Skip to contentSkip to conversation

Apply the PG_sensitive flag to audit subsystem

PATCH In review Larry H. 3 participants 4 messages May 20, 2009 – May 30, 2009

Subsystems (2)
lore
Share & tools
lore

Larry H.

author
#0 lore raw

PATCH 3/5 Apply the PG_sensitive flag to audit subsystem

This patch deploys the use of the PG_sensitive page allocator flag within the audit subsystem. It's not necessarily a high profile target for use of this flag, but could be expected to contain potentially sensitive information under some circumstances.

---
 kernel/audit.c |    3 +++
 1 file changed, 3 insertions(+)
kernel/audit.c +3 -0
Index: linux-2.6/kernel/audit.c
===================================================================
@@ -1061,6 +1061,9 @@ static struct audit_buffer * audit_buffe
 	}
 	spin_unlock_irqrestore(&audit_freelist_lock, flags);
 
+	if (!(gfp_mask & GFP_SENSITIVE))
+		gfp_mask |= GFP_SENSITIVE;
+
 	if (!ab) {
 		ab = kmalloc(sizeof(*ab), gfp_mask);
 		if (!ab)

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

#1 lore raw

kernel/audit.c static struct audit_buffer * audit_buffe · line 1066

Quoted 3 lines
kernel/audit.c +3 -0
+	if (!(gfp_mask & GFP_SENSITIVE))
+		gfp_mask |= GFP_SENSITIVE;
+

The condition is unnecessary. Just do the |= in all cases.

Olaf

#2 lore raw
In reply to Larry H.

kernel/audit.c static struct audit_buffer * audit_buffe · line 1065

Quoted 3 lines · Larry H.
kernel/audit.c +2 -0
+	if (!(gfp_mask & GFP_SENSITIVE))
+		gfp_mask |= GFP_SENSITIVE;

WTF?

Why is this different from just "gfp_mask |= GFP_SENSITIVE;"

Linus

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

author
#3 lore raw
Quoted 8 lines · Linus Torvalds

kernel/audit.c static struct audit_buffer * audit_buffe · line 1065

Quoted 3 lines · Larry H.
kernel/audit.c +2 -0
+	if (!(gfp_mask & GFP_SENSITIVE))
+		gfp_mask |= GFP_SENSITIVE;

WTF?

Indeed.

Quoted 1 line

Why is this different from just "gfp_mask |= GFP_SENSITIVE;"

Blame anal retentiveness at the time of writing that. Surely the test should be ditched. Looking back at that, I honestly think there might be a place to plug the flag (in the caller) instead of doing that. I don't think there are many places to do it, so this particular patch from the set can be ditched and rewritten (if you want to take the selective clearing road...)

Larry

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>