Re: SPEC failures

[Andi Kleen <ak@suse.de>]

On Wed, Aug 15, 2001 at 10:30:46AM -0500, Steve Lord wrote:
> Andrew had the approach of setting the umask of the nfsd process to 0 at
> startup, there was some other reason for this not being popular.

He did it for init_task, which disturbed all other kernel threads too
and opened tons of security holes.
The right way IMHO is to give nfsd an own fs_struct and set umask there,
as in this patch.

-Andi


--- linux-xfs/fs/nfsd/nfssvc.c-NFSUMASK	Wed Jun 20 13:08:49 2001
+++ linux-xfs/fs/nfsd/nfssvc.c	Wed Aug 15 19:32:11 2001
@@ -136,6 +136,18 @@
 	}
 }
 
+int fork_fsstruct(void)
+{
+	struct fs_struct *oldfs = current->fs, *newfs;
+	newfs = copy_fs_struct(oldfs); 
+	if (newfs) { 
+		current->fs = newfs;
+		put_fs_struct(oldfs);	
+		return 0;
+	} 
+	return -1;		
+} 
+
 /*
  * This is the NFS server kernel thread
  */
@@ -150,6 +162,8 @@
 	MOD_INC_USE_COUNT;
 	lock_kernel();
 	daemonize();
+	if (!fork_fsstruct())
+		current->fs->umask = 0;
 	sprintf(current->comm, "nfsd");
 	current->rlim[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY; 
 
--- linux-xfs/kernel/ksyms.c-NFSUMASK	Tue Aug 14 02:10:34 2001
+++ linux-xfs/kernel/ksyms.c	Wed Aug 15 19:34:36 2001
@@ -298,6 +298,8 @@
 EXPORT_SYMBOL(buffermem_pages);
 EXPORT_SYMBOL(nr_free_pages);
 EXPORT_SYMBOL(page_cache_size);
+EXPORT_SYMBOL(copy_fs_struct);
+EXPORT_SYMBOL(put_fs_struct);
 
 /* for stackable file systems (lofs, wrapfs, cryptfs, etc.) */
 EXPORT_SYMBOL(default_llseek);