CVE-2024-46733

In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> btrfs: fix qgroup reserve leaks in cow_file_range<br /> <br /> In the buffered write path, the dirty page owns the qgroup reserve until<br /> it creates an ordered_extent.<br /> <br /> Therefore, any errors that occur before the ordered_extent is created<br /> must free that reservation, or else the space is leaked. The fstest<br /> generic/475 exercises various IO error paths, and is able to trigger<br /> errors in cow_file_range where we fail to get to allocating the ordered<br /> extent. Note that because we *do* clear delalloc, we are likely to<br /> remove the inode from the delalloc list, so the inodes/pages to not have<br /> invalidate/launder called on them in the commit abort path.<br /> <br /> This results in failures at the unmount stage of the test that look like:<br /> <br /> BTRFS: error (device dm-8 state EA) in cleanup_transaction:2018: errno=-5 IO failure<br /> BTRFS: error (device dm-8 state EA) in btrfs_replace_file_extents:2416: errno=-5 IO failure<br /> BTRFS warning (device dm-8 state EA): qgroup 0/5 has unreleased space, type 0 rsv 28672<br /> ------------[ cut here ]------------<br /> WARNING: CPU: 3 PID: 22588 at fs/btrfs/disk-io.c:4333 close_ctree+0x222/0x4d0 [btrfs]<br /> Modules linked in: btrfs blake2b_generic libcrc32c xor zstd_compress raid6_pq<br /> CPU: 3 PID: 22588 Comm: umount Kdump: loaded Tainted: G W 6.10.0-rc7-gab56fde445b8 #21<br /> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Arch Linux 1.16.3-1-1 04/01/2014<br /> RIP: 0010:close_ctree+0x222/0x4d0 [btrfs]<br /> RSP: 0018:ffffb4465283be00 EFLAGS: 00010202<br /> RAX: 0000000000000001 RBX: ffffa1a1818e1000 RCX: 0000000000000001<br /> RDX: 0000000000000000 RSI: ffffb4465283bbe0 RDI: ffffa1a19374fcb8<br /> RBP: ffffa1a1818e13c0 R08: 0000000100028b16 R09: 0000000000000000<br /> R10: 0000000000000003 R11: 0000000000000003 R12: ffffa1a18ad7972c<br /> R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000<br /> FS: 00007f9168312b80(0000) GS:ffffa1a4afcc0000(0000) knlGS:0000000000000000<br /> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033<br /> CR2: 00007f91683c9140 CR3: 000000010acaa000 CR4: 00000000000006f0<br /> Call Trace:<br /> <br /> ? close_ctree+0x222/0x4d0 [btrfs]<br /> ? __warn.cold+0x8e/0xea<br /> ? close_ctree+0x222/0x4d0 [btrfs]<br /> ? report_bug+0xff/0x140<br /> ? handle_bug+0x3b/0x70<br /> ? exc_invalid_op+0x17/0x70<br /> ? asm_exc_invalid_op+0x1a/0x20<br /> ? close_ctree+0x222/0x4d0 [btrfs]<br /> generic_shutdown_super+0x70/0x160<br /> kill_anon_super+0x11/0x40<br /> btrfs_kill_super+0x11/0x20 [btrfs]<br /> deactivate_locked_super+0x2e/0xa0<br /> cleanup_mnt+0xb5/0x150<br /> task_work_run+0x57/0x80<br /> syscall_exit_to_user_mode+0x121/0x130<br /> do_syscall_64+0xab/0x1a0<br /> entry_SYSCALL_64_after_hwframe+0x77/0x7f<br /> RIP: 0033:0x7f916847a887<br /> ---[ end trace 0000000000000000 ]---<br /> BTRFS error (device dm-8 state EA): qgroup reserved space leaked<br /> <br /> Cases 2 and 3 in the out_reserve path both pertain to this type of leak<br /> and must free the reserved qgroup data. Because it is already an error<br /> path, I opted not to handle the possible errors in<br /> btrfs_free_qgroup_data.