lkml.org 
[lkml]   [2019]   [Sep]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] driver core: ensure a device has valid node id in device_add()
On Mon 09-09-19 14:04:23, Yunsheng Lin wrote:
> Currently a device does not belong to any of the numa nodes
> (dev->numa_node is NUMA_NO_NODE) when the node id is neither
> specified by fw nor by virtual device layer and the device has
> no parent device.
>
> According to discussion in [1]:

Please do not reference important parts of the justification via a link.
Just quote the relevant part to the changelog. It is just too easy that
external links die - not to mention lkml.org.

> Even if a device's numa node is not specified, the device really
> does belong to a node.

What does this mean?

> This patch sets the device node to node 0 in device_add() if the
> device's node id is not specified and it either has no parent
> device, or the parent device also does not have a valid node id.

Why is node 0 special? I have seen platforms with node 0 missing or
being memory less. The changelog also lacks an actual problem
descripton. Why do we even care about NUMA_NO_NODE? E.g. the page
allocator interprets NUMA_NO_NODE as the closest node with a memory.
And by closest it really means to the CPU which is performing the
allocation.

> [1] https://lkml.org/lkml/2019/9/2/466
>
> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
> ---
> Changelog RFC -> v1:
> 1. Drop log error message and use a "if" instead of "? :".
> 2. Drop the RFC tag.
> ---
> drivers/base/core.c | 10 +++++++---
> include/linux/numa.h | 2 ++
> 2 files changed, 9 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/base/core.c b/drivers/base/core.c
> index 1669d41..f79ad20 100644
> --- a/drivers/base/core.c
> +++ b/drivers/base/core.c
> @@ -2107,9 +2107,13 @@ int device_add(struct device *dev)
> if (kobj)
> dev->kobj.parent = kobj;
>
> - /* use parent numa_node */
> - if (parent && (dev_to_node(dev) == NUMA_NO_NODE))
> - set_dev_node(dev, dev_to_node(parent));
> + /* use parent numa_node or default node 0 */
> + if (!numa_node_valid(dev_to_node(dev))) {
> + if (parent && numa_node_valid(dev_to_node(parent)))
> + set_dev_node(dev, dev_to_node(parent));
> + else
> + set_dev_node(dev, 0);
> + }
>
> /* first, register with generic layer. */
> /* we require the name to be set before, and pass NULL */
> diff --git a/include/linux/numa.h b/include/linux/numa.h
> index 110b0e5..eccc757 100644
> --- a/include/linux/numa.h
> +++ b/include/linux/numa.h
> @@ -13,4 +13,6 @@
>
> #define NUMA_NO_NODE (-1)
>
> +#define numa_node_valid(node) ((unsigned int)(node) < nr_node_ids)
> +
> #endif /* _LINUX_NUMA_H */
> --
> 2.8.1

--
Michal Hocko
SUSE Labs

\
 
 \ /
  Last update: 2019-09-09 20:51    [W:0.077 / U:1.780 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site