On Tue, 2015-02-10 at 10:41 +, Andrew Cooper wrote:
> On 09/02/15 20:04, Boris Ostrovsky wrote:
> > diff --git a/xen/arch/x86/srat.c b/xen/arch/x86/srat.c
> > index 29fc724..4dfa1c3 100644
> > --- a/xen/arch/x86/srat.c
> > +++ b/xen/arch/x86/srat.c
> > @@ -21,13 +21,16 @@
> > #include
> > #
On 02/10/2015 06:39 AM, Jan Beulich wrote:
On 09.02.15 at 21:04, wrote:
--- a/xen/include/asm-x86/numa.h
+++ b/xen/include/asm-x86/numa.h
@@ -21,7 +21,7 @@ struct node {
extern int compute_hash_shift(struct node *nodes, int numnodes,
int *nodeids);
-extern i
>>> On 09.02.15 at 21:04, wrote:
> --- a/xen/include/asm-x86/numa.h
> +++ b/xen/include/asm-x86/numa.h
> @@ -21,7 +21,7 @@ struct node {
>
> extern int compute_hash_shift(struct node *nodes, int numnodes,
> int *nodeids);
> -extern int pxm_to_node(int nid);
> +extern
On 09/02/15 20:04, Boris Ostrovsky wrote:
> Various pieces of code test whether node value is NUMA_NO_NODE even
> though pxm_to_node() may return (int)-1 for an invalid node.
>
> Make pxm_to_node() and setup_node() return u8 and have them return
> NUMA_NO_NODE when necessary.
>
> Adjust code that t
Various pieces of code test whether node value is NUMA_NO_NODE even
though pxm_to_node() may return (int)-1 for an invalid node.
Make pxm_to_node() and setup_node() return u8 and have them return
NUMA_NO_NODE when necessary.
Adjust code that tests for (node == -1).
Signed-off-by: Boris Ostrovsky
