[RFC Patch V1 21/30] mm, irqchip: Use cpu_to_mem()/numa_mem_id() to support memoryless node

[RFC Patch V1 21/30] mm, irqchip: Use cpu_to_mem()/numa_mem_id() to support memoryless node

[Jiang Liu]

When CONFIG_HAVE_MEMORYLESS_NODES is enabled, cpu_to_node()/numa_node_id()
may return a node without memory, and later cause system failure/panic
when calling kmalloc_node() and friends with returned node id.
So use cpu_to_mem()/numa_mem_id() instead to get the nearest node with
memory for the/current cpu.

If CONFIG_HAVE_MEMORYLESS_NODES is disabled, cpu_to_mem()/numa_mem_id()
is the same as cpu_to_node()/numa_node_id().

Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com>
---
 drivers/irqchip/irq-clps711x.c |    2 +-
 drivers/irqchip/irq-gic.c      |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/irqchip/irq-clps711x.c b/drivers/irqchip/irq-clps711x.c
index 33340dc97d1d..b0acf8b32a1a 100644
--- a/drivers/irqchip/irq-clps711x.c
+++ b/drivers/irqchip/irq-clps711x.c
@@ -186,7 +186,7 @@ static int __init _clps711x_intc_init(struct device_node *np,
 	writel_relaxed(0, clps711x_intc->intmr[1]);
 	writel_relaxed(0, clps711x_intc->intmr[2]);
 
-	err = irq_alloc_descs(-1, 0, ARRAY_SIZE(clps711x_irqs), numa_node_id());
+	err = irq_alloc_descs(-1, 0, ARRAY_SIZE(clps711x_irqs), numa_mem_id());
 	if (IS_ERR_VALUE(err))
 		goto out_iounmap;
 
diff --git a/drivers/irqchip/irq-gic.c b/drivers/irqchip/irq-gic.c
index 7e11c9d6ae8c..a7e6c043d823 100644
--- a/drivers/irqchip/irq-gic.c
+++ b/drivers/irqchip/irq-gic.c
@@ -1005,7 +1005,7 @@ void __init gic_init_bases(unsigned int gic_nr, int irq_start,
 	if (of_property_read_u32(node, "arm,routable-irqs",
 				 &nr_routable_irqs)) {
 		irq_base = irq_alloc_descs(irq_start, 16, gic_irqs,
-					   numa_node_id());
+					   numa_mem_id());
 		if (IS_ERR_VALUE(irq_base)) {
 			WARN(1, "Cannot allocate irq_descs @ IRQ%d, assuming pre-allocated\n",
 			     irq_start);
-- 
1.7.10.4

[RFC Patch V1 21/30] mm, irqchip: Use cpu_to_mem()/numa_mem_id() to support memoryless node

[Jason Cooper]

On Fri, Jul 11, 2014 at 03:37:38PM +0800, Jiang Liu wrote:
> When CONFIG_HAVE_MEMORYLESS_NODES is enabled, cpu_to_node()/numa_node_id()
> may return a node without memory, and later cause system failure/panic
> when calling kmalloc_node() and friends with returned node id.
> So use cpu_to_mem()/numa_mem_id() instead to get the nearest node with
> memory for the/current cpu.
> 
> If CONFIG_HAVE_MEMORYLESS_NODES is disabled, cpu_to_mem()/numa_mem_id()
> is the same as cpu_to_node()/numa_node_id().
> 
> Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com>
> ---
>  drivers/irqchip/irq-clps711x.c |    2 +-
>  drivers/irqchip/irq-gic.c      |    2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)

Do you have anything depending on this?  Can apply it to irqchip?  If
you need to keep it with other changes,

Acked-by: Jason Cooper <jason@lakedaemon.net>

But please do let me know if I can take it.

thx,

Jason.

[RFC Patch V1 21/30] mm, irqchip: Use cpu_to_mem()/numa_mem_id() to support memoryless node

[Jiang Liu]

Hi Jason,
	Thanks for your review. According to review comments,
we need to rework the patch set in another direction and will
give up this patch.
Regards!
Gerry

On 2014/7/18 20:40, Jason Cooper wrote:
> On Fri, Jul 11, 2014 at 03:37:38PM +0800, Jiang Liu wrote:
>> When CONFIG_HAVE_MEMORYLESS_NODES is enabled, cpu_to_node()/numa_node_id()
>> may return a node without memory, and later cause system failure/panic
>> when calling kmalloc_node() and friends with returned node id.
>> So use cpu_to_mem()/numa_mem_id() instead to get the nearest node with
>> memory for the/current cpu.
>>
>> If CONFIG_HAVE_MEMORYLESS_NODES is disabled, cpu_to_mem()/numa_mem_id()
>> is the same as cpu_to_node()/numa_node_id().
>>
>> Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com>
>> ---
>>  drivers/irqchip/irq-clps711x.c |    2 +-
>>  drivers/irqchip/irq-gic.c      |    2 +-
>>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> Do you have anything depending on this?  Can apply it to irqchip?  If
> you need to keep it with other changes,
> 
> Acked-by: Jason Cooper <jason@lakedaemon.net>
> 
> But please do let me know if I can take it.
> 
> thx,
> 
> Jason.
>