From 1ffc8f5f7751f91fe6af527d426a723231b741a6 Mon Sep 17 00:00:00 2001 From: Jaegeuk Kim Date: Wed, 14 Jul 2021 16:14:02 -0700 Subject: f2fs: let's keep writing IOs on SBI_NEED_FSCK SBI_NEED_FSCK is an indicator that fsck.f2fs needs to be triggered, so it is not fully critical to stop any IO writes. So, let's allow to write data instead of reporting EIO forever given SBI_NEED_FSCK, but do keep OPU. Fixes: 955772787667 ("f2fs: drop inplace IO if fs status is abnormal") Cc: # v5.13+ Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/f2fs/segment.c') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 15cc89eef28d..f9b7fb785e1d 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -3563,7 +3563,7 @@ int f2fs_inplace_write_data(struct f2fs_io_info *fio) goto drop_bio; } - if (is_sbi_flag_set(sbi, SBI_NEED_FSCK) || f2fs_cp_error(sbi)) { + if (f2fs_cp_error(sbi)) { err = -EIO; goto drop_bio; } -- cgit From dc675a97129c4d9d5af55a3d7f23d7e092b8e032 Mon Sep 17 00:00:00 2001 From: Laibin Qiu Date: Sat, 31 Jul 2021 11:26:46 +0800 Subject: f2fs: fix min_seq_blocks can not make sense in some scenes. F2FS have dirty page count control for batched sequential write in writepages, and get the value of min_seq_blocks by blocks_per_seg * segs_per_sec(segs_per_sec defaults to 1). But in some scenes we set a lager section size, Min_seq_blocks will become too large to achieve the expected effect(eg. 4thread sequential write, the number of merge requests will be reduced). Signed-off-by: Laibin Qiu Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/f2fs/segment.c') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index f9b7fb785e1d..0f976cefe425 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -5159,7 +5159,7 @@ int f2fs_build_segment_manager(struct f2fs_sb_info *sbi) sm_info->ipu_policy = 1 << F2FS_IPU_FSYNC; sm_info->min_ipu_util = DEF_MIN_IPU_UTIL; sm_info->min_fsync_blocks = DEF_MIN_FSYNC_BLOCKS; - sm_info->min_seq_blocks = sbi->blocks_per_seg * sbi->segs_per_sec; + sm_info->min_seq_blocks = sbi->blocks_per_seg; sm_info->min_hot_blocks = DEF_MIN_HOT_BLOCKS; sm_info->min_ssr_sections = reserved_sections(sbi); -- cgit From 4f993264fe2965c344f223d854ccbb549b16ed71 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Tue, 3 Aug 2021 08:15:43 +0800 Subject: f2fs: introduce discard_unit mount option As James Z reported in bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=213877 [1.] One-line summary of the problem: Mount multiple SMR block devices exceed certain number cause system non-response [2.] Full description of the problem/report: Created some F2FS on SMR devices (mkfs.f2fs -m), then mounted in sequence. Each device is the same Model: HGST HSH721414AL (Size 14TB). Empirically, found that when the amount of SMR device * 1.5Gb > System RAM, the system ran out of memory and hung. No dmesg output. For example, 24 SMR Disk need 24*1.5GB = 36GB. A system with 32G RAM can only mount 21 devices, the 22nd device will be a reproducible cause of system hang. The number of SMR devices with other FS mounted on this system does not interfere with the result above. [3.] Keywords (i.e., modules, networking, kernel): F2FS, SMR, Memory [4.] Kernel information [4.1.] Kernel version (uname -a): Linux 5.13.4-200.fc34.x86_64 #1 SMP Tue Jul 20 20:27:29 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux [4.2.] Kernel .config file: Default Fedora 34 with f2fs-tools-1.14.0-2.fc34.x86_64 [5.] Most recent kernel version which did not have the bug: None [6.] Output of Oops.. message (if applicable) with symbolic information resolved (see Documentation/admin-guide/oops-tracing.rst) None [7.] A small shell script or example program which triggers the problem (if possible) mount /dev/sdX /mnt/0X [8.] Memory consumption With 24 * 14T SMR Block device with F2FS free -g total used free shared buff/cache available Mem: 46 36 0 0 10 10 Swap: 0 0 0 With 3 * 14T SMR Block device with F2FS free -g total used free shared buff/cache available Mem: 7 5 0 0 1 1 Swap: 7 0 7 The root cause is, there are three bitmaps: - cur_valid_map - ckpt_valid_map - discard_map and each of them will cost ~500MB memory, {cur, ckpt}_valid_map are necessary, but discard_map is optional, since this bitmap will only be useful in mountpoint that small discard is enabled. For a blkzoned device such as SMR or ZNS devices, f2fs will only issue discard for a section(zone) when all blocks of that section are invalid, so, for such device, we don't need small discard functionality at all. This patch introduces a new mountoption "discard_unit=block|segment| section" to support issuing discard with different basic unit which is aligned to block, segment or section, so that user can specify "discard_unit=segment" or "discard_unit=section" to disable small discard functionality. Note that this mount option can not be changed by remount() due to related metadata need to be initialized during mount(). In order to save memory, let's use "discard_unit=section" for blkzoned device by default. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 82 +++++++++++++++++++++++++++++++++++-------------------- 1 file changed, 53 insertions(+), 29 deletions(-) (limited to 'fs/f2fs/segment.c') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 0f976cefe425..80f26158e304 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -1893,7 +1893,8 @@ static int f2fs_issue_discard(struct f2fs_sb_info *sbi, se = get_seg_entry(sbi, GET_SEGNO(sbi, i)); offset = GET_BLKOFF_FROM_SEG0(sbi, i); - if (!f2fs_test_and_set_bit(offset, se->discard_map)) + if (f2fs_block_unit_discard(sbi) && + !f2fs_test_and_set_bit(offset, se->discard_map)) sbi->discard_blks--; } @@ -1918,7 +1919,8 @@ static bool add_discard_addrs(struct f2fs_sb_info *sbi, struct cp_control *cpc, struct list_head *head = &SM_I(sbi)->dcc_info->entry_list; int i; - if (se->valid_blocks == max_blocks || !f2fs_hw_support_discard(sbi)) + if (se->valid_blocks == max_blocks || !f2fs_hw_support_discard(sbi) || + !f2fs_block_unit_discard(sbi)) return false; if (!force) { @@ -2003,14 +2005,18 @@ void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi, unsigned int start = 0, end = -1; unsigned int secno, start_segno; bool force = (cpc->reason & CP_DISCARD); - bool need_align = f2fs_lfs_mode(sbi) && __is_large_section(sbi); + bool section_alignment = F2FS_OPTION(sbi).discard_unit == + DISCARD_UNIT_SECTION; + + if (f2fs_lfs_mode(sbi) && __is_large_section(sbi)) + section_alignment = true; mutex_lock(&dirty_i->seglist_lock); while (1) { int i; - if (need_align && end != -1) + if (section_alignment && end != -1) end--; start = find_next_bit(prefree_map, MAIN_SEGS(sbi), end + 1); if (start >= MAIN_SEGS(sbi)) @@ -2018,7 +2024,7 @@ void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi, end = find_next_zero_bit(prefree_map, MAIN_SEGS(sbi), start + 1); - if (need_align) { + if (section_alignment) { start = rounddown(start, sbi->segs_per_sec); end = roundup(end, sbi->segs_per_sec); } @@ -2056,6 +2062,9 @@ next: } mutex_unlock(&dirty_i->seglist_lock); + if (!f2fs_block_unit_discard(sbi)) + goto wakeup; + /* send small discards */ list_for_each_entry_safe(entry, this, head, list) { unsigned int cur_pos = 0, next_pos, len, total_len = 0; @@ -2089,6 +2098,7 @@ skip: dcc->nr_discards -= total_len; } +wakeup: wake_up_discard_thread(sbi, false); } @@ -2108,6 +2118,11 @@ static int create_discard_cmd_control(struct f2fs_sb_info *sbi) return -ENOMEM; dcc->discard_granularity = DEFAULT_DISCARD_GRANULARITY; + if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_SEGMENT) + dcc->discard_granularity = sbi->blocks_per_seg; + else if (F2FS_OPTION(sbi).discard_unit == DISCARD_UNIT_SECTION) + dcc->discard_granularity = BLKS_PER_SEC(sbi); + INIT_LIST_HEAD(&dcc->entry_list); for (i = 0; i < MAX_PLIST_NUM; i++) INIT_LIST_HEAD(&dcc->pend_list[i]); @@ -2255,7 +2270,8 @@ static void update_sit_entry(struct f2fs_sb_info *sbi, block_t blkaddr, int del) del = 0; } - if (!f2fs_test_and_set_bit(offset, se->discard_map)) + if (f2fs_block_unit_discard(sbi) && + !f2fs_test_and_set_bit(offset, se->discard_map)) sbi->discard_blks--; /* @@ -2297,7 +2313,8 @@ static void update_sit_entry(struct f2fs_sb_info *sbi, block_t blkaddr, int del) } } - if (f2fs_test_and_clear_bit(offset, se->discard_map)) + if (f2fs_block_unit_discard(sbi) && + f2fs_test_and_clear_bit(offset, se->discard_map)) sbi->discard_blks++; } if (!f2fs_test_bit(offset, se->ckpt_valid_map)) @@ -4282,6 +4299,7 @@ static int build_sit_info(struct f2fs_sb_info *sbi) unsigned int sit_segs, start; char *src_bitmap, *bitmap; unsigned int bitmap_size, main_bitmap_size, sit_bitmap_size; + unsigned int discard_map = f2fs_block_unit_discard(sbi) ? 1 : 0; /* allocate memory for SIT information */ sit_i = f2fs_kzalloc(sbi, sizeof(struct sit_info), GFP_KERNEL); @@ -4304,9 +4322,9 @@ static int build_sit_info(struct f2fs_sb_info *sbi) return -ENOMEM; #ifdef CONFIG_F2FS_CHECK_FS - bitmap_size = MAIN_SEGS(sbi) * SIT_VBLOCK_MAP_SIZE * 4; + bitmap_size = MAIN_SEGS(sbi) * SIT_VBLOCK_MAP_SIZE * (3 + discard_map); #else - bitmap_size = MAIN_SEGS(sbi) * SIT_VBLOCK_MAP_SIZE * 3; + bitmap_size = MAIN_SEGS(sbi) * SIT_VBLOCK_MAP_SIZE * (2 + discard_map); #endif sit_i->bitmap = f2fs_kvzalloc(sbi, bitmap_size, GFP_KERNEL); if (!sit_i->bitmap) @@ -4326,8 +4344,10 @@ static int build_sit_info(struct f2fs_sb_info *sbi) bitmap += SIT_VBLOCK_MAP_SIZE; #endif - sit_i->sentries[start].discard_map = bitmap; - bitmap += SIT_VBLOCK_MAP_SIZE; + if (discard_map) { + sit_i->sentries[start].discard_map = bitmap; + bitmap += SIT_VBLOCK_MAP_SIZE; + } } sit_i->tmp_map = f2fs_kzalloc(sbi, SIT_VBLOCK_MAP_SIZE, GFP_KERNEL); @@ -4489,17 +4509,19 @@ static int build_sit_entries(struct f2fs_sb_info *sbi) if (IS_NODESEG(se->type)) total_node_blocks += se->valid_blocks; - /* build discard map only one time */ - if (is_set_ckpt_flags(sbi, CP_TRIMMED_FLAG)) { - memset(se->discard_map, 0xff, - SIT_VBLOCK_MAP_SIZE); - } else { - memcpy(se->discard_map, - se->cur_valid_map, - SIT_VBLOCK_MAP_SIZE); - sbi->discard_blks += - sbi->blocks_per_seg - - se->valid_blocks; + if (f2fs_block_unit_discard(sbi)) { + /* build discard map only one time */ + if (is_set_ckpt_flags(sbi, CP_TRIMMED_FLAG)) { + memset(se->discard_map, 0xff, + SIT_VBLOCK_MAP_SIZE); + } else { + memcpy(se->discard_map, + se->cur_valid_map, + SIT_VBLOCK_MAP_SIZE); + sbi->discard_blks += + sbi->blocks_per_seg - + se->valid_blocks; + } } if (__is_large_section(sbi)) @@ -4535,13 +4557,15 @@ static int build_sit_entries(struct f2fs_sb_info *sbi) if (IS_NODESEG(se->type)) total_node_blocks += se->valid_blocks; - if (is_set_ckpt_flags(sbi, CP_TRIMMED_FLAG)) { - memset(se->discard_map, 0xff, SIT_VBLOCK_MAP_SIZE); - } else { - memcpy(se->discard_map, se->cur_valid_map, - SIT_VBLOCK_MAP_SIZE); - sbi->discard_blks += old_valid_blocks; - sbi->discard_blks -= se->valid_blocks; + if (f2fs_block_unit_discard(sbi)) { + if (is_set_ckpt_flags(sbi, CP_TRIMMED_FLAG)) { + memset(se->discard_map, 0xff, SIT_VBLOCK_MAP_SIZE); + } else { + memcpy(se->discard_map, se->cur_valid_map, + SIT_VBLOCK_MAP_SIZE); + sbi->discard_blks += old_valid_blocks; + sbi->discard_blks -= se->valid_blocks; + } } if (__is_large_section(sbi)) { -- cgit From 91803392c732c43b5cf440e885ea89be7f5fecef Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Wed, 4 Aug 2021 08:38:38 +0800 Subject: f2fs: fix to stop filesystem update once CP failed During f2fs_write_checkpoint(), once we failed in f2fs_flush_nat_entries() or do_checkpoint(), metadata of filesystem such as prefree bitmap, nat/sit version bitmap won't be recovered, it may cause f2fs image to be inconsistent, let's just set CP error flag to avoid further updates until we figure out a scheme to rollback all metadatas in such condition. Reported-by: Yangtao Li Signed-off-by: Yangtao Li Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) (limited to 'fs/f2fs/segment.c') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 80f26158e304..ca9876a6d396 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -776,11 +776,22 @@ int f2fs_flush_device_cache(struct f2fs_sb_info *sbi) return 0; for (i = 1; i < sbi->s_ndevs; i++) { + int count = DEFAULT_RETRY_IO_COUNT; + if (!f2fs_test_bit(i, (char *)&sbi->dirty_device)) continue; - ret = __submit_flush_wait(sbi, FDEV(i).bdev); - if (ret) + + do { + ret = __submit_flush_wait(sbi, FDEV(i).bdev); + if (ret) + congestion_wait(BLK_RW_ASYNC, + DEFAULT_IO_TIMEOUT); + } while (ret && --count); + + if (ret) { + f2fs_stop_checkpoint(sbi, false); break; + } spin_lock(&sbi->dev_lock); f2fs_clear_bit(i, (char *)&sbi->dirty_device); -- cgit From 324105775c1982aacbd2972b7024d8cc06474abe Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Mon, 9 Aug 2021 08:24:48 +0800 Subject: f2fs: support fault injection for f2fs_kmem_cache_alloc() This patch supports to inject fault into f2fs_kmem_cache_alloc(). Usage: a) echo 32768 > /sys/fs/f2fs//inject_type or b) mount -o fault_type=32768 Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'fs/f2fs/segment.c') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index ca9876a6d396..b4dd22134a73 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -188,7 +188,8 @@ void f2fs_register_inmem_page(struct inode *inode, struct page *page) set_page_private_atomic(page); - new = f2fs_kmem_cache_alloc(inmem_entry_slab, GFP_NOFS); + new = f2fs_kmem_cache_alloc(inmem_entry_slab, + GFP_NOFS, true, NULL); /* add atomic page indices to the list */ new->page = page; @@ -1001,7 +1002,7 @@ static struct discard_cmd *__create_discard_cmd(struct f2fs_sb_info *sbi, pend_list = &dcc->pend_list[plist_idx(len)]; - dc = f2fs_kmem_cache_alloc(discard_cmd_slab, GFP_NOFS); + dc = f2fs_kmem_cache_alloc(discard_cmd_slab, GFP_NOFS, true, NULL); INIT_LIST_HEAD(&dc->list); dc->bdev = bdev; dc->lstart = lstart; @@ -1962,7 +1963,7 @@ static bool add_discard_addrs(struct f2fs_sb_info *sbi, struct cp_control *cpc, if (!de) { de = f2fs_kmem_cache_alloc(discard_entry_slab, - GFP_F2FS_ZERO); + GFP_F2FS_ZERO, true, NULL); de->start_blkaddr = START_BLOCK(sbi, cpc->trim_start); list_add_tail(&de->list, head); } @@ -4099,7 +4100,8 @@ static struct page *get_next_sit_page(struct f2fs_sb_info *sbi, static struct sit_entry_set *grab_sit_entry_set(void) { struct sit_entry_set *ses = - f2fs_kmem_cache_alloc(sit_entry_set_slab, GFP_NOFS); + f2fs_kmem_cache_alloc(sit_entry_set_slab, + GFP_NOFS, true, NULL); ses->entry_cnt = 0; INIT_LIST_HEAD(&ses->set_list); -- cgit From 521187439abfb3e1c946796dc2187c443e5457ab Mon Sep 17 00:00:00 2001 From: Daeho Jeong Date: Thu, 19 Aug 2021 20:52:28 -0700 Subject: f2fs: separate out iostat feature Added F2FS_IOSTAT config option to support getting IO statistics through sysfs and printing out periodic IO statistics tracepoint events and moved I/O statistics related codes into separate files for better maintenance. Signed-off-by: Daeho Jeong Reviewed-by: Chao Yu [Jaegeuk Kim: set default=y] Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 1 + 1 file changed, 1 insertion(+) (limited to 'fs/f2fs/segment.c') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index b4dd22134a73..73abec9988e9 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -20,6 +20,7 @@ #include "segment.h" #include "node.h" #include "gc.h" +#include "iostat.h" #include #define __reverse_ffz(x) __reverse_ffs(~(x)) -- cgit From 4d67490498acb4ffcef5ba7bc44990d46e66a44c Mon Sep 17 00:00:00 2001 From: Fengnan Chang Date: Thu, 19 Aug 2021 16:02:37 +0800 Subject: f2fs: Don't create discard thread when device doesn't support realtime discard Don't create discard thread when device doesn't support realtime discard or user specifies nodiscard mount option. Signed-off-by: Fengnan Chang Signed-off-by: Yangtao Li Reviewed-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/segment.c | 25 +++++++++++++++++++------ 1 file changed, 19 insertions(+), 6 deletions(-) (limited to 'fs/f2fs/segment.c') diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index 73abec9988e9..a135d2247415 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -2115,9 +2115,25 @@ wakeup: wake_up_discard_thread(sbi, false); } -static int create_discard_cmd_control(struct f2fs_sb_info *sbi) +int f2fs_start_discard_thread(struct f2fs_sb_info *sbi) { dev_t dev = sbi->sb->s_bdev->bd_dev; + struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info; + int err = 0; + + if (!f2fs_realtime_discard_enable(sbi)) + return 0; + + dcc->f2fs_issue_discard = kthread_run(issue_discard_thread, sbi, + "f2fs_discard-%u:%u", MAJOR(dev), MINOR(dev)); + if (IS_ERR(dcc->f2fs_issue_discard)) + err = PTR_ERR(dcc->f2fs_issue_discard); + + return err; +} + +static int create_discard_cmd_control(struct f2fs_sb_info *sbi) +{ struct discard_cmd_control *dcc; int err = 0, i; @@ -2155,13 +2171,10 @@ static int create_discard_cmd_control(struct f2fs_sb_info *sbi) init_waitqueue_head(&dcc->discard_wait_queue); SM_I(sbi)->dcc_info = dcc; init_thread: - dcc->f2fs_issue_discard = kthread_run(issue_discard_thread, sbi, - "f2fs_discard-%u:%u", MAJOR(dev), MINOR(dev)); - if (IS_ERR(dcc->f2fs_issue_discard)) { - err = PTR_ERR(dcc->f2fs_issue_discard); + err = f2fs_start_discard_thread(sbi); + if (err) { kfree(dcc); SM_I(sbi)->dcc_info = NULL; - return err; } return err; -- cgit