On 29.06.21 г. 16:59, Josef Bacik wrote: > We're going to remove sync_inode, so migrate to filemap_fdatawrite_wbc > instead. > > Signed-off-by: Josef Bacik <josef@xxxxxxxxxxxxxx> > --- > fs/9p/vfs_file.c | 7 +------ > 1 file changed, 1 insertion(+), 6 deletions(-) > > diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c > index 59c32c9b799f..6b64e8391f30 100644 > --- a/fs/9p/vfs_file.c > +++ b/fs/9p/vfs_file.c > @@ -625,12 +625,7 @@ static void v9fs_mmap_vm_close(struct vm_area_struct *vma) > p9_debug(P9_DEBUG_VFS, "9p VMA close, %p, flushing", vma); > > inode = file_inode(vma->vm_file); > - > - if (!mapping_can_writeback(inode->i_mapping)) > - wbc.nr_to_write = 0; > - > - might_sleep(); Not a 9p expert but we are losing the might_sleep check and do_writepages can sleep due to cond_resched or congestion_wait > - sync_inode(inode, &wbc); > + filemap_fdatawrite_wbc(inode->i_mapping, &wbc); > } > > >
