diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2018-06-11 10:16:13 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2018-06-11 10:16:13 -0700 |
commit | d54d35c501bcbd57b9722a6b371c0608b5d34199 (patch) | |
tree | 3dcbc5b272a7c18634a238f7503af452cbed1fd8 /fs/f2fs/gc.c | |
parent | a2225d931f75ddd3c39f4d0d195fad99dfd68671 (diff) | |
parent | dfa742803fbbd8b82790c152600e09c5ad99caca (diff) |
Merge tag 'f2fs-for-4.18' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs
Pull f2fs updates from Jaegeuk Kim:
"In this round, we've mainly focused on discard, aka unmap, control
along with fstrim for Android-specific usage model. In addition, we've
fixed writepage flow which returned EAGAIN previously resulting in EIO
of fsync(2) due to mapping's error state. In order to avoid old MM bug
[1], we decided not to use __GFP_ZERO for the mapping for node and
meta page caches. As always, we've cleaned up many places for future
fsverity and symbol conflicts.
Enhancements:
- do discard/fstrim in lower priority considering fs utilization
- split large discard commands into smaller ones for better responsiveness
- add more sanity checks to address syzbot reports
- add a mount option, fsync_mode=nobarrier, which can reduce # of cache flushes
- clean up symbol namespace with modified function names
- be strict on block allocation and IO control in corner cases
Bug fixes:
- don't use __GFP_ZERO for mappings
- fix error reports in writepage to avoid fsync() failure
- avoid selinux denial on CAP_RESOURCE on resgid/resuid
- fix some subtle race conditions in GC/atomic writes/shutdown
- fix overflow bugs in sanity_check_raw_super
- fix missing bits on get_flags
Clean-ups:
- prepare the generic flow for future fsverity integration
- fix some broken coding standard"
[1] https://lkml.org/lkml/2018/4/8/661
* tag 'f2fs-for-4.18' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (79 commits)
f2fs: fix to clear FI_VOLATILE_FILE correctly
f2fs: let sync node IO interrupt async one
f2fs: don't change wbc->sync_mode
f2fs: fix to update mtime correctly
fs: f2fs: insert space around that ':' and ', '
fs: f2fs: add missing blank lines after declarations
fs: f2fs: changed variable type of offset "unsigned" to "loff_t"
f2fs: clean up symbol namespace
f2fs: make set_de_type() static
f2fs: make __f2fs_write_data_pages() static
f2fs: fix to avoid accessing cross the boundary
f2fs: fix to let caller retry allocating block address
disable loading f2fs module on PAGE_SIZE > 4KB
f2fs: fix error path of move_data_page
f2fs: don't drop dentry pages after fs shutdown
f2fs: fix to avoid race during access gc_thread pointer
f2fs: clean up with clear_radix_tree_dirty_tag
f2fs: fix to don't trigger writeback during recovery
f2fs: clear discard_wake earlier
f2fs: let discard thread wait a little longer if dev is busy
...
Diffstat (limited to 'fs/f2fs/gc.c')
-rw-r--r-- | fs/f2fs/gc.c | 183 |
1 files changed, 101 insertions, 82 deletions
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index 9327411fd93b..9093be6e7a7d 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -76,7 +76,7 @@ static int gc_thread_func(void *data) * invalidated soon after by user update or deletion. * So, I'd like to wait some time to collect dirty segments. */ - if (gc_th->gc_urgent) { + if (sbi->gc_mode == GC_URGENT) { wait_ms = gc_th->urgent_sleep_time; mutex_lock(&sbi->gc_mutex); goto do_gc; @@ -114,7 +114,7 @@ next: return 0; } -int start_gc_thread(struct f2fs_sb_info *sbi) +int f2fs_start_gc_thread(struct f2fs_sb_info *sbi) { struct f2fs_gc_kthread *gc_th; dev_t dev = sbi->sb->s_bdev->bd_dev; @@ -131,8 +131,6 @@ int start_gc_thread(struct f2fs_sb_info *sbi) gc_th->max_sleep_time = DEF_GC_THREAD_MAX_SLEEP_TIME; gc_th->no_gc_sleep_time = DEF_GC_THREAD_NOGC_SLEEP_TIME; - gc_th->gc_idle = 0; - gc_th->gc_urgent = 0; gc_th->gc_wake= 0; sbi->gc_thread = gc_th; @@ -148,7 +146,7 @@ out: return err; } -void stop_gc_thread(struct f2fs_sb_info *sbi) +void f2fs_stop_gc_thread(struct f2fs_sb_info *sbi) { struct f2fs_gc_kthread *gc_th = sbi->gc_thread; if (!gc_th) @@ -158,21 +156,19 @@ void stop_gc_thread(struct f2fs_sb_info *sbi) sbi->gc_thread = NULL; } -static int select_gc_type(struct f2fs_gc_kthread *gc_th, int gc_type) +static int select_gc_type(struct f2fs_sb_info *sbi, int gc_type) { int gc_mode = (gc_type == BG_GC) ? GC_CB : GC_GREEDY; - if (!gc_th) - return gc_mode; - - if (gc_th->gc_idle) { - if (gc_th->gc_idle == 1) - gc_mode = GC_CB; - else if (gc_th->gc_idle == 2) - gc_mode = GC_GREEDY; - } - if (gc_th->gc_urgent) + switch (sbi->gc_mode) { + case GC_IDLE_CB: + gc_mode = GC_CB; + break; + case GC_IDLE_GREEDY: + case GC_URGENT: gc_mode = GC_GREEDY; + break; + } return gc_mode; } @@ -187,7 +183,7 @@ static void select_policy(struct f2fs_sb_info *sbi, int gc_type, p->max_search = dirty_i->nr_dirty[type]; p->ofs_unit = 1; } else { - p->gc_mode = select_gc_type(sbi->gc_thread, gc_type); + p->gc_mode = select_gc_type(sbi, gc_type); p->dirty_segmap = dirty_i->dirty_segmap[DIRTY]; p->max_search = dirty_i->nr_dirty[DIRTY]; p->ofs_unit = sbi->segs_per_sec; @@ -195,7 +191,7 @@ static void select_policy(struct f2fs_sb_info *sbi, int gc_type, /* we need to check every dirty segments in the FG_GC case */ if (gc_type != FG_GC && - (sbi->gc_thread && !sbi->gc_thread->gc_urgent) && + (sbi->gc_mode != GC_URGENT) && p->max_search > sbi->max_victim_search) p->max_search = sbi->max_victim_search; @@ -234,10 +230,6 @@ static unsigned int check_bg_victims(struct f2fs_sb_info *sbi) for_each_set_bit(secno, dirty_i->victim_secmap, MAIN_SECS(sbi)) { if (sec_usage_check(sbi, secno)) continue; - - if (no_fggc_candidate(sbi, secno)) - continue; - clear_bit(secno, dirty_i->victim_secmap); return GET_SEG_FROM_SEC(sbi, secno); } @@ -377,9 +369,6 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi, goto next; if (gc_type == BG_GC && test_bit(secno, dirty_i->victim_secmap)) goto next; - if (gc_type == FG_GC && p.alloc_mode == LFS && - no_fggc_candidate(sbi, secno)) - goto next; cost = get_gc_cost(sbi, segno, &p); @@ -440,7 +429,7 @@ static void add_gc_inode(struct gc_inode_list *gc_list, struct inode *inode) iput(inode); return; } - new_ie = f2fs_kmem_cache_alloc(inode_entry_slab, GFP_NOFS); + new_ie = f2fs_kmem_cache_alloc(f2fs_inode_entry_slab, GFP_NOFS); new_ie->inode = inode; f2fs_radix_tree_insert(&gc_list->iroot, inode->i_ino, new_ie); @@ -454,7 +443,7 @@ static void put_gc_inode(struct gc_inode_list *gc_list) radix_tree_delete(&gc_list->iroot, ie->inode->i_ino); iput(ie->inode); list_del(&ie->list); - kmem_cache_free(inode_entry_slab, ie); + kmem_cache_free(f2fs_inode_entry_slab, ie); } } @@ -484,12 +473,16 @@ static void gc_node_segment(struct f2fs_sb_info *sbi, block_t start_addr; int off; int phase = 0; + bool fggc = (gc_type == FG_GC); start_addr = START_BLOCK(sbi, segno); next_step: entry = sum; + if (fggc && phase == 2) + atomic_inc(&sbi->wb_sync_req[NODE]); + for (off = 0; off < sbi->blocks_per_seg; off++, entry++) { nid_t nid = le32_to_cpu(entry->nid); struct page *node_page; @@ -503,39 +496,42 @@ next_step: continue; if (phase == 0) { - ra_meta_pages(sbi, NAT_BLOCK_OFFSET(nid), 1, + f2fs_ra_meta_pages(sbi, NAT_BLOCK_OFFSET(nid), 1, META_NAT, true); continue; } if (phase == 1) { - ra_node_page(sbi, nid); + f2fs_ra_node_page(sbi, nid); continue; } /* phase == 2 */ - node_page = get_node_page(sbi, nid); + node_page = f2fs_get_node_page(sbi, nid); if (IS_ERR(node_page)) continue; - /* block may become invalid during get_node_page */ + /* block may become invalid during f2fs_get_node_page */ if (check_valid_map(sbi, segno, off) == 0) { f2fs_put_page(node_page, 1); continue; } - get_node_info(sbi, nid, &ni); + f2fs_get_node_info(sbi, nid, &ni); if (ni.blk_addr != start_addr + off) { f2fs_put_page(node_page, 1); continue; } - move_node_page(node_page, gc_type); + f2fs_move_node_page(node_page, gc_type); stat_inc_node_blk_count(sbi, 1, gc_type); } if (++phase < 3) goto next_step; + + if (fggc) + atomic_dec(&sbi->wb_sync_req[NODE]); } /* @@ -545,7 +541,7 @@ next_step: * as indirect or double indirect node blocks, are given, it must be a caller's * bug. */ -block_t start_bidx_of_node(unsigned int node_ofs, struct inode *inode) +block_t f2fs_start_bidx_of_node(unsigned int node_ofs, struct inode *inode) { unsigned int indirect_blks = 2 * NIDS_PER_BLOCK + 4; unsigned int bidx; @@ -576,11 +572,11 @@ static bool is_alive(struct f2fs_sb_info *sbi, struct f2fs_summary *sum, nid = le32_to_cpu(sum->nid); ofs_in_node = le16_to_cpu(sum->ofs_in_node); - node_page = get_node_page(sbi, nid); + node_page = f2fs_get_node_page(sbi, nid); if (IS_ERR(node_page)) return false; - get_node_info(sbi, nid, dni); + f2fs_get_node_info(sbi, nid, dni); if (sum->version != dni->version) { f2fs_msg(sbi->sb, KERN_WARNING, @@ -603,7 +599,7 @@ static bool is_alive(struct f2fs_sb_info *sbi, struct f2fs_summary *sum, * This can be used to move blocks, aka LBAs, directly on disk. */ static void move_data_block(struct inode *inode, block_t bidx, - unsigned int segno, int off) + int gc_type, unsigned int segno, int off) { struct f2fs_io_info fio = { .sbi = F2FS_I_SB(inode), @@ -614,6 +610,7 @@ static void move_data_block(struct inode *inode, block_t bidx, .op_flags = 0, .encrypted_page = NULL, .in_list = false, + .retry = false, }; struct dnode_of_data dn; struct f2fs_summary sum; @@ -621,6 +618,7 @@ static void move_data_block(struct inode *inode, block_t bidx, struct page *page; block_t newaddr; int err; + bool lfs_mode = test_opt(fio.sbi, LFS); /* do not read out */ page = f2fs_grab_cache_page(inode->i_mapping, bidx, false); @@ -630,8 +628,11 @@ static void move_data_block(struct inode *inode, block_t bidx, if (!check_valid_map(F2FS_I_SB(inode), segno, off)) goto out; - if (f2fs_is_atomic_file(inode)) + if (f2fs_is_atomic_file(inode)) { + F2FS_I(inode)->i_gc_failures[GC_FAILURE_ATOMIC]++; + F2FS_I_SB(inode)->skipped_atomic_files[gc_type]++; goto out; + } if (f2fs_is_pinned_file(inode)) { f2fs_pin_file_control(inode, true); @@ -639,7 +640,7 @@ static void move_data_block(struct inode *inode, block_t bidx, } set_new_dnode(&dn, inode, NULL, NULL, 0); - err = get_dnode_of_data(&dn, bidx, LOOKUP_NODE); + err = f2fs_get_dnode_of_data(&dn, bidx, LOOKUP_NODE); if (err) goto out; @@ -654,14 +655,17 @@ static void move_data_block(struct inode *inode, block_t bidx, */ f2fs_wait_on_page_writeback(page, DATA, true); - get_node_info(fio.sbi, dn.nid, &ni); + f2fs_get_node_info(fio.sbi, dn.nid, &ni); set_summary(&sum, dn.nid, dn.ofs_in_node, ni.version); /* read page */ fio.page = page; fio.new_blkaddr = fio.old_blkaddr = dn.data_blkaddr; - allocate_data_block(fio.sbi, NULL, fio.old_blkaddr, &newaddr, + if (lfs_mode) + down_write(&fio.sbi->io_order_lock); + + f2fs_allocate_data_block(fio.sbi, NULL, fio.old_blkaddr, &newaddr, &sum, CURSEG_COLD_DATA, NULL, false); fio.encrypted_page = f2fs_pagecache_get_page(META_MAPPING(fio.sbi), @@ -693,6 +697,7 @@ static void move_data_block(struct inode *inode, block_t bidx, dec_page_count(fio.sbi, F2FS_DIRTY_META); set_page_writeback(fio.encrypted_page); + ClearPageError(page); /* allocate block address */ f2fs_wait_on_page_writeback(dn.node_page, NODE, true); @@ -700,8 +705,8 @@ static void move_data_block(struct inode *inode, block_t bidx, fio.op = REQ_OP_WRITE; fio.op_flags = REQ_SYNC; fio.new_blkaddr = newaddr; - err = f2fs_submit_page_write(&fio); - if (err) { + f2fs_submit_page_write(&fio); + if (fio.retry) { if (PageWriteback(fio.encrypted_page)) end_page_writeback(fio.encrypted_page); goto put_page_out; @@ -716,8 +721,10 @@ static void move_data_block(struct inode *inode, block_t bidx, put_page_out: f2fs_put_page(fio.encrypted_page, 1); recover_block: + if (lfs_mode) + up_write(&fio.sbi->io_order_lock); if (err) - __f2fs_replace_block(fio.sbi, &sum, newaddr, fio.old_blkaddr, + f2fs_do_replace_block(fio.sbi, &sum, newaddr, fio.old_blkaddr, true, true); put_out: f2fs_put_dnode(&dn); @@ -730,15 +737,18 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type, { struct page *page; - page = get_lock_data_page(inode, bidx, true); + page = f2fs_get_lock_data_page(inode, bidx, true); if (IS_ERR(page)) return; if (!check_valid_map(F2FS_I_SB(inode), segno, off)) goto out; - if (f2fs_is_atomic_file(inode)) + if (f2fs_is_atomic_file(inode)) { + F2FS_I(inode)->i_gc_failures[GC_FAILURE_ATOMIC]++; + F2FS_I_SB(inode)->skipped_atomic_files[gc_type]++; goto out; + } if (f2fs_is_pinned_file(inode)) { if (gc_type == FG_GC) f2fs_pin_file_control(inode, true); @@ -772,15 +782,20 @@ retry: f2fs_wait_on_page_writeback(page, DATA, true); if (clear_page_dirty_for_io(page)) { inode_dec_dirty_pages(inode); - remove_dirty_inode(inode); + f2fs_remove_dirty_inode(inode); } set_cold_data(page); - err = do_write_data_page(&fio); - if (err == -ENOMEM && is_dirty) { - congestion_wait(BLK_RW_ASYNC, HZ/50); - goto retry; + err = f2fs_do_write_data_page(&fio); + if (err) { + clear_cold_data(page); + if (err == -ENOMEM) { + congestion_wait(BLK_RW_ASYNC, HZ/50); + goto retry; + } + if (is_dirty) + set_page_dirty(page); } } out: @@ -824,13 +839,13 @@ next_step: continue; if (phase == 0) { - ra_meta_pages(sbi, NAT_BLOCK_OFFSET(nid), 1, + f2fs_ra_meta_pages(sbi, NAT_BLOCK_OFFSET(nid), 1, META_NAT, true); continue; } if (phase == 1) { - ra_node_page(sbi, nid); + f2fs_ra_node_page(sbi, nid); continue; } @@ -839,7 +854,7 @@ next_step: continue; if (phase == 2) { - ra_node_page(sbi, dni.ino); + f2fs_ra_node_page(sbi, dni.ino); continue; } @@ -850,23 +865,23 @@ next_step: if (IS_ERR(inode) || is_bad_inode(inode)) continue; - /* if encrypted inode, let's go phase 3 */ - if (f2fs_encrypted_file(inode)) { + /* if inode uses special I/O path, let's go phase 3 */ + if (f2fs_post_read_required(inode)) { add_gc_inode(gc_list, inode); continue; } if (!down_write_trylock( - &F2FS_I(inode)->dio_rwsem[WRITE])) { + &F2FS_I(inode)->i_gc_rwsem[WRITE])) { iput(inode); continue; } - start_bidx = start_bidx_of_node(nofs, inode); - data_page = get_read_data_page(inode, + start_bidx = f2fs_start_bidx_of_node(nofs, inode); + data_page = f2fs_get_read_data_page(inode, start_bidx + ofs_in_node, REQ_RAHEAD, true); - up_write(&F2FS_I(inode)->dio_rwsem[WRITE]); + up_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); if (IS_ERR(data_page)) { iput(inode); continue; @@ -884,11 +899,11 @@ next_step: bool locked = false; if (S_ISREG(inode->i_mode)) { - if (!down_write_trylock(&fi->dio_rwsem[READ])) + if (!down_write_trylock(&fi->i_gc_rwsem[READ])) continue; if (!down_write_trylock( - &fi->dio_rwsem[WRITE])) { - up_write(&fi->dio_rwsem[READ]); + &fi->i_gc_rwsem[WRITE])) { + up_write(&fi->i_gc_rwsem[READ]); continue; } locked = true; @@ -897,17 +912,18 @@ next_step: inode_dio_wait(inode); } - start_bidx = start_bidx_of_node(nofs, inode) + start_bidx = f2fs_start_bidx_of_node(nofs, inode) + ofs_in_node; - if (f2fs_encrypted_file(inode)) - move_data_block(inode, start_bidx, segno, off); + if (f2fs_post_read_required(inode)) + move_data_block(inode, start_bidx, gc_type, + segno, off); else move_data_page(inode, start_bidx, gc_type, segno, off); if (locked) { - up_write(&fi->dio_rwsem[WRITE]); - up_write(&fi->dio_rwsem[READ]); + up_write(&fi->i_gc_rwsem[WRITE]); + up_write(&fi->i_gc_rwsem[READ]); } stat_inc_data_blk_count(sbi, 1, gc_type); @@ -946,12 +962,12 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi, /* readahead multi ssa blocks those have contiguous address */ if (sbi->segs_per_sec > 1) - ra_meta_pages(sbi, GET_SUM_BLOCK(sbi, segno), + f2fs_ra_meta_pages(sbi, GET_SUM_BLOCK(sbi, segno), sbi->segs_per_sec, META_SSA, true); /* reference all summary page */ while (segno < end_segno) { - sum_page = get_sum_page(sbi, segno++); + sum_page = f2fs_get_sum_page(sbi, segno++); unlock_page(sum_page); } @@ -1017,6 +1033,8 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync, .ilist = LIST_HEAD_INIT(gc_list.ilist), .iroot = RADIX_TREE_INIT(gc_list.iroot, GFP_NOFS), }; + unsigned long long last_skipped = sbi->skipped_atomic_files[FG_GC]; + unsigned int skipped_round = 0, round = 0; trace_f2fs_gc_begin(sbi->sb, sync, background, get_pages(sbi, F2FS_DIRTY_NODES), @@ -1045,7 +1063,7 @@ gc_more: * secure free segments which doesn't need fggc any more. */ if (prefree_segments(sbi)) { - ret = write_checkpoint(sbi, &cpc); + ret = f2fs_write_checkpoint(sbi, &cpc); if (ret) goto stop; } @@ -1068,17 +1086,27 @@ gc_more: sec_freed++; total_freed += seg_freed; + if (gc_type == FG_GC) { + if (sbi->skipped_atomic_files[FG_GC] > last_skipped) + skipped_round++; + last_skipped = sbi->skipped_atomic_files[FG_GC]; + round++; + } + if (gc_type == FG_GC) sbi->cur_victim_sec = NULL_SEGNO; if (!sync) { if (has_not_enough_free_secs(sbi, sec_freed, 0)) { + if (skipped_round > MAX_SKIP_ATOMIC_COUNT && + skipped_round * 2 >= round) + f2fs_drop_inmem_pages_all(sbi, true); segno = NULL_SEGNO; goto gc_more; } if (gc_type == FG_GC) - ret = write_checkpoint(sbi, &cpc); + ret = f2fs_write_checkpoint(sbi, &cpc); } stop: SIT_I(sbi)->last_victim[ALLOC_NEXT] = 0; @@ -1102,19 +1130,10 @@ stop: return ret; } -void build_gc_manager(struct f2fs_sb_info *sbi) +void f2fs_build_gc_manager(struct f2fs_sb_info *sbi) { - u64 main_count, resv_count, ovp_count; - DIRTY_I(sbi)->v_ops = &default_v_ops; - /* threshold of # of valid blocks in a section for victims of FG_GC */ - main_count = SM_I(sbi)->main_segments << sbi->log_blocks_per_seg; - resv_count = SM_I(sbi)->reserved_segments << sbi->log_blocks_per_seg; - ovp_count = SM_I(sbi)->ovp_segments << sbi->log_blocks_per_seg; - - sbi->fggc_threshold = div64_u64((main_count - ovp_count) * - BLKS_PER_SEC(sbi), (main_count - resv_count)); sbi->gc_pin_file_threshold = DEF_GC_FAILED_PINNED_FILES; /* give warm/cold data area from slower device */ |