[PATCH] ppc64 mf_proc file position fix

[PATCH] ppc64 mf_proc file position fix

[Dave C Boutcher]

Andrew, 

arch/ppc64/kernel/mf_proc.c uses a bad interface for moving 
along file position in a proc_write routine.  This quit working
altogether in 2.6.8.  Patch to fix.  And I did a quick scan of the
kernel to see if anyone else was similarly broken...apparantly not :-)

Fixes a broken update of f_pos in a proc file write routine.

Signed off by: Dave Boutcher <sleddog@us.ibm.com>

--- linux-2.6.8.1/arch/ppc64/kernel/mf_proc.c.orig	2004-08-20 10:37:09.000000000 -0500
+++ linux-2.6.8.1/arch/ppc64/kernel/mf_proc.c	2004-08-20 14:16:01.000000000 -0500
@@ -168,22 +168,29 @@
 	return count;			
 }
 
-static int proc_mf_change_vmlinux(struct file *file, const char *buffer,
-		unsigned long count, void *data)
+static ssize_t proc_mf_change_vmlinux(struct file *file, 
+				      const char __user *buf,
+				      size_t count, loff_t *ppos)
 {
+	struct inode * inode = file->f_dentry->d_inode;
+	struct proc_dir_entry * dp = PDE(inode);
 	int rc;
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
 
-	rc = mf_setVmlinuxChunk(buffer, count, file->f_pos, (u64)data);
+	rc = mf_setVmlinuxChunk(buf, count, *ppos, (u64)dp->data);
 	if (rc < 0)
 		return rc;
 
-	file->f_pos += count;
+	*ppos += count;
 
 	return count;			
 }
 
+static struct file_operations proc_vmlinux_operations = {
+	.write		= proc_mf_change_vmlinux,
+};
+
 static int __init mf_proc_init(void)
 {
 	struct proc_dir_entry *mf_proc_root;
@@ -219,20 +226,7 @@
 			return 1;
 		ent->nlink = 1;
 		ent->data = (void *)(long)i;
-#if 0
-		if (i == 3) {
-			/*
-			 * if we had a 'D' vmlinux entry, it would only
-			 * be readable.
-			 */
-			ent->read_proc = proc_mf_dump_vmlinux;
-			ent->write_proc = NULL;
-		} else
-#endif
-		{
-			ent->write_proc = proc_mf_change_vmlinux;
-			ent->read_proc = NULL;
-		}
+		ent->proc_fops = &proc_vmlinux_operations;
 	}
 
 	ent = create_proc_entry("side", S_IFREG|S_IRUSR|S_IWUSR, mf_proc_root);




-- 
Dave Boutcher

Re: [PATCH] ppc64 mf_proc file position fix

[Christoph Hellwig]

On Fri, Aug 20, 2004 at 03:10:32PM -0500, Dave C Boutcher wrote:
> Andrew, 
> 
> arch/ppc64/kernel/mf_proc.c uses a bad interface for moving 
> along file position in a proc_write routine.  This quit working
> altogether in 2.6.8.  Patch to fix.  And I did a quick scan of the
> kernel to see if anyone else was similarly broken...apparantly not :-)
> 
> Fixes a broken update of f_pos in a proc file write routine.

What about moving on to seq_file while you're at it?  Switching from
one deprecated interface to another doesn't really sound like a worthwile
effort.

Re: [PATCH] ppc64 mf_proc file position fix

[Dave C Boutcher]

On Mon, Aug 23, 2004 at 06:54:23PM +0100, Christoph Hellwig wrote:
> On Fri, Aug 20, 2004 at 03:10:32PM -0500, Dave C Boutcher wrote:
> > Andrew, 
> > 
> > arch/ppc64/kernel/mf_proc.c uses a bad interface for moving 
> > along file position in a proc_write routine.  This quit working
> > altogether in 2.6.8.  Patch to fix.  And I did a quick scan of the
> > kernel to see if anyone else was similarly broken...apparantly not :-)
> > 
> > Fixes a broken update of f_pos in a proc file write routine.
> 
> What about moving on to seq_file while you're at it?  Switching from
> one deprecated interface to another doesn't really sound like a worthwile
> effort.

The specific problem was on the write side, not the read side, so AFAIK,
seq_file wasn't going to help out.  I started re-writing the whole file
to use seq_file for all the read stuff, and after a few hours into it I
decided to just fix the broken bit rather than gratuitously rewriting
the whole thing.

-- 
Dave Boutcher