[syzbot] [mm?] BUG: unable to handle kernel paging request in __pte_offset_map_lock

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: 78124b0c1d10 Merge branch 'for-next/core' into for-kernelci
git tree: git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux.git for-kernelci
console output: https://syzkaller.appspot.com/x/log.txt?x=111b0e71680000
kernel config: https://syzkaller.appspot.com/x/.config?x=f27cd6e68911e026
dashboard link: https://syzkaller.appspot.com/bug?extid=89edd67979b52675ddec
compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
userspace arch: arm64
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=16b8e671680000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=125a9df5680000

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/bd512de820ae/disk-78124b0c.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/a47a437b1d4f/vmlinux-78124b0c.xz
kernel image: https://storage.googleapis.com/syzbot-assets/3ae8b966bcd7/Image-78124b0c.gz.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+89edd6...@syzkaller.appspotmail.com

Unable to handle kernel paging request at virtual address dfff800000000004
KASAN: null-ptr-deref in range [0x0000000000000020-0x0000000000000027]
Mem abort info:
ESR = 0x0000000096000005
EC = 0x25: DABT (current EL), IL = 32 bits
SET = 0, FnV = 0
EA = 0, S1PTW = 0
FSC = 0x05: level 1 translation fault
Data abort info:
ISV = 0, ISS = 0x00000005, ISS2 = 0x00000000
CM = 0, WnR = 0, TnD = 0, TagAccess = 0
GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[dfff800000000004] address between user and kernel address ranges
Internal error: Oops: 0000000096000005 [#1] PREEMPT SMP
Modules linked in:
CPU: 0 PID: 7952 Comm: syz-executor682 Not tainted 6.6.0-rc6-syzkaller-g78124b0c1d10 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/06/2023
pstate: 804000c5 (Nzcv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
pc : __lock_acquire+0x104/0x75e8 kernel/locking/lockdep.c:5004
lr : lock_acquire+0x23c/0x71c kernel/locking/lockdep.c:5753
sp : ffff800098f26d40
x29: ffff800098f27000 x28: ffff8000808df4bc x27: ffff7000131e4e18
x26: 1ffff00011c340b9 x25: 0000000000000000 x24: 0000000000000000
x23: ffff7000131e4dd0 x22: 0000000000000000 x21: 0000000000000000
x20: 0000000000000000 x19: 0000000000000022 x18: ffff800098f27750
x17: 0000ffff833dafff x16: ffff80008a632120 x15: 0000000000000001
x14: ffff80008e1a05d0 x13: ffff800098f26e80 x12: dfff800000000000
x11: ffff800080319468 x10: ffff80008e1a05cc x9 : 00000000000000f3
x8 : 0000000000000004 x7 : ffff8000808df4bc x6 : 0000000000000000
x5 : 0000000000000000 x4 : 0000000000000001 x3 : 0000000000000000
x2 : 0000000000000000 x1 : 0000000000000000 x0 : 0000000000000022
Call trace:
__lock_acquire+0x104/0x75e8 kernel/locking/lockdep.c:5004
lock_acquire+0x23c/0x71c kernel/locking/lockdep.c:5753
__raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
_raw_spin_lock+0x48/0x60 kernel/locking/spinlock.c:154
spin_lock include/linux/spinlock.h:351 [inline]
__pte_offset_map_lock+0x154/0x360 mm/pgtable-generic.c:373
pte_offset_map_lock include/linux/mm.h:2939 [inline]
filemap_map_pages+0x698/0x11f0 mm/filemap.c:3582
do_fault_around mm/memory.c:4525 [inline]
do_read_fault mm/memory.c:4558 [inline]
do_fault mm/memory.c:4705 [inline]
do_pte_missing mm/memory.c:3669 [inline]
handle_pte_fault mm/memory.c:4978 [inline]
__handle_mm_fault mm/memory.c:5119 [inline]
handle_mm_fault+0x326c/0x49fc mm/memory.c:5284
faultin_page mm/gup.c:956 [inline]
__get_user_pages+0x3e0/0xa24 mm/gup.c:1239
populate_vma_page_range+0x254/0x328 mm/gup.c:1666
__mm_populate+0x240/0x3d8 mm/gup.c:1775
mm_populate include/linux/mm.h:3305 [inline]
vm_mmap_pgoff+0x2bc/0x3d4 mm/util.c:551
ksys_mmap_pgoff+0xd0/0x5b0 mm/mmap.c:1400
__do_sys_mmap arch/arm64/kernel/sys.c:28 [inline]
__se_sys_mmap arch/arm64/kernel/sys.c:21 [inline]
__arm64_sys_mmap+0xf8/0x110 arch/arm64/kernel/sys.c:21
__invoke_syscall arch/arm64/kernel/syscall.c:37 [inline]
invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:51
el0_svc_common+0x130/0x23c arch/arm64/kernel/syscall.c:136
do_el0_svc+0x48/0x58 arch/arm64/kernel/syscall.c:155
el0_svc+0x54/0x158 arch/arm64/kernel/entry-common.c:678
el0t_64_sync_handler+0x84/0xfc arch/arm64/kernel/entry-common.c:696
el0t_64_sync+0x190/0x194 arch/arm64/kernel/entry.S:595
Code: b006f948 b943a108 34000208 d343fe68 (386c6908)
---[ end trace 0000000000000000 ]---
----------------
Code disassembly (best guess):
0: b006f948 adrp x8, 0xdf29000
4: b943a108 ldr w8, [x8, #928]
8: 34000208 cbz w8, 0x48
c: d343fe68 lsr x8, x19, #3
* 10: 386c6908 ldrb w8, [x8, x12] <-- trapping instruction


---
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 syzk...@googlegroups.com.

syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

If the bug is already fixed, 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 bug's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)

If the bug is a duplicate of another bug, reply with:
#syz dup: exact-subject-of-another-report

If you want to undo deduplication, reply with:
#syz undup

[Hugh Dickins]

In hope, redirecting this report to linux-arm-kernel

I've spent a while worrying over this report, but have not been able
glean much from it: I'm not at all familiar with arm64 debugging, so
cannot deduce anything from the registers shown, though suspect they
would shed good light on it; but it may just be a waste of time, since
it was on a transient 6.6-rc6-based for-kernelci branch from last week.

If I read right, the reproducer is exercising MADV_PAGEOUT (splitting
huge pages) and MADV_COLLAPSE (assembling huge pages), on mmaps
MAP_FIXED MAP_SHARED MAP_ANONYMOUS i.e. shmem.

Suspicion falls on my 6.6-rc1 mm/khugepaged.c changes; but I don't see
what's wrong, and shall probably give up and ignore this - unless an
arm64 expert can take it further, or syzbot reproduces it on x86 on a
known tree.

Hugh

[Matthew Wilcox]

On Thu, Oct 26, 2023 at 11:07:35PM -0700, Hugh Dickins wrote:
> I've spent a while worrying over this report, but have not been able
> glean much from it: I'm not at all familiar with arm64 debugging, so
> cannot deduce anything from the registers shown, though suspect they
> would shed good light on it; but it may just be a waste of time, since
> it was on a transient 6.6-rc6-based for-kernelci branch from last week.
>
> If I read right, the reproducer is exercising MADV_PAGEOUT (splitting
> huge pages) and MADV_COLLAPSE (assembling huge pages), on mmaps
> MAP_FIXED MAP_SHARED MAP_ANONYMOUS i.e. shmem.
>
> Suspicion falls on my 6.6-rc1 mm/khugepaged.c changes; but I don't see
> what's wrong, and shall probably give up and ignore this - unless an
> arm64 expert can take it further, or syzbot reproduces it on x86 on a
> known tree.

Just to tie the two threads together ... it looks to me like what's
happening is __pte_offset_map_lock() is racing with pagetable_pte_dtor().
That is, we're walking the page tables, find a pmd, look up its
page/ptdesc, but because CONFIG_LOCKDEP is enabled, ptdesc->ptl is a
pointer to a lock, and that pointer is NULL.

More discussion here:
https://lore.kernel.org/linux-mm/ZVUWLgFg...@casper.infradead.org/T/#t

[Hugh Dickins]

On Thu, 26 Oct 2023, syzbot wrote:

Okay, let's try again by replying to the full orginal report
(see thread Matthew linked to for discussion leading to this):

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git b85ea95d086471afb4ad062012a4d73cd328fa86

Subject: [PATCH] mm/pgtable: smp_rmb() to match smp_wmb() in pmd_install()

Not-Yet-Signed-off-by: Hugh Dickins <hu...@google.com>
---
mm/memory.c | 2 ++
mm/pgtable-generic.c | 5 +++++
2 files changed, 7 insertions(+)

diff --git a/mm/memory.c b/mm/memory.c
index 1f18ed4a5497..8939357f1509 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -425,6 +425,8 @@ void pmd_install(struct mm_struct *mm, pmd_t *pmd, pgtable_t *pte)
* being the notable exception) will already guarantee loads are
* seen in-order. See the alpha page table accessors for the
* smp_rmb() barriers in page table walking code.
+ *
+ * See __pte_offset_map() for the smp_rmb() at the pte level.
*/
smp_wmb(); /* Could be smp_wmb__xxx(before|after)_spin_lock */
pmd_populate(mm, pmd, *pte);
diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c
index 4fcd959dcc4d..3330b666e9c3 100644
--- a/mm/pgtable-generic.c
+++ b/mm/pgtable-generic.c
@@ -297,6 +297,11 @@ pte_t *__pte_offset_map(pmd_t *pmd, unsigned long addr, pmd_t *pmdvalp)
pmd_clear_bad(pmd);
goto nomap;
}
+ /*
+ * Pair with the smp_wmb() in pmd_install(): make sure that the
+ * page table lock and page table contents are visibly initialized.
+ */
+ smp_rmb();
return __pte_map(&pmdval, addr);
nomap:
rcu_read_unlock();
--
2.35.3

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
BUG: unable to handle kernel paging request in __pte_offset_map_lock

Unable to handle kernel paging request at virtual address dfff800000000004
KASAN: null-ptr-deref in range [0x0000000000000020-0x0000000000000027]
Mem abort info:
ESR = 0x0000000096000005
EC = 0x25: DABT (current EL), IL = 32 bits
SET = 0, FnV = 0
EA = 0, S1PTW = 0
FSC = 0x05: level 1 translation fault
Data abort info:
ISV = 0, ISS = 0x00000005, ISS2 = 0x00000000
CM = 0, WnR = 0, TnD = 0, TagAccess = 0
GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[dfff800000000004] address between user and kernel address ranges
Internal error: Oops: 0000000096000005 [#1] PREEMPT SMP
Modules linked in:

CPU: 0 PID: 6731 Comm: syz-executor.5 Not tainted 6.7.0-rc1-syzkaller-dirty #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/09/2023

pstate: 804000c5 (Nzcv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
pc : __lock_acquire+0x104/0x75e8 kernel/locking/lockdep.c:5004
lr : lock_acquire+0x23c/0x71c kernel/locking/lockdep.c:5753

sp : ffff800097066d40
x29: ffff800097067000 x28: ffff8000808f70ac x27: ffff700012e0ce18
x26: 1ffff00011c64088 x25: 0000000000000000 x24: 0000000000000000
x23: ffff700012e0cdd0 x22: 0000000000000000 x21: 0000000000000000
x20: 0000000000000000 x19: 0000000000000022 x18: ffff800097067750
x17: ffff80008e31d000 x16: ffff80008a73b6ac x15: 0000000000000001
x14: ffff80008e320448 x13: ffff800097066e80 x12: dfff800000000000
x11: ffff80008031ef10 x10: ffff80008e320444 x9 : 00000000000000f3
x8 : 0000000000000004 x7 : ffff8000808f70ac x6 : 0000000000000000

x5 : 0000000000000000 x4 : 0000000000000001 x3 : 0000000000000000
x2 : 0000000000000000 x1 : 0000000000000000 x0 : 0000000000000022
Call trace:
__lock_acquire+0x104/0x75e8 kernel/locking/lockdep.c:5004
lock_acquire+0x23c/0x71c kernel/locking/lockdep.c:5753
__raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
_raw_spin_lock+0x48/0x60 kernel/locking/spinlock.c:154
spin_lock include/linux/spinlock.h:351 [inline]

__pte_offset_map_lock+0x154/0x360 mm/pgtable-generic.c:378
pte_offset_map_lock include/linux/mm.h:2946 [inline]
filemap_map_pages+0x5cc/0x112c mm/filemap.c:3531
do_fault_around mm/memory.c:4588 [inline]
do_read_fault mm/memory.c:4621 [inline]
do_fault mm/memory.c:4764 [inline]
do_pte_missing mm/memory.c:3732 [inline]
handle_pte_fault mm/memory.c:5040 [inline]
__handle_mm_fault mm/memory.c:5181 [inline]
handle_mm_fault+0x35ec/0x49f8 mm/memory.c:5346

faultin_page mm/gup.c:956 [inline]
__get_user_pages+0x3e0/0xa24 mm/gup.c:1239

populate_vma_page_range+0x254/0x328 mm/gup.c:1677
__mm_populate+0x240/0x3d8 mm/gup.c:1786
mm_populate include/linux/mm.h:3379 [inline]
vm_mmap_pgoff+0x2bc/0x3d4 mm/util.c:551
ksys_mmap_pgoff+0xd0/0x5b0 mm/mmap.c:1425

__do_sys_mmap arch/arm64/kernel/sys.c:28 [inline]
__se_sys_mmap arch/arm64/kernel/sys.c:21 [inline]
__arm64_sys_mmap+0xf8/0x110 arch/arm64/kernel/sys.c:21
__invoke_syscall arch/arm64/kernel/syscall.c:37 [inline]
invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:51
el0_svc_common+0x130/0x23c arch/arm64/kernel/syscall.c:136
do_el0_svc+0x48/0x58 arch/arm64/kernel/syscall.c:155
el0_svc+0x54/0x158 arch/arm64/kernel/entry-common.c:678
el0t_64_sync_handler+0x84/0xfc arch/arm64/kernel/entry-common.c:696
el0t_64_sync+0x190/0x194 arch/arm64/kernel/entry.S:595

Code: 90070528 b9424108 34000208 d343fe68 (386c6908)

---[ end trace 0000000000000000 ]---
----------------
Code disassembly (best guess):

0: 90070528 adrp x8, 0xe0a4000
4: b9424108 ldr w8, [x8, #576]

8: 34000208 cbz w8, 0x48
c: d343fe68 lsr x8, x19, #3
* 10: 386c6908 ldrb w8, [x8, x12] <-- trapping instruction

Tested on:

commit: b85ea95d Linux 6.7-rc1
git tree: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
console output: https://syzkaller.appspot.com/x/log.txt?x=14cf3388e80000
kernel config: https://syzkaller.appspot.com/x/.config?x=fffc11e84313b7c6

dashboard link: https://syzkaller.appspot.com/bug?extid=89edd67979b52675ddec
compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
userspace arch: arm64

patch: https://syzkaller.appspot.com/x/patch.diff?x=136db347680000

[Hugh Dickins]

On Thu, 26 Oct 2023, syzbot wrote:

I expect syzbot to approve of this one...

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git b85ea95d086471afb4ad062012a4d73cd328fa86

[PATCH] mm: fix oops when filemap_map_pmd() without prealloc_pte

[text to be filled in a little later]

Fixes: f9ce0be71d1f ("mm: Cleanup faultaround and finish_fault() codepaths")
Signed-off-by: Hugh Dickins <hu...@google.com>
---
mm/filemap.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/filemap.c b/mm/filemap.c
index 9710f43a89ac..3d4dae9d1070 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -3371,7 +3371,7 @@ static bool filemap_map_pmd(struct vm_fault *vmf, struct folio *folio,
}
}

- if (pmd_none(*vmf->pmd))
+ if (pmd_none(*vmf->pmd) && vmf->prealloc_pte)
pmd_install(mm, vmf->pmd, &vmf->prealloc_pte);

return false;
--
2.35.3

[syzbot]

Hello,

syzbot has tested the proposed patch and the reproducer did not trigger any issue:

Reported-and-tested-by: syzbot+89edd6...@syzkaller.appspotmail.com

Tested on:

commit: b85ea95d Linux 6.7-rc1
git tree: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git

console output: https://syzkaller.appspot.com/x/log.txt?x=14e7e350e80000
kernel config: https://syzkaller.appspot.com/x/.config?x=fffc11e84313b7c6

dashboard link: https://syzkaller.appspot.com/bug?extid=89edd67979b52675ddec
compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
userspace arch: arm64

patch: https://syzkaller.appspot.com/x/patch.diff?x=1586ca14e80000

Note: testing is done by a robot and is best-effort only.