* [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2)
@ 2023-11-12 5:32 syzbot
2023-11-13 11:00 ` [syzbot] [PATCH] Test oob in squashfs readahead syzbot
` (12 more replies)
0 siblings, 13 replies; 19+ messages in thread
From: syzbot @ 2023-11-12 5:32 UTC (permalink / raw)
To: akpm, linux-fsdevel, linux-kernel, phillip, squashfs-devel,
syzkaller-bugs
Hello,
syzbot found the following issue on:
HEAD commit: 13d88ac54ddd Merge tag 'vfs-6.7.fsid' of git://git.kernel...
git tree: upstream
console+strace: https://syzkaller.appspot.com/x/log.txt?x=121965ef680000
kernel config: https://syzkaller.appspot.com/x/.config?x=beb32a598fd79db9
dashboard link: https://syzkaller.appspot.com/bug?extid=604424eb051c2f696163
compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=12b40c7b680000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=10f691ef680000
Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/9e81dc4903c2/disk-13d88ac5.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/f40fd7326b3f/vmlinux-13d88ac5.xz
kernel image: https://storage.googleapis.com/syzbot-assets/2f399cd6ff7d/bzImage-13d88ac5.xz
mounted in repro: https://storage.googleapis.com/syzbot-assets/8e127c645a04/mount_0.gz
The issue was bisected to:
commit f268eedddf3595e85f8883dc50aed29654785696
Author: Phillip Lougher <phillip@squashfs.org.uk>
Date: Sat Jun 11 03:21:32 2022 +0000
squashfs: extend "page actor" to handle missing pages
bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=1252b717680000
final oops: https://syzkaller.appspot.com/x/report.txt?x=1152b717680000
console output: https://syzkaller.appspot.com/x/log.txt?x=1652b717680000
IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+604424eb051c2f696163@syzkaller.appspotmail.com
Fixes: f268eedddf35 ("squashfs: extend "page actor" to handle missing pages")
SQUASHFS error: Unable to read metadata cache entry [6fa]
SQUASHFS error: Unable to read metadata cache entry [6fa]
SQUASHFS error: Unable to read metadata cache entry [6fa]
==================================================================
BUG: KASAN: slab-out-of-bounds in __readahead_batch include/linux/pagemap.h:1364 [inline]
BUG: KASAN: slab-out-of-bounds in squashfs_readahead+0x9a6/0x20d0 fs/squashfs/file.c:569
Write of size 8 at addr ffff88801e393648 by task syz-executor100/5067
CPU: 1 PID: 5067 Comm: syz-executor100 Not tainted 6.6.0-syzkaller-15156-g13d88ac54ddd #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/09/2023
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x1e7/0x2d0 lib/dump_stack.c:106
print_address_description mm/kasan/report.c:364 [inline]
print_report+0x163/0x540 mm/kasan/report.c:475
kasan_report+0x142/0x170 mm/kasan/report.c:588
__readahead_batch include/linux/pagemap.h:1364 [inline]
squashfs_readahead+0x9a6/0x20d0 fs/squashfs/file.c:569
read_pages+0x183/0x830 mm/readahead.c:160
page_cache_ra_unbounded+0x68e/0x7c0 mm/readahead.c:269
page_cache_sync_readahead include/linux/pagemap.h:1266 [inline]
filemap_get_pages+0x49c/0x2080 mm/filemap.c:2497
filemap_read+0x42b/0x10b0 mm/filemap.c:2593
__kernel_read+0x425/0x8b0 fs/read_write.c:428
integrity_kernel_read+0xb0/0xf0 security/integrity/iint.c:221
ima_calc_file_hash_tfm security/integrity/ima/ima_crypto.c:485 [inline]
ima_calc_file_shash security/integrity/ima/ima_crypto.c:516 [inline]
ima_calc_file_hash+0xad1/0x1b30 security/integrity/ima/ima_crypto.c:573
ima_collect_measurement+0x554/0xb30 security/integrity/ima/ima_api.c:290
process_measurement+0x1373/0x21c0 security/integrity/ima/ima_main.c:359
ima_file_check+0xf1/0x170 security/integrity/ima/ima_main.c:557
do_open fs/namei.c:3624 [inline]
path_openat+0x2893/0x3280 fs/namei.c:3779
do_filp_open+0x234/0x490 fs/namei.c:3809
do_sys_openat2+0x13e/0x1d0 fs/open.c:1440
do_sys_open fs/open.c:1455 [inline]
__do_sys_open fs/open.c:1463 [inline]
__se_sys_open fs/open.c:1459 [inline]
__x64_sys_open+0x225/0x270 fs/open.c:1459
do_syscall_x64 arch/x86/entry/common.c:51 [inline]
do_syscall_64+0x44/0x110 arch/x86/entry/common.c:82
entry_SYSCALL_64_after_hwframe+0x63/0x6b
RIP: 0033:0x7f0e73d6c5f9
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 61 17 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffd516ff158 EFLAGS: 00000246 ORIG_RAX: 0000000000000002
RAX: ffffffffffffffda RBX: 0031656c69662f2e RCX: 00007f0e73d6c5f9
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 00000000200000c0
RBP: 00007f0e73ddf610 R08: 0000000000000225 R09: 0000000000000000
R10: 00007ffd516ff020 R11: 0000000000000246 R12: 0000000000000001
R13: 00007ffd516ff328 R14: 0000000000000001 R15: 0000000000000001
</TASK>
Allocated by task 5067:
kasan_save_stack mm/kasan/common.c:45 [inline]
kasan_set_track+0x4f/0x70 mm/kasan/common.c:52
____kasan_kmalloc mm/kasan/common.c:374 [inline]
__kasan_kmalloc+0x98/0xb0 mm/kasan/common.c:383
kasan_kmalloc include/linux/kasan.h:198 [inline]
__do_kmalloc_node mm/slab_common.c:1007 [inline]
__kmalloc+0xb9/0x230 mm/slab_common.c:1020
kmalloc_array include/linux/slab.h:637 [inline]
squashfs_readahead+0x30c/0x20d0 fs/squashfs/file.c:552
read_pages+0x183/0x830 mm/readahead.c:160
page_cache_ra_unbounded+0x68e/0x7c0 mm/readahead.c:269
page_cache_sync_readahead include/linux/pagemap.h:1266 [inline]
filemap_get_pages+0x49c/0x2080 mm/filemap.c:2497
filemap_read+0x42b/0x10b0 mm/filemap.c:2593
__kernel_read+0x425/0x8b0 fs/read_write.c:428
integrity_kernel_read+0xb0/0xf0 security/integrity/iint.c:221
ima_calc_file_hash_tfm security/integrity/ima/ima_crypto.c:485 [inline]
ima_calc_file_shash security/integrity/ima/ima_crypto.c:516 [inline]
ima_calc_file_hash+0xad1/0x1b30 security/integrity/ima/ima_crypto.c:573
ima_collect_measurement+0x554/0xb30 security/integrity/ima/ima_api.c:290
process_measurement+0x1373/0x21c0 security/integrity/ima/ima_main.c:359
ima_file_check+0xf1/0x170 security/integrity/ima/ima_main.c:557
do_open fs/namei.c:3624 [inline]
path_openat+0x2893/0x3280 fs/namei.c:3779
do_filp_open+0x234/0x490 fs/namei.c:3809
do_sys_openat2+0x13e/0x1d0 fs/open.c:1440
do_sys_open fs/open.c:1455 [inline]
__do_sys_open fs/open.c:1463 [inline]
__se_sys_open fs/open.c:1459 [inline]
__x64_sys_open+0x225/0x270 fs/open.c:1459
do_syscall_x64 arch/x86/entry/common.c:51 [inline]
do_syscall_64+0x44/0x110 arch/x86/entry/common.c:82
entry_SYSCALL_64_after_hwframe+0x63/0x6b
The buggy address belongs to the object at ffff88801e393640
which belongs to the cache kmalloc-8 of size 8
The buggy address is located 0 bytes to the right of
allocated 8-byte region [ffff88801e393640, ffff88801e393648)
The buggy address belongs to the physical page:
page:ffffea000078e4c0 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff88801e393230 pfn:0x1e393
anon flags: 0xfff00000000800(slab|node=0|zone=1|lastcpupid=0x7ff)
page_type: 0xffffffff()
raw: 00fff00000000800 ffff888012c41280 0000000000000000 dead000000000001
raw: ffff88801e393230 0000000080660063 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 0, migratetype Unmovable, gfp_mask 0x12cc0(GFP_KERNEL|__GFP_NOWARN|__GFP_NORETRY), pid 1, tgid 1 (swapper/0), ts 10649202635, free_ts 9358727270
set_page_owner include/linux/page_owner.h:31 [inline]
post_alloc_hook+0x1e6/0x210 mm/page_alloc.c:1537
prep_new_page mm/page_alloc.c:1544 [inline]
get_page_from_freelist+0x339a/0x3530 mm/page_alloc.c:3312
__alloc_pages+0x255/0x670 mm/page_alloc.c:4568
alloc_pages_mpol+0x3de/0x640 mm/mempolicy.c:2133
alloc_slab_page+0x6a/0x160 mm/slub.c:1870
allocate_slab mm/slub.c:2017 [inline]
new_slab+0x84/0x2f0 mm/slub.c:2070
___slab_alloc+0xc85/0x1310 mm/slub.c:3223
__slab_alloc mm/slub.c:3322 [inline]
__slab_alloc_node mm/slub.c:3375 [inline]
slab_alloc_node mm/slub.c:3468 [inline]
__kmem_cache_alloc_node+0x21d/0x300 mm/slub.c:3517
__do_kmalloc_node mm/slab_common.c:1006 [inline]
__kmalloc_node_track_caller+0xa5/0x230 mm/slab_common.c:1027
kstrdup+0x3a/0x70 mm/util.c:62
__kernfs_new_node+0x9d/0x870 fs/kernfs/dir.c:611
kernfs_new_node+0x99/0x170 fs/kernfs/dir.c:679
kernfs_create_link+0xa5/0x1f0 fs/kernfs/symlink.c:39
sysfs_do_create_link_sd+0x85/0x100 fs/sysfs/symlink.c:44
device_create_sys_dev_entry+0x10f/0x170 drivers/base/core.c:3451
device_add+0x8cf/0xd30 drivers/base/core.c:3595
page last free stack trace:
reset_page_owner include/linux/page_owner.h:24 [inline]
free_pages_prepare mm/page_alloc.c:1137 [inline]
free_unref_page_prepare+0x92a/0xa50 mm/page_alloc.c:2347
free_unref_page+0x37/0x3f0 mm/page_alloc.c:2487
vfree+0x186/0x2e0 mm/vmalloc.c:2842
delayed_vfree_work+0x56/0x80 mm/vmalloc.c:2763
process_one_work kernel/workqueue.c:2630 [inline]
process_scheduled_works+0x90f/0x1400 kernel/workqueue.c:2703
worker_thread+0xa5f/0xff0 kernel/workqueue.c:2784
kthread+0x2d3/0x370 kernel/kthread.c:388
ret_from_fork+0x48/0x80 arch/x86/kernel/process.c:147
ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:242
Memory state around the buggy address:
ffff88801e393500: 00 fc fc fc fc fa fc fc fc fc 05 fc fc fc fc 05
ffff88801e393580: fc fc fc fc 00 fc fc fc fc 05 fc fc fc fc fa fc
>ffff88801e393600: fc fc fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc
^
ffff88801e393680: fc fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc
ffff88801e393700: fc 00 fc fc fc fc fb fc fc fc fc 00 fc fc fc fc
==================================================================
---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzkaller@googlegroups.com.
syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
For information about bisection process see: https://goo.gl/tpsmEJ#bisection
If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title
If you want syzbot to run the reproducer, reply with:
#syz test: git://repo/address.git branch-or-commit-hash
If you attach or paste a git patch, syzbot will apply it before testing.
If you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)
If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-report
If you want to undo deduplication, reply with:
#syz undup
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
@ 2023-11-13 11:00 ` syzbot
2023-11-13 12:10 ` syzbot
` (11 subsequent siblings)
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-13 11:00 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..c91a37be5589 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -553,6 +553,9 @@ static void squashfs_readahead(struct readahead_control *ractl)
if (!pages)
return;
+ printk("p: %p, bs: %d, bl: %d, mp: %d, start: %d, fe: %d, %s\n",
+ pages, msblk->block_size, msblk->block_log,
+ max_pages, start, file_end, __func__);
for (;;) {
pgoff_t index;
int res, bsize;
@@ -566,6 +569,8 @@ static void squashfs_readahead(struct readahead_control *ractl)
max_pages = (expected + PAGE_SIZE - 1) >> PAGE_SHIFT;
+ printk("p: %p, mp: %d, e: %d, idx: %d, s: %d, %s\n", pages, max_pages,
+ expected, pages[0]->index, start, __func__);
nr_pages = __readahead_batch(ractl, pages, max_pages);
if (!nr_pages)
break;
diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c
index 22e812808e5c..aad49a62b5ce 100644
--- a/fs/squashfs/super.c
+++ b/fs/squashfs/super.c
@@ -275,6 +275,8 @@ static int squashfs_fill_super(struct super_block *sb, struct fs_context *fc)
if (msblk->block_log > SQUASHFS_FILE_MAX_LOG)
goto failed_mount;
+ printk("bs: %d, bl: %d, %s\n", msblk->block_size, msblk->block_log, __func__);
+
/* Check that block_size and block_log match */
if (msblk->block_size != (1 << msblk->block_log))
goto insanity;
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
2023-11-13 11:00 ` [syzbot] [PATCH] Test oob in squashfs readahead syzbot
@ 2023-11-13 12:10 ` syzbot
2023-11-13 15:27 ` [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) Phillip Lougher
` (10 subsequent siblings)
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-13 12:10 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..a60c2be4447e 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -553,6 +553,9 @@ static void squashfs_readahead(struct readahead_control *ractl)
if (!pages)
return;
+ printk("p: %p, bs: %d, bl: %d, mp: %d, start: %d, fe: %d, %s\n",
+ pages, msblk->block_size, msblk->block_log,
+ max_pages, start, file_end, __func__);
for (;;) {
pgoff_t index;
int res, bsize;
@@ -566,6 +569,8 @@ static void squashfs_readahead(struct readahead_control *ractl)
max_pages = (expected + PAGE_SIZE - 1) >> PAGE_SHIFT;
+ printk("p: %p, mp: %d, e: %d,isr: %d, start: %d, %s\n", pages, max_pages,
+ expected, i_size_read(inode), start, __func__);
nr_pages = __readahead_batch(ractl, pages, max_pages);
if (!nr_pages)
break;
diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c
index 22e812808e5c..aad49a62b5ce 100644
--- a/fs/squashfs/super.c
+++ b/fs/squashfs/super.c
@@ -275,6 +275,8 @@ static int squashfs_fill_super(struct super_block *sb, struct fs_context *fc)
if (msblk->block_log > SQUASHFS_FILE_MAX_LOG)
goto failed_mount;
+ printk("bs: %d, bl: %d, %s\n", msblk->block_size, msblk->block_log, __func__);
+
/* Check that block_size and block_log match */
if (msblk->block_size != (1 << msblk->block_log))
goto insanity;
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2)
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
2023-11-13 11:00 ` [syzbot] [PATCH] Test oob in squashfs readahead syzbot
2023-11-13 12:10 ` syzbot
@ 2023-11-13 15:27 ` Phillip Lougher
2023-11-14 0:22 ` [syzbot] [PATCH] Test oob in squashfs readahead syzbot
` (9 subsequent siblings)
12 siblings, 0 replies; 19+ messages in thread
From: Phillip Lougher @ 2023-11-13 15:27 UTC (permalink / raw)
To: syzbot
Cc: akpm, linux-fsdevel, linux-kernel, phillip, squashfs-devel,
syzkaller-bugs
On Sun, Nov 12, 2023 at 5:32 AM syzbot
<syzbot+604424eb051c2f696163@syzkaller.appspotmail.com> wrote:
>
> Hello,
>
> syzbot found the following issue on:
>
> HEAD commit: 13d88ac54ddd Merge tag 'vfs-6.7.fsid' of git://git.kernel...
> git tree: upstream
> console+strace: https://syzkaller.appspot.com/x/log.txt?x=121965ef680000
> kernel config: https://syzkaller.appspot.com/x/.config?x=beb32a598fd79db9
> dashboard link: https://syzkaller.appspot.com/bug?extid=604424eb051c2f696163
> compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
> syz repro: https://syzkaller.appspot.com/x/repro.syz?x=12b40c7b680000
> C reproducer: https://syzkaller.appspot.com/x/repro.c?x=10f691ef680000
>
> Downloadable assets:
> disk image: https://storage.googleapis.com/syzbot-assets/9e81dc4903c2/disk-13d88ac5.raw.xz
> vmlinux: https://storage.googleapis.com/syzbot-assets/f40fd7326b3f/vmlinux-13d88ac5.xz
> kernel image: https://storage.googleapis.com/syzbot-assets/2f399cd6ff7d/bzImage-13d88ac5.xz
> mounted in repro: https://storage.googleapis.com/syzbot-assets/8e127c645a04/mount_0.gz
>
> The issue was bisected to:
>
> commit f268eedddf3595e85f8883dc50aed29654785696
> Author: Phillip Lougher <phillip@squashfs.org.uk>
> Date: Sat Jun 11 03:21:32 2022 +0000
>
> squashfs: extend "page actor" to handle missing pages
Fixed the issue, and will send a patch shortly.
Phillip
>
> bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=1252b717680000
> final oops: https://syzkaller.appspot.com/x/report.txt?x=1152b717680000
> console output: https://syzkaller.appspot.com/x/log.txt?x=1652b717680000
>
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+604424eb051c2f696163@syzkaller.appspotmail.com
> Fixes: f268eedddf35 ("squashfs: extend "page actor" to handle missing pages")
>
> SQUASHFS error: Unable to read metadata cache entry [6fa]
> SQUASHFS error: Unable to read metadata cache entry [6fa]
> SQUASHFS error: Unable to read metadata cache entry [6fa]
> ==================================================================
> BUG: KASAN: slab-out-of-bounds in __readahead_batch include/linux/pagemap.h:1364 [inline]
> BUG: KASAN: slab-out-of-bounds in squashfs_readahead+0x9a6/0x20d0 fs/squashfs/file.c:569
> Write of size 8 at addr ffff88801e393648 by task syz-executor100/5067
>
> CPU: 1 PID: 5067 Comm: syz-executor100 Not tainted 6.6.0-syzkaller-15156-g13d88ac54ddd #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/09/2023
> Call Trace:
> <TASK>
> __dump_stack lib/dump_stack.c:88 [inline]
> dump_stack_lvl+0x1e7/0x2d0 lib/dump_stack.c:106
> print_address_description mm/kasan/report.c:364 [inline]
> print_report+0x163/0x540 mm/kasan/report.c:475
> kasan_report+0x142/0x170 mm/kasan/report.c:588
> __readahead_batch include/linux/pagemap.h:1364 [inline]
> squashfs_readahead+0x9a6/0x20d0 fs/squashfs/file.c:569
> read_pages+0x183/0x830 mm/readahead.c:160
> page_cache_ra_unbounded+0x68e/0x7c0 mm/readahead.c:269
> page_cache_sync_readahead include/linux/pagemap.h:1266 [inline]
> filemap_get_pages+0x49c/0x2080 mm/filemap.c:2497
> filemap_read+0x42b/0x10b0 mm/filemap.c:2593
> __kernel_read+0x425/0x8b0 fs/read_write.c:428
> integrity_kernel_read+0xb0/0xf0 security/integrity/iint.c:221
> ima_calc_file_hash_tfm security/integrity/ima/ima_crypto.c:485 [inline]
> ima_calc_file_shash security/integrity/ima/ima_crypto.c:516 [inline]
> ima_calc_file_hash+0xad1/0x1b30 security/integrity/ima/ima_crypto.c:573
> ima_collect_measurement+0x554/0xb30 security/integrity/ima/ima_api.c:290
> process_measurement+0x1373/0x21c0 security/integrity/ima/ima_main.c:359
> ima_file_check+0xf1/0x170 security/integrity/ima/ima_main.c:557
> do_open fs/namei.c:3624 [inline]
> path_openat+0x2893/0x3280 fs/namei.c:3779
> do_filp_open+0x234/0x490 fs/namei.c:3809
> do_sys_openat2+0x13e/0x1d0 fs/open.c:1440
> do_sys_open fs/open.c:1455 [inline]
> __do_sys_open fs/open.c:1463 [inline]
> __se_sys_open fs/open.c:1459 [inline]
> __x64_sys_open+0x225/0x270 fs/open.c:1459
> do_syscall_x64 arch/x86/entry/common.c:51 [inline]
> do_syscall_64+0x44/0x110 arch/x86/entry/common.c:82
> entry_SYSCALL_64_after_hwframe+0x63/0x6b
> RIP: 0033:0x7f0e73d6c5f9
> Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 61 17 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48
> RSP: 002b:00007ffd516ff158 EFLAGS: 00000246 ORIG_RAX: 0000000000000002
> RAX: ffffffffffffffda RBX: 0031656c69662f2e RCX: 00007f0e73d6c5f9
> RDX: 0000000000000000 RSI: 0000000000000000 RDI: 00000000200000c0
> RBP: 00007f0e73ddf610 R08: 0000000000000225 R09: 0000000000000000
> R10: 00007ffd516ff020 R11: 0000000000000246 R12: 0000000000000001
> R13: 00007ffd516ff328 R14: 0000000000000001 R15: 0000000000000001
> </TASK>
>
> Allocated by task 5067:
> kasan_save_stack mm/kasan/common.c:45 [inline]
> kasan_set_track+0x4f/0x70 mm/kasan/common.c:52
> ____kasan_kmalloc mm/kasan/common.c:374 [inline]
> __kasan_kmalloc+0x98/0xb0 mm/kasan/common.c:383
> kasan_kmalloc include/linux/kasan.h:198 [inline]
> __do_kmalloc_node mm/slab_common.c:1007 [inline]
> __kmalloc+0xb9/0x230 mm/slab_common.c:1020
> kmalloc_array include/linux/slab.h:637 [inline]
> squashfs_readahead+0x30c/0x20d0 fs/squashfs/file.c:552
> read_pages+0x183/0x830 mm/readahead.c:160
> page_cache_ra_unbounded+0x68e/0x7c0 mm/readahead.c:269
> page_cache_sync_readahead include/linux/pagemap.h:1266 [inline]
> filemap_get_pages+0x49c/0x2080 mm/filemap.c:2497
> filemap_read+0x42b/0x10b0 mm/filemap.c:2593
> __kernel_read+0x425/0x8b0 fs/read_write.c:428
> integrity_kernel_read+0xb0/0xf0 security/integrity/iint.c:221
> ima_calc_file_hash_tfm security/integrity/ima/ima_crypto.c:485 [inline]
> ima_calc_file_shash security/integrity/ima/ima_crypto.c:516 [inline]
> ima_calc_file_hash+0xad1/0x1b30 security/integrity/ima/ima_crypto.c:573
> ima_collect_measurement+0x554/0xb30 security/integrity/ima/ima_api.c:290
> process_measurement+0x1373/0x21c0 security/integrity/ima/ima_main.c:359
> ima_file_check+0xf1/0x170 security/integrity/ima/ima_main.c:557
> do_open fs/namei.c:3624 [inline]
> path_openat+0x2893/0x3280 fs/namei.c:3779
> do_filp_open+0x234/0x490 fs/namei.c:3809
> do_sys_openat2+0x13e/0x1d0 fs/open.c:1440
> do_sys_open fs/open.c:1455 [inline]
> __do_sys_open fs/open.c:1463 [inline]
> __se_sys_open fs/open.c:1459 [inline]
> __x64_sys_open+0x225/0x270 fs/open.c:1459
> do_syscall_x64 arch/x86/entry/common.c:51 [inline]
> do_syscall_64+0x44/0x110 arch/x86/entry/common.c:82
> entry_SYSCALL_64_after_hwframe+0x63/0x6b
>
> The buggy address belongs to the object at ffff88801e393640
> which belongs to the cache kmalloc-8 of size 8
> The buggy address is located 0 bytes to the right of
> allocated 8-byte region [ffff88801e393640, ffff88801e393648)
>
> The buggy address belongs to the physical page:
> page:ffffea000078e4c0 refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff88801e393230 pfn:0x1e393
> anon flags: 0xfff00000000800(slab|node=0|zone=1|lastcpupid=0x7ff)
> page_type: 0xffffffff()
> raw: 00fff00000000800 ffff888012c41280 0000000000000000 dead000000000001
> raw: ffff88801e393230 0000000080660063 00000001ffffffff 0000000000000000
> page dumped because: kasan: bad access detected
> page_owner tracks the page as allocated
> page last allocated via order 0, migratetype Unmovable, gfp_mask 0x12cc0(GFP_KERNEL|__GFP_NOWARN|__GFP_NORETRY), pid 1, tgid 1 (swapper/0), ts 10649202635, free_ts 9358727270
> set_page_owner include/linux/page_owner.h:31 [inline]
> post_alloc_hook+0x1e6/0x210 mm/page_alloc.c:1537
> prep_new_page mm/page_alloc.c:1544 [inline]
> get_page_from_freelist+0x339a/0x3530 mm/page_alloc.c:3312
> __alloc_pages+0x255/0x670 mm/page_alloc.c:4568
> alloc_pages_mpol+0x3de/0x640 mm/mempolicy.c:2133
> alloc_slab_page+0x6a/0x160 mm/slub.c:1870
> allocate_slab mm/slub.c:2017 [inline]
> new_slab+0x84/0x2f0 mm/slub.c:2070
> ___slab_alloc+0xc85/0x1310 mm/slub.c:3223
> __slab_alloc mm/slub.c:3322 [inline]
> __slab_alloc_node mm/slub.c:3375 [inline]
> slab_alloc_node mm/slub.c:3468 [inline]
> __kmem_cache_alloc_node+0x21d/0x300 mm/slub.c:3517
> __do_kmalloc_node mm/slab_common.c:1006 [inline]
> __kmalloc_node_track_caller+0xa5/0x230 mm/slab_common.c:1027
> kstrdup+0x3a/0x70 mm/util.c:62
> __kernfs_new_node+0x9d/0x870 fs/kernfs/dir.c:611
> kernfs_new_node+0x99/0x170 fs/kernfs/dir.c:679
> kernfs_create_link+0xa5/0x1f0 fs/kernfs/symlink.c:39
> sysfs_do_create_link_sd+0x85/0x100 fs/sysfs/symlink.c:44
> device_create_sys_dev_entry+0x10f/0x170 drivers/base/core.c:3451
> device_add+0x8cf/0xd30 drivers/base/core.c:3595
> page last free stack trace:
> reset_page_owner include/linux/page_owner.h:24 [inline]
> free_pages_prepare mm/page_alloc.c:1137 [inline]
> free_unref_page_prepare+0x92a/0xa50 mm/page_alloc.c:2347
> free_unref_page+0x37/0x3f0 mm/page_alloc.c:2487
> vfree+0x186/0x2e0 mm/vmalloc.c:2842
> delayed_vfree_work+0x56/0x80 mm/vmalloc.c:2763
> process_one_work kernel/workqueue.c:2630 [inline]
> process_scheduled_works+0x90f/0x1400 kernel/workqueue.c:2703
> worker_thread+0xa5f/0xff0 kernel/workqueue.c:2784
> kthread+0x2d3/0x370 kernel/kthread.c:388
> ret_from_fork+0x48/0x80 arch/x86/kernel/process.c:147
> ret_from_fork_asm+0x11/0x20 arch/x86/entry/entry_64.S:242
>
> Memory state around the buggy address:
> ffff88801e393500: 00 fc fc fc fc fa fc fc fc fc 05 fc fc fc fc 05
> ffff88801e393580: fc fc fc fc 00 fc fc fc fc 05 fc fc fc fc fa fc
> >ffff88801e393600: fc fc fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc
> ^
> ffff88801e393680: fc fc 00 fc fc fc fc 00 fc fc fc fc 00 fc fc fc
> ffff88801e393700: fc 00 fc fc fc fc fb fc fc fc fc 00 fc fc fc fc
> ==================================================================
>
>
> ---
> This report is generated by a bot. It may contain errors.
> See https://goo.gl/tpsmEJ for more information about syzbot.
> syzbot engineers can be reached at syzkaller@googlegroups.com.
>
> syzbot will keep track of this issue. See:
> https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
> For information about bisection process see: https://goo.gl/tpsmEJ#bisection
>
> If the report is already addressed, let syzbot know by replying with:
> #syz fix: exact-commit-title
>
> If you want syzbot to run the reproducer, reply with:
> #syz test: git://repo/address.git branch-or-commit-hash
> If you attach or paste a git patch, syzbot will apply it before testing.
>
> If you want to overwrite report's subsystems, reply with:
> #syz set subsystems: new-subsystem
> (See the list of subsystem names on the web dashboard)
>
> If the report is a duplicate of another one, reply with:
> #syz dup: exact-subject-of-another-report
>
> If you want to undo deduplication, reply with:
> #syz undup
>
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
` (2 preceding siblings ...)
2023-11-13 15:27 ` [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) Phillip Lougher
@ 2023-11-14 0:22 ` syzbot
2023-11-14 1:04 ` syzbot
` (8 subsequent siblings)
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-14 0:22 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..cb3753782d85 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -566,6 +566,8 @@ static void squashfs_readahead(struct readahead_control *ractl)
max_pages = (expected + PAGE_SIZE - 1) >> PAGE_SHIFT;
+ if (!max_pages)
+ break;
nr_pages = __readahead_batch(ractl, pages, max_pages);
if (!nr_pages)
break;
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
` (3 preceding siblings ...)
2023-11-14 0:22 ` [syzbot] [PATCH] Test oob in squashfs readahead syzbot
@ 2023-11-14 1:04 ` syzbot
2023-11-14 1:55 ` syzbot
` (7 subsequent siblings)
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-14 1:04 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..0a586a7f470e 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -547,6 +547,9 @@ static void squashfs_readahead(struct readahead_control *ractl)
int i, file_end = i_size_read(inode) >> msblk->block_log;
unsigned int max_pages = 1UL << shift;
+ if (!file_end && !start)
+ return;
+
readahead_expand(ractl, start, (len | mask) + 1);
pages = kmalloc_array(max_pages, sizeof(void *), GFP_KERNEL);
diff --git a/fs/squashfs/file_direct.c b/fs/squashfs/file_direct.c
index f1ccad519e28..9c167df9248d 100644
--- a/fs/squashfs/file_direct.c
+++ b/fs/squashfs/file_direct.c
@@ -35,6 +35,9 @@ int squashfs_readpage_block(struct page *target_page, u64 block, int bsize,
struct squashfs_page_actor *actor;
void *pageaddr;
+ if (!file_end)
+ return -EINVAL;
+
if (end_index > file_end)
end_index = file_end;
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
` (4 preceding siblings ...)
2023-11-14 1:04 ` syzbot
@ 2023-11-14 1:55 ` syzbot
2023-11-14 3:33 ` syzbot
` (6 subsequent siblings)
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-14 1:55 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..e6dc2a5a8bcd 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -461,6 +461,11 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
page->index, squashfs_i(inode)->start);
+ if (!file_end) {
+ res = -EINVAL;
+ goto out;
+ }
+
if (page->index >= ((i_size_read(inode) + PAGE_SIZE - 1) >>
PAGE_SHIFT))
goto out;
@@ -547,6 +552,9 @@ static void squashfs_readahead(struct readahead_control *ractl)
int i, file_end = i_size_read(inode) >> msblk->block_log;
unsigned int max_pages = 1UL << shift;
+ if (!file_end && !start)
+ return;
+
readahead_expand(ractl, start, (len | mask) + 1);
pages = kmalloc_array(max_pages, sizeof(void *), GFP_KERNEL);
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
` (5 preceding siblings ...)
2023-11-14 1:55 ` syzbot
@ 2023-11-14 3:33 ` syzbot
2023-11-14 4:06 ` syzbot
` (5 subsequent siblings)
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-14 3:33 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..b54d6b993357 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -461,6 +461,12 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
page->index, squashfs_i(inode)->start);
+ if (!file_end) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ res = -EINVAL;
+ goto out;
+ }
+
if (page->index >= ((i_size_read(inode) + PAGE_SIZE - 1) >>
PAGE_SHIFT))
goto out;
@@ -547,6 +553,11 @@ static void squashfs_readahead(struct readahead_control *ractl)
int i, file_end = i_size_read(inode) >> msblk->block_log;
unsigned int max_pages = 1UL << shift;
+ if (!file_end && !start) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ return;
+ }
+
readahead_expand(ractl, start, (len | mask) + 1);
pages = kmalloc_array(max_pages, sizeof(void *), GFP_KERNEL);
diff --git a/fs/squashfs/inode.c b/fs/squashfs/inode.c
index aa3411354e66..680d8b302e60 100644
--- a/fs/squashfs/inode.c
+++ b/fs/squashfs/inode.c
@@ -125,6 +125,7 @@ int squashfs_read_inode(struct inode *inode, long long ino)
offset = SQUASHFS_INODE_OFFSET(ino);
type = le16_to_cpu(sqshb_ino->inode_type);
+ printk("in: %p, fs: %d, it: %d, %s\n", inode, le32_to_cpu(sqsh_ino->file_size), type, __func__);
switch (type) {
case SQUASHFS_REG_TYPE: {
unsigned int frag_offset, frag;
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
` (6 preceding siblings ...)
2023-11-14 3:33 ` syzbot
@ 2023-11-14 4:06 ` syzbot
2023-11-14 6:35 ` syzbot
` (4 subsequent siblings)
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-14 4:06 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..b54d6b993357 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -461,6 +461,12 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
page->index, squashfs_i(inode)->start);
+ if (!file_end) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ res = -EINVAL;
+ goto out;
+ }
+
if (page->index >= ((i_size_read(inode) + PAGE_SIZE - 1) >>
PAGE_SHIFT))
goto out;
@@ -547,6 +553,11 @@ static void squashfs_readahead(struct readahead_control *ractl)
int i, file_end = i_size_read(inode) >> msblk->block_log;
unsigned int max_pages = 1UL << shift;
+ if (!file_end && !start) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ return;
+ }
+
readahead_expand(ractl, start, (len | mask) + 1);
pages = kmalloc_array(max_pages, sizeof(void *), GFP_KERNEL);
diff --git a/fs/squashfs/inode.c b/fs/squashfs/inode.c
index aa3411354e66..e7c2ccd6a382 100644
--- a/fs/squashfs/inode.c
+++ b/fs/squashfs/inode.c
@@ -403,9 +403,11 @@ int squashfs_read_inode(struct inode *inode, long long ino)
} else
squashfs_i(inode)->xattr_count = 0;
+ printk("in: %p, fs: %d, it: %d, %s\n", inode, inode->i_size, type, __func__);
return 0;
failed_read:
+ printk("in: %p, fs: %d, it: %d, %s\n", inode, inode->i_size, type, __func__);
ERROR("Unable to read inode 0x%llx\n", ino);
return err;
}
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
` (7 preceding siblings ...)
2023-11-14 4:06 ` syzbot
@ 2023-11-14 6:35 ` syzbot
2023-11-14 7:53 ` syzbot
` (3 subsequent siblings)
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-14 6:35 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c
index 581ce9519339..7db8f2920c9e 100644
--- a/fs/squashfs/block.c
+++ b/fs/squashfs/block.c
@@ -287,6 +287,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
if (index + 2 > msblk->bytes_used) {
res = -EIO;
+ printk("srd1: i:%d, bu:%d \n", index, msblk->bytes_used);
goto out;
}
res = squashfs_bio_read(sb, index, 2, &bio, &offset);
@@ -295,6 +296,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
if (WARN_ON_ONCE(!bio_next_segment(bio, &iter_all))) {
res = -EIO;
+ printk("srd2 \n");
goto out_free_bio;
}
/* Extract the length of the metadata block */
@@ -305,6 +307,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
} else {
if (WARN_ON_ONCE(!bio_next_segment(bio, &iter_all))) {
res = -EIO;
+ printk("srd3 \n");
goto out_free_bio;
}
data = bvec_virt(bvec);
@@ -324,6 +327,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
if (length < 0 || length > output->length ||
(index + length) > msblk->bytes_used) {
res = -EIO;
+ printk("srd4: l:%d, ol: %d, bu: %d \n", length, output->length, msblk->bytes_used);
goto out;
}
@@ -336,10 +340,12 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
if (compressed) {
if (!msblk->stream) {
+ printk("srd5 \n");
res = -EIO;
goto out_free_bio;
}
res = msblk->thread_ops->decompress(msblk, bio, offset, length, output);
+ printk("srd6: r: %d \n", res);
} else {
res = copy_bio_to_actor(bio, output, offset, length);
}
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..b54d6b993357 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -461,6 +461,12 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
page->index, squashfs_i(inode)->start);
+ if (!file_end) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ res = -EINVAL;
+ goto out;
+ }
+
if (page->index >= ((i_size_read(inode) + PAGE_SIZE - 1) >>
PAGE_SHIFT))
goto out;
@@ -547,6 +553,11 @@ static void squashfs_readahead(struct readahead_control *ractl)
int i, file_end = i_size_read(inode) >> msblk->block_log;
unsigned int max_pages = 1UL << shift;
+ if (!file_end && !start) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ return;
+ }
+
readahead_expand(ractl, start, (len | mask) + 1);
pages = kmalloc_array(max_pages, sizeof(void *), GFP_KERNEL);
diff --git a/fs/squashfs/inode.c b/fs/squashfs/inode.c
index aa3411354e66..e7c2ccd6a382 100644
--- a/fs/squashfs/inode.c
+++ b/fs/squashfs/inode.c
@@ -403,9 +403,11 @@ int squashfs_read_inode(struct inode *inode, long long ino)
} else
squashfs_i(inode)->xattr_count = 0;
+ printk("in: %p, fs: %d, it: %d, %s\n", inode, inode->i_size, type, __func__);
return 0;
failed_read:
+ printk("in: %p, fs: %d, it: %d, %s\n", inode, inode->i_size, type, __func__);
ERROR("Unable to read inode 0x%llx\n", ino);
return err;
}
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
` (8 preceding siblings ...)
2023-11-14 6:35 ` syzbot
@ 2023-11-14 7:53 ` syzbot
2023-11-14 8:58 ` syzbot
` (2 subsequent siblings)
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-14 7:53 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c
index 581ce9519339..1c7c5500206b 100644
--- a/fs/squashfs/block.c
+++ b/fs/squashfs/block.c
@@ -314,9 +314,11 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
bio_uninit(bio);
kfree(bio);
+ printk("datal: %d \n", length);
compressed = SQUASHFS_COMPRESSED(length);
length = SQUASHFS_COMPRESSED_SIZE(length);
index += 2;
+ printk("datal2: %d, c:%d, i:%d \n", length, compressed, index);
TRACE("Block @ 0x%llx, %scompressed size %d\n", index - 2,
compressed ? "" : "un", length);
@@ -324,6 +326,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
if (length < 0 || length > output->length ||
(index + length) > msblk->bytes_used) {
res = -EIO;
+ printk("srd: l:%d, ol: %d, bu: %d \n", length, output->length, msblk->bytes_used);
goto out;
}
@@ -340,6 +343,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
goto out_free_bio;
}
res = msblk->thread_ops->decompress(msblk, bio, offset, length, output);
+ printk("srd6: r: %d \n", res);
} else {
res = copy_bio_to_actor(bio, output, offset, length);
}
diff --git a/fs/squashfs/cache.c b/fs/squashfs/cache.c
index 5062326d0efb..dac9eedea868 100644
--- a/fs/squashfs/cache.c
+++ b/fs/squashfs/cache.c
@@ -340,6 +340,7 @@ int squashfs_read_metadata(struct super_block *sb, void *buffer,
if (unlikely(length < 0))
return -EIO;
+ printk("srm: %d\n", length);
while (length) {
entry = squashfs_cache_get(sb, msblk->block_cache, *block, 0);
if (entry->error) {
@@ -381,6 +382,7 @@ struct squashfs_cache_entry *squashfs_get_fragment(struct super_block *sb,
{
struct squashfs_sb_info *msblk = sb->s_fs_info;
+ printk("sgf: %d\n", length);
return squashfs_cache_get(sb, msblk->fragment_cache, start_block,
length);
}
@@ -396,6 +398,7 @@ struct squashfs_cache_entry *squashfs_get_datablock(struct super_block *sb,
{
struct squashfs_sb_info *msblk = sb->s_fs_info;
+ printk("sgd: %d\n", length);
return squashfs_cache_get(sb, msblk->read_page, start_block, length);
}
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..b54d6b993357 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -461,6 +461,12 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
page->index, squashfs_i(inode)->start);
+ if (!file_end) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ res = -EINVAL;
+ goto out;
+ }
+
if (page->index >= ((i_size_read(inode) + PAGE_SIZE - 1) >>
PAGE_SHIFT))
goto out;
@@ -547,6 +553,11 @@ static void squashfs_readahead(struct readahead_control *ractl)
int i, file_end = i_size_read(inode) >> msblk->block_log;
unsigned int max_pages = 1UL << shift;
+ if (!file_end && !start) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ return;
+ }
+
readahead_expand(ractl, start, (len | mask) + 1);
pages = kmalloc_array(max_pages, sizeof(void *), GFP_KERNEL);
diff --git a/fs/squashfs/inode.c b/fs/squashfs/inode.c
index aa3411354e66..f3b0111e6fbd 100644
--- a/fs/squashfs/inode.c
+++ b/fs/squashfs/inode.c
@@ -175,6 +175,7 @@ int squashfs_read_inode(struct inode *inode, long long ino)
u64 frag_blk;
struct squashfs_lreg_inode *sqsh_ino = &squashfs_ino.lreg;
+ printk("in0: %p, fs: %d, it: %d, %s\n", inode, inode->i_size, type, __func__);
err = squashfs_read_metadata(sb, sqsh_ino, &block, &offset,
sizeof(*sqsh_ino));
if (err < 0)
@@ -403,6 +404,7 @@ int squashfs_read_inode(struct inode *inode, long long ino)
} else
squashfs_i(inode)->xattr_count = 0;
+ printk("in: %p, fs: %d, it: %d, %s\n", inode, inode->i_size, type, __func__);
return 0;
failed_read:
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
` (9 preceding siblings ...)
2023-11-14 7:53 ` syzbot
@ 2023-11-14 8:58 ` syzbot
2023-11-15 2:21 ` syzbot
2023-11-15 4:05 ` [PATCH] squashfs: fix oob in squashfs_readahead Edward Adam Davis
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-14 8:58 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c
index 581ce9519339..ba4f53c3725d 100644
--- a/fs/squashfs/block.c
+++ b/fs/squashfs/block.c
@@ -321,11 +321,16 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
TRACE("Block @ 0x%llx, %scompressed size %d\n", index - 2,
compressed ? "" : "un", length);
}
- if (length < 0 || length > output->length ||
- (index + length) > msblk->bytes_used) {
+
+ if (length < 0) {
res = -EIO;
goto out;
}
+ if (length > output->length || (index + length) > msblk->bytes_used) {
+ printk("srd: l:%d, ol: %d, bu: %d,i:%d, c:%d \n", length,
+ output->length, msblk->bytes_used, index, compressed);
+ length = min_t(int, msblk->bytes_used - index, output->length);
+ }
if (next_index)
*next_index = index + length;
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..b54d6b993357 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -461,6 +461,12 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
page->index, squashfs_i(inode)->start);
+ if (!file_end) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ res = -EINVAL;
+ goto out;
+ }
+
if (page->index >= ((i_size_read(inode) + PAGE_SIZE - 1) >>
PAGE_SHIFT))
goto out;
@@ -547,6 +553,11 @@ static void squashfs_readahead(struct readahead_control *ractl)
int i, file_end = i_size_read(inode) >> msblk->block_log;
unsigned int max_pages = 1UL << shift;
+ if (!file_end && !start) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ return;
+ }
+
readahead_expand(ractl, start, (len | mask) + 1);
pages = kmalloc_array(max_pages, sizeof(void *), GFP_KERNEL);
diff --git a/fs/squashfs/inode.c b/fs/squashfs/inode.c
index aa3411354e66..f3b0111e6fbd 100644
--- a/fs/squashfs/inode.c
+++ b/fs/squashfs/inode.c
@@ -175,6 +175,7 @@ int squashfs_read_inode(struct inode *inode, long long ino)
u64 frag_blk;
struct squashfs_lreg_inode *sqsh_ino = &squashfs_ino.lreg;
+ printk("in0: %p, fs: %d, it: %d, %s\n", inode, inode->i_size, type, __func__);
err = squashfs_read_metadata(sb, sqsh_ino, &block, &offset,
sizeof(*sqsh_ino));
if (err < 0)
@@ -403,6 +404,7 @@ int squashfs_read_inode(struct inode *inode, long long ino)
} else
squashfs_i(inode)->xattr_count = 0;
+ printk("in: %p, fs: %d, it: %d, %s\n", inode, inode->i_size, type, __func__);
return 0;
failed_read:
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [syzbot] [PATCH] Test oob in squashfs readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
` (10 preceding siblings ...)
2023-11-14 8:58 ` syzbot
@ 2023-11-15 2:21 ` syzbot
2023-11-15 4:05 ` [PATCH] squashfs: fix oob in squashfs_readahead Edward Adam Davis
12 siblings, 0 replies; 19+ messages in thread
From: syzbot @ 2023-11-15 2:21 UTC (permalink / raw)
To: linux-kernel
For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.
***
Subject: [PATCH] Test oob in squashfs readahead
Author: eadavis@qq.com
please test squashfs readahead oob
#syz test https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 13d88ac54ddd
diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c
index 581ce9519339..d335f28c822c 100644
--- a/fs/squashfs/block.c
+++ b/fs/squashfs/block.c
@@ -323,7 +323,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
}
if (length < 0 || length > output->length ||
(index + length) > msblk->bytes_used) {
- res = -EIO;
+ res = length < 0 ? -EIO : -EFBIG;
goto out;
}
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..5472ddd3596c 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -461,6 +461,11 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
page->index, squashfs_i(inode)->start);
+ if (!file_end) {
+ res = -EINVAL;
+ goto out;
+ }
+
if (page->index >= ((i_size_read(inode) + PAGE_SIZE - 1) >>
PAGE_SHIFT))
goto out;
@@ -547,6 +552,9 @@ static void squashfs_readahead(struct readahead_control *ractl)
int i, file_end = i_size_read(inode) >> msblk->block_log;
unsigned int max_pages = 1UL << shift;
+ if (!file_end)
+ return;
+
readahead_expand(ractl, start, (len | mask) + 1);
pages = kmalloc_array(max_pages, sizeof(void *), GFP_KERNEL);
^ permalink raw reply related [flat|nested] 19+ messages in thread
* [PATCH] squashfs: fix oob in squashfs_readahead
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
` (11 preceding siblings ...)
2023-11-15 2:21 ` syzbot
@ 2023-11-15 4:05 ` Edward Adam Davis
2023-11-15 22:39 ` Andrew Morton
` (2 more replies)
12 siblings, 3 replies; 19+ messages in thread
From: Edward Adam Davis @ 2023-11-15 4:05 UTC (permalink / raw)
To: syzbot+604424eb051c2f696163
Cc: akpm, linux-fsdevel, linux-kernel, phillip, squashfs-devel,
syzkaller-bugs
[Syz log]
SQUASHFS error: Failed to read block 0x6fc: -5
SQUASHFS error: Unable to read metadata cache entry [6fa]
SQUASHFS error: Unable to read metadata cache entry [6fa]
SQUASHFS error: Unable to read metadata cache entry [6fa]
==================================================================
BUG: KASAN: slab-out-of-bounds in __readahead_batch include/linux/pagemap.h:1364 [inline]
BUG: KASAN: slab-out-of-bounds in squashfs_readahead+0x9a6/0x20d0 fs/squashfs/file.c:569
Write of size 8 at addr ffff88801e393648 by task syz-executor100/5067
CPU: 1 PID: 5067 Comm: syz-executor100 Not tainted 6.6.0-syzkaller-15156-g13d88ac54ddd #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/09/2023
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x1e7/0x2d0 lib/dump_stack.c:106
print_address_description mm/kasan/report.c:364 [inline]
print_report+0x163/0x540 mm/kasan/report.c:475
kasan_report+0x142/0x170 mm/kasan/report.c:588
__readahead_batch include/linux/pagemap.h:1364 [inline]
squashfs_readahead+0x9a6/0x20d0 fs/squashfs/file.c:569
read_pages+0x183/0x830 mm/readahead.c:160
page_cache_ra_unbounded+0x68e/0x7c0 mm/readahead.c:269
page_cache_sync_readahead include/linux/pagemap.h:1266 [inline]
filemap_get_pages+0x49c/0x2080 mm/filemap.c:2497
filemap_read+0x42b/0x10b0 mm/filemap.c:2593
__kernel_read+0x425/0x8b0 fs/read_write.c:428
integrity_kernel_read+0xb0/0xf0 security/integrity/iint.c:221
ima_calc_file_hash_tfm security/integrity/ima/ima_crypto.c:485 [inline]
ima_calc_file_shash security/integrity/ima/ima_crypto.c:516 [inline]
ima_calc_file_hash+0xad1/0x1b30 security/integrity/ima/ima_crypto.c:573
ima_collect_measurement+0x554/0xb30 security/integrity/ima/ima_api.c:290
process_measurement+0x1373/0x21c0 security/integrity/ima/ima_main.c:359
ima_file_check+0xf1/0x170 security/integrity/ima/ima_main.c:557
do_open fs/namei.c:3624 [inline]
path_openat+0x2893/0x3280 fs/namei.c:3779
[Bug]
path_openat() called open_last_lookups() before calling do_open() and
open_last_lookups() will eventually call squashfs_read_inode() to set
inode->i_size, but before setting i_size, it is necessary to obtain file_size
from the disk.
However, during the value retrieval process, the length of the value retrieved
from the disk was greater than output->length, resulting(-EIO) in the failure of
squashfs_read_data(), further leading to i_size has not been initialized,
i.e. its value is 0.
This resulted in the failure of squashfs_read_data(), where "SQUASHFS error:
Failed to read block 0x6fc: -5" was output in the syz log.
This also resulted in the failure of squashfs_cache_get(), outputting "SQUASHFS
error: Unable to read metadata cache entry [6fa]" in the syz log.
[Fix]
Before performing a read ahead operation in squashfs_read_folio() and
squashfs_readahead(), check if i_size is not 0 before continuing.
Optimize the return value of squashfs_read_data() and return -EFBIG when the
length is greater than output->length(or (index + length) >
msblk->bytes_used).
Reported-and-tested-by: syzbot+604424eb051c2f696163@syzkaller.appspotmail.com
Fixes: f268eedddf35 ("squashfs: extend "page actor" to handle missing pages")
Signed-off-by: Edward Adam Davis <eadavis@qq.com>
---
fs/squashfs/block.c | 2 +-
fs/squashfs/file.c | 8 ++++++++
2 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c
index 581ce9519339..d335f28c822c 100644
--- a/fs/squashfs/block.c
+++ b/fs/squashfs/block.c
@@ -323,7 +323,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
}
if (length < 0 || length > output->length ||
(index + length) > msblk->bytes_used) {
- res = -EIO;
+ res = length < 0 ? -EIO : -EFBIG;
goto out;
}
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index 8ba8c4c50770..5472ddd3596c 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -461,6 +461,11 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
page->index, squashfs_i(inode)->start);
+ if (!file_end) {
+ res = -EINVAL;
+ goto out;
+ }
+
if (page->index >= ((i_size_read(inode) + PAGE_SIZE - 1) >>
PAGE_SHIFT))
goto out;
@@ -547,6 +552,9 @@ static void squashfs_readahead(struct readahead_control *ractl)
int i, file_end = i_size_read(inode) >> msblk->block_log;
unsigned int max_pages = 1UL << shift;
+ if (!file_end)
+ return;
+
readahead_expand(ractl, start, (len | mask) + 1);
pages = kmalloc_array(max_pages, sizeof(void *), GFP_KERNEL);
--
2.25.1
^ permalink raw reply related [flat|nested] 19+ messages in thread
* Re: [PATCH] squashfs: fix oob in squashfs_readahead
2023-11-15 4:05 ` [PATCH] squashfs: fix oob in squashfs_readahead Edward Adam Davis
@ 2023-11-15 22:39 ` Andrew Morton
2023-11-16 15:14 ` Phillip Lougher
[not found] ` <CGME20231117131718eucas1p13328b32942cce99a99197eb28e14a981@eucas1p1.samsung.com>
2 siblings, 0 replies; 19+ messages in thread
From: Andrew Morton @ 2023-11-15 22:39 UTC (permalink / raw)
To: Edward Adam Davis
Cc: syzbot+604424eb051c2f696163, linux-fsdevel, linux-kernel,
phillip, squashfs-devel, syzkaller-bugs
On Wed, 15 Nov 2023 12:05:35 +0800 Edward Adam Davis <eadavis@qq.com> wrote:
> Before performing a read ahead operation in squashfs_read_folio() and
> squashfs_readahead(), check if i_size is not 0 before continuing.
I'll merge this for testing, pending Phillip's review. One thing:
> --- a/fs/squashfs/block.c
> +++ b/fs/squashfs/block.c
> @@ -323,7 +323,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
> }
> if (length < 0 || length > output->length ||
> (index + length) > msblk->bytes_used) {
> - res = -EIO;
> + res = length < 0 ? -EIO : -EFBIG;
> goto out;
> }
Seems a bit ugly to test `length' twice for the same thing. How about
if (length < 0) {
res = -EIO;
got out;
}
if (length > output->length || (index + length) > msblk->bytes_used) {
res = -EFBIG;
goto out;
}
?
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] squashfs: fix oob in squashfs_readahead
2023-11-15 4:05 ` [PATCH] squashfs: fix oob in squashfs_readahead Edward Adam Davis
2023-11-15 22:39 ` Andrew Morton
@ 2023-11-16 15:14 ` Phillip Lougher
2023-11-18 2:12 ` Edward Adam Davis
[not found] ` <CGME20231117131718eucas1p13328b32942cce99a99197eb28e14a981@eucas1p1.samsung.com>
2 siblings, 1 reply; 19+ messages in thread
From: Phillip Lougher @ 2023-11-16 15:14 UTC (permalink / raw)
To: eadavis
Cc: akpm, linux-fsdevel, linux-kernel, phillip, squashfs-devel,
syzbot+604424eb051c2f696163, syzkaller-bugs
> [Bug]
> path_openat() called open_last_lookups() before calling do_open() and
> open_last_lookups() will eventually call squashfs_read_inode() to set
> inode->i_size, but before setting i_size, it is necessary to obtain file_size
> from the disk.
>
> However, during the value retrieval process, the length of the value retrieved
> from the disk was greater than output->length, resulting(-EIO) in the failure of
> squashfs_read_data(), further leading to i_size has not been initialized,
> i.e. its value is 0.
>
NACK
This analysis is completely *wrong*. First, if there was I/O error reading
the inode it would never be created, and squasfs_readahead() would
never be called on it, because it will never exist.
Second i_size isn't unintialised and it isn't 0 in value. Where
you got this bogus information from is because in your test patches,
i.e.
https://lore.kernel.org/all/000000000000bb74b9060a14717c@google.com/
You have
+ if (!file_end) {
+ printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
+ res = -EINVAL;
+ goto out;
+ }
+
You have used %d, and the result of i_size_read(inode) overflows, giving the
bogus 0 value.
The actual value is 1407374883553280, or 0x5000000000000, which is
too big to fit into an unsigned int.
> This resulted in the failure of squashfs_read_data(), where "SQUASHFS error:
> Failed to read block 0x6fc: -5" was output in the syz log.
> This also resulted in the failure of squashfs_cache_get(), outputting "SQUASHFS
> error: Unable to read metadata cache entry [6fa]" in the syz log.
>
NO, *that* is caused by the failure to read some other inodes which
as a result are correctly not created. Nothing to do with the oops here.
> [Fix]
> Before performing a read ahead operation in squashfs_read_folio() and
> squashfs_readahead(), check if i_size is not 0 before continuing.
>
A third NO, it is only 0 because the variable overflowed.
Additionally, let's look at your "fix" here.
> @@ -461,6 +461,11 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
> TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
> page->index, squashfs_i(inode)->start);
>
> + if (!file_end) {
> + res = -EINVAL;
> + goto out;
> + }
> +
file_end is computed by
int file_end = i_size_read(inode) >> msblk->block_log;
So your "fix" will reject *any* file less than msblk->block_log in
size as invalid, including perfectly valid zero size files (empty
files are valid too).
I already identified the cause and send a fix patch here:
https://lore.kernel.org/all/20231113160901.6444-1-phillip@squashfs.org.uk/
NACK
Phillip
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] squashfs: fix oob in squashfs_readahead
[not found] ` <CGME20231117131718eucas1p13328b32942cce99a99197eb28e14a981@eucas1p1.samsung.com>
@ 2023-11-17 13:17 ` Marek Szyprowski
2023-11-17 15:48 ` Andrew Morton
0 siblings, 1 reply; 19+ messages in thread
From: Marek Szyprowski @ 2023-11-17 13:17 UTC (permalink / raw)
To: Edward Adam Davis, syzbot+604424eb051c2f696163
Cc: akpm, linux-fsdevel, linux-kernel, phillip, squashfs-devel,
syzkaller-bugs
Hi All,
On 15.11.2023 05:05, Edward Adam Davis wrote:
> [Syz log]
> SQUASHFS error: Failed to read block 0x6fc: -5
> SQUASHFS error: Unable to read metadata cache entry [6fa]
> SQUASHFS error: Unable to read metadata cache entry [6fa]
> SQUASHFS error: Unable to read metadata cache entry [6fa]
> ==================================================================
> BUG: KASAN: slab-out-of-bounds in __readahead_batch include/linux/pagemap.h:1364 [inline]
> BUG: KASAN: slab-out-of-bounds in squashfs_readahead+0x9a6/0x20d0 fs/squashfs/file.c:569
> Write of size 8 at addr ffff88801e393648 by task syz-executor100/5067
>
> CPU: 1 PID: 5067 Comm: syz-executor100 Not tainted 6.6.0-syzkaller-15156-g13d88ac54ddd #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/09/2023
> Call Trace:
> <TASK>
> __dump_stack lib/dump_stack.c:88 [inline]
> dump_stack_lvl+0x1e7/0x2d0 lib/dump_stack.c:106
> print_address_description mm/kasan/report.c:364 [inline]
> print_report+0x163/0x540 mm/kasan/report.c:475
> kasan_report+0x142/0x170 mm/kasan/report.c:588
> __readahead_batch include/linux/pagemap.h:1364 [inline]
> squashfs_readahead+0x9a6/0x20d0 fs/squashfs/file.c:569
> read_pages+0x183/0x830 mm/readahead.c:160
> page_cache_ra_unbounded+0x68e/0x7c0 mm/readahead.c:269
> page_cache_sync_readahead include/linux/pagemap.h:1266 [inline]
> filemap_get_pages+0x49c/0x2080 mm/filemap.c:2497
> filemap_read+0x42b/0x10b0 mm/filemap.c:2593
> __kernel_read+0x425/0x8b0 fs/read_write.c:428
> integrity_kernel_read+0xb0/0xf0 security/integrity/iint.c:221
> ima_calc_file_hash_tfm security/integrity/ima/ima_crypto.c:485 [inline]
> ima_calc_file_shash security/integrity/ima/ima_crypto.c:516 [inline]
> ima_calc_file_hash+0xad1/0x1b30 security/integrity/ima/ima_crypto.c:573
> ima_collect_measurement+0x554/0xb30 security/integrity/ima/ima_api.c:290
> process_measurement+0x1373/0x21c0 security/integrity/ima/ima_main.c:359
> ima_file_check+0xf1/0x170 security/integrity/ima/ima_main.c:557
> do_open fs/namei.c:3624 [inline]
> path_openat+0x2893/0x3280 fs/namei.c:3779
>
> [Bug]
> path_openat() called open_last_lookups() before calling do_open() and
> open_last_lookups() will eventually call squashfs_read_inode() to set
> inode->i_size, but before setting i_size, it is necessary to obtain file_size
> from the disk.
>
> However, during the value retrieval process, the length of the value retrieved
> from the disk was greater than output->length, resulting(-EIO) in the failure of
> squashfs_read_data(), further leading to i_size has not been initialized,
> i.e. its value is 0.
>
> This resulted in the failure of squashfs_read_data(), where "SQUASHFS error:
> Failed to read block 0x6fc: -5" was output in the syz log.
> This also resulted in the failure of squashfs_cache_get(), outputting "SQUASHFS
> error: Unable to read metadata cache entry [6fa]" in the syz log.
>
> [Fix]
> Before performing a read ahead operation in squashfs_read_folio() and
> squashfs_readahead(), check if i_size is not 0 before continuing.
>
> Optimize the return value of squashfs_read_data() and return -EFBIG when the
> length is greater than output->length(or (index + length) >
> msblk->bytes_used).
>
> Reported-and-tested-by: syzbot+604424eb051c2f696163@syzkaller.appspotmail.com
> Fixes: f268eedddf35 ("squashfs: extend "page actor" to handle missing pages")
> Signed-off-by: Edward Adam Davis <eadavis@qq.com>
This patch, merged to linux-next as commit 1ff947abe24a ("squashfs: fix
oob in squashfs_readahead"), breaks mounting squashfs volumes on all my
test systems. Let me know if you need more information to debug this issue.
> ---
> fs/squashfs/block.c | 2 +-
> fs/squashfs/file.c | 8 ++++++++
> 2 files changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c
> index 581ce9519339..d335f28c822c 100644
> --- a/fs/squashfs/block.c
> +++ b/fs/squashfs/block.c
> @@ -323,7 +323,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
> }
> if (length < 0 || length > output->length ||
> (index + length) > msblk->bytes_used) {
> - res = -EIO;
> + res = length < 0 ? -EIO : -EFBIG;
> goto out;
> }
>
> diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
> index 8ba8c4c50770..5472ddd3596c 100644
> --- a/fs/squashfs/file.c
> +++ b/fs/squashfs/file.c
> @@ -461,6 +461,11 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
> TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
> page->index, squashfs_i(inode)->start);
>
> + if (!file_end) {
> + res = -EINVAL;
> + goto out;
> + }
> +
> if (page->index >= ((i_size_read(inode) + PAGE_SIZE - 1) >>
> PAGE_SHIFT))
> goto out;
> @@ -547,6 +552,9 @@ static void squashfs_readahead(struct readahead_control *ractl)
> int i, file_end = i_size_read(inode) >> msblk->block_log;
> unsigned int max_pages = 1UL << shift;
>
> + if (!file_end)
> + return;
> +
> readahead_expand(ractl, start, (len | mask) + 1);
>
> pages = kmalloc_array(max_pages, sizeof(void *), GFP_KERNEL);
Best regards
--
Marek Szyprowski, PhD
Samsung R&D Institute Poland
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] squashfs: fix oob in squashfs_readahead
2023-11-17 13:17 ` Marek Szyprowski
@ 2023-11-17 15:48 ` Andrew Morton
0 siblings, 0 replies; 19+ messages in thread
From: Andrew Morton @ 2023-11-17 15:48 UTC (permalink / raw)
To: Marek Szyprowski
Cc: Edward Adam Davis, syzbot+604424eb051c2f696163, linux-fsdevel,
linux-kernel, phillip, squashfs-devel, syzkaller-bugs
On Fri, 17 Nov 2023 14:17:17 +0100 Marek Szyprowski <m.szyprowski@samsung.com> wrote:
> > Reported-and-tested-by: syzbot+604424eb051c2f696163@syzkaller.appspotmail.com
> > Fixes: f268eedddf35 ("squashfs: extend "page actor" to handle missing pages")
> > Signed-off-by: Edward Adam Davis <eadavis@qq.com>
>
> This patch, merged to linux-next as commit 1ff947abe24a ("squashfs: fix
> oob in squashfs_readahead"), breaks mounting squashfs volumes on all my
> test systems. Let me know if you need more information to debug this issue.
Thanks. The patch has been dropped.
^ permalink raw reply [flat|nested] 19+ messages in thread
* Re: [PATCH] squashfs: fix oob in squashfs_readahead
2023-11-16 15:14 ` Phillip Lougher
@ 2023-11-18 2:12 ` Edward Adam Davis
0 siblings, 0 replies; 19+ messages in thread
From: Edward Adam Davis @ 2023-11-18 2:12 UTC (permalink / raw)
To: phillip
Cc: akpm, eadavis, linux-fsdevel, linux-kernel, squashfs-devel,
syzbot+604424eb051c2f696163, syzkaller-bugs
On Thu, 16 Nov 2023 15:14:24 +0000, Phillip Lougher wrote:
> > [Bug]
> > path_openat() called open_last_lookups() before calling do_open() and
> > open_last_lookups() will eventually call squashfs_read_inode() to set
> > inode->i_size, but before setting i_size, it is necessary to obtain file_size
> > from the disk.
> >
> > However, during the value retrieval process, the length of the value retrieved
> > from the disk was greater than output->length, resulting(-EIO) in the failure of
> > squashfs_read_data(), further leading to i_size has not been initialized,
> > i.e. its value is 0.
> >
>
> NACK
>
> This analysis is completely *wrong*. First, if there was I/O error reading
> the inode it would never be created, and squasfs_readahead() would
> never be called on it, because it will never exist.
>
> Second i_size isn't unintialised and it isn't 0 in value. Where
> you got this bogus information from is because in your test patches,
> i.e.
[There is my debuging patch]
diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c
index 581ce9519339..1c7c5500206b 100644
--- a/fs/squashfs/block.c
+++ b/fs/squashfs/block.c
@@ -314,9 +314,11 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
bio_uninit(bio);
kfree(bio);
+ printk("datal: %d \n", length);
compressed = SQUASHFS_COMPRESSED(length);
length = SQUASHFS_COMPRESSED_SIZE(length);
index += 2;
+ printk("datal2: %d, c:%d, i:%d \n", length, compressed, index);
TRACE("Block @ 0x%llx, %scompressed size %d\n", index - 2,
compressed ? "" : "un", length);
@@ -324,6 +326,7 @@ int squashfs_read_data(struct super_block *sb, u64 index, int length,
if (length < 0 || length > output->length ||
(index + length) > msblk->bytes_used) {
res = -EIO;
+ printk("srd: l:%d, ol: %d, bu: %d \n", length, output->length, msblk->bytes_used);
goto out;
}
patch link: https://syzkaller.appspot.com/text?tag=Patch&x=1142f82f680000
[There is my test log]
[ 457.030754][ T8879] datal: 65473
[ 457.034334][ T8879] datal2: 32705, c:0, i:1788
[ 457.039253][ T8879] srd: l:32705, ol: 8192, bu: 1870
[ 457.044513][ T8879] SQUASHFS error: Failed to read block 0x6fc: -5
[ 457.052034][ T8879] SQUASHFS error: Unable to read metadata cache entry [6fa]
log link: https://syzkaller.appspot.com/x/log.txt?x=137b0270e80000
[Answer your doubts]
Based on the above test, it can be clearly determined that length=32705 is
greater than the maximum metadata size of 8192, resulting in squashfs_read_data() failed.
This will further lead to squashfs_cache_get() returns "entry->error=entry->length=-EIO",
followed by squashfs_read_metadata() failed, which will ultimately result in i_size not
being initialized in squashfs_read_inode().
The following are the relevant call stacks:
23 const struct inode_operations squashfs_dir_inode_ops = {
24 .lookup = squashfs_lookup,
25 .listxattr = squashfs_listxattr
26 };
NORMAL +0 ~0 -0 fs/squashfs/namei.c
path_openat()->open_last_lookups()->lookup_open()
1 if (d_in_lookup(dentry)) {
3455 struct dentry *res = dir_inode->i_op->lookup(dir_inode, dentry, 1 nd->flags);
squashfs_lookup()->
squashfs_iget()->
squashfs_read_inode()->
init inode->i_size, example: inode->i_size = sqsh_ino->file_size;
>
> https://lore.kernel.org/all/000000000000bb74b9060a14717c@google.com/
>
> You have
>
> + if (!file_end) {
> + printk("i:%p, is:%d, %s\n", inode, i_size_read(inode), __func__);
> + res = -EINVAL;
> + goto out;
> + }
> +
>
> You have used %d, and the result of i_size_read(inode) overflows, giving the
> bogus 0 value.
>
> The actual value is 1407374883553280, or 0x5000000000000, which is
> too big to fit into an unsigned int.
>
> > This resulted in the failure of squashfs_read_data(), where "SQUASHFS error:
> > Failed to read block 0x6fc: -5" was output in the syz log.
> > This also resulted in the failure of squashfs_cache_get(), outputting "SQUASHFS
> > error: Unable to read metadata cache entry [6fa]" in the syz log.
> >
>
> NO, *that* is caused by the failure to read some other inodes which
> as a result are correctly not created. Nothing to do with the oops here.
>
> > [Fix]
> > Before performing a read ahead operation in squashfs_read_folio() and
> > squashfs_readahead(), check if i_size is not 0 before continuing.
> >
>
> A third NO, it is only 0 because the variable overflowed.
>
> Additionally, let's look at your "fix" here.
>
> > @@ -461,6 +461,11 @@ static int squashfs_read_folio(struct file *file, struct folio *folio)
> > TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
> > page->index, squashfs_i(inode)->start);
> >
> > + if (!file_end) {
> > + res = -EINVAL;
> > + goto out;
> > + }
> > +
>
> file_end is computed by
>
> int file_end = i_size_read(inode) >> msblk->block_log;
>
> So your "fix" will reject *any* file less than msblk->block_log in
> size as invalid, including perfectly valid zero size files (empty
> files are valid too).
edward
^ permalink raw reply related [flat|nested] 19+ messages in thread
end of thread, other threads:[~2023-11-18 2:12 UTC | newest]
Thread overview: 19+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-11-12 5:32 [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) syzbot
2023-11-13 11:00 ` [syzbot] [PATCH] Test oob in squashfs readahead syzbot
2023-11-13 12:10 ` syzbot
2023-11-13 15:27 ` [syzbot] [squashfs?] KASAN: slab-out-of-bounds Write in squashfs_readahead (2) Phillip Lougher
2023-11-14 0:22 ` [syzbot] [PATCH] Test oob in squashfs readahead syzbot
2023-11-14 1:04 ` syzbot
2023-11-14 1:55 ` syzbot
2023-11-14 3:33 ` syzbot
2023-11-14 4:06 ` syzbot
2023-11-14 6:35 ` syzbot
2023-11-14 7:53 ` syzbot
2023-11-14 8:58 ` syzbot
2023-11-15 2:21 ` syzbot
2023-11-15 4:05 ` [PATCH] squashfs: fix oob in squashfs_readahead Edward Adam Davis
2023-11-15 22:39 ` Andrew Morton
2023-11-16 15:14 ` Phillip Lougher
2023-11-18 2:12 ` Edward Adam Davis
[not found] ` <CGME20231117131718eucas1p13328b32942cce99a99197eb28e14a981@eucas1p1.samsung.com>
2023-11-17 13:17 ` Marek Szyprowski
2023-11-17 15:48 ` Andrew Morton
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.