From: Yufeng Wang wangyufeng@kylinos.cn
There is an issue in the kernel: if inode is NULL pointer. the function "inode_lock_nested" (or function "inode_lock" before) a crash will happen at code "&inode->i_rwsem". [292618.520532] BUG: unable to handle kernel NULL pointer dereference at 00000000000000a0 [...] [292618.560398] RIP: 0010:down_write+0x12/0x30 [292618.565580] Code: 83 f8 01 74 08 48 c7 47 20 01 00 00 00 f3 c3 66 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 ba 01 00 00 00 ff ff ff ff 48 89 f8 <f0> 48 0f c1 10 85 d2 74 05 e8 00 43 ff ff 65 48 8b 04 25 80 5c 01 [292618.587219] RSP: 0018:ffffb898dc86fc20 EFLAGS: 00010246 [292618.593666] RAX: 00000000000000a0 RBX: ffff94c84f363950 RCX: ffffff8000000000 [292618.602255] RDX: ffffffff00000001 RSI: 0000000000000063 RDI: 00000000000000a0 [292618.610844] RBP: ffffb898dc86fc78 R08: 0000000000000000 R09: 0000000000000000 [292618.619434] R10: ffffb898dc86fca8 R11: 0000000000000000 R12: 0000000000000000 [292618.628022] R13: ffff94c84f362a20 R14: ffff954d3f2fb4a0 R15: ffff954c3afa5010 [292618.636612] FS: 0000555555989cc0(0000) GS:ffff956dbf900000(0000) knlGS:0000000000000000 [292618.646271] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [292618.653300] CR2: 00000000000000a0 CR3: 000000fc7f25a000 CR4: 00000000003406e0 [292618.661888] Call Trace: [292618.665225] simple_recursive_removal+0x4f/0x230 [292618.670994] ? debug_fill_super+0xe0/0xe0 [292618.676079] debugfs_remove+0x40/0x60 [292618.680799] kvm_vcpu_release+0x19/0x30 [kvm]
Fixes: a3d1e7eb5abe ("simple_recursive_removal(): kernel-side rm -rf for ramfs-style filesystems") Cc: stable@vger.kernel.org
Signed-off-by: Yufeng Wang wangyufeng@kylinos.cn --- fs/libfs.c | 2 ++ 1 file changed, 2 insertions(+)
diff --git a/fs/libfs.c b/fs/libfs.c index 1661dcb7d983..9090cd1f97c4 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -611,6 +611,8 @@ static void __simple_recursive_removal(struct dentry *dentry, struct dentry *victim = NULL, *child; struct inode *inode = this->d_inode;
+ if (!inode) + return; inode_lock_nested(inode, I_MUTEX_CHILD); if (d_is_dir(this)) inode->i_flags |= S_DEAD;
On Thu, Nov 13, 2025 at 01:23:57PM +0800, rom.wang wrote:
From: Yufeng Wang wangyufeng@kylinos.cn
There is an issue in the kernel: if inode is NULL pointer. the function "inode_lock_nested" (or function "inode_lock" before) a crash will happen at code "&inode->i_rwsem".
How is inode NULL? What is causing that?
[292618.520532] BUG: unable to handle kernel NULL pointer dereference at 00000000000000a0 [...] [292618.560398] RIP: 0010:down_write+0x12/0x30 [292618.565580] Code: 83 f8 01 74 08 48 c7 47 20 01 00 00 00 f3 c3 66 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 ba 01 00 00 00 ff ff ff ff 48 89 f8 <f0> 48 0f c1 10 85 d2 74 05 e8 00 43 ff ff 65 48 8b 04 25 80 5c 01 [292618.587219] RSP: 0018:ffffb898dc86fc20 EFLAGS: 00010246 [292618.593666] RAX: 00000000000000a0 RBX: ffff94c84f363950 RCX: ffffff8000000000 [292618.602255] RDX: ffffffff00000001 RSI: 0000000000000063 RDI: 00000000000000a0 [292618.610844] RBP: ffffb898dc86fc78 R08: 0000000000000000 R09: 0000000000000000 [292618.619434] R10: ffffb898dc86fca8 R11: 0000000000000000 R12: 0000000000000000 [292618.628022] R13: ffff94c84f362a20 R14: ffff954d3f2fb4a0 R15: ffff954c3afa5010 [292618.636612] FS: 0000555555989cc0(0000) GS:ffff956dbf900000(0000) knlGS:0000000000000000 [292618.646271] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [292618.653300] CR2: 00000000000000a0 CR3: 000000fc7f25a000 CR4: 00000000003406e0 [292618.661888] Call Trace: [292618.665225] simple_recursive_removal+0x4f/0x230 [292618.670994] ? debug_fill_super+0xe0/0xe0 [292618.676079] debugfs_remove+0x40/0x60 [292618.680799] kvm_vcpu_release+0x19/0x30 [kvm]
Is the kvm code doing something wrong here? debugfs shouldn't be trying to remove an inode that is already removed, so please fix the root cause, do not paper over it.
thanks,
greg k-h
On Thu, Nov 13, 2025 at 08:25:32AM -0500, Greg KH wrote:
On Thu, Nov 13, 2025 at 01:23:57PM +0800, rom.wang wrote:
From: Yufeng Wang wangyufeng@kylinos.cn
There is an issue in the kernel: if inode is NULL pointer. the function "inode_lock_nested" (or function "inode_lock" before) a crash will happen at code "&inode->i_rwsem".
How is inode NULL? What is causing that?
[292618.520532] BUG: unable to handle kernel NULL pointer dereference at 00000000000000a0 [...] [292618.560398] RIP: 0010:down_write+0x12/0x30 [292618.565580] Code: 83 f8 01 74 08 48 c7 47 20 01 00 00 00 f3 c3 66 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 ba 01 00 00 00 ff ff ff ff 48 89 f8 <f0> 48 0f c1 10 85 d2 74 05 e8 00 43 ff ff 65 48 8b 04 25 80 5c 01 [292618.587219] RSP: 0018:ffffb898dc86fc20 EFLAGS: 00010246 [292618.593666] RAX: 00000000000000a0 RBX: ffff94c84f363950 RCX: ffffff8000000000 [292618.602255] RDX: ffffffff00000001 RSI: 0000000000000063 RDI: 00000000000000a0 [292618.610844] RBP: ffffb898dc86fc78 R08: 0000000000000000 R09: 0000000000000000 [292618.619434] R10: ffffb898dc86fca8 R11: 0000000000000000 R12: 0000000000000000 [292618.628022] R13: ffff94c84f362a20 R14: ffff954d3f2fb4a0 R15: ffff954c3afa5010 [292618.636612] FS: 0000555555989cc0(0000) GS:ffff956dbf900000(0000) knlGS:0000000000000000 [292618.646271] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [292618.653300] CR2: 00000000000000a0 CR3: 000000fc7f25a000 CR4: 00000000003406e0 [292618.661888] Call Trace: [292618.665225] simple_recursive_removal+0x4f/0x230 [292618.670994] ? debug_fill_super+0xe0/0xe0 [292618.676079] debugfs_remove+0x40/0x60 [292618.680799] kvm_vcpu_release+0x19/0x30 [kvm]
Is the kvm code doing something wrong here? debugfs shouldn't be trying to remove an inode that is already removed, so please fix the root cause, do not paper over it.
That's... interesting. It's not just inode that is already removed, it's very likely a dentry that has already beed freed (dentry with refcount greater than 1 can not become negative).
Yes, thank you, i aggree with your suggest dentry has already beed freed, and i can got the dentry.d_name.name info is "vcpu126" from debug vmcore, and it's parent dentry is "xxxpid_xx" which's d_inode is NULL too., but the parent of "xxxpid_xx" dentry is "kvm" which's d_inode is not NULL. and after checking the source, i guest the root reason is kvm->debugfs_dentry has been freed before freeing vcpu->debugfs_dentry. and the vmcore kernel version is 4.19.
[Patch that introduced the issue] In commit 3d75b8aa5c2 ("KVM: Always flush async #PF workqueue when vCPU is being destroyed"), the kvm_get_kvm(work->vcpu->kvm) call was removed from kvm_setup_async_pf, but the corresponding kvm_put_kvm(work->vcpu->kvm) call was not removed.
[Patch that fixes the issue] Commit 7863e346e108 ("KVM: async_pf: Cleanup kvm_setup_async_pf()") will remove the kvm_put_kvm(work->vcpu->kvm) call.
if cherry-pick Commit 7863e346e108 ("KVM: async_pf: Cleanup kvm_setup_async_pf()") to 4.19 brach can fix this issue.
linux-stable-mirror@lists.linaro.org
-
Al Viro -
Greg KH -
rom.wang