Re: Oops in rpc_clnt_debugfs_register() from debugfs change

[Greg Kroah-Hartman]

On Tue, Feb 12, 2019 at 03:42:14PM +0100, Greg Kroah-Hartman wrote:
> On Tue, Feb 12, 2019 at 03:37:20PM +0100, Greg Kroah-Hartman wrote:
> > On Tue, Feb 12, 2019 at 02:31:14PM +0000, David Howells wrote:
> > > I've bisected an oops that occurs in rpc_clnt_debugfs_register() trying to
> > > dereference a pointer with -EACCES in it.  This is the causing commit, though
> > > I suspect the bug is in sunrpc expecting to see NULL rather than an error.
> > > 
> > > ff9fb72bc07705c00795ca48631f7fffe24d2c6b is the first bad commit
> > > commit ff9fb72bc07705c00795ca48631f7fffe24d2c6b
> > > Author: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
> > > Date:   Wed Jan 23 11:28:14 2019 +0100
> > > 
> > >     debugfs: return error values, not NULL
> > >     
> > >     When an error happens, debugfs should return an error pointer value, not
> > >     NULL.  This will prevent the totally theoretical error where a debugfs
> > >     call fails due to lack of memory, returning NULL, and that dentry value
> > >     is then passed to another debugfs call, which would end up succeeding,
> > >     creating a file at the root of the debugfs tree, but would then be
> > >     impossible to remove (because you can not remove the directory NULL).
> > >     
> > >     So, to make everyone happy, always return errors, this makes the users
> > >     of debugfs much simpler (they do not have to ever check the return
> > >     value), and everyone can rest easy.
> > >     ...
> > > 
> > > The attached oops occurs during boot from the gssproxy process in
> > > rpc_clnt_debugfs_register().  The code at this point is:
> > > 
> > >    0xffffffff8195cbdd <+450>:   mov    0x50(%rax),%rcx   <--- oopsing
> > >    0xffffffff8195cbe1 <+454>:   mov    $0xffffffff821cc8ba,%rdx
> > >    0xffffffff8195cbe8 <+461>:   mov    $0x18,%esi
> > >    0xffffffff8195cbed <+466>:   lea    -0x30(%rbp),%rdi
> > >    0xffffffff8195cbf1 <+470>:   callq  0xffffffff819db773 <snprintf>
> > > 
> > > RAX is -EACCES.
> > > 
> > > Looking in the source:
> > > 
> > > 	len = snprintf(name, sizeof(name), "../../rpc_xprt/%s",
> > > 			xprt->debugfs->d_name.name);
> > > 
> > > I think xprt->debugfs is the value in RAX.
> > > 
> > > 	(gdb) p &((struct dentry *)0)->d_name.name
> > > 	$5 = (const unsigned char **) 0x50 <irq_stack_union+80>
> > > 
> > > which matches the offset on the oopsing MOV instruction.
> > > 
> > > This is with linus/master (aa0c38cf39de73bf7360a3da8f1707601261e518).
> > 
> > Ugh, yeah, I see the problem, sorry about that.
> > 
> > I wonder why the debugfs call is always failing, that's not good...
> > 
> > let me dig and see if I already have a patch for this...
> 
> I have a much larger cleanup patch for this code, but this single line
> change should solve the issue for now.  Can you test it to verify?
> 
> thanks,
> 
> greg k-h
> 
> ------------------
> 
> diff --git a/net/sunrpc/debugfs.c b/net/sunrpc/debugfs.c
> index 45a033329cd4..19bb356230ed 100644
> --- a/net/sunrpc/debugfs.c
> +++ b/net/sunrpc/debugfs.c
> @@ -146,7 +146,7 @@ rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
>  	rcu_read_lock();
>  	xprt = rcu_dereference(clnt->cl_xprt);
>  	/* no "debugfs" dentry? Don't bother with the symlink. */
> -	if (!xprt->debugfs) {
> +	if (IS_ERR_OR_NULL(xprt->debugfs)) {
>  		rcu_read_unlock();
>  		return;
>  	}


And, if you want my larger fix that I will be sending to netdev one of
these days, here's that one.  It includes the above patch as part of it.

thanks,

greg k-h

---------------

commit 8d885c486153d1731c14a6a435774a4e9ccd1ebc
Author: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
Date:   Fri Jan 4 13:40:56 2019 +0100

    sunrpc: fix changelog

diff --git a/net/sunrpc/debugfs.c b/net/sunrpc/debugfs.c
index 45a033329cd4..ca63f6ed873f 100644
--- a/net/sunrpc/debugfs.c
+++ b/net/sunrpc/debugfs.c
@@ -135,18 +135,15 @@ rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
 
 	/* make the per-client dir */
 	clnt->cl_debugfs = debugfs_create_dir(name, rpc_clnt_dir);
-	if (!clnt->cl_debugfs)
-		return;
 
 	/* make tasks file */
-	if (!debugfs_create_file("tasks", S_IFREG | 0400, clnt->cl_debugfs,
-				 clnt, &tasks_fops))
-		goto out_err;
+	debugfs_create_file("tasks", S_IFREG | 0400, clnt->cl_debugfs, clnt,
+			    &tasks_fops);
 
 	rcu_read_lock();
 	xprt = rcu_dereference(clnt->cl_xprt);
 	/* no "debugfs" dentry? Don't bother with the symlink. */
-	if (!xprt->debugfs) {
+	if (IS_ERR_OR_NULL(xprt->debugfs)) {
 		rcu_read_unlock();
 		return;
 	}
@@ -157,8 +154,7 @@ rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
 	if (len >= sizeof(name))
 		goto out_err;
 
-	if (!debugfs_create_symlink("xprt", clnt->cl_debugfs, name))
-		goto out_err;
+	debugfs_create_symlink("xprt", clnt->cl_debugfs, name);
 
 	return;
 out_err:
@@ -237,15 +233,10 @@ rpc_xprt_debugfs_register(struct rpc_xprt *xprt)
 
 	/* make the per-client dir */
 	xprt->debugfs = debugfs_create_dir(name, rpc_xprt_dir);
-	if (!xprt->debugfs)
-		return;
 
 	/* make tasks file */
-	if (!debugfs_create_file("info", S_IFREG | 0400, xprt->debugfs,
-				 xprt, &xprt_info_fops)) {
-		debugfs_remove_recursive(xprt->debugfs);
-		xprt->debugfs = NULL;
-	}
+	debugfs_create_file("info", S_IFREG | 0400, xprt->debugfs, xprt,
+			    &xprt_info_fops);
 
 	atomic_set(&xprt->inject_disconnect, rpc_inject_disconnect);
 }
@@ -308,22 +299,6 @@ static const struct file_operations fault_disconnect_fops = {
 	.release	= fault_release,
 };
 
-static struct dentry *
-inject_fault_dir(struct dentry *topdir)
-{
-	struct dentry *faultdir;
-
-	faultdir = debugfs_create_dir("inject_fault", topdir);
-	if (!faultdir)
-		return NULL;
-
-	if (!debugfs_create_file("disconnect", S_IFREG | 0400, faultdir,
-				 NULL, &fault_disconnect_fops))
-		return NULL;
-
-	return faultdir;
-}
-
 void __exit
 sunrpc_debugfs_exit(void)
 {
@@ -338,25 +313,13 @@ void __init
 sunrpc_debugfs_init(void)
 {
 	topdir = debugfs_create_dir("sunrpc", NULL);
-	if (!topdir)
-		return;
-
-	rpc_fault_dir = inject_fault_dir(topdir);
-	if (!rpc_fault_dir)
-		goto out_remove;
 
 	rpc_clnt_dir = debugfs_create_dir("rpc_clnt", topdir);
-	if (!rpc_clnt_dir)
-		goto out_remove;
 
 	rpc_xprt_dir = debugfs_create_dir("rpc_xprt", topdir);
-	if (!rpc_xprt_dir)
-		goto out_remove;
 
-	return;
-out_remove:
-	debugfs_remove_recursive(topdir);
-	topdir = NULL;
-	rpc_fault_dir = NULL;
-	rpc_clnt_dir = NULL;
+	rpc_fault_dir = debugfs_create_dir("inject_fault", topdir);
+
+	debugfs_create_file("disconnect", S_IFREG | 0400, rpc_fault_dir, NULL,
+			    &fault_disconnect_fops);
 }