f2fs: support passing down write hints given by users to block layer

Add the 'whint_mode' mount option that controls which write
hints are passed down to block layer. There are "off" and
"user-based" mode. The default mode is "off".

1) whint_mode=off. F2FS only passes down WRITE_LIFE_NOT_SET.

2) whint_mode=user-based. F2FS tries to pass down hints given
by users.

User                  F2FS                     Block
----                  ----                     -----
                      META                     WRITE_LIFE_NOT_SET
                      HOT_NODE                 "
                      WARM_NODE                "
                      COLD_NODE                "
ioctl(COLD)           COLD_DATA                WRITE_LIFE_EXTREME
extension list        "                        "

-- buffered io
WRITE_LIFE_EXTREME    COLD_DATA                WRITE_LIFE_EXTREME
WRITE_LIFE_SHORT      HOT_DATA                 WRITE_LIFE_SHORT
WRITE_LIFE_NOT_SET    WARM_DATA                WRITE_LIFE_NOT_SET
WRITE_LIFE_NONE       "                        "
WRITE_LIFE_MEDIUM     "                        "
WRITE_LIFE_LONG       "                        "

-- direct io
WRITE_LIFE_EXTREME    COLD_DATA                WRITE_LIFE_EXTREME
WRITE_LIFE_SHORT      HOT_DATA                 WRITE_LIFE_SHORT
WRITE_LIFE_NOT_SET    WARM_DATA                WRITE_LIFE_NOT_SET
WRITE_LIFE_NONE       "                        WRITE_LIFE_NONE
WRITE_LIFE_MEDIUM     "                        WRITE_LIFE_MEDIUM
WRITE_LIFE_LONG       "                        WRITE_LIFE_LONG

Many thanks to Chao Yu and Jaegeuk Kim for comments to
implement this patch.

Signed-off-by: Hyunchul Lee <cheol.lee@lge.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
[Jaegeuk Kim: avoid build warning]
[Chao Yu: fix to restore whint_mode in ->remount_fs]
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index d603f40..4c29be8 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -173,15 +173,22 @@
  */
 static struct bio *__bio_alloc(struct f2fs_sb_info *sbi, block_t blk_addr,
 				struct writeback_control *wbc,
-				int npages, bool is_read)
+				int npages, bool is_read,
+				enum page_type type, enum temp_type temp)
 {
 	struct bio *bio;
 
 	bio = f2fs_bio_alloc(sbi, npages, true);
 
 	f2fs_target_device(sbi, blk_addr, bio);
-	bio->bi_end_io = is_read ? f2fs_read_end_io : f2fs_write_end_io;
-	bio->bi_private = is_read ? NULL : sbi;
+	if (is_read) {
+		bio->bi_end_io = f2fs_read_end_io;
+		bio->bi_private = NULL;
+	} else {
+		bio->bi_end_io = f2fs_write_end_io;
+		bio->bi_private = sbi;
+		bio->bi_write_hint = io_type_to_rw_hint(sbi, type, temp);
+	}
 	if (wbc)
 		wbc_init_bio(wbc, bio);
 
@@ -380,7 +387,7 @@
 
 	/* Allocate a new bio */
 	bio = __bio_alloc(fio->sbi, fio->new_blkaddr, fio->io_wbc,
-				1, is_read_io(fio->op));
+				1, is_read_io(fio->op), fio->type, fio->temp);
 
 	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
 		bio_put(bio);
@@ -443,7 +450,8 @@
 			goto out_fail;
 		}
 		io->bio = __bio_alloc(sbi, fio->new_blkaddr, fio->io_wbc,
-						BIO_MAX_PAGES, false);
+						BIO_MAX_PAGES, false,
+						fio->type, fio->temp);
 		io->fio = *fio;
 	}
 
@@ -867,8 +875,7 @@
 	map.m_seg_type = NO_CHECK_TYPE;
 
 	if (direct_io) {
-		/* map.m_seg_type = rw_hint_to_seg_type(iocb->ki_hint); */
-		map.m_seg_type = rw_hint_to_seg_type(WRITE_LIFE_NOT_SET);
+		map.m_seg_type = rw_hint_to_seg_type(iocb->ki_hint);
 		flag = __force_buffered_io(inode, WRITE) ?
 					F2FS_GET_BLOCK_PRE_AIO :
 					F2FS_GET_BLOCK_PRE_DIO;
@@ -1150,8 +1157,7 @@
 	return __get_data_block(inode, iblock, bh_result, create,
 						F2FS_GET_BLOCK_DEFAULT, NULL,
 						rw_hint_to_seg_type(
-							WRITE_LIFE_NOT_SET));
-						/* inode->i_write_hint)); */
+							inode->i_write_hint));
 }
 
 static int get_data_block_bmap(struct inode *inode, sector_t iblock,
@@ -2292,10 +2298,13 @@
 {
 	struct address_space *mapping = iocb->ki_filp->f_mapping;
 	struct inode *inode = mapping->host;
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	size_t count = iov_iter_count(iter);
 	loff_t offset = iocb->ki_pos;
 	int rw = iov_iter_rw(iter);
 	int err;
+	enum rw_hint hint = iocb->ki_hint;
+	int whint_mode = sbi->whint_mode;
 
 	err = check_direct_IO(inode, iter, offset);
 	if (err)
@@ -2306,11 +2315,16 @@
 
 	trace_f2fs_direct_IO_enter(inode, offset, count, rw);
 
+	if (rw == WRITE && whint_mode == WHINT_MODE_OFF)
+		iocb->ki_hint = WRITE_LIFE_NOT_SET;
+
 	down_read(&F2FS_I(inode)->dio_rwsem[rw]);
 	err = blockdev_direct_IO(iocb, inode, iter, get_data_block_dio);
 	up_read(&F2FS_I(inode)->dio_rwsem[rw]);
 
 	if (rw == WRITE) {
+		if (whint_mode == WHINT_MODE_OFF)
+			iocb->ki_hint = hint;
 		if (err > 0) {
 			f2fs_update_iostat(F2FS_I_SB(inode), APP_DIRECT_IO,
 									err);
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 1942b53..1fc80ba 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -1035,6 +1035,11 @@
 	MAX_TIME,
 };
 
+enum {
+	WHINT_MODE_OFF,		/* not pass down write hints */
+	WHINT_MODE_USER,	/* try to pass down hints given by users */
+};
+
 struct f2fs_sb_info {
 	struct super_block *sb;			/* pointer to VFS super block */
 	struct proc_dir_entry *s_proc;		/* proc entry */
@@ -1218,6 +1223,8 @@
 	char *s_qf_names[MAXQUOTAS];
 	int s_jquota_fmt;			/* Format of quota to use */
 #endif
+	/* For which write hints are passed down to block layer */
+	int whint_mode;
 };
 
 #ifdef CONFIG_F2FS_FAULT_INJECTION
@@ -2507,15 +2514,6 @@
 	return ret;
 }
 
-enum rw_hint {
-	WRITE_LIFE_NOT_SET	= 0,
-	WRITE_LIFE_NONE		= 1, /* RWH_WRITE_LIFE_NONE */
-	WRITE_LIFE_SHORT	= 2, /* RWH_WRITE_LIFE_SHORT */
-	WRITE_LIFE_MEDIUM	= 3, /* RWH_WRITE_LIFE_MEDIUM */
-	WRITE_LIFE_LONG		= 4, /* RWH_WRITE_LIFE_LONG */
-	WRITE_LIFE_EXTREME	= 5, /* RWH_WRITE_LIFE_EXTREME */
-};
-
 static inline int wbc_to_write_flags(struct writeback_control *wbc)
 {
 	if (wbc->sync_mode == WB_SYNC_ALL)
@@ -2806,6 +2804,8 @@
 int __init create_segment_manager_caches(void);
 void destroy_segment_manager_caches(void);
 int rw_hint_to_seg_type(enum rw_hint hint);
+enum rw_hint io_type_to_rw_hint(struct f2fs_sb_info *sbi, enum page_type type,
+				enum temp_type temp);
 
 /*
  * checkpoint.c
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 915984a..fbb5539 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -2456,6 +2456,62 @@
 	}
 }
 
+/* This returns write hints for each segment type. This hints will be
+ * passed down to block layer. There are mapping tables which depend on
+ * the mount option 'whint_mode'.
+ *
+ * 1) whint_mode=off. F2FS only passes down WRITE_LIFE_NOT_SET.
+ *
+ * 2) whint_mode=user-based. F2FS tries to pass down hints given by users.
+ *
+ * User                  F2FS                     Block
+ * ----                  ----                     -----
+ *                       META                     WRITE_LIFE_NOT_SET
+ *                       HOT_NODE                 "
+ *                       WARM_NODE                "
+ *                       COLD_NODE                "
+ * ioctl(COLD)           COLD_DATA                WRITE_LIFE_EXTREME
+ * extension list        "                        "
+ *
+ * -- buffered io
+ * WRITE_LIFE_EXTREME    COLD_DATA                WRITE_LIFE_EXTREME
+ * WRITE_LIFE_SHORT      HOT_DATA                 WRITE_LIFE_SHORT
+ * WRITE_LIFE_NOT_SET    WARM_DATA                WRITE_LIFE_NOT_SET
+ * WRITE_LIFE_NONE       "                        "
+ * WRITE_LIFE_MEDIUM     "                        "
+ * WRITE_LIFE_LONG       "                        "
+ *
+ * -- direct io
+ * WRITE_LIFE_EXTREME    COLD_DATA                WRITE_LIFE_EXTREME
+ * WRITE_LIFE_SHORT      HOT_DATA                 WRITE_LIFE_SHORT
+ * WRITE_LIFE_NOT_SET    WARM_DATA                WRITE_LIFE_NOT_SET
+ * WRITE_LIFE_NONE       "                        WRITE_LIFE_NONE
+ * WRITE_LIFE_MEDIUM     "                        WRITE_LIFE_MEDIUM
+ * WRITE_LIFE_LONG       "                        WRITE_LIFE_LONG
+ *
+ */
+
+enum rw_hint io_type_to_rw_hint(struct f2fs_sb_info *sbi,
+				enum page_type type, enum temp_type temp)
+{
+	if (sbi->whint_mode == WHINT_MODE_USER) {
+		if (type == DATA) {
+			switch (temp) {
+			case COLD:
+				return WRITE_LIFE_EXTREME;
+			case HOT:
+				return WRITE_LIFE_SHORT;
+			default:
+				return WRITE_LIFE_NOT_SET;
+			}
+		} else {
+			return WRITE_LIFE_NOT_SET;
+		}
+	} else {
+		return WRITE_LIFE_NOT_SET;
+	}
+}
+
 static int __get_segment_type_2(struct f2fs_io_info *fio)
 {
 	if (fio->type == DATA)
@@ -2644,6 +2700,7 @@
 	struct f2fs_io_info fio = {
 		.sbi = sbi,
 		.type = META,
+		.temp = HOT,
 		.op = REQ_OP_WRITE,
 		.op_flags = REQ_SYNC | REQ_META | REQ_PRIO,
 		.old_blkaddr = page->index,
@@ -2692,6 +2749,8 @@
 	int err;
 
 	fio->new_blkaddr = fio->old_blkaddr;
+	/* i/o temperature is needed for passing down write hints */
+	__get_segment_type(fio);
 	stat_inc_inplace_blocks(fio->sbi);
 
 	err = f2fs_submit_page_bio(fio);
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 6afb5e2..b5e6fd1 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -129,6 +129,7 @@
 	Opt_jqfmt_vfsold,
 	Opt_jqfmt_vfsv0,
 	Opt_jqfmt_vfsv1,
+	Opt_whint,
 	Opt_err,
 };
 
@@ -182,6 +183,7 @@
 	{Opt_jqfmt_vfsold, "jqfmt=vfsold"},
 	{Opt_jqfmt_vfsv0, "jqfmt=vfsv0"},
 	{Opt_jqfmt_vfsv1, "jqfmt=vfsv1"},
+	{Opt_whint, "whint_mode=%s"},
 	{Opt_err, NULL},
 };
 
@@ -679,6 +681,22 @@
 					"quota operations not supported");
 			break;
 #endif
+		case Opt_whint:
+			name = match_strdup(&args[0]);
+			if (!name)
+				return -ENOMEM;
+			if (strlen(name) == 10 &&
+					!strncmp(name, "user-based", 10)) {
+				sbi->whint_mode = WHINT_MODE_USER;
+			} else if (strlen(name) == 3 &&
+					!strncmp(name, "off", 3)) {
+				sbi->whint_mode = WHINT_MODE_OFF;
+			} else {
+				kfree(name);
+				return -EINVAL;
+			}
+			kfree(name);
+			break;
 		default:
 			f2fs_msg(sb, KERN_ERR,
 				"Unrecognized mount option \"%s\" or missing value",
@@ -722,6 +740,12 @@
 			return -EINVAL;
 		}
 	}
+
+	/* Not pass down write hints if the number of active logs is lesser
+	 * than NR_CURSEG_TYPE.
+	 */
+	if (sbi->active_logs != NR_CURSEG_TYPE)
+		sbi->whint_mode = WHINT_MODE_OFF;
 	return 0;
 }
 
@@ -1232,6 +1256,8 @@
 		seq_puts(seq, ",prjquota");
 #endif
 	f2fs_show_quota_options(seq, sbi->sb);
+	if (sbi->whint_mode == WHINT_MODE_USER)
+		seq_printf(seq, ",whint_mode=%s", "user-based");
 
 	return 0;
 }
@@ -1241,6 +1267,7 @@
 	/* init some FS parameters */
 	sbi->active_logs = NR_CURSEG_TYPE;
 	sbi->inline_xattr_size = DEFAULT_INLINE_XATTR_ADDRS;
+	sbi->whint_mode = WHINT_MODE_OFF;
 
 	set_opt(sbi, BG_GC);
 	set_opt(sbi, INLINE_XATTR);
@@ -1281,6 +1308,7 @@
 	bool need_restart_gc = false;
 	bool need_stop_gc = false;
 	bool no_extent_cache = !test_opt(sbi, EXTENT_CACHE);
+	int old_whint_mode = sbi->whint_mode;
 #ifdef CONFIG_F2FS_FAULT_INJECTION
 	struct f2fs_fault_info ffi = sbi->fault_info;
 #endif
@@ -1380,7 +1408,7 @@
 		need_stop_gc = true;
 	}
 
-	if (*flags & MS_RDONLY) {
+	if (*flags & MS_RDONLY || sbi->whint_mode != old_whint_mode) {
 		writeback_inodes_sb(sb, WB_REASON_SYNC);
 		sync_inodes_sb(sb);
 
@@ -1430,6 +1458,7 @@
 		sbi->s_qf_names[i] = s_qf_names[i];
 	}
 #endif
+	sbi->whint_mode = old_whint_mode;
 	sbi->mount_opt = org_mount_opt;
 	sbi->active_logs = active_logs;
 	sb->s_flags = old_sb_flags;
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index cd395ec..99ad610 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -25,6 +25,7 @@
 struct bio {
 	struct bio		*bi_next;	/* request queue link */
 	struct block_device	*bi_bdev;
+	unsigned short		bi_write_hint;
 	int			bi_error;
 	unsigned int		bi_opf;		/* bottom bits req flags,
 						 * top bits REQ_OP. Use
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 825e61a..d384b50 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -20,6 +20,7 @@
 #include <linux/rwsem.h>
 #include <linux/capability.h>
 #include <linux/semaphore.h>
+#include <linux/fcntl.h>
 #include <linux/fiemap.h>
 #include <linux/rculist_bl.h>
 #include <linux/atomic.h>
@@ -315,6 +316,18 @@
 struct address_space;
 struct writeback_control;
 
+/*
+ * Write life time hint values.
+ */
+enum rw_hint {
+	WRITE_LIFE_NOT_SET	= 0,
+	WRITE_LIFE_NONE		= RWH_WRITE_LIFE_NONE,
+	WRITE_LIFE_SHORT	= RWH_WRITE_LIFE_SHORT,
+	WRITE_LIFE_MEDIUM	= RWH_WRITE_LIFE_MEDIUM,
+	WRITE_LIFE_LONG		= RWH_WRITE_LIFE_LONG,
+	WRITE_LIFE_EXTREME	= RWH_WRITE_LIFE_EXTREME,
+};
+
 #define IOCB_EVENTFD		(1 << 0)
 #define IOCB_APPEND		(1 << 1)
 #define IOCB_DIRECT		(1 << 2)
@@ -329,6 +342,7 @@
 	void (*ki_complete)(struct kiocb *iocb, long ret, long ret2);
 	void			*private;
 	int			ki_flags;
+	enum rw_hint		ki_hint;
 };
 
 static inline bool is_sync_kiocb(struct kiocb *kiocb)
@@ -642,6 +656,7 @@
 	spinlock_t		i_lock;	/* i_blocks, i_bytes, maybe i_size */
 	unsigned short          i_bytes;
 	unsigned int		i_blkbits;
+	enum rw_hint		i_write_hint;
 	blkcnt_t		i_blocks;
 
 #ifdef __NEED_I_SIZE_ORDERED
@@ -1069,8 +1084,6 @@
 #define OFFT_OFFSET_MAX	INT_LIMIT(off_t)
 #endif
 
-#include <linux/fcntl.h>
-
 extern void send_sigio(struct fown_struct *fown, int fd, int band);
 
 /*
diff --git a/include/uapi/linux/fcntl.h b/include/uapi/linux/fcntl.h
index beed138..f85ed3a 100644
--- a/include/uapi/linux/fcntl.h
+++ b/include/uapi/linux/fcntl.h
@@ -43,6 +43,27 @@
 /* (1U << 31) is reserved for signed error codes */
 
 /*
+ * Set/Get write life time hints. {GET,SET}_RW_HINT operate on the
+ * underlying inode, while {GET,SET}_FILE_RW_HINT operate only on
+ * the specific file.
+ */
+#define F_GET_RW_HINT		(F_LINUX_SPECIFIC_BASE + 11)
+#define F_SET_RW_HINT		(F_LINUX_SPECIFIC_BASE + 12)
+#define F_GET_FILE_RW_HINT	(F_LINUX_SPECIFIC_BASE + 13)
+#define F_SET_FILE_RW_HINT	(F_LINUX_SPECIFIC_BASE + 14)
+
+/*
+ * Valid hint values for F_{GET,SET}_RW_HINT. 0 is "not set", or can be
+ * used to clear any hints previously set.
+ */
+#define RWF_WRITE_LIFE_NOT_SET	0
+#define RWH_WRITE_LIFE_NONE	1
+#define RWH_WRITE_LIFE_SHORT	2
+#define RWH_WRITE_LIFE_MEDIUM	3
+#define RWH_WRITE_LIFE_LONG	4
+#define RWH_WRITE_LIFE_EXTREME	5
+
+/*
  * Types of directory notifications that may be requested.
  */
 #define DN_ACCESS	0x00000001	/* File accessed */