On 02/07/2014 05:28 AM, David Rientjes wrote:
On Thu, 6 Feb 2014, David Rientjes wrote:
+#define MAX_REMOTE_READAHEAD 4096UL
Normally it wouldn't matter because there's no significant downside to it
racing, things like mempolicies which use numa_node_id() extensively would
result in, oops, a page allocation on the wrong node.
This stands out to me, though, because you're expecting the calculation to
be correct for a specific node.
The patch is still wrong, though, it should just do
int node = ACCESS_ONCE(numa_mem_id());
return min(nr, (node_page_state(node, NR_INACTIVE_FILE) +
node_page_state(node, NR_FREE_PAGES)) / 2);
since we want to readahead based on the cpu's local node, the comment
saying we're reading ahead onto "remote memory" is wrong since a
memoryless node has local affinity to numa_mem_id().
Oops, forgot about the MAX_REMOTE_READAHEAD which needs to be factored in
as well, but this handles the bound on local node's statistics.
So following discussion TODO for my patch is:
1) Update the changelog with user visible impact of the patch.
(Andrew's suggestion)
2) Add ACCESS_ONCE to numa_node_id().
3) Change the "readahead into remote memory" part of the documentation
which is misleading.
( I feel no need to add numa_mem_id() since we would specifically limit
the readahead with MAX_REMOTE_READAHEAD in memoryless cpu cases).
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>