Re: x86_64: potential critical issue with quicklists and page table pages

[Christoph Lameter <clameter@sgi.com>]

On Fri, 21 Sep 2007, Siddha, Suresh B wrote:

> Essentially quicklist free routines are doing something like
> __quicklist_free()
> 	...
> 	if (unlikely(nid != numa_node_id())) {
> 		__free_page(page);
> 		...
> 	}
> 
> 	....
> 
> 
> Now this will potentially cause a problem, if a cpu in someother node starts
> using this page, while the corresponding TLB entries are still alive
> in the original cpu which is still freeing the page table pages.

Hmmm... Yes could be.

> This potentially can cause SW failures and hard to debug issues like
> http://www.ussg.iu.edu/hypermail/linux/kernel/0205.2/1254.html

Cannot get to this page:

Not Found

The requested URL /hypermail/linux/kernel/0205.2/1254.htm was not found on 
this server.


> Can we revert this commit for 2.6.23 and look at this code post 2.6.23?

We can remove this piece alone since it was an optimization.

---
 include/linux/quicklist.h |    7 -------
 1 file changed, 7 deletions(-)

Index: linux-2.6/include/linux/quicklist.h
===================================================================
--- linux-2.6.orig/include/linux/quicklist.h	2007-09-21 11:46:44.000000000 -0700
+++ linux-2.6/include/linux/quicklist.h	2007-09-21 11:47:01.000000000 -0700
@@ -58,13 +58,6 @@ static inline void __quicklist_free(int 
 	struct quicklist *q;
 	int nid = page_to_nid(page);
 
-	if (unlikely(nid != numa_node_id())) {
-		if (dtor)
-			dtor(p);
-		__free_page(page);
-		return;
-	}
-
 	q = &get_cpu_var(quicklist)[nr];
 	*(void **)p = q->page;
 	q->page = p;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/