Message-ID: <d5a1fcfa-24b6-5c18-c899-1de19292d3bb@huawei.com>
Date:   Mon, 14 Aug 2023 11:22:37 +0800
From:   Baokun Li <libaokun1@...wei.com>
To:     Yikebaer Aizezi <yikebaer61@...il.com>, <adilger.kernel@...ger.ca>,
        <linux-ext4@...r.kernel.org>, <tytso@....edu>
CC:     <linux-kernel@...r.kernel.org>, Baokun Li <libaokun1@...wei.com>
Subject: Re: KASAN: slab-use-after-free Read in ext4_es_insert_extent

Hello!

On 2023/8/14 9:59, Yikebaer Aizezi wrote:
> Hello,
>
> When using Healer to fuzz the Linux-6.5-rc5,  the following crash
> was triggered.
>
> HEAD commit: 52a93d39b17dc7eb98b6aa3edb93943248e03b2f (tag: v6.5-rc5)
> git tree: upstream
>
> console output:
> https://drive.google.com/file/d/1yKtvQct90Q7xY09N28iIwqAUSjq2KQPs/view?usp=drive_link
> kernel config:https://drive.google.com/file/d/1hClF9kiDlmdnocuMCe1WZezKlhuOCq9A/view?usp=drive_link
> C reproducer:https://drive.google.com/file/d/1yfIE42YP4YKIeJ3VxJTRMLZn3b83cs8A/view?usp=drive_link
> Syzlang reproducer:https://drive.google.com/file/d/1afZPMtWGcZMvSR8AfleA-lDn_bj-aWm1/view?usp=drive_link
>
>
> If you fix this issue, please add the following tag to the commit:
> Reported-by: Yikebaer Aizezi <yikebaer61@...il.com>
>
>
> FAULT_INJECTION: forcing a failure.
> name failslab, interval 1, probability 0, space 0, times 0
> CPU: 1 PID: 8438 Comm: syz-executor Not tainted 6.5.0-rc5 #1
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
> rel-1.12.0-59-gc9ba5276e321-prebuilt.qemu.org 04/01/2014
> Call Trace:
>   <TASK>
>   __dump_stack lib/dump_stack.c:88 [inline]
>   dump_stack_lvl+0x132/0x150 lib/dump_stack.c:106
>   fail_dump lib/fault-inject.c:52 [inline]
>   should_fail_ex+0x49f/0x5b0 lib/fault-inject.c:153
>   should_failslab+0x5/0x10 mm/slab_common.c:1471
>   slab_pre_alloc_hook mm/slab.h:711 [inline]
>   slab_alloc_node mm/slub.c:3452 [inline]
>   slab_alloc mm/slub.c:3478 [inline]
>   __kmem_cache_alloc_lru mm/slub.c:3485 [inline]
>   kmem_cache_alloc+0x5e/0x390 mm/slub.c:3494
>   __es_alloc_extent fs/ext4/extents_status.c:467 [inline]
>   __es_alloc_extent fs/ext4/extents_status.c:464 [inline]
>   __es_insert_extent+0xde9/0x1440 fs/ext4/extents_status.c:815
>   __es_remove_extent+0x73b/0x16f0 fs/ext4/extents_status.c:1383
>   ext4_es_insert_extent+0x2a1/0xcb0 fs/ext4/extents_status.c:878
>   ext4_map_blocks+0x92a/0x16f0 fs/ext4/inode.c:680
>   ext4_alloc_file_blocks.isra.0+0x2df/0xb70 fs/ext4/extents.c:4462
>   ext4_zero_range fs/ext4/extents.c:4622 [inline]
>   ext4_fallocate+0x251c/0x3ce0 fs/ext4/extents.c:4721
>   vfs_fallocate+0x492/0xed0 fs/open.c:324
>   ksys_fallocate fs/open.c:347 [inline]
>   __do_sys_fallocate fs/open.c:355 [inline]
>   __se_sys_fallocate fs/open.c:353 [inline]
>   __x64_sys_fallocate+0xce/0x140 fs/open.c:353
>   do_syscall_x64 arch/x86/entry/common.c:50 [inline]
>   do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
>   entry_SYSCALL_64_after_hwframe+0x63/0xcd
> RIP: 0033:0x47959d
> Code: 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa 48 89 f8 48
> 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08
> RSP: 002b:00007fbdfe383068 EFLAGS: 00000246 ORIG_RAX: 000000000000011d
> RAX: ffffffffffffffda RBX: 000000000059c0a0 RCX: 000000000047959d
> RDX: 000000000000000f RSI: 0000000000000010 RDI: 0000000000000003
> RBP: 0000000000000001 R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000008000 R11: 0000000000000246 R12: 000000000059c0ac
> R13: 000000000000000b R14: 0000000000437250 R15: 00007fbdfe363000
>   </TASK>
> ==================================================================
> BUG: KASAN: slab-use-after-free in ext4_es_insert_extent+0xc68/0xcb0
> fs/ext4/extents_status.c:894
> Read of size 4 at addr ffff888112ecc1a4 by task syz-executor/8438
>
> CPU: 1 PID: 8438 Comm: syz-executor Not tainted 6.5.0-rc5 #1
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
> rel-1.12.0-59-gc9ba5276e321-prebuilt.qemu.org 04/01/2014
> Call Trace:
>   <TASK>
>   __dump_stack lib/dump_stack.c:88 [inline]
>   dump_stack_lvl+0xd5/0x150 lib/dump_stack.c:106
>   print_address_description mm/kasan/report.c:364 [inline]
>   print_report+0xc1/0x5e0 mm/kasan/report.c:475
>   kasan_report+0xba/0xf0 mm/kasan/report.c:588
>   ext4_es_insert_extent+0xc68/0xcb0 fs/ext4/extents_status.c:894
>   ext4_map_blocks+0x92a/0x16f0 fs/ext4/inode.c:680
>   ext4_alloc_file_blocks.isra.0+0x2df/0xb70 fs/ext4/extents.c:4462
>   ext4_zero_range fs/ext4/extents.c:4622 [inline]
>   ext4_fallocate+0x251c/0x3ce0 fs/ext4/extents.c:4721
>   vfs_fallocate+0x492/0xed0 fs/open.c:324
>   ksys_fallocate fs/open.c:347 [inline]
>   __do_sys_fallocate fs/open.c:355 [inline]
>   __se_sys_fallocate fs/open.c:353 [inline]
>   __x64_sys_fallocate+0xce/0x140 fs/open.c:353
>   do_syscall_x64 arch/x86/entry/common.c:50 [inline]
>   do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
>   entry_SYSCALL_64_after_hwframe+0x63/0xcd
> RIP: 0033:0x47959d
> Code: 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa 48 89 f8 48
> 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08
> RSP: 002b:00007fbdfe383068 EFLAGS: 00000246 ORIG_RAX: 000000000000011d
> RAX: ffffffffffffffda RBX: 000000000059c0a0 RCX: 000000000047959d
> RDX: 000000000000000f RSI: 0000000000000010 RDI: 0000000000000003
> RBP: 0000000000000001 R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000008000 R11: 0000000000000246 R12: 000000000059c0ac
> R13: 000000000000000b R14: 0000000000437250 R15: 00007fbdfe363000
>   </TASK>
>
> Allocated by task 8438:
>   kasan_save_stack+0x1e/0x40 mm/kasan/common.c:45
>   kasan_set_track+0x21/0x30 mm/kasan/common.c:52
>   __kasan_slab_alloc+0x7b/0x80 mm/kasan/common.c:328
>   kasan_slab_alloc include/linux/kasan.h:186 [inline]
>   slab_post_alloc_hook mm/slab.h:762 [inline]
>   slab_alloc_node mm/slub.c:3470 [inline]
>   slab_alloc mm/slub.c:3478 [inline]
>   __kmem_cache_alloc_lru mm/slub.c:3485 [inline]
>   kmem_cache_alloc+0x16b/0x390 mm/slub.c:3494
>   kmem_cache_zalloc include/linux/slab.h:693 [inline]
>   __es_alloc_extent fs/ext4/extents_status.c:469 [inline]
>   ext4_es_insert_extent+0x672/0xcb0 fs/ext4/extents_status.c:873
>   ext4_map_blocks+0x92a/0x16f0 fs/ext4/inode.c:680
>   ext4_alloc_file_blocks.isra.0+0x2df/0xb70 fs/ext4/extents.c:4462
>   ext4_zero_range fs/ext4/extents.c:4622 [inline]
>   ext4_fallocate+0x251c/0x3ce0 fs/ext4/extents.c:4721
>   vfs_fallocate+0x492/0xed0 fs/open.c:324
>   ksys_fallocate fs/open.c:347 [inline]
>   __do_sys_fallocate fs/open.c:355 [inline]
>   __se_sys_fallocate fs/open.c:353 [inline]
>   __x64_sys_fallocate+0xce/0x140 fs/open.c:353
>   do_syscall_x64 arch/x86/entry/common.c:50 [inline]
>   do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
>   entry_SYSCALL_64_after_hwframe+0x63/0xcd
>
> Freed by task 8438:
>   kasan_save_stack+0x1e/0x40 mm/kasan/common.c:45
>   kasan_set_track+0x21/0x30 mm/kasan/common.c:52
>   kasan_save_free_info+0x27/0x40 mm/kasan/generic.c:522
>   ____kasan_slab_free mm/kasan/common.c:236 [inline]
>   ____kasan_slab_free+0x161/0x1c0 mm/kasan/common.c:200
>   kasan_slab_free include/linux/kasan.h:162 [inline]
>   slab_free_hook mm/slub.c:1792 [inline]
>   slab_free_freelist_hook+0x89/0x1c0 mm/slub.c:1818
>   slab_free mm/slub.c:3801 [inline]
>   kmem_cache_free+0xec/0x490 mm/slub.c:3823
>   ext4_es_try_to_merge_right fs/ext4/extents_status.c:593 [inline]
>   __es_insert_extent+0x9f4/0x1440 fs/ext4/extents_status.c:802
>   ext4_es_insert_extent+0x2ca/0xcb0 fs/ext4/extents_status.c:882
>   ext4_map_blocks+0x92a/0x16f0 fs/ext4/inode.c:680
>   ext4_alloc_file_blocks.isra.0+0x2df/0xb70 fs/ext4/extents.c:4462
>   ext4_zero_range fs/ext4/extents.c:4622 [inline]
>   ext4_fallocate+0x251c/0x3ce0 fs/ext4/extents.c:4721
>   vfs_fallocate+0x492/0xed0 fs/open.c:324
>   ksys_fallocate fs/open.c:347 [inline]
>   __do_sys_fallocate fs/open.c:355 [inline]
>   __se_sys_fallocate fs/open.c:353 [inline]
>   __x64_sys_fallocate+0xce/0x140 fs/open.c:353
>   do_syscall_x64 arch/x86/entry/common.c:50 [inline]
>   do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
>   entry_SYSCALL_64_after_hwframe+0x63/0xcd
>
> The buggy address belongs to the object at ffff888112ecc188
>   which belongs to the cache extent_status of size 40
> The buggy address is located 28 bytes inside of
>   freed 40-byte region [ffff888112ecc188, ffff888112ecc1b0)
>
> The buggy address belongs to the physical page:
> page:ffffea00044bb300 refcount:1 mapcount:0 mapping:0000000000000000
> index:0x0 pfn:0x112ecc
> flags: 0x57ff00000000200(slab|node=1|zone=2|lastcpupid=0x7ff)
> page_type: 0xffffffff()
> raw: 057ff00000000200 ffff888014391500 dead000000000122 0000000000000000
> raw: 0000000000000000 0000000080490049 00000001ffffffff 0000000000000000
> page dumped because: kasan: bad access detected
> page_owner tracks the page as allocated
> page last allocated via order 0, migratetype Reclaimable, gfp_mask
> 0x12830(GFP_ATOMIC|__GFP_NOWARN|__GFP_NORETRY|__GFP_RECLAIM
>   prep_new_page mm/page_alloc.c:1577 [inline]
>   get_page_from_freelist+0xfe0/0x2b80 mm/page_alloc.c:3221
>   __alloc_pages+0x1c7/0x490 mm/page_alloc.c:4477
>   alloc_pages+0x1a6/0x270 mm/mempolicy.c:2292
>   alloc_slab_page mm/slub.c:1862 [inline]
>   allocate_slab+0x25f/0x390 mm/slub.c:2009
>   new_slab mm/slub.c:2062 [inline]
>   ___slab_alloc+0xbc6/0x15c0 mm/slub.c:3215
>   __slab_alloc.constprop.0+0x56/0xa0 mm/slub.c:3314
>   __slab_alloc_node mm/slub.c:3367 [inline]
>   slab_alloc_node mm/slub.c:3460 [inline]
>   slab_alloc mm/slub.c:3478 [inline]
>   __kmem_cache_alloc_lru mm/slub.c:3485 [inline]
>   kmem_cache_alloc+0x369/0x390 mm/slub.c:3494
>   __es_alloc_extent fs/ext4/extents_status.c:467 [inline]
>   __es_alloc_extent fs/ext4/extents_status.c:464 [inline]
>   __es_insert_extent+0xde9/0x1440 fs/ext4/extents_status.c:815
>   ext4_es_cache_extent+0x2cb/0x480 fs/ext4/extents_status.c:937
>   ext4_cache_extents+0x13e/0x2d0 fs/ext4/extents.c:541
>   ext4_find_extent+0xac0/0xd20 fs/ext4/extents.c:925
>   ext4_ext_map_blocks+0x241/0x5980 fs/ext4/extents.c:4101
>   ext4_map_blocks+0xa27/0x16f0 fs/ext4/inode.c:548
>   ext4_mpage_readpages+0xd7d/0x1970 fs/ext4/readpage.c:297
>   ext4_readahead+0x102/0x140 fs/ext4/inode.c:3104
>   read_pages+0x1a2/0xd40 mm/readahead.c:160
> page_owner free stack trace missing
>
> Memory state around the buggy address:
>   ffff888112ecc080: 00 00 00 fc fc 00 00 00 00 00 fc fc 00 00 00 00
>   ffff888112ecc100: 00 fc fc 00 00 00 00 00 fc fc 00 00 00 00 00 fc
>> ffff888112ecc180: fc fa fb fb fb fb fc fc 00 00 00 00 00 fc fc 00
>                                 ^
>   ffff888112ecc200: 00 00 00 00 fc fc 00 00 00 00 00 fc fc 00 00 00
>   ffff888112ecc280: 00 00 fc fc 00 00 00 00 00 fc fc 00 00 00 00 00
> ==================================================================
>
>   ext4_es_try_to_merge_right fs/ext4/extents_status.c:593 [inline]
>   __es_insert_extent+0x9f4/0x1440 fs/ext4/extents_status.c:802
>   ext4_es_insert_extent+0x2ca/0xcb0 fs/ext4/extents_status.c:882
>   ext4_map_blocks+0x92a/0x16f0 fs/ext4/inode.c:680
>   ext4_alloc_file_blocks.isra.0+0x2df/0xb70 fs/ext4/extents.c:4462
>   ext4_zero_range fs/ext4/extents.c:4622 [inline]
>   ext4_fallocate+0x251c/0x3ce0 fs/ext4/extents.c:4721
>   vfs_fallocate+0x492/0xed0 fs/open.c:324
>   ksys_fallocate fs/open.c:347 [inline]
>   __do_sys_fallocate fs/open.c:355 [inline]
>   __se_sys_fallocate fs/open.c:353 [inline]
>   __x64_sys_fallocate+0xce/0x140 fs/open.c:353
>   do_syscall_x64 arch/x86/entry/common.c:50 [inline]
>   do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
>   entry_SYSCALL_64_after_hwframe+0x63/0xcd
>
> The buggy address belongs to the object at ffff888112ecc188
>   which belongs to the cache extent_status of size 40
> The buggy address is located 28 bytes inside of
>   freed 40-byte region [ffff888112ecc188, ffff888112ecc1b0)
>
> The buggy address belongs to the physical page:
> page:ffffea00044bb300 refcount:1 mapcount:0 mapping:0000000000000000
> index:0x0 pfn:0x112ecc
> flags: 0x57ff00000000200(slab|node=1|zone=2|lastcpupid=0x7ff)
> page_type: 0xffffffff()
> raw: 057ff00000000200 ffff888014391500 dead000000000122 0000000000000000
> raw: 0000000000000000 0000000080490049 00000001ffffffff 0000000000000000
> page dumped because: kasan: bad access detected
> page_owner tracks the page as allocated
> page last allocated via order 0, migratetype Reclaimable, gfp_mask
> 0x12830(GFP_ATOMIC|__GFP_NOWARN|__GFP_NORETRY|__GFP_RECLAIM
>   prep_new_page mm/page_alloc.c:1577 [inline]
>   get_page_from_freelist+0xfe0/0x2b80 mm/page_alloc.c:3221
>   __alloc_pages+0x1c7/0x490 mm/page_alloc.c:4477
>   alloc_pages+0x1a6/0x270 mm/mempolicy.c:2292
>   alloc_slab_page mm/slub.c:1862 [inline]
>   allocate_slab+0x25f/0x390 mm/slub.c:2009
>   new_slab mm/slub.c:2062 [inline]
>   ___slab_alloc+0xbc6/0x15c0 mm/slub.c:3215
>   __slab_alloc.constprop.0+0x56/0xa0 mm/slub.c:3314
>   __slab_alloc_node mm/slub.c:3367 [inline]
>   slab_alloc_node mm/slub.c:3460 [inline]
>   slab_alloc mm/slub.c:3478 [inline]
>   __kmem_cache_alloc_lru mm/slub.c:3485 [inline]
>   kmem_cache_alloc+0x369/0x390 mm/slub.c:3494
>   __es_alloc_extent fs/ext4/extents_status.c:467 [inline]
>   __es_alloc_extent fs/ext4/extents_status.c:464 [inline]
>   __es_insert_extent+0xde9/0x1440 fs/ext4/extents_status.c:815
>   ext4_es_cache_extent+0x2cb/0x480 fs/ext4/extents_status.c:937
>   ext4_cache_extents+0x13e/0x2d0 fs/ext4/extents.c:541
>   ext4_find_extent+0xac0/0xd20 fs/ext4/extents.c:925
>   ext4_ext_map_blocks+0x241/0x5980 fs/ext4/extents.c:4101
>   ext4_map_blocks+0xa27/0x16f0 fs/ext4/inode.c:548
>   ext4_mpage_readpages+0xd7d/0x1970 fs/ext4/readpage.c:297
>   ext4_readahead+0x102/0x140 fs/ext4/inode.c:3104
>   read_pages+0x1a2/0xd40 mm/readahead.c:160
> page_owner free stack trace missing
>
> Memory state around the buggy address:
>   ffff888112ecc080: 00 00 00 fc fc 00 00 00 00 00 fc fc 00 00 00 00
>   ffff888112ecc100: 00 fc fc 00 00 00 00 00 fc fc 00 00 00 00 00 fc
>> ffff888112ecc180: fc fa fb fb fb fb fc fc 00 00 00 00 00 fc fc 00
>                                 ^
>   ffff888112ecc200: 00 00 00 00 fc fc 00 00 00 00 00 fc fc 00 00 00
>   ffff888112ecc280: 00 00 fc fc 00 00 00 00 00 fc fc 00 00 00 00 00
> ==================================================================
> Kernel panic - not syncing: KASAN: panic_on_warn set ...
> CPU: 1 PID: 8438 Comm: syz-executor Not tainted 6.5.0-rc5 #1
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
> rel-1.12.0-59-gc9ba5276e321-prebuilt.qemu.org 04/01/2014
> Call Trace:
>   <TASK>
>   __dump_stack lib/dump_stack.c:88 [inline]
>   dump_stack_lvl+0xd5/0x150 lib/dump_stack.c:106
>   panic+0x67e/0x730 kernel/panic.c:340
>   check_panic_on_warn+0xad/0xb0 kernel/panic.c:236
>   end_report+0x108/0x150 mm/kasan/report.c:225
>   kasan_report+0xca/0xf0 mm/kasan/report.c:590
>   ext4_es_insert_extent+0xc68/0xcb0 fs/ext4/extents_status.c:894
>   ext4_map_blocks+0x92a/0x16f0 fs/ext4/inode.c:680
>   ext4_alloc_file_blocks.isra.0+0x2df/0xb70 fs/ext4/extents.c:4462
>   ext4_zero_range fs/ext4/extents.c:4622 [inline]
>   ext4_fallocate+0x251c/0x3ce0 fs/ext4/extents.c:4721
>   vfs_fallocate+0x492/0xed0 fs/open.c:324
>   ksys_fallocate fs/open.c:347 [inline]
>   __do_sys_fallocate fs/open.c:355 [inline]
>   __se_sys_fallocate fs/open.c:353 [inline]
>   __x64_sys_fallocate+0xce/0x140 fs/open.c:353
>   do_syscall_x64 arch/x86/entry/common.c:50 [inline]
>   do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
>   entry_SYSCALL_64_after_hwframe+0x63/0xcd
> RIP: 0033:0x47959d
> Code: 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa 48 89 f8 48
> 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08
> RSP: 002b:00007fbdfe383068 EFLAGS: 00000246 ORIG_RAX: 000000000000011d
> RAX: ffffffffffffffda RBX: 000000000059c0a0 RCX: 000000000047959d
> RDX: 000000000000000f RSI: 0000000000000010 RDI: 0000000000000003
> RBP: 0000000000000001 R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000008000 R11: 0000000000000246 R12: 000000000059c0ac
> R13: 000000000000000b R14: 0000000000437250 R15: 00007fbdfe363000
>   </TASK>
> Dumping ftrace buffer:
>     (ftrace buffer empty)
> Kernel Offset: disabled
> Rebooting in 1 seconds..
I'm very sorry that
     2a69c450083d ("ext4: using nofail preallocation in 
ext4_es_insert_extent()")
introduced this issue. The flow of issue triggering is as follows:

1. the raw es to update
|------------------------|

2. remove
   es    toremove     es1
|----|------------|------|

3. insert
   es     newes       es1
|----|------------|------|

es merges with newes, then merges with es1, frees es1, then determines
if es1->es_len is 0 and triggers a UAF.

The code flow is as follows:
ext4_es_insert_extent
   es1 = __es_alloc_extent(true);
   es2 = __es_alloc_extent(true);
   __es_remove_extent(inode, lblk, end, NULL, es1)
     __es_insert_extent(inode, &newes, es1) ---> insert es1 to es tree
   __es_insert_extent(inode, &newes, es2)
     ext4_es_try_to_merge_right
       ext4_es_free_extent(inode, es1) --->  es1 is freed
   if (es1 && !es1->es_len)
     // Trigger UAF by determining if es1 is used.

What's strange here is why the extent status is exactly the same before
and after ext4_es_insert_extent() is executed, and we still call
ext4_es_insert_extent() to perform the update.

But the problem is obvious and I will send a patch later.

With Best Regards,
-- 
Baokun Li
.

Powered by blists - more mailing lists