Re: [PATCH] nfsd: set acl_access/acl_default after getting successful

[Li Lingfeng <lilingfeng3@huawei.com>]

在 2024/11/28 9:17, zhangjian (CG) 写道:
> there is one case when disk error cause get_inode_acl(inode, 
> ACL_TYPE_DEFAULT) return EIO,
> resp->acl_access will not be null. 
> posix_acl_release(resp->acl_default) will trigger this warning.
>
Yes, in the problematic scenario, resp->acl_access would not be NULL.
The memory pointed to by resp->acl_access would be freed and then accessed
again, thereby triggering the issue.
>
>> If getting acl_default fails, acl_access and acl_default will be 
>> released
>> simultaneously. However, acl_access will still retain a pointer pointing
>> to the released posix_acl, which will trigger a WARNING in
>> nfs3svc_release_getacl like this:
>>
>> ------------[ cut here ]------------
>> refcount_t: underflow; use-after-free.
>> WARNING: CPU: 26 PID: 3199 at lib/refcount.c:28
>> refcount_warn_saturate+0xb5/0x170
>> Modules linked in:
>> CPU: 26 UID: 0 PID: 3199 Comm: nfsd Not tainted
>> 6.12.0-rc6-00079-g04ae226af01f-dirty #8
>> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
>> 1.16.1-2.fc37 04/01/2014
>> RIP: 0010:refcount_warn_saturate+0xb5/0x170
>> Code: cc cc 0f b6 1d b3 20 a5 03 80 fb 01 0f 87 65 48 d8 00 83 e3 01 75
>> e4 48 c7 c7 c0 3b 9b 85 c6 05 97 20 a5 03 01 e8 fb 3e 30 ff <0f> 0b eb
>> cd 0f b6 1d 8a3
>> RSP: 0018:ffffc90008637cd8 EFLAGS: 00010282
>> RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffffff83904fde
>> RDX: dffffc0000000000 RSI: 0000000000000008 RDI: ffff88871ed36380
>> RBP: ffff888158beeb40 R08: 0000000000000001 R09: fffff520010c6f56
>> R10: ffffc90008637ab7 R11: 0000000000000001 R12: 0000000000000001
>> R13: ffff888140e77400 R14: ffff888140e77408 R15: ffffffff858b42c0
>> FS:  0000000000000000(0000) GS:ffff88871ed00000(0000)
>> knlGS:0000000000000000
>> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> CR2: 0000562384d32158 CR3: 000000055cc6a000 CR4: 00000000000006f0
>> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>> Call Trace:
>>   <TASK>
>>   ? refcount_warn_saturate+0xb5/0x170
>>   ? __warn+0xa5/0x140
>>   ? refcount_warn_saturate+0xb5/0x170
>>   ? report_bug+0x1b1/0x1e0
>>   ? handle_bug+0x53/0xa0
>>   ? exc_invalid_op+0x17/0x40
>>   ? asm_exc_invalid_op+0x1a/0x20
>>   ? tick_nohz_tick_stopped+0x1e/0x40
>>   ? refcount_warn_saturate+0xb5/0x170
>>   ? refcount_warn_saturate+0xb5/0x170
>>   nfs3svc_release_getacl+0xc9/0xe0
>>   svc_process_common+0x5db/0xb60
>>   ? __pfx_svc_process_common+0x10/0x10
>>   ? __rcu_read_unlock+0x69/0xa0
>>   ? __pfx_nfsd_dispatch+0x10/0x10
>>   ? svc_xprt_received+0xa1/0x120
>>   ? xdr_init_decode+0x11d/0x190
>>   svc_process+0x2a7/0x330
>>   svc_handle_xprt+0x69d/0x940
>>   svc_recv+0x180/0x2d0
>>   nfsd+0x168/0x200
>>   ? __pfx_nfsd+0x10/0x10
>>   kthread+0x1a2/0x1e0
>>   ? kthread+0xf4/0x1e0
>>   ? __pfx_kthread+0x10/0x10
>>   ret_from_fork+0x34/0x60
>>   ? __pfx_kthread+0x10/0x10
>>   ret_from_fork_asm+0x1a/0x30
>>   </TASK>
>> Kernel panic - not syncing: kernel: panic_on_warn set ...
>>
>> Clear acl_access/acl_default first and set both of them only when both
>> ACLs are successfully obtained.
>>
>> Fixes: a257cdd0e217 ("[PATCH] NFSD: Add server support for NFSv3 ACLs.")
>> Signed-off-by: Li Lingfeng <lilingfeng@huaweicloud.com>
>> ---
>>   fs/nfsd/nfs3acl.c | 14 ++++++++------
>>   1 file changed, 8 insertions(+), 6 deletions(-)
>>
>> diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c
>> index 5e34e98db969..17579a032a5b 100644
>> --- a/fs/nfsd/nfs3acl.c
>> +++ b/fs/nfsd/nfs3acl.c
>> @@ -29,10 +29,12 @@ static __be32 nfsd3_proc_getacl(struct svc_rqst 
>> *rqstp)
>>   {
>>       struct nfsd3_getaclargs *argp = rqstp->rq_argp;
>>       struct nfsd3_getaclres *resp = rqstp->rq_resp;
>> -    struct posix_acl *acl;
>> +    struct posix_acl *acl = NULL, *dacl = NULL;
>>       struct inode *inode;
>>       svc_fh *fh;
>>   +    resp->acl_access = NULL;
>> +    resp->acl_default = NULL;
>>       fh = fh_copy(&resp->fh, &argp->fh);
>>       resp->status = fh_verify(rqstp, &resp->fh, 0, NFSD_MAY_NOP);
>>       if (resp->status != nfs_ok)
>> @@ -56,19 +58,19 @@ static __be32 nfsd3_proc_getacl(struct svc_rqst 
>> *rqstp)
>>               resp->status = nfserrno(PTR_ERR(acl));
>>               goto fail;
>>           }
>> -        resp->acl_access = acl;
>>       }
>>       if (resp->mask & (NFS_DFACL|NFS_DFACLCNT)) {
>>           /* Check how Solaris handles requests for the Default ACL
>>              of a non-directory! */
>> -        acl = get_inode_acl(inode, ACL_TYPE_DEFAULT);
>> -        if (IS_ERR(acl)) {
>> -            resp->status = nfserrno(PTR_ERR(acl));
>> +        dacl = get_inode_acl(inode, ACL_TYPE_DEFAULT);
>> +        if (IS_ERR(dacl)) {
>> +            resp->status = nfserrno(PTR_ERR(dacl));
>>               goto fail;
>>           }
>> -        resp->acl_default = acl;
>>       }
>>   +    resp->acl_access = acl;
>> +    resp->acl_default = dacl;
>>       /* resp->acl_{access,default} are released in 
>> nfs3svc_release_getacl. */
>>   out:
>>       return rpc_success;