"value":"In the Linux kernel, the following vulnerability has been resolved:\n\nnilfs2: do not force clear folio if buffer is referenced\n\nPatch series \"nilfs2: protect busy buffer heads from being force-cleared\".\n\nThis series fixes the buffer head state inconsistency issues reported by\nsyzbot that occurs when the filesystem is corrupted and falls back to\nread-only, and the associated buffer head use-after-free issue.\n\n\nThis patch (of 2):\n\nSyzbot has reported that after nilfs2 detects filesystem corruption and\nfalls back to read-only, inconsistencies in the buffer state may occur.\n\nOne of the inconsistencies is that when nilfs2 calls mark_buffer_dirty()\nto set a data or metadata buffer as dirty, but it detects that the buffer\nis not in the uptodate state:\n\n WARNING: CPU: 0 PID: 6049 at fs/buffer.c:1177 mark_buffer_dirty+0x2e5/0x520\n fs/buffer.c:1177\n ...\n Call Trace:\n <TASK>\n nilfs_palloc_commit_alloc_entry+0x4b/0x160 fs/nilfs2/alloc.c:598\n nilfs_ifile_create_inode+0x1dd/0x3a0 fs/nilfs2/ifile.c:73\n nilfs_new_inode+0x254/0x830 fs/nilfs2/inode.c:344\n nilfs_mkdir+0x10d/0x340 fs/nilfs2/namei.c:218\n vfs_mkdir+0x2f9/0x4f0 fs/namei.c:4257\n do_mkdirat+0x264/0x3a0 fs/namei.c:4280\n __do_sys_mkdirat fs/namei.c:4295 [inline]\n __se_sys_mkdirat fs/namei.c:4293 [inline]\n __x64_sys_mkdirat+0x87/0xa0 fs/namei.c:4293\n do_syscall_x64 arch/x86/entry/common.c:52 [inline]\n do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83\n entry_SYSCALL_64_after_hwframe+0x77/0x7f\n\nThe other is when nilfs_btree_propagate(), which propagates the dirty\nstate to the ancestor nodes of a b-tree that point to a dirty buffer,\ndetects that the origin buffer is not dirty, even though it should be:\n\n WARNING: CPU: 0 PID: 5245 at fs/nilfs2/btree.c:2089\n nilfs_btree_propagate+0xc79/0xdf0 fs/nilfs2/btree.c:2089\n ...\n Call Trace:\n <TASK>\n nilfs_bmap_propagate+0x75/0x120 fs/nilfs2/bmap.c:345\n nilfs_collect_file_data+0x4d/0xd0 fs/nilfs2/segment.c:587\n nilfs_segctor_apply_buffers+0x184/0x340 fs/nilfs2/segment.c:1006\n nilfs_segctor_scan_file+0x28c/0xa50 fs/nilfs2/segment.c:1045\n nilfs_segctor_collect_blocks fs/nilfs2/segment.c:1216 [inline]\n nilfs_segctor_collect fs/nilfs2/segment.c:1540 [inline]\n nilfs_segctor_do_construct+0x1c28/0x6b90 fs/nilfs2/segment.c:2115\n nilfs_segctor_construct+0x181/0x6b0 fs/nilfs2/segment.c:2479\n nilfs_segctor_thread_construct fs/nilfs2/segment.c:2587 [inline]\n nilfs_segctor_thread+0x69e/0xe80 fs/nilfs2/segment.c:2701\n kthread+0x2f0/0x390 kernel/kthread.c:389\n ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147\n ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244\n </TASK>\n\nBoth of these issues are caused by the callbacks that handle the\npage/folio write requests, forcibly clear various states, including the\nworking state of the buffers they hold, at unexpected times when they\ndetect read-only fallback.\n\nFix these issues by checking if the buffer is referenced before clearing\nthe page/folio state, and skipping the clear if it is."
