LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [patch] iov_iter_advance fix
@ 2008-03-06  5:39 Nick Piggin
  2008-03-06 15:15 ` Kevin Coffman
  0 siblings, 1 reply; 2+ messages in thread
From: Nick Piggin @ 2008-03-06  5:39 UTC (permalink / raw)
  To: Andrew Morton
  Cc: stable, Kevin Coffman, Alexey Dobriyan, Linux Kernel Mailing List


iov_iter_advance skips over zero-length iovecs, however it does not properly
terminate at the end of the iovec array. Fix this by checking against
i->count before we skip a zero-length iov.

The bug was reproduced with a test program that continually randomly creates
iovs to writev. The fix was also verified with the same program and also it
could verify that the correct data was contained in the file after each
writev.

2.6.25 and 2.6.24.stable requires this fix.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: stable@kernel.org
---
Index: linux-2.6/mm/filemap.c
===================================================================
--- linux-2.6.orig/mm/filemap.c	2008-03-06 08:46:38.000000000 +1100
+++ linux-2.6/mm/filemap.c	2008-03-06 08:48:00.000000000 +1100
@@ -1743,21 +1743,27 @@
 }
 EXPORT_SYMBOL(iov_iter_copy_from_user);
 
-static void __iov_iter_advance_iov(struct iov_iter *i, size_t bytes)
+void iov_iter_advance(struct iov_iter *i, size_t bytes)
 {
+	BUG_ON(i->count < bytes);
+
 	if (likely(i->nr_segs == 1)) {
 		i->iov_offset += bytes;
+		i->count -= bytes;
 	} else {
 		const struct iovec *iov = i->iov;
 		size_t base = i->iov_offset;
 
 		/*
 		 * The !iov->iov_len check ensures we skip over unlikely
-		 * zero-length segments.
+		 * zero-length segments (without overruning the iovec).
 		 */
-		while (bytes || !iov->iov_len) {
-			int copy = min(bytes, iov->iov_len - base);
+		while (bytes || unlikely(!iov->iov_len && i->count)) {
+			int copy;
 
+			copy = min(bytes, iov->iov_len - base);
+			BUG_ON(!i->count || i->count < copy);
+			i->count -= copy;
 			bytes -= copy;
 			base += copy;
 			if (iov->iov_len == base) {
@@ -1769,14 +1775,6 @@
 		i->iov_offset = base;
 	}
 }
-
-void iov_iter_advance(struct iov_iter *i, size_t bytes)
-{
-	BUG_ON(i->count < bytes);
-
-	__iov_iter_advance_iov(i, bytes);
-	i->count -= bytes;
-}
 EXPORT_SYMBOL(iov_iter_advance);
 
 /*

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [patch] iov_iter_advance fix
  2008-03-06  5:39 [patch] iov_iter_advance fix Nick Piggin
@ 2008-03-06 15:15 ` Kevin Coffman
  0 siblings, 0 replies; 2+ messages in thread
From: Kevin Coffman @ 2008-03-06 15:15 UTC (permalink / raw)
  To: Nick Piggin
  Cc: Andrew Morton, stable, Alexey Dobriyan,
	Linux Kernel Mailing List, Kevin Coffman

Thanks Nick,

I've verified this fixes the problem I was seeing.

On Thu, Mar 6, 2008 at 12:39 AM, Nick Piggin <npiggin@suse.de> wrote:
>
>  iov_iter_advance skips over zero-length iovecs, however it does not properly
>  terminate at the end of the iovec array. Fix this by checking against
>  i->count before we skip a zero-length iov.
>
>  The bug was reproduced with a test program that continually randomly creates
>  iovs to writev. The fix was also verified with the same program and also it
>  could verify that the correct data was contained in the file after each
>  writev.
>
>  2.6.25 and 2.6.24.stable requires this fix.
>
>  Signed-off-by: Nick Piggin <npiggin@suse.de>
>  Cc: stable@kernel.org

Tested-by: Kevin Coffman <kwc@citi.umich.edu>

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2008-03-06 15:16 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2008-03-06  5:39 [patch] iov_iter_advance fix Nick Piggin
2008-03-06 15:15 ` Kevin Coffman

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).