On Wed, 8 Feb 2006, Andrew Morton wrote:
> > I think it should be put into 2.6.16. Andrew?
>
> Does every single caller of __alloc_pages(__GFP_FS) correctly handle a NULL
> return? I doubt it, in which case this patch will cause oopses and hangs.
I sent you a patch with static inline..... But I am having second thoughts
about this patch. Paul is partially right. Maybe we can move the logic
into the out_of_memory handler for now? That would allow us to implement
more sophisticated things later (for example page migration would allow us
to move memory of processes that can also allocate on other nodes from the
nodes where we lack memory) and Paul may put something in there to
address his concerns.
---
Terminate process that fails on a constrained allocation
Some allocations are restricted to a limited set of nodes (due to memory
policies or cpuset constraints). If the page allocator is not able to find
enough memory then that does not mean that overall system memory is low.
In particular going postal and more or less randomly shooting at processes
is not likely going to help the situation but may just lead to suicide (the
whole system coming down).
It is better to signal to the process that no memory exists given the
constraints that the process (or the configuration of the process) has
placed on the allocation behavior. The process may be killed but then the
sysadmin or developer can investigate the situation. The solution is similar
to what we do when running out of hugepages.
This patch adds a check before the out of memory killer is invoked. At that
point performance considerations do not matter much so we just scan the zonelist
and reconstruct a list of nodes. If the list of nodes does not contain all
online nodes then this is a constrained allocation and we should not calle
the OOM killer.
Signed-off-by: Christoph Lameter <[email protected]>
Index: linux-2.6.16-rc2/mm/page_alloc.c
===================================================================
--- linux-2.6.16-rc2.orig/mm/page_alloc.c 2006-02-02 22:03:08.000000000 -0800
+++ linux-2.6.16-rc2/mm/page_alloc.c 2006-02-08 14:06:12.000000000 -0800
@@ -1011,8 +1012,10 @@ rebalance:
if (page)
goto got_pg;
- out_of_memory(gfp_mask, order);
- goto restart;
+ if (out_of_memory(zonelist, gfp_mask, order))
+ goto restart;
+
+ return NULL;
}
/*
Index: linux-2.6.16-rc2/include/linux/swap.h
===================================================================
--- linux-2.6.16-rc2.orig/include/linux/swap.h 2006-02-02 22:03:08.000000000 -0800
+++ linux-2.6.16-rc2/include/linux/swap.h 2006-02-08 14:10:02.000000000 -0800
@@ -147,7 +147,7 @@ struct swap_list_t {
#define vm_swap_full() (nr_swap_pages*2 < total_swap_pages)
/* linux/mm/oom_kill.c */
-extern void out_of_memory(gfp_t gfp_mask, int order);
+extern int out_of_memory(struct zonelist *zl, gfp_t gfp_mask, int order);
/* linux/mm/memory.c */
extern void swapin_readahead(swp_entry_t, unsigned long, struct vm_area_struct *);
Index: linux-2.6.16-rc2/mm/oom_kill.c
===================================================================
--- linux-2.6.16-rc2.orig/mm/oom_kill.c 2006-02-02 22:03:08.000000000 -0800
+++ linux-2.6.16-rc2/mm/oom_kill.c 2006-02-08 14:09:43.000000000 -0800
@@ -131,6 +131,27 @@ unsigned long badness(struct task_struct
}
/*
+ * check if a given zonelist allows allocation over all the nodes
+ * in the system.
+ */
+static noinline int zonelist_incomplete(struct zonelist *zonelist, gfp_t gfp_mask)
+{
+#ifdef CONFIG_NUMA
+ struct zone **z;
+ nodemask_t nodes;
+
+ nodes = node_online_map;
+ for (z = zonelist->zones; *z; z++)
+ if (cpuset_zone_allowed(*z, gfp_mask))
+ node_clear((*z)->zone_pgdat->node_id,
+ nodes);
+ return !nodes_empty(nodes);
+#else
+ return 0;
+#endif
+}
+
+/*
* Simple selection loop. We chose the process with the highest
* number of 'points'. We expect the caller will lock the tasklist.
*
@@ -262,12 +283,23 @@ static struct mm_struct *oom_kill_proces
* killing a random task (bad), letting the system crash (worse)
* OR try to be smart about which process to kill. Note that we
* don't have to be perfect here, we just have to be good.
+ *
+ * Function returns 1 if the allocation should be retried.
+ * zero if the allocation should fail.
*/
-void out_of_memory(gfp_t gfp_mask, int order)
+int out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
{
struct mm_struct *mm = NULL;
task_t * p;
+ /*
+ * Simply fail an allocation that does not allow
+ * allocation on all nodes. We may want to have
+ * more sophisticated logic here in the future.
+ */
+ if (zonelist_incomplete(zonelist, gfp_mask))
+ return 0;
+
if (printk_ratelimit()) {
printk("oom-killer: gfp_mask=0x%x, order=%d\n",
gfp_mask, order);
@@ -306,4 +338,5 @@ retry:
*/
if (!test_thread_flag(TIF_MEMDIE))
schedule_timeout_interruptible(1);
+ return 1;
}
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]
