From bb06664a534ba4833373f19ae018efff91ab2908 Mon Sep 17 00:00:00 2001 From: Chao Yu Date: Fri, 3 Nov 2017 10:21:05 +0800 Subject: [PATCH] f2fs: avoid race in between GC and block exchange During block exchange in {insert,collapse,move}_range, page-block mapping is unstable due to mapping moving or recovery, so there should be no concurrent cache read operation rely on such mapping, nor cache write operation to mess up block exchange. So this patch let background GC be aware of that. Signed-off-by: Chao Yu Signed-off-by: Jaegeuk Kim --- fs/f2fs/file.c | 26 +++++++++++++++++++++----- fs/f2fs/gc.c | 7 +++++++ 2 files changed, 28 insertions(+), 5 deletions(-) diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index ee12bd42af13..b4cc860fdd01 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -1189,11 +1189,14 @@ static int f2fs_collapse_range(struct inode *inode, loff_t offset, loff_t len) if (ret) goto out; + /* avoid gc operation during block exchange */ + down_write(&F2FS_I(inode)->dio_rwsem[WRITE]); + truncate_pagecache(inode, offset); ret = f2fs_do_collapse(inode, pg_start, pg_end); if (ret) - goto out; + goto out_unlock; /* write out all moved pages, if possible */ filemap_write_and_wait_range(inode->i_mapping, offset, LLONG_MAX); @@ -1205,7 +1208,8 @@ static int f2fs_collapse_range(struct inode *inode, loff_t offset, loff_t len) ret = truncate_blocks(inode, new_size, true); if (!ret) f2fs_i_size_write(inode, new_size); - +out_unlock: + up_write(&F2FS_I(inode)->dio_rwsem[WRITE]); out: up_write(&F2FS_I(inode)->i_mmap_sem); return ret; @@ -1388,6 +1392,9 @@ static int f2fs_insert_range(struct inode *inode, loff_t offset, loff_t len) if (ret) goto out; + /* avoid gc operation during block exchange */ + down_write(&F2FS_I(inode)->dio_rwsem[WRITE]); + truncate_pagecache(inode, offset); pg_start = offset >> PAGE_SHIFT; @@ -1415,6 +1422,8 @@ static int f2fs_insert_range(struct inode *inode, loff_t offset, loff_t len) if (!ret) f2fs_i_size_write(inode, new_size); + + up_write(&F2FS_I(inode)->dio_rwsem[WRITE]); out: up_write(&F2FS_I(inode)->i_mmap_sem); return ret; @@ -2277,9 +2286,13 @@ static int f2fs_move_file_range(struct file *file_in, loff_t pos_in, } inode_lock(src); + down_write(&F2FS_I(src)->dio_rwsem[WRITE]); if (src != dst) { - if (!inode_trylock(dst)) { - ret = -EBUSY; + ret = -EBUSY; + if (!inode_trylock(dst)) + goto out; + if (!down_write_trylock(&F2FS_I(dst)->dio_rwsem[WRITE])) { + inode_unlock(dst); goto out; } } @@ -2339,9 +2352,12 @@ static int f2fs_move_file_range(struct file *file_in, loff_t pos_in, } f2fs_unlock_op(sbi); out_unlock: - if (src != dst) + if (src != dst) { + up_write(&F2FS_I(dst)->dio_rwsem[WRITE]); inode_unlock(dst); + } out: + up_write(&F2FS_I(src)->dio_rwsem[WRITE]); inode_unlock(src); return ret; } diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index ff8f0012888d..5d5bba462f26 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -832,10 +832,17 @@ static void gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum, continue; } + if (!down_write_trylock( + &F2FS_I(inode)->dio_rwsem[WRITE])) { + iput(inode); + continue; + } + start_bidx = start_bidx_of_node(nofs, inode); data_page = get_read_data_page(inode, start_bidx + ofs_in_node, REQ_RAHEAD, true); + up_write(&F2FS_I(inode)->dio_rwsem[WRITE]); if (IS_ERR(data_page)) { iput(inode); continue;