *busyp;
e84661aa CH	219	int error = 0;
	220
	221	list_for_each_entry(busyp, list, list) {
	222	trace_xfs_discard_extent(mp, busyp->agno, busyp->bno,
	223	busyp->length);
	224
	225	error = -blkdev_issue_discard(mp->m_ddev_targp->bt_bdev,
	226	XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno),
	227	XFS_FSB_TO_BB(mp, busyp->length),
	228	GFP_NOFS, 0);
	229	if (error && error != EOPNOTSUPP) {
	230	xfs_info(mp,
	231	"discard failed for extent [0x%llu,%u], error %d",
	232	(unsigned long long)busyp->bno,
	233	busyp->length,
	234	error);
	235	return error;
	236	}
	237	}
	238
	239	return 0;
	240	}