f2fs: refactor ->page_mkwrite() flow

Thread A				Thread B
- f2fs_vm_page_mkwrite
					- f2fs_setattr
					 - down_write(i_mmap_sem)
					 - truncate_setsize
					 - f2fs_truncate
					 - up_write(i_mmap_sem)
 - f2fs_reserve_block
 reserve NEW_ADDR
 - skip dirty page due to truncation

1. we don't need to rserve new block address for a truncated page.
2. dn.data_blkaddr is used out of node page lock coverage.

Refactor ->page_mkwrite() flow to fix above issues:
- use __do_map_lock() to avoid racing checkpoint()
- lock data page in prior to dnode page
- cover f2fs_reserve_block with i_mmap_sem lock
- wait page writeback before zeroing page

Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index ea16cadd416e4b0a2431ea617f26d98a03083242..3f01bc2d73eb95a755c6c99665deee985efc5e98 100644 (file)
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -972,7 +972,7 @@ map_blocks:
        return err;
 }
 
-static inline void __do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock)
+void __do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock)
 {
        if (flag == F2FS_GET_BLOCK_PRE_AIO) {
                if (lock)

diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index f633662bf6459b47a32dedb6975d2c2463fa1211..05217aad4c0b37ba0b6a41faf508eedc27e1d385 100644 (file)
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -3055,6 +3055,7 @@ struct page *f2fs_get_lock_data_page(struct inode *inode, pgoff_t index,
 struct page *f2fs_get_new_data_page(struct inode *inode,
                        struct page *ipage, pgoff_t index, bool new_i_size);
 int f2fs_do_write_data_page(struct f2fs_io_info *fio);
+void __do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock);
 int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
                        int create, int flag);
 int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,

diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 357422a4c3197cf6e0d5c01fa86f2926c51ab938..6db808c808135c44db4417728637f4d92b901ba7 100644 (file)
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -47,7 +47,7 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf)
        struct page *page = vmf->page;
        struct inode *inode = file_inode(vmf->vma->vm_file);
        struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-       struct dnode_of_data dn;
+       struct dnode_of_data dn = { .node_changed = false };
        int err;
 
        if (unlikely(f2fs_cp_error(sbi))) {
@@ -59,19 +59,6 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf)
 
        f2fs_bug_on(sbi, f2fs_has_inline_data(inode));
 
-       /* block allocation */
-       f2fs_lock_op(sbi);
-       set_new_dnode(&dn, inode, NULL, NULL, 0);
-       err = f2fs_reserve_block(&dn, page->index);
-       if (err) {
-               f2fs_unlock_op(sbi);
-               goto out;
-       }
-       f2fs_put_dnode(&dn);
-       f2fs_unlock_op(sbi);
-
-       f2fs_balance_fs(sbi, dn.node_changed);
-
        file_update_time(vmf->vma->vm_file);
        down_read(&F2FS_I(inode)->i_mmap_sem);
        lock_page(page);
@@ -83,11 +70,28 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf)
                goto out_sem;
        }
 
+       /* block allocation */
+       __do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true);
+       set_new_dnode(&dn, inode, NULL, NULL, 0);
+       err = f2fs_get_block(&dn, page->index);
+       f2fs_put_dnode(&dn);
+       __do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false);
+       if (err) {
+               unlock_page(page);
+               goto out_sem;
+       }
+
+       /* fill the page */
+       f2fs_wait_on_page_writeback(page, DATA, false);
+
+       /* wait for GCed page writeback via META_MAPPING */
+       f2fs_wait_on_block_writeback(inode, dn.data_blkaddr);
+
        /*
         * check to see if the page is mapped already (no holes)
         */
        if (PageMappedToDisk(page))
-               goto mapped;
+               goto out_sem;
 
        /* page is wholly or partially inside EOF */
        if (((loff_t)(page->index + 1) << PAGE_SHIFT) >
@@ -104,16 +108,11 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf)
        f2fs_update_iostat(sbi, APP_MAPPED_IO, F2FS_BLKSIZE);
 
        trace_f2fs_vm_page_mkwrite(page, DATA);
-mapped:
-       /* fill the page */
-       f2fs_wait_on_page_writeback(page, DATA, false);
-
-       /* wait for GCed page writeback via META_MAPPING */
-       f2fs_wait_on_block_writeback(inode, dn.data_blkaddr);
-
 out_sem:
        up_read(&F2FS_I(inode)->i_mmap_sem);
-out:
+
+       f2fs_balance_fs(sbi, dn.node_changed);
+
        sb_end_pagefault(inode->i_sb);
        f2fs_update_time(sbi, REQ_TIME);
 err: