Re: A null-ptr-deref bug in f2fs_write_end_io

From: butt3rflyh4ck
Date: Sun May 21 2023 - 23:58:34 EST

Next message: Miaohe Lin: "[PATCH] sched/deadline: simplify dl_bw_cpus() using cpumask_weight_and()"
Previous message: Huacai Chen: "Re: [PATCH] LoongArch: Add support to clone a time namespace"
In reply to: Chao Yu: "Re: A null-ptr-deref bug in f2fs_write_end_io"
Next in thread: Chao Yu: "Re: A null-ptr-deref bug in f2fs_write_end_io"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

OK, the attachment is a reproducer.

Regards.
butt3rflyh4ck.

On Fri, May 19, 2023 at 11:24 PM Chao Yu <chao@xxxxxxxxxx> wrote:
>
> Hi,
>
> Thanks for the report, it will be helpful if you can provide a
> reproducer.
>
> On 2023/5/19 15:13, butt3rflyh4ck wrote:
> > Hi, there is a null-ptr-deref bug in f2fs_write_end_io in
> > fs/f2fs/data.c, I reproduce it in the latest kernel too.
> >
> > #Quick description
> > When a thread always calls F2FS_IOC_RESIZE_FS to resize fs, if resize
> > fs is failed, f2fs kernel thread would invoke callback function to
> > update
> > f2fs io info, it would call f2fs_write_end_io and may trigger
> > null-ptr-deref in NODE_MAPPING.
> > ```
> > static inline struct address_space *NODE_MAPPING(struct f2fs_sb_info *sbi)
> > {
> > return sbi->node_inode->i_mapping;
> > }
> > ```
> > there is deref in sbi.
> >
> > #crash log
> > ----------------------------------------
> > general protection fault, probably for non-canonical address
> > 0xdffffc0000000006: 0000 [#1] PREEMPT SMP KASAN
> > KASAN: null-ptr-deref in range [0x0000000000000030-0x0000000000000037]
> > CPU: 0 PID: 17 Comm: ksoftirqd/0 Not tainted 6.4.0-rc1 #18
> > Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
> > RIP: 0010:NODE_MAPPING fs/f2fs/f2fs.h:1972 [inline]
> > RIP: 0010:f2fs_write_end_io+0x727/0x1050 fs/f2fs/data.c:370
> > Code: 00 00 48 89 f8 48 c1 e8 03 80 3c 18 00 0f 85 b3 07 00 00 48 8b
> > 44 24 08 4c 8b a8 60 01 00 00 49 8d 7d 30 48 89 f8 48 c1 e8 03 <80> 3c
> > 18 00 0f 85 9c 07 00 00 4d 3b 75 30 0f 84 10 04 00 00 e8 10
> > RSP: 0018:ffffc9000042fc78 EFLAGS: 00010216
> > RAX: 0000000000000006 RBX: dffffc0000000000 RCX: 0000000000000100
> > RDX: ffff888013d18000 RSI: ffffffff83a93a4d RDI: 0000000000000030
> > RBP: ffffea00009e6900 R08: 0000000000000001 R09: 0000000000000000
> > R10: 0000000000000000 R11: 0000000000000000 R12: ffffea00009e6900
> > R13: 0000000000000000 R14: ffff88802a3cec48 R15: 0000000000000000
> > FS: 0000000000000000(0000) GS:ffff88802ca00000(0000) knlGS:0000000000000000
> > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > CR2: 0000555faf194000 CR3: 00000000251bd000 CR4: 0000000000750ef0
> > PKRU: 55555554
> > Call Trace:
> > <TASK>
> > bio_endio+0x5af/0x6c0 block/bio.c:1608
> > req_bio_endio block/blk-mq.c:761 [inline]
> > blk_update_request+0x5cc/0x1690 block/blk-mq.c:906
> > blk_mq_end_request+0x59/0x4c0 block/blk-mq.c:1023
> > lo_complete_rq+0x1c6/0x280 drivers/block/loop.c:370
> > blk_complete_reqs+0xad/0xe0 block/blk-mq.c:1101
> > __do_softirq+0x1d4/0x8ef kernel/softirq.c:571
> > run_ksoftirqd kernel/softirq.c:939 [inline]
> > run_ksoftirqd+0x31/0x60 kernel/softirq.c:931
> > smpboot_thread_fn+0x659/0x9e0 kernel/smpboot.c:164
> > kthread+0x33e/0x440 kernel/kthread.c:379
> > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:308
> > </TASK>
> > Modules linked in:
> > ---[ end trace 0000000000000000 ]---
> > RIP: 0010:NODE_MAPPING fs/f2fs/f2fs.h:1972 [inline]
> > RIP: 0010:f2fs_write_end_io+0x727/0x1050 fs/f2fs/data.c:370
> > Code: 00 00 48 89 f8 48 c1 e8 03 80 3c 18 00 0f 85 b3 07 00 00 48 8b
> > 44 24 08 4c 8b a8 60 01 00 00 49 8d 7d 30 48 89 f8 48 c1 e8 03 <80> 3c
> > 18 00 0f 85 9c 07 00 00 4d 3b 75 30 0f 84 10 04 00 00 e8 10
> > RSP: 0018:ffffc9000042fc78 EFLAGS: 00010216
> > RAX: 0000000000000006 RBX: dffffc0000000000 RCX: 0000000000000100
> > RDX: ffff888013d18000 RSI: ffffffff83a93a4d RDI: 0000000000000030
> > RBP: ffffea00009e6900 R08: 0000000000000001 R09: 0000000000000000
> > R10: 0000000000000000 R11: 0000000000000000 R12: ffffea00009e6900
> > R13: 0000000000000000 R14: ffff88802a3cec48 R15: 0000000000000000
> > FS: 0000000000000000(0000) GS:ffff88802ca00000(0000) knlGS:0000000000000000
> > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > CR2: 0000555faf194000 CR3: 00000000251bd000 CR4: 0000000000750ef0
> > PKRU: 55555554
> > -----------------------
> >
> > # new crash log in latest kernel
> > ---------------------
> > [ 193.695164][ T8174] loop0: detected capacity change from 0 to 264192
> > [ 193.696467][ T8174] F2FS-fs (loop0): Magic Mismatch,
> > valid(0xf2f52010) - read(0x0)
> > [ 193.696875][ T8174] F2FS-fs (loop0): Can't find valid F2FS
> > filesystem in 2th superblock
> > [ 193.698363][ T8174] F2FS-fs (loop0): invalid crc_offset: 0
> > [ 193.700454][ T8174] F2FS-fs (loop0): Disable nat_bits due to
> > incorrect cp_ver (4542359912962316977, 0)
> > [ 193.716594][ T8174] F2FS-fs (loop0): Try to recover 2th superblock, ret: 0
> > [ 193.717102][ T8174] F2FS-fs (loop0): Mounted with checkpoint
> > version = 3e17dab1
> > [ 193.743330][ T8174] F2FS-fs (loop0): For resize: curseg of type 0: 46 ==> 4
> > [ 193.743904][ T8174] F2FS-fs (loop0): For resize: curseg of type 3: 52 ==> 6
> > [ 193.745690][ T8174] F2FS-fs (loop0): For resize: curseg of type 4: 50 ==> 8
> > [ 193.746108][ T8174] F2FS-fs (loop0): For resize: curseg of type 5: 48 ==> 10
> > [ 193.751857][ T8174] F2FS-fs (loop0): resize_fs failed, should run
> > fsck to repair!
> > [ 193.780283][ C0] general protection fault, probably for
> > non-canonical address 0xdffffc0000000006: 0000 [#1] PREEMPT SMP KASAN
> > [ 193.781027][ C0] KASAN: null-ptr-deref in range
> > [0x0000000000000030-0x0000000000000037]
> > [ 193.781572][ C0] CPU: 0 PID: 17 Comm: ksoftirqd/0 Not tainted
> > 6.4.0-rc2-00163-g2d1bcbc6cd70-dirty #17
> > [ 193.782201][ C0] Hardware name: QEMU Standard PC (i440FX + PIIX,
> > 1996), BIOS 1.15.0-1 04/01/2014
> > [ 193.782727][ C0] RIP: 0010:f2fs_write_end_io+0x727/0x1050
> > [ 193.783083][ C0] Code: 00 00 48 89 f8 48 c1 e8 03 80 3c 18 00 0f
> > 85 b3 07 00 00 48 8b 44 24 08 4c 8b a8 60 01 00 00 49 8d 7d 30 48 89
> > f8 48 c1 e8 03 <80> 3c 18 00 0f 85 9c 07 00 00 4d 3b 75 30 0f 0
> > [ 193.784268][ C0] RSP: 0018:ffffc9000042fc78 EFLAGS: 00010216
> > [ 193.784629][ C0] RAX: 0000000000000006 RBX: dffffc0000000000
> > RCX: 0000000000000100
> > [ 193.785109][ C0] RDX: ffff888013d18000 RSI: ffffffff83a9588d
> > RDI: 0000000000000030
> > [ 193.785576][ C0] RBP: ffffea000143e740 R08: 0000000000000001
> > R09: 0000000000000000
> > [ 193.786051][ C0] R10: 0000000000000000 R11: 0000000000000000
> > R12: ffffea000143e740
> > [ 193.786571][ C0] R13: 0000000000000000 R14: ffff888041423738
> > R15: 0000000000000000
> > [ 193.787055][ C0] FS: 0000000000000000(0000)
> > GS:ffff88802ca00000(0000) knlGS:0000000000000000
> > [ 193.787620][ C0] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > [ 193.787997][ C0] CR2: 000056056fd87000 CR3: 000000001b546000
> > CR4: 00000000000006f0
> > [ 193.788585][ C0] Call Trace:
> > [ 193.788863][ C0] <TASK>
> > [ 193.789115][ C0] ? bio_uninit+0x1b7/0x410
> > [ 193.789509][ C0] ? f2fs_write_end+0xa80/0xa80
> > [ 193.790053][ C0] bio_endio+0x5af/0x6c0
> > [ 193.790522][ C0] blk_update_request+0x5cc/0x1690
> > [ 193.791171][ C0] blk_mq_end_request+0x59/0x4c0
> > [ 193.791695][ C0] lo_complete_rq+0x1c6/0x280
> > [ 193.792247][ C0] blk_complete_reqs+0xad/0xe0
> > [ 193.792759][ C0] __do_softirq+0x1d4/0x8ef
> > [ 193.793312][ C0] ? __irq_exit_rcu+0x190/0x190
> > [ 193.793805][ C0] run_ksoftirqd+0x31/0x60
> > [ 193.794183][ C0] smpboot_thread_fn+0x659/0x9e0
> > [ 193.794576][ C0] ? sort_range+0x30/0x30
> > [ 193.794900][ C0] kthread+0x33e/0x440
> > [ 193.795263][ C0] ? kthread_complete_and_exit+0x40/0x40
> > [ 193.795907][ C0] ret_from_fork+0x1f/0x30
> > [ 193.796324][ C0] </TASK>
> > [ 193.796689][ C0] Modules linked in:
> > [ 193.797189][ C0] ---[ end trace 0000000000000000 ]---
> > [ 193.797635][ C0] RIP: 0010:f2fs_write_end_io+0x727/0x1050
> > [ 193.798182][ C0] Code: 00 00 48 89 f8 48 c1 e8 03 80 3c 18 00 0f
> > 85 b3 07 00 00 48 8b 44 24 08 4c 8b a8 60 01 00 00 49 8d 7d 30 48 89
> > f8 48 c1 e8 03 <80> 3c 18 00 0f 85 9c 07 00 00 4d 3b 75 30 0f 0
> > [ 193.799559][ C0] RSP: 0018:ffffc9000042fc78 EFLAGS: 00010216
> > [ 193.799945][ C0] RAX: 0000000000000006 RBX: dffffc0000000000
> > RCX: 0000000000000100
> > [ 193.800329][ C0] RDX: ffff888013d18000 RSI: ffffffff83a9588d
> > RDI: 0000000000000030
> > [ 193.800666][ C0] RBP: ffffea000143e740 R08: 0000000000000001
> > R09: 0000000000000000
> > [ 193.801047][ C0] R10: 0000000000000000 R11: 0000000000000000
> > R12: ffffea000143e740
> > [ 193.801503][ C0] R13: 0000000000000000 R14: ffff888041423738
> > R15: 0000000000000000
> > [ 193.802069][ C0] FS: 0000000000000000(0000)
> > GS:ffff88802ca00000(0000) knlGS:0000000000000000
> > [ 193.802804][ C0] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > [ 193.803356][ C0] CR2: 000056056fd87000 CR3: 000000001b546000
> > CR4: 00000000000006f0
> > [ 193.804046][ C0] Kernel panic - not syncing: Fatal exception in interrupt
> > [ 193.804748][ C0] Kernel Offset: disabled
> > [ 193.805086][ C0] Rebooting in 86400 seconds..
> > ----------------------------
> >
> > If needed, I would provide reproduce.
> >
> > Regards,
> > butt3rflyh4ck.
> >
> >
> >

--
Active Defense Lab of Venustech

Attachment: repro.cprog
Description: Binary data

Next message: Miaohe Lin: "[PATCH] sched/deadline: simplify dl_bw_cpus() using cpumask_weight_and()"
Previous message: Huacai Chen: "Re: [PATCH] LoongArch: Add support to clone a time namespace"
In reply to: Chao Yu: "Re: A null-ptr-deref bug in f2fs_write_end_io"
Next in thread: Chao Yu: "Re: A null-ptr-deref bug in f2fs_write_end_io"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]