lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Sun, 21 Jun 2009 14:02:43 +0100 (BST)
From:	Hugh Dickins <hugh.dickins@...cali.co.uk>
To:	Andrew Morton <akpm@...ux-foundation.org>
cc:	Johannes Weiner <hannes@...xchg.org>,
	Andrea Arcangeli <aarcange@...hat.com>,
	Izik Eidus <ieidus@...hat.com>, Rik van Riel <riel@...hat.com>,
	Nick Piggin <nickpiggin@...oo.com.au>,
	Minchan Kim <minchan.kim@...il.com>,
	linux-kernel@...r.kernel.org
Subject: [PATCH] mm: pass mm to grab_swap_token

If a kthread happens to use get_user_pages() on an mm (as KSM does),
there's a chance that it will end up trying to read in a swap page,
then oops in grab_swap_token() because the kthread has no mm: GUP
passes down the right mm, so grab_swap_token() ought to be using it.

We have not identified a stronger case than KSM's daemon (not yet
in mainline), but the issue must have come up before, since RHEL has
included a fix for this for years (though a different fix, they just
back out of grab_swap_token if current->mm is unset: which is what
we first proposed, but using the right mm here seems more correct).

Reported-by: Izik Eidus <ieidus@...hat.com>
Signed-off-by: Johannes Weiner <hannes@...xchg.org>
Signed-off-by: Hugh Dickins <hugh.dickins@...cali.co.uk>
Acked-by: Rik van Riel <riel@...hat.com>
---
Includes a small swap.h tidyup inspired by checkpatch.pl.
Due to crossed mails and crossed wires, there's a competing
patch-pair from Hannes:

http://lkml.org/lkml/2009/6/16/623
http://lkml.org/lkml/2009/6/16/624

We're all agreed that either this or those should go into 2.6.31:
personally, I don't rate arg typechecking of null functions when
!CONFIG_SWAP very highly, and prefer the one-line macros of Rik's
original; but you may well disagree, and prefer to take Hannes's.

 include/linux/swap.h |   12 ++++++------
 mm/memory.c          |    2 +-
 mm/thrash.c          |   32 +++++++++++++++-----------------
 3 files changed, 22 insertions(+), 24 deletions(-)

--- 2.6.30-git17/include/linux/swap.h	2009-06-21 07:48:18.000000000 +0100
+++ linux/include/linux/swap.h	2009-06-21 09:35:51.000000000 +0100
@@ -298,8 +298,8 @@ extern int try_to_free_swap(struct page
 struct backing_dev_info;
 
 /* linux/mm/thrash.c */
-extern struct mm_struct * swap_token_mm;
-extern void grab_swap_token(void);
+extern struct mm_struct *swap_token_mm;
+extern void grab_swap_token(struct mm_struct *);
 extern void __put_swap_token(struct mm_struct *);
 
 static inline int has_swap_token(struct mm_struct *mm)
@@ -419,10 +419,10 @@ static inline swp_entry_t get_swap_page(
 }
 
 /* linux/mm/thrash.c */
-#define put_swap_token(x) do { } while(0)
-#define grab_swap_token()  do { } while(0)
-#define has_swap_token(x) 0
-#define disable_swap_token() do { } while(0)
+#define put_swap_token(mm)	do { } while (0)
+#define grab_swap_token(mm)	do { } while (0)
+#define has_swap_token(mm)	0
+#define disable_swap_token()	do { } while (0)
 
 static inline void
 mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent)
--- 2.6.30-git17/mm/memory.c	2009-06-21 07:48:19.000000000 +0100
+++ linux/mm/memory.c	2009-06-21 09:35:51.000000000 +0100
@@ -2516,7 +2516,7 @@ static int do_swap_page(struct mm_struct
 	delayacct_set_flag(DELAYACCT_PF_SWAPIN);
 	page = lookup_swap_cache(entry);
 	if (!page) {
-		grab_swap_token(); /* Contend for token _before_ read-in */
+		grab_swap_token(mm); /* Contend for token _before_ read-in */
 		page = swapin_readahead(entry,
 					GFP_HIGHUSER_MOVABLE, vma, address);
 		if (!page) {
--- 2.6.30-git17/mm/thrash.c	2007-07-09 00:32:17.000000000 +0100
+++ linux/mm/thrash.c	2009-06-21 09:35:51.000000000 +0100
@@ -26,47 +26,45 @@ static DEFINE_SPINLOCK(swap_token_lock);
 struct mm_struct *swap_token_mm;
 static unsigned int global_faults;
 
-void grab_swap_token(void)
+void grab_swap_token(struct mm_struct *mm)
 {
 	int current_interval;
 
 	global_faults++;
 
-	current_interval = global_faults - current->mm->faultstamp;
+	current_interval = global_faults - mm->faultstamp;
 
 	if (!spin_trylock(&swap_token_lock))
 		return;
 
 	/* First come first served */
 	if (swap_token_mm == NULL) {
-		current->mm->token_priority = current->mm->token_priority + 2;
-		swap_token_mm = current->mm;
+		mm->token_priority = mm->token_priority + 2;
+		swap_token_mm = mm;
 		goto out;
 	}
 
-	if (current->mm != swap_token_mm) {
-		if (current_interval < current->mm->last_interval)
-			current->mm->token_priority++;
+	if (mm != swap_token_mm) {
+		if (current_interval < mm->last_interval)
+			mm->token_priority++;
 		else {
-			if (likely(current->mm->token_priority > 0))
-				current->mm->token_priority--;
+			if (likely(mm->token_priority > 0))
+				mm->token_priority--;
 		}
 		/* Check if we deserve the token */
-		if (current->mm->token_priority >
-				swap_token_mm->token_priority) {
-			current->mm->token_priority += 2;
-			swap_token_mm = current->mm;
+		if (mm->token_priority > swap_token_mm->token_priority) {
+			mm->token_priority += 2;
+			swap_token_mm = mm;
 		}
 	} else {
 		/* Token holder came in again! */
-		current->mm->token_priority += 2;
+		mm->token_priority += 2;
 	}
 
 out:
-	current->mm->faultstamp = global_faults;
-	current->mm->last_interval = current_interval;
+	mm->faultstamp = global_faults;
+	mm->last_interval = current_interval;
 	spin_unlock(&swap_token_lock);
-return;
 }
 
 /* Called on process exit. */
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ